Adding a spindle to your LongMill?

Over the last few months, I’ve been playing around with using a spindle on the LongMill MK2. Originally, we didn’t recommend for customers to use spindles on their LongMills due to the overall cost and complexity, as well as because we hadn’t done much testing on how a heavier spindle would behave on the LongMill.

I recently wrote about working with Andy McTaggart, one of our beta testers in one of our posts. There I mentioned that at the speed and cutting depth he was running his project, I could audibly hear his Makita router struggle. Although the project was completed without incident, I also realized that now with the LongMill MK2 bringing significant rigidity improvements, there were a few more areas in which we could push the boundaries of how hard we can run these machines.

Now with the overall rigidity improvements on the LongMill MK2, we are more confident in recommending installing a spindle for some customers who might benefit from the extra power and features a spindle can offer, which we’ll discuss in this article. We’ll also talk generally about installing a spindle and some of the things I recommend watching out for.

Disclaimer

I have done my best to make sure the information in this article is useful, accurate, and relevant. However, I do not take any responsibility for any issues, injuries, or damage arising from this. We do not provide direct company support for spindles and VFDs, so we cannot help you with your specific setup or installation. If you have any inquiries or questions, please direct them to the manufacturer of your spindle or VFD.

This article is designed to provide some general information, not a step-by-step instructions for adding a spindle to your machine. Installation will vary significantly depending on what hardware you are using.

We want to create a plug-and-play option for a spindle for the LongMill in the future. However, at this time, there is no timeline or plans for this to be available anytime soon. If you wish to stay up to date on future development, please sign up for our mailing list: https://sienci.us13.list-manage.com/subscribe/post?u=720f140325aaa668a8aa09916&id=e18d59a085

Who would I recommend spindles for?

I’ve had the chance to work with spindles in both industrial and hobby settings, as well as play around with a variety of different types of spindles over the years. I’ve also spent many years using the Makita RT0701 router which we recommend for the LongMill as well.

The biggest and main reason I don’t recommend using a spindle is because spindles and VFDs are much more complicated than routers. Yes, a spindle comes with a lot of advantages, but for most beginners, I don’t think the benefits outweigh the potential cost and headaches of setting one and using one brings. Although there are now some plug-and-play spindle kits available for hobbyists, such as from PwnCNC* that can take some of the guesswork out, there are a lot of settings, wiring, and other technical details that may confuse users. From my experience, cheaper spindle kits that you can find on Amazon and Aliexpress have many quality issues and come with pretty much no documentation and support. Most also don’t come with any cables and require additional cables and soldering to set up. If you are just starting out with CNC and don’t want to make your life more complicated, using a small router is still an excellent choice.

*I have not used the Spindle Kit from PwnCNC and cannot vouch for the product. This is not a recommendation or endorsement.

Over the years, I have played around with several different spindles from different vendors which are all “budget focused”. Here are some common things I learned:

  • They don’t come with any wiring, so you’ll have to source your own.
  • VFDs are not pre-programmed out of the box to work. Running the spindle without the right settings can cause damage.
  • Some come with a very brief instruction manual which requires a lot of research to decipher. There are usually a lot of different variations of VFDs and finding manuals online can be difficult
  • It is hard to judge the actual quality without taking the spindle apart or with special tools. I would take any specifications posted for each spindle with a grain of salt.
Anemic wiring on a 1.5KW air-cooled spindle

I think it’s also important to give credit to how good the Makita RT0701 actually is. Although 1.25HP (0.9KW) doesn’t sound like a lot when you compare it to a 1.5KW or 2.2KW spindle, because of the constant speed control under load feature, the router will keep a constant RPM even at high loads. On the other hand, 3 phase motors in general typically require to spin within a certain range to provide a certain level of torque, but may not provide the full torque potential of the motor. This means that the power rating of the spindle may not represent the actual usable power at the RPM ranges you want to work in. With the exception of certain types of high-load jobs, such as surfacing and bowl cutting, the Makita 0701 should be able to keep up without issues. We’ve used the Makita router for cutting steel and aluminum and it’s survived if that’s worth anything.

Here is a general list of pros and cons with a spindle compared to a router:

Pros:

  • More overall power
  • Quieter
  • More durable, as it does not need brushes
  • Less runout
  • Allows for speed control using gcode or the computer interface

Cons:

  • More expensive, typically $300ish on the low end, and $1000+ on the high end
  • Complicated to set up and use
  • Additional electrical installation to handle added current loads may be needed
  • Safety concerns with dealing with mains voltages
  • Higher chance of user error and damage
  • Higher chance of having EMF issues

Choosing a spindle and VFD

The two main components of the spindle system are the spindle and VFD. The spindle is the motor part, which holds the bit and spins it. The VFD (variable frequency drive), is an electronic driver or controller that controls the frequency and current of the electricity going into the spindle to adjust its power and speed. We’ll be talking about both of these components and what to look for when selecting them.

Common example of a spindle
Common example of a VFD

Power

Most spindles in the size category will either be 0.8KW, 1.5KW, or 2.2KW. VFDs can be matched with the spindle as required. The larger the diameter of the spindle the more power it usually has. Most 65mm spindles will generally have a power rating of 0.8KW to 1.5KW. 80mm spindles will generally have a power rating of 1.5KW to 2.2KW. Larger and smaller spindles do exist, but for the purpose of this article, we won’t get into them.

It’s sort of difficult to suggest one power rating over another because cutting loads vary a lot based on material and tooling used, but for context, the Makita RT0701 is about 0.9KW. A 1.5KW spindle theoretically would have up to 67% more power and a 2.2KW would have up to 144% more power.

I would recommend going with the 2.2KW if your power outlets can handle it, but if you are on 110V/120V with standard 15A breakers, you’ll be limited to a 1.5KW model. A 2.2KW spindle on 120V will peak at 18A and a 1.5KW spindle on 120V will peak at 12.5A. On my setup, I am using a 2.2KW spindle and VFD but limited the max current to around 10A to prevent the breaker from going off.

Size and weight

CNC spindles for this type of CNC use typically come in 65mm and 80mm diameter sizes. We sell a 65mm and 80mm that works with any LongMill. The larger the diameter of the spindle the more power it usually has. Most 65mm spindles will generally have a power rating of 0.8KW to 1.5KW. 80mm spindles will generally have a power rating of 1.5KW to 2.2KW. Larger and smaller spindles do exist, but for the purpose of this article, we won’t get into them.

It’s also good to note that spindle size also generally determines what collet sizes you can get with it as well. Most 65mm spindles will have an ER11 or ER16 system. 80mm spindles usually use an ER16 or ER20 system. The number of the ER collet dictates the largest shank that the system can take plus one millimeter. So an ER16 system can hold up to a 17mm shank.

An ER11 collet

The next thing to consider is the weight of your spindle. I don’t have exact weights for the different sizes, but this 80mm spindle weighs about 10lbs. A 65mm spindle would obviously be lighter. The one below is an air-cooled model, which has some extra fins and bits for heat dissipation, and based on some cursory online research, a water-cooled spindle of the same diameter and power should be slightly lighter. A lighter spindle makes it easier for the machine to control the acceleration of the spindle and puts less stress and wear on the overall machine, but in my testing, using the 80mm air-cooled spindle was totally fine with default LongMill settings.

Water-cooled vs air-cooled

Spindles are available as water-cooled or air-cooled. Each has its pros and cons. I would preface to say that I don’t have any experience using water-cooled spindles, as I chose to go with air-cooled ones due to their simplicity. This part will come from general research done online plus some of my experience using air-cooled spindles.

So the first major difference is in sound level. Because air-cooled spindles need to have air flowing through them, a sound is generated in this process. Water-cooled spindles are generally quieter since it uses a liquid flowing through the body to cool the spindle.

When comparing the air-cooled spindle to the Makita, the air-cooled spindle is much quieter. It may be worth noting that during cutting, the sound of the bit cutting is much louder than the spindle itself, so I am guessing that the overall difference in real-life use isn’t too large. I chose to go with an air-cooled to avoid needing to deal with coolant lines and such. Once you add in the sound of your dust collection as well, my opinion is that I would expect that the difference would be minimal.

The second big difference and the reason I chose to get an air-cooled spindle is regards to the fact that coolant lines and a bunch of other parts are not needed. Water-cooled spindles need coolant, lines, a reservoir, and a pump to keep the spindle cool. Although not particularly complicated to set up, I wanted to avoid the clutter. I also wanted to avoid dealing with coolant and the chance of it leaking, spilling, and making a mess. It’s important to note that since air-cooled spindles use ambient air to cool themselves if you are in a high-temperature environment, a water-cooled spindle may be more suitable.

Voltage and phases

In a spindle system, you’ll need to concern yourself with the voltages and phase count of both the VFD and the spindle itself. Most will be 110V or 220V and accept it in single phase or three phase. Usually, VFDs can accept a range of voltages within their base working voltage. For example, if you have a 110V VFD, it should work within 100V and 130V. Although most VFDs are 3-phase, VFDs that have different numbers of phases also exist.

Using a higher voltage, such as 220V over 110V, typically makes it easier to transfer more power with the same cable thickness. When you start wiring your spindle, you’ll have to consider the wire gauge and power requirements of your system. The limit to the power you can carry on any given wire generally comes down to the amount of electrical current you are carrying and the resistance of the wire itself. A thicker wire has lower resistance and thus can transfer the same amount of current while generating less heat. If the heat generated is more than the amount that the wire can handle, you will have a meltdown, and lots of bad things happen. Note that these formulas are assuming DC instead of AC, and are simplified for sake of ease of explanation. For more info on calculating with AC, please check out this article.

Heat in watts =(Current in amps^2) x (Wire resistance in Ohms)

Power in watts = Voltage in volts x Current in amps

Based on these formulas, current is inversely proportional to voltage. This means that a 220V circuit requires half of the current to carry the same amount of power as a 110V circuit. A 220V circuit will also generate less heat flowing the same amount of power through a wire. This concept will be important when balancing choosing your input and output voltages of your VFD.

Here’s a pretty cheap on I found on Amazon

To choose your input voltage and phase, confirm how you’re planning to power your VFD. Most North American households will have access to 120V, single-phase outlets. In this case, you’ll want to either choose a VFD that accepts 110V or use a transformer to change the voltage to the VFD you have.

Next, you’ll want to select the output voltage of your VFD. You’ll need to match this based on the voltage rating on your spindle. I’ve found that 220V tends to be the most common and is probably the one you want.

If I were to make a recommendation it would be:

  • If you have 220V power available to you, to get a 220V VFD and 220V spindle.
  • If you have 110V power only, find a VFD that has an input voltage of 110V in single phase and an output voltage of 220V in 3-phase.

Frequency

Most VFDs and spindle motors have a rated frequency and speed range. Most VFDs for CNC use will usually be rated for 0-400HZ, but it’s important to check that the working frequency range can support the spindle speeds you want. Spindles will have a speed range with the max RPM being the speed the motor can run at its rated frequency. Most that I’ve seen have a range from 10,000 to 24,000RPM.

Set up

Wiring

Setting up a VFD from scratch involves a lot of wiring. If you have a pre-configured kit that comes with everything you might be able to skip this step.

First, we’ll talk about the AC input. In my, I cut a spare power cord to expose the green, black, and white wires. The green wire is connected to the ground and the white wire to the “L”. On the “N” terminal, I’ve wired in-line with the black wire an E-stop switch for a bit of extra safety. If you don’t have a E-stop, your wire color will probably be black, but in my case the E-stop wire is red.

Just as a side note, please make sure to check the gauge and current carrying capacity of your AC cable. In my case I am using a 14AWG cable good for 15A. I have experienced AC power cables melt from being used beyond their current limits. There should be a rating stamped or printed on the side of the cable for you to double-check.

Next, we’ll want to wire up the three-phase size to the spindle which is denoted by “W”, “V”, and “U”. This is where the lack of documentation makes things a bit more complicated. Some VFDs will say “U”, “V”, “W” instead as well.

In my case, the single sheet of paper that came in the box instructed me to wire Pin 1 on the provided aviation plug with “U”, Pin 2 with “V”, and Pin 3 with “W”. This involved doing some soldering to get the wires onto the aviation plug. If you have everything wired up correctly, the spindle should turn clockwise when looking from the top.

I would mention that the spindle did not come with any cables. I am assuming that the user is supposed to source their own. Spindles generate a lot of EMF, and so proper shielding is also important, but for some reason the paper manual also said to not ground the spindle and the cable. I did open up the spindle at the top cover and indeed the body of the spindle was not grounded.

In any case, if you have a shielded cable, you can ground the shielding and be on your merry way. I haven’t run into any interference issues yet, but your results may vary.

It is possible to purchase spindle-specific cables. The one I’m holding is one from an Ebay seller, which is probably the best type to use. However, because of the small size of the plug that came with the kit, I used a thinner, less durable 4 conductor security cable. Since it’s an 18GA wire, it’ll probably be ok for 10-15A, but it’s not ideal since this isn’t specifically designed to get bent and moved around that much.

A proper 3 phase spindle cable
4 conductor shielded security cable

Programming

I’ve found this to be the trickiest part of the setup, because there are a lot of parameters to select before running the VFD. We’ll go through some of the basic settings and talk about what they do, but it’s likely that the parameters and names of each are going to vary depending on the model you have. Having the wrong setting might fry your VFD or spindle so make sure to keep track of what settings you are changing and write notes down if you need to. I’m going to write down the name of the parameter and the description of the setting, but they may not be the same for your VFD.

First, get into “Programming mode”. There is probably a PROG button or something similar on the main panel. This will bring up each setting and you can navigate them using the arrows. You can choose to modify the setting with another button (“FUNCT/DATA” in my case) and save it. Make sure to double-check your settings persist regularly to make sure your settings are staying.

These are some of the settings that I feel like are most important. However, you should double-check all of them.

P00 Maximum voltage: Output voltage setting, or what we want to voltage going to the spindle to be. 220V.

P01 Reference frequency: This is the incoming voltage. For our case, it should be 60Hz since that’s the frequency our grid uses.

P02 Intermediate voltage: This is the incoming voltage, which in our case is 120V.

P07 Minimum operating frequency: This is the minimum frequency you can set for your spindle. In my case, the air-cooled spindle may overheat if it goes too slowly, so it may be good to set this at 166Hz, or a minimum RPM of 10,000.

P10 Working frequency source: This chooses where you want to get your speed control from. You can either control it directly on the control panel manually, but it’s likely you’ll want to be able to control it in gcode or software. If you have all of the other wiring set up, you can choose to use an external signal (in our case, a “2, external analog signal“) to control the speed of the spindle.

P11 Start/Stop control source: You can also choose how you want to turn on and off the spindle. The best way to set this up is to have it turn on with an external signal, such as the signal controlling the speed of the spindle.

P50 to P55 Multi-function binding post: This setting allows us to choose what turning on one of the input terminals does. In our case, we have it set up to “wire forward operation” because we want the spindle to turn when the terminal is active.

P62 Display options: You can choose what to display on the panel, such as RPM, current, operating frequency, etc. In my case, I just wanted to see the RPM so I have it set to “2 revolution”

Installation

As we discussed earlier, most spindles will come as a 65mm or 80mm body size. You’ll need a mount that fits this. If you already have a Makita RT0701, you can probably use the same original router mount as it is also 65mm, but if you are going with a larger 80mm body, you’ll have to order a new one. Router mounts can be purchased from our store.

From this point, you should be able to mount your spindle and route the cable back to the VFD through the drag chains in the same way as the Makita router.

Firmware and gSender settings

If you wish to have control over your spindle speed through gcode or gSender, you’ll have to check a few different settings for your machine. Some of the added features include:

  • You can have the spindle turn on and off automatically. For example, you can have your spindle turn on and spin up for 10 seconds before starting your cut, and then turn off the spindle automatically after the job is complete.
  • You can change spindle speed directly in your gcode. If you want to start the cut with a fast spindle speed, then slow down later in the job, you can do that directly with the code.
  • You can change your spindle speed on the fly with a few clicks, rather than fiddling with the knob on top of the Makita.

First, we’ll do a bit of wiring. Start by adding two leads from the SpinPWM output terminal from your LongBoard and wiring it to the input on your VFD. If you’re running a laser as well, you can have them in parallel as long as your laser is off. It should also be noted that you may need to change your min and max intensity values on your laser to match with your spindles min and max RPM so you don’t have to keep changing them in your firmware.

For more details about the LongBoard and stuff you can do with it, please check out our resources.

If your firmware settings using gSender, you’ll need to select your minimum and maximum spindle speed settings. In my case, I’ve selected 24,000RPM for the max spindle speed and 0 for the minimum. When the controller outputs a PWM signal, it will set the PWM duty cycle to 100% at 24,000RPM or higher, and between 0 and 0.4% duty cycle at 0RPM. Don’t forget to press “Apply New Settings” to have the settings propagate.

It’s very important for us to discuss the difference between analog input and PWM input. They are different and need to be taken into account when wiring your VFD and controller. I’ve talked to a lot of folks adding accessories that have been confused about this. The LongBoard controller and most GRBL controllers will have a PWM output. This means that the controller produces an on-off signal very quickly. Depending on the percentage time it is on, or the duty cycle, determines the speed or intensity that we want to have in controlling a device. This means regardless of the actual voltage being output, a PWM signal can represent the intensity accurately.

However, most VFDs use an analog voltage control. This means that the higher the voltage, the faster your spindle will run. Most have a 0-10V range, although some can be configured for 0-5V. This means that simply plugging in a PWM signal to a VFD that uses analog control may not work. If you are using an analog input VFD, you may want to find a digital to analog converter like the one below.

With the specific VFD that I’m using, I was able to set the voltage range to 0-5V. I have the PWM signal lines connected to the analog inputs directly and I am able to control the speed this way. This only works for a small and very specific set of reasons:

  • The output voltage of the PWM signal is close enough to 5V, or the max input voltage so that when the PWM signal is at 100% duty cycle, the spindle speed is also set to 100% speed.
  • The way that the VFD measures the voltage is by taking the average voltage over a certain period of time. So running at 50% duty cycle means that it thinks the input voltage is 2.5V.

This may not work for you and I don’t recommend setting things up like this, as factors such as your PWM voltage and the way your VDF interprets the incoming signal may vary. The most ideal way to set things up would be to find a VFD that can accept a PWM input.

Also, disable “laser-mode”. Again, if you have a laser you may need to change these settings back when you use your laser again.

Next, by clicking on the gSender’s setting button (the gear icon at the upper right corner of the interface), you can toggle on Spindle/Laser interface and the max and minimum spindle speeds. In this case, I have it set to 10,000 to 24,000RPM.

Once you exit out of the settings, you’ll be able to find the Spindle/Laser controls in your gSender interface.

From here, you can run your machine clockwise with the “CW (M3)” button and stop the spindle using the “Stop (M5) button”. If you have a VFD that can input a signal to run the spindle counterclockwise, you can also wire this with the “SpinDirection” pin and another terminal on the VFD. We won’t get into this since I don’t most folks will need this feature.

Another important thing I want to touch on the spindle dropping down due to its weight and inertia. The lead screw on your LongMill may not have enough drag to keep it in place when the motor is powered off. You can combat this by:

  • Setting the $1 Step Idle Delay to 255, to hold your steppers.
  • Adding a counterweight or using the lightest spindle possible.

To change your Step Idle Delay, you can find it in the Firmware tool again. Changing it to 255 means that the stepper motors will hold their position when they are not moving. Otherwise, they will power off after a small delay which allows them to move freely.

It’s important to note that setting the steppers to hold their place means that power continues to go to the motors, which may cause them to get hot. I would recommend shutting off the machine or changing the step idle delay back to the default if you aren’t using the machine. Although you shouldn’t have any issues or damage with regular use, you do run a larger risk for fire, which is why I try to avoid having the steppers hold.

I’ve created some macros to basically hold and unhold the stepper motors, which makes it easy to get around this issue. If you want to download and install them yourself, here’s the code:

To install it, just download the file and upload it into your macro section.

Alternatively, you can counteract having the spindle fall with a counterweight or springs, which we won’t get into here.

Using your spindle

If you have your spindle set up to run manually, for example, so that you turn on and off the spindle with the button on the front panel and speed with the potentiometer, then you can use your spindle by adjusting these.

However, if you’ve wired everything up to control directly through your computer and controller, you’ll be able to control your spindle directly through gSender. I’m assuming most users will want to do this as this is one of the most convenient parts of having a spindle in the first place.

First, you can control and test your spindle using the interface at the bottom left. Simply click the “CW (M3)” button to run the spindle, and set your RPM with the slider. A small note, you may need to reclick the “CW (M3)” button again to have the speed update. You can turn on and off your spindle using these controls. If you’re wondering what “CCW (M4)” does, this is the command to run the spindle counterclockwise, which sends the same PWM signal to control the speed but also a high signal on the “spindirection” output terminal on the control board to indicate the VFD to turn the other direction. You probably won’t need to worry about this one, unless you’re doing some really advanced stuff. Finally, the “Stop (M5)” command stops the spindle.

If you want to control spindle speed in your gcode, you will need to include it in your CAM. Find the setting that selects your spindle speed in your CAM software. The setting may be specific to each tool, or be a global option for your whole job. Another note is that once your spindle is running at the speed set in gcode, you can use the feedrate override controls to change the speed of your spindle.

And lastly and my personal favorite, is using the “Start/Stop Gcode” feature in gSender. This basically adds gcode at the start and end of every job. So when you press “Start Job”, it’ll run some code first. For my setup, I’ve made it spin up to 24,000RPM (M3 S24000) and then have a dwell (G4 P10) for 10 seconds to give it a moment to get up to speed. At the end, it sends an M5 command to turn off the spindle. You can adjust and change the code to fit whatever works in your system.

If spindles are a bit too confusing to you but you want to control accessories like your Makita router or vacuum with a relay, please check out our page about IOT relays.

Conclusion

As our community continues to grow and folks continue to push their LongMills further and further, I’m excited to test and share what we’ve learned to add more capabilities to our machines. When I’m using the LongMill for my personal projects, I’ve found myself running the machine harder and with more confidence as well, and I believe that upgrades like these can provide significant improvement to the machine’s productivity.

I found that setting up spindles and VFDs are pretty complicated. However, I’m hoping that as the hobby CNC market expands, we’ll see more third parties create plug-and-play options to eliminate the confusion that exists with budget setups. In the meantime, if you have a spindle set up with your machine, I encourage you to share your knowledge and setup! I’ve already found a lot of great info on our forum, if you’re looking for what’s already out there.

Happy making!

April 2023 Production Updates

April fools

LongMill MK2

LongMills continue to ship out as usual. We received another batch of controllers after being out for another few days.

Batch 9 production continues and we continue to focus on getting some of the new things such as:

  • Spring-loaded anti-backlash nut
  • Injection molded feet
  • SuperLongboard

This also includes existing changes and improvements that already exist on the LongMills that are currently shipping.

  • Motor to leadscrew couplers using M5 hardware
  • New ACME locking nuts

To reflect the changes to the LongMill MK2, we will be calling this update the LongMill MK2.5.

We are expecting Batch 9 to start in June. At this point, we will increase pricing for the LongMill MK2 to the MK2.5 to reflect the addition of primarily the SuperLongBoard and other additions. Pricing for this new version is to increase by $150CAD/$110USD approximately.

LaserBeam and Vortex

LaserBeam and Vortex orders are shipping out as usual within a few days.

Ikenna and Abeku have been working on a magnetic mounting system for the LaserBeam to allow for faster and easier mounting and dismounting of the laser, and folks should expect more information to come out soon near the end of April. They also let me know that while the mount is suitable for the LongMill, they are continuing to work to improve the stability of the mount to work with AltMill and higher speeds.

The magnetic mounting system should work on all mounts either on the right or left side, or just on the front, depending on the version. More info to come soon.

Prototype magnetic mount

AltMill

The AltMill can now be ordered on the AltMill product page.

Check out the Launch live stream below:

A couple of things going on with production:

  • Due a random failure with one of the closed-loop steppers used in testing, we have started testing motors from 4 different companies to nail down the highest quality motor for the AltMill and ensure that we can understand the reasons for the motor to stop working.
  • The gantries for the AltMill are done and have gone through an anodization process to make them black. This should improve the aesthetics of the machine and make it look more polished
  • We are testing and manufacturing the dust covers for the linear guides on the Y-axis. Based on our testing results, this may become a standard included part of the AltMill
  • We are working on testing a 4KW spindle to push the limits of the AltMill
  • Testing on the relay and power distribution control board are being tested now. This control board allows us to distribute power from the power supply to the motors and kill power in an emergency setting, improving the machine’s safety

Additional parts for the first 50 AltMills are expected to arrive mid-April and we will start assembly as soon as they arrive.

Test fit of new gantries

Sienci Router

The Sienci Router project continues to move along. We have now received two controller boards, but have not been satisfied by the level of speed response we’ve gotten. For a bit of context, traditional motors will slow down when under load. In the case for CNC milling, we don’t want our router or spindle speed to go down, as it increases the chipload. If an end mill rotates at a set speed, each rotation takes a certain amount of cut per pass. If the end mill rotates slower, then each cut has to remove a large amount of material, which can overload the bit and cause a crash.

With the first two boards, because the motor slows down too much under load, it would not be optimal for CNC. We believe that there were some communication issues and misunderstandings with the motor manufacturer for this requirement, and so after a lot of back and forth and some group testing on a video call, we were able to sort everything out and are expecting to test the third version soon.

What is exciting is that the manufacturer has been able to get a response time of around 40ms, which is faster than the original Makita RT0701 and its brushless motor counterpart. Although in practice, this probably won’t make too much of a difference since users are not likely to load their motors to the extent we had in testing, this in theory means that the cutting speeds and loads will stay more consistent.

There are still some details to iron out for the motor, however, primarily in the additional tuning of the motor.

We’ve started doing some testing of loading the motor a certain amount and releasing the load quickly. You can see there is an initial amount of time where the motor slows down at first, and then when the load is released, it speeds up for a moment before returning to its original speed. The main reason we suspect this is happening is because the tuning of the motor was originally done using a motor without the larger shaft holding the collet. Because the final assembly of the router has more inertia, the tuning of the motor is not correct.

To fix this issue, we’ve sent one of our prototypes to the manufacturer to do additional tuning to reduce this variation.

We believe that there may be some limitations in the technology on how quickly the motor can react to changes in load, and so we are exploring other methods such as using an encoder or speed sensor in line with the motor. However, we do believe that with proper tuning, the motor will be able to perform within the scope of this project.

In the meantime, our co-op students are building a bit of a makeshift dyno to test the routers and spindles.

Dyno project
RPM logging

SuperLongBoard

Check out the completed SLB box, which will be the ones reaching the first 475 users soon! We have received our first batch of SLBs and are prepping them for shipping.

Due to some delays on the die for the box and the e-stop PCBs, there may be a few days we are waiting next week, but we are working on shipping the first 100 SLB before the end of the week if possible.

Testing jig for SLB

CO2 Laser

I’m excited to have witnessed the first firing and testing of the CO2 Laser currently in development. Ikenna will be making an update post soon so keep an eye out for that!

First burn tests and focusing on the UltraBeam

Everything you need to know about the AltMill

Hey guys, it’s Andy here. We’re excited to be launching a new CNC machine, the AltMill! This article is designed to tell you everything about the AltMill that we possibly can.

Dragon and Phoenix Carving

What is the AltMill?

The AltMill name is derived from “alternative”, “alter”, and “mill”. We believe that “alternative” fits the namesake of the AltMill because this machine represents a different way of approaching things in engineering, technology, and the way we build our company as a whole. We believe “alter” fits the purpose of the machine, which is to take materials and alter them into new forms, uses, and purposes. And of course, since we use milling as our way of altering the material, we can put these ideas and words together to form the name “AltMill”.

The AltMill is a culmination of new technologies, hardware, experience in manufacturing, and customer feedback. Our goal was to take all of these things and build a machine we felt brought the most value to the CNC user by incorporating our ability to engineer high-performance, quality machines at scale.

Differences and Similarities Between the AltMill and the LongMill

The focus for the LongMill was to be a medium-format, hobby-focused CNC machine. Most users typically are looking to make things with a CNC in their spare time. Because of this, we made the LongMill the most affordable option possible in this size category, while being useful and effective enough to handle just about any CNC woodworking project you want to throw at it. We define LongMill’s success in its ability to make CNCing accessible to the average person.

The AltMill is designed to have the most “effective value” as possible. Its success is defined by how much value it can create versus the cost of the machine itself. In simple terms, we wanted the machine that would bring the highest ROI and productivity. The AltMill focuses on the intersection where we believe the performance versus cost ratio is the highest it can be for these users.

Here are some spec comparisons to highlight the differences between AltMill and LongMill:

  • We’ve tested the AltMill to over 830IPM (although regular cutting speeds are in the 250 to 500IPM range) in cutting speed, while most projects we run on the LongMill run at 100IPM.
  • Based on our deflection testing, the AltMill is approximately 8x more rigid than the LongMill.
  • While the LongMill’s main bottleneck in cutting speed is in motor power and rigidity, the AltMill’s bottlenecks are in spindle power and the strength of the endmills. We estimate that the AltMill could handle upwards of 6KW of spindle power (although it’s not likely we’ll dive into that anytime soon because there are a lot more expensive electrical requirements we need to tackle to get this set up for the average user).

That being said, there is an overlap between these two machines. The process of using the AltMill is basically the same as the LongMill, which means that our resources and education translate over between the two machines, and the ease of use and process of using of both is similar. Additionally, both machines are designed primarily for woodworking use, and the types of projects we expect users to create will be similar as well.

Should I get the AltMill or the LongMill?

This depends primarily on your budget and what you want to do with the machine.

Here’s why you would want a LongMill:

  • You are budget-conscious. The total setup cost for the LongMill is roughly half the cost of an AltMill. Additionally, replacement parts and maintenance is cheaper overall.
  • You have limited power in your shop. We recommend having two breakers to power your LongMill and accessories such as the computer and dust collection, but in some cases, you can get away with one. On the other hand, you may need to connect the machine, dust collection, and spindle on separate breakers depending on power availability in your shop.

On the other hand, here’s why you would want an AltMill:

  • You need speed. The AltMill can cut significantly faster than the LongMill. We regularly run the AltMill over 5x the recommended milling speeds compared to the LongMill. The productivity of the AltMill is much higher.
  • You need more working area. The largest version of the LongMill has a working area of 48×30″, whereas the AltMill can cut 48×48″. This means you can process half of a standard 4×8′ sheet in one setup.
  • You need more precision. With ball screws and linear guides, the overall precision of the AltMill is better. In practice this may not matter as much with woodworking projects, you may see a bigger difference in the accuracy of parts for materials like aluminum.

AltMill Budgeting

Of course, the higher performance of the AltMill comes at a cost. We discussed the breakdown of costs one should expect with the LongMill in our article here. Here are some rough estimates on what you should budget (in USD).

Based on my estimates, you should budget around $4500USD to fully set up an AltMill from scratch, about double the cost of the LongMill. You may have some of these items already which will lower your costs.

#image_title

The machine itself ($2950USD)

This is the machine and all the doo-dads to have a working machine, minus the spindle, wasteboard, and computer.

Unlike the LongMill, the AltMill comes default with table legs, so you don’t need to build a bench for it. Additionally, the AltMill comes with inductive homing sensors by default.

Based on our rough estimates, shipping within US and Canada should cost between $150 to $200USD.

Spindle and dust shoe ($515USD)

We recommend users to use a spindle with their machine, because running the Makita router on the AltMill will make the life of the router very very short and unenjoyable. We expect the majority of users to order the Spindle and Dust Shoe Kit which we offer as an add-on kit.

For users wanting to add their own spindle, we expect after all of the parts, wiring, and extra things you’ll need with the machine, it’ll run around $500USD on the low end, but could cost up to a few thousand dollars for some really high end spindles.

We are offering a 1.5KW 110V spindle at this time because we believe it will accompany the most number of people at the launch of the AltMill. However, we will likely offer higher-powered spindles requiring 220V in the future. Users planning to implement higher power spindles should also budget the cost of hiring an electrician for any extra work.

Dust collection ($300 to $1000USD)

Assuming you get the Spindle and Dust Shoe Kit from us, the dust shoe comes with a 4″ hose mount. To get the full performance from the AltMill dust shoe and to keep up with the cutting rate of the machine, a dedicated dust collector should be used. A regular shop-vac can be used, but may not be able to keep up with the waste material generated by the AltMill when running quickly.

Computer ($250 to $800USD)

The AltMill and the LongMill share the same system requirements, which can be found on the System Requirements page in our Resources.

At minimum, you will need a computer to run gSender to control the machine.

Wasteboard ($50USD)

Like the LongMill, you will need to mount a wasteboard to the machine. A 4x4ft sheet of 3/4″ is recommended as the base. On the AltMill, the wasteboard is mounted to it’s frame using screws. You can additionally purchase t-tracks if you wish to add some workholding to the wasteboard directly.

Tooling ($100-200USD)

You’ll also probably want to order some endmills and bits for your machine. You can use the same tools between the LongMill and AltMill, although we will likely expand our range of endmills to accommodate the larger collet sizes we can use with the AltMill spindle. Tooling costs vary widely, but we sell affordable end mills on our store.

Software ($0USD to $699USD)

The AltMill works with our free gSender for all of your machine control needs, along with additional features such as machine calibration, firmware changes and updates, surfacing g-code generation, and more.

For CAM, all of the software that works with the LongMill also works with the AltMill. You can learn more from our Software Resources. There are both many free and paid software options for the AltMill.

The most popular software, and the one we recommend frequently is VCarve Pro, which offers advanced 2D and some 3D carving features, flip-milling set-ups, 4th axis support, and more. This software has a one-time cost of $699USD.

What to Expect When You Get an AltMill

Please note that at the time of writing, we are starting the first production run of the AltMill. You may find relevant information about the Order Status page or in our Production Updates in the blog.

Shipping and Delivery

The AltMill comes in three large boxes. They have been specially designed to fit compactly to reduce shipping costs as much as possible. We estimate each box will weigh about 70lbs. With UPS or other courier shipping, you should expect transit times within the US and Canada to take around 1 week.

Packaging mockup

Based on some general shipping cost calculations, customers should expect to pay around $150-250USD for shipping within the US and Canada for an AltMill.

Set up and Assembly

Large portions of the machine will come pre-assembled, but final assembly of joining the sub-assemblies will be necessary. Most of the machine is put together with M5 and M6 screws, and a set of metric bits or allen keys are required to finish the assembly. Additionally, you will need to provide and mount your own wasteboard. We recommend using 3/4″ MDF.

We will have resources to help users set up and assemble their AltMill on our Resources in a similar style to the LongMill and other products we’ve created. Users should allocate between 3-5 hours for the assembly and set up.

Users will also need to provide a computer that can run gSender. More information about downloading gSender and details about its setup can be found in the gSender Resources.

Operation

The AltMill works just like the LongMill and most other CNC machines. We’ll provide more guides on setting up the machine for different types of work. Beginner users may find content in the LongMill Resources relevant. We expect this group of users to be a bit more advanced, and some of the content and information will reflect this. Additionally, users should expect revised content such as updated feed and speeds and installation for the add-ons to come soon.

Maintenance

The AltMill maintenance is simple, but needs to be done regularly to prevent damage to some of its components and have optimal performance and longevity.

Here are some estimates for bearing maintenance:

  • Ball screw and supporting bearing components should be greased every 3000-5000 hours, or a small amount of greasing done more frequently
  • Linear guides should be wiped and oiled for every 100km of use. With full-time use, users should oil their linear guides every 3 to 4 weeks.

We will provide thorough resources and guides to perform proper maintenance for the AltMill.

Engineering

Extrusion Design

The process of designing the LongMill extrusions gave us a lot of experience in designing extrusions for high precision and rigidity critical applications. Since then, we’ve used this knowledge to manufacture parts such as the t-tracks and the enclosures for the SLB. This experience was able to help us understand the quirks and strengths of aluminum extrusion in the construction of the AltMill.

The AltMill extrusions are big, especially compared to the LongMill, which lends to its strength and rigidity. If you want to check out the video where we talk about this more, see it below:

Clip from the video

To make sure that the frame and axis of the AltMill are straight, a few additional steps have been incorporated to make extract as much performance and accuracy from the extrusion as possible.

First is in our process of extruding and milling the critical surfaces flat. Once the extrusion is pressed, it gets annealed and straightened, which allows for fairly high tolerances. However, there can still be sub-tenths of a millimetre deviations in the tolerance and straightness of the extrusion. To eliminate this, mounting surfaces for the linear guides and the ends of the extrusions have been machined to ensure they are flat and parallel.

Second, additional extrusions that make up the base of the machine are designed to help keep the Y-axis rails straight and parallel. Each of the bottom rail ends have been machined to make sure they are of an exact length, so the distance between the two Y rails is extremely accurate. This also provides additional support and rigidity to the machine, plus having mounting points to make it easy to add a wasteboard to the machine.

Test assembly of the AltMill “bench”

Users will find that the extrusions also come with many different design aspects, such as t-tracks and mounting points for motor components, legs, drag chains, and more, allowing the machine to be easier to assemble and maintain while saving costs by reducing the number of parts needed to build the machine.

Design for Shipping

One of the challenges that come from getting a CNC machine, especially a large one, is getting it from the factory to the door. Most machines in this range do need to be shipped in multiple boxes and the AltMill is no exception.

To make it as inexpensive as possible to ship, we have made a lot of considerations such as:

  • How can we maximize the travel of all of the axis while keeping the rails as short as possible?
  • How can we keep the weight of the machine low while being structurally rigid?
  • What items are we going to make default or optional, and how will we design the packaging around these considerations?
  • What sort of weight and size can the packages be and how will it affect handling?

The AltMill is designed to come in 3 large boxes weighing around 60 to 70lbs each.

Design for Assembly

Because of the more complicated procedure for assembly of the AltMill, a lot of consideration was made to make the machine as easy to assemble as possible internally. Here are some considerations for the design for assembly:

  • Mounting surfaces for high-precision components like the linear guides and bearing blocks are machined and toleranced to help initial assembly onto the rails and gantries pre-aligned.
  • We’ve worked with our fasteners manufacturer to use custom make screws with pre-applied thread locker, reducing the chance of screws coming out from vibration and avoid the need to apply thread locker manually
  • Reutilization of linear guide assembly procedure, testing, and torquing of screws to ensure smooth movement of the assembly used in the LongMill Z-axis.
Machining tolerances for the AltMill X-axis
Back of the X-axis gantry where the linear guide blocks mount

Linear Guides and Ball Screws

One of the biggest differentiators between the LongMill and the AltMill comes from the use of linear guides and ball screws. While the LongMill’s v-wheels and lead screws offer an affordable, simple, and forgiving linear motion experience, ball screws and linear guides offer another step above in precision and rigidity.

Our ability to integrate linear guides and ball screws comes from a few different areas of expertise we’ve developed over the years and additional research and testing we’ve done in this project.

Linear guides and ball screws are much more expensive than v-wheels and lead screws. However, we have been able to source them at a lower cost because we are able to order them in larger quantities. For context, ordering them at wholesale can cost 1/3 or less than retail prices, and we can pass those savings onto the customer. Second is our ability to understand the differences and context of the cost and in-practice differences between different linear motion components. The AltMill doesn’t use the highest-end, most expensive components because we know that for our users, they won’t experience or notice the benefits of it. By being able to balance cost and performance, we’re able to choose the right components for our application.

We’ve also had the chance to work with different linear motion components in the production of the LongMill Z-axis, as well as through other internal projects. Through this experience, we’ve been able to better understand the process of assembling and working with different linear motion components at scale, see firsthand the results of long term use and lack of maintence through LongMill users, and test linear motion parts from different designs and manufacturers.

We do expect some new challenges with working with linear guides and ball screws, mainly in educating users on proper maintenance of these components. While these components can last a very long time, proper cleaning and lubrication is critical to make sure they perform properly. From our experience, the biggest factor in the longevity of these components is in the proper lubrication. It should be noted that one of the main downsides of using bearing-based linear motion components is that lack of lubrication can dramatically accelerate wear and cause catastrophic damage. To mitigate this, we plan to provide proper instruction and lubrication supplies to make it easy for users to perform proper maintenance for their machines.

Tramming Functionality

For those who aren’t familiar, tramming is the process of adjusting the position of the router or spindle, typically by tiling it forwards and backwards, and/or left to right. This ensures that the cutting tool stays aligned with the Z-axis. Having a machine out-of-tram can result in things such as ridges or artifacts, especially on surfacing operations with wider bits.

The AltMill, as far as we’ve been able to find, is the first hobby CNC machine to have nod adjustment, allowing for the X-axis rail to tilt slightly forwards and backwards, eliminating the need to shim the spindle mount. Additionally, the AltMill offers tramming on the spindle mount, which allows the user to move the mount left and right slightly as well.

A combination of an eccentric bushing and shoulder bolts are used in the Y gantry and X-axis to provide nod adjustment (bolt removed for clarity)
Similar to the nod adjustment, the router mount uses an eccentric bushing to adjust left-right alignment

I should note that the eccentric bushings are only used for tramming. In the regular set-up, they have a bolt securing through them, or a shoulder bolt which aligns them to the gantry by default. Also, I should note that currently the tramming for the Z-axis is only built into the default 80mm mount. We do have mounting options for the LongMill 65mm mount, but this does not have tramming.

It should be noted that having the machine assembled in the default position will be within spec enough for virtually all users. Most of the relevant parts are machined to very high tolerances, and there are also many reference mounting points and hardware to help have the machine aligned and straight out of the factory in the assembly process. It is unlikely users will be able to easily make meaningful adjustments without tools like a tramming tool or gauge and a surface plate, and we recommend keeping the machine at its default position.

Why did we add this as a feature? To be completely honest, I hadn’t even thought about it, but one day when I asked Daniel if he’d thought about it, he just shrugged and said he designed the whole thing just for the heck of it.

SuperLongBoard

While the SLB initially started off as a project for the LongMill, many of the technologies and improvements will be present in the AltMill as well. We are currently in the development of a new version of the SLB (code named ALT-SLB or SLT-EXT) that will allow the use of external drivers that the AltMill uses for its motion system.

SLB-EXT, with plugs for external drivers

The main focus of the SLB was to improve stability and reliability of the LongMill’s motion control. We believe the development in this area will translate to rock-solid connection between gSender and the controller. Additionally, on-going development to have a dedicated, on-board computer will be able to be used for both the AltMill and LongMill.

There are some other important aspects and features that we believe are very relevant for the AltMill, including*:

  • RS485 Spindle control
  • Communication between the controller and computer using Ethernet and USB-C
  • Programmable physical buttons to trigger g-code commands and functions such as homing, moving the machine to a corner, and more
  • Tool Length Sensor and Tool Changer support
  • Ability to manage more pulses per second for higher motor speeds
  • Squaring and calibration functionality
  • Full independent 4th axis support

*Please note that some of these features are in-progress and are not ready for release yet.

Along with the “brains” of the board, we are also working on an external power-switching system to distribute power to the motors and interface with the e-stop to ensure the power is cut completely in the case of an emergency.

It should be noted that the SLB and the SLB-EXT are not interchangeable. The SLB is primarily focused on being a controller designed to be plug-and-play with the existing motors on the LongMill, and does not support external drivers for the X,Y, and Z axis. The SLB-EXT does not have integrated stepper drivers on board and must use external drivers.

48V Power Supply

During the pandemic, due to the chip shortage, the original power supply that we were using for the LongMill became extremely expensive due to a specific chip needed for manufacturing. Because of this, we spend a lot of time looking into alternative power supply designs. By switching to a design more commonly used in lighting applications, not only were we able to decrease the cost of the stepper motors, we were able to improve the reliability of the power supply because we were able to internally encapsulate the components to protect it against moisture and vibration. Additionally, we were able to more easily make larger power supplies because of the improved heat dissipation properties of the new power supply.

With this change, we’ve seen a dramatic decrease in power supply-related issues for the LongMill while being able to provide more than 20% more power than the previous design. The AltMill power supply offers nearly double the power of the LongMill power supply at 48V, which allows for additional benefits.

The main benefit of a 48V system is the ability to run the motors faster. Stepper motors lose torque the faster they turn, which means that the faster the machine moves, the more likely it is to lose steps. Increasing the voltage allows stepper motors to “flatten” the torque curve, allowing more torque at higher speeds. With our initial testing with open-loop steppers, we were able to see some small gains in motor speed by about 20%.

Closed-loop stepper motors

When we first started prototyping and testing the AltMill, we quickly found out that one of our bottlenecks was the speed of the open-loop steppers we were using. Over the last few years, the technologies behind motor control has greatly improved, and the cost of these components has come down as well. Additionally, because of our ability to purchase the components at a higher volume, we’re able to also make the machine as a whole more affordable as well.

Closed-loop steppers are motors with an encoder that provide the driver with feedback on the position of the motor. With this information, the driver can correct the position of the motor in real-time, compensating for lost steps and adjusting power consumption based on speed and load.

With the LongMill open loop steppers at 48V, we were able to hit speeds of around 4000-5000mm/min. With the new closed-loop motors, we were able move the machine at around 25,000mm/min before the ball screws start vibrating and jam up the machine. We were also able to push the AltMill all they way to 5000mm/min^2 acceleration rates, 6-7 times higher than the LongMill’s defaults.

There are a number of additional features that make the closed-loop steppers an exciting part of the AltMill because:

  • The motors can detect crashes and jams, then send an alarm to the controller to automatically shut down the power and spindle down, making the machine safer
  • Because of the higher efficiency, they can run cooler and a smaller power supply than what an equivalent open-loop stepper system would need
  • Noise and vibration are generally lower

Spindles and VFDs

Given how powerful the AltMill is, having a spindle is a must. Looking into different spindle options has continued to allow us to look into many different considerations when it comes to picking the right one.

Testing deflection on a spindle

As I talked about in the past in this article, the quality of different spindles varies widely. Because of this, we’ve been hesitant in offering a spindle option since we knew there would need to be a lot of work involved in finding the right manufacturer for it. However, now that we need to get one for the AltMill, we’ve learned a lot about them. Here are some of the considerations about the spindle sourcing including:

  • The bearing quality, size, and configuration and how it affects the longevity of the spindle
  • The deflection from the spindle itself
  • Cost and availability
  • Cable quality and durability
  • EMI
  • Communication protocol like PWM, analog 0-10V, and RS485
  • VFD efficiency and control technologies like vector control
  • Programmability and ease of use of the VFD
  • Collet quality and size
  • Runout

We have a pile of VFDs and spindles in the testing phase. We haven’t settled on a specific combination yet, but we are planning on choosing one soon.

At launch, we plan to offer a 1.5KW 110V spindle that can run off a regular North American outlet. We are currently in the process of testing a 4KW spindle option that will require more power.

Table Legs

One of the main reasons why we didn’t end up making a 4×4′ LongMill is because we would need to move away from the “mount the machine to a big piece of MDF” design. Basically, since the machine needs to be larger than it’s working area. If we wanted to go for a 4×4′ working area, the material the LongMill sits on would need to be a bit larger (probably around 6×6′).

I should also add here that the main reason we chose a 4×4′ working area for the AltMill is because most standard sheet goods like plywood and MDF come in 4×8′ sizes, so the AltMill can process half the sheet at a time. I should also note that because of the back end of the machine is open, you can pass materials through the back, so you can put the full sheet on, cut into it, and move it out the other end without cutting it in half.

AltMill table mockup

If we were to have the AltMill come without legs, users would need to have a larger bench, also about 6ft wide and long, to accommodate. If you took two 6ft wide workbenches from your hardware store and put two of them together, you’d probably be paying around $800 for the pair. The extra legs on the otherhand, cost about $150USD and are made from bent and welded sheet metal in a local shop near us.

Additionally, the AltMill carries a lot more inertia when it moves, which means that the structure it sits on needs to be as solid as possible. This is why a lot of industrial machines use cast steel and welded steel frames, since the mass of its structure dampens the movement and reduces vibration. Of course, we’re not getting into that realm, but we knew that if we built our own table legs, we could make sure it was to the specifications and level of stability we would want to see users have by default. Additionally, we wanted to take the chance to build in a few extra features including:

  • The ability to add extra shelves underneath by screwing in some extra 2x4s to the pre-cut holes. These shelves are also designed to be 4ft wide at the front, so you can keep half sheets of material under the machine. Having the extra cross bracing adds more stability to the structure as well.
  • Add leveling feet, to make sure that each side of the table is contacting the ground, to avoid wobbling. These could potentially be swapped for casters, so the machine can get moved around.
  • Add mounting points for other items such as the VFD and e-stop (still in progress to solidify exact positioning)

We’ve found that at some really high accelerations and fast movements, the machine can “walk”, and so we’ve been considering sandbagging the machine to see if we can up the speeds even further. It’s probably not practical, but something of consideration.

Performance

One of the things I sort of realized is that for most hobbyists and beginner-level users, it’s hard to get context on how much “performance” a CNC machine has. So We figured that the best way to show it was to film videos on the machine doing stuff. So here are some videos below.

Here’s also another non-scientific test.

Daniel standing on the Z axis of the AltMill

Here are also some of Daniel’s notes about the rigidity, which should provide some context on where the AltMill stands based on his estimates. I should note that these are just rough estimates based on calculations from tests done by other users, so they may be incorrect.

Market and Competition

I see a lot of similarities between the launch of the original LongMill and the original AltMill in a number of ways. When the LongMill was first launched, it competed directly against machines such as the Shapeoko 3 and the XCarve. At that time, there were a lot fewer hobby CNC companies, and those were the most popular machines at the time. At this stage, cost was a really important barrier for people to get into the hobby, and we knew we needed to design something robust enough for people to use the machine to make valuable products while being affordable enough to get that ROI back as soon as possible.

An interview video with one of our first beta testers for the first generation of the LongMill

First, we made the LongMill design as simple as possible. This allowed us to have fewer parts, which not only reduced the cost overall, but made it easier to pack, assemble, and ship the machines. Second, we tackled the most common complaint and limiting factor that users (including Keith) complained about, which was the use of belts. By using leadscrews on the LongMill, we made this machine an even more compelling option. Even though our company was not very well known at this point, I believe that a combination of our approach to designing a better machine at nearly half the cost of the competition got a lot of new users on board into the hobby.

The idea of a 4×4′ CNC router table isn’t unique. In this market, the AltMill competes directly with machines like the Onefinity Elite Series, Shapeoko 5 Pro, Shapeoko HDM, and XCarve Pro Series as a mid to high-end CNC hobby router. However, by reading and studying the feedback from users of all of these machines, we’re able to focus on addressing these things in the AltMill design itself. I won’t go into them here in this article specifically, but some examples include:

  • Using closed-loop steppers to eliminate the issue of losing steps
  • SLB and SLB-EX development to address static and EMI-related control issues
  • Plug and play 4th-axis option

Additionally, although the price difference between the AltMill and it’s direct competitors is not as dramatic as it was for the LongMill, it is still priced extremely competitively and minimum $1000USD less expensive than it’s closest priced counterparts.

Moving beyond that, we believe that the AltMill may have some customers looking for high-end hobby to semi-industrial machines who are also considering the Avid PRO4848, Phantom SC44 and StepCraft Q.404CNC. These are machines in the $8000-13,000USD price range.

Just to be clear, the AltMill is not a direct competitor or replacement for a semi-industrial or industrial machine. A lot of these higher-end machines may have things like welded steel frames, which is technically better for high-speed cutting. In practice though, there is still a lot of overlap because both machines can do the same projects and run at pretty high speeds.

I sent an email to Avid to ask about the specs for the PRO4848 and this was the response:

In any case, Daniel and I had a long discussion about where the AltMill falls. It feels like the AltMill with its speed, power, and price, blows all of the hobbyist-level machines out of the water. When comparing the AltMill to some of the upper-level machines, the distinction between them also starts to shrink.

Here are some of my final thoughts:

  • If you’re looking for a hobby machine in the $3000-6000USD range, the AltMill has the highest performance in this category
  • If you’re considering a semi-industrial machine above the $6000USD range, all the way to around $15,000USD, an AltMill might still work for your application, or you can buy more than one to double your production ability.

There are a few areas I do feel like we need to work on to make the AltMill even more compelling and competitive which include:

  • An onboard computer and touch screen, similar to the Masso controller. At this current time, we are working on our own computer and touchscreen setup, and expect to have more news in the coming months
  • A more powerful spindle option around 3-5KW
  • Toolchanger
  • More advanced work holding system, such as a vacuum table.
  • Larger machine options

Future Goals for the AltMill

Engineering Trickle up and Trickle down

There is a lot of engineering that can be transferred between the lower end of our product line (the LongMill) from the AltMill to make improvements, as well a lot of things we learned from the LongMill in the manufacturing of the AltMill.

Some AltMill development that may trickle into LongMill include:

  • Toolchanger
  • 3 phase spindle options
  • Z-axis and router mount designs
  • Bench/table designs

Some LongMill developments that may trickle into, or have already impacted the AltMill feature set include:

  • SLB features, such as USB-C and Ethernet, RS485, tool length sensor input, physical macro button support, external relay control, improved laser rastering, and more
  • Extrusion manufacturing and quality control processes
  • Customer service and resource development processes

Additionally, as we make more volume of each new product and technology, we can bring the price point and improve accessiblity because we can leverage economies of scale and the fact that the work of each project impacts more people overall.

New Markets and Verticals

So a bit of backstory. Sometime last year, we started rediscussing the potential of designing and developing a new machine. From this discussion, we had three contenders, the AltMill, the CO2 laser, and some sort of mill dedicated to milling metals. The decision was made that we should try to make all of them, one way or another. Both the AltMill and the CO2 laser are in development now, which leaves the metal milling machine project still up in the air.

Just for a bit of context, the engineering team had a general idea of making something similar to the Langmuir MR1, or basically an affordable, hobby-level gantry mill.

The plan was to make the AltMill first, and shrink the AltMill down to make our a gantry mill using the same core parts and electronics. Based on Daniel’s testing on material removal rates, it looks pretty promising we can build a competitive machine in this realm.

In my travels in the past year to Brazil and China, one of the gaps that I’ve seen has been in the way CNC machines such as VMCs are used for production. While large industrial machines have their place in making high-precision parts, their running costs are high, even if the parts are small, of lower value, or don’t require small tolerances. The idea is, rather than using expensive big machines all the time, smaller, less expensive machines could be used for milling batches of smaller, lower tolerance parts. Additionally, factories install and use several machines for the space and cost of one large machine to scale up their throughput.

The team and I have a lot of interest in exploring this avenue next, I expect the things we learn from AltMill design and production will help us tackle this new vertical.

That’s not to say we have some other things on the docket of considerations, such as:

  • Making a machine to target hobbists looking for a machine option below the LongMill’s pricepoint, size, and capability like the Mill One Plus.
  • A larger version of the AltMill for 4×8′ cutting
  • New accessories for the AltMill and LongMill, as well as support for other machines outside of our ecosystem

We’ll probably worry more about this after the launch and first production of the AltMill. But if you want to share your thoughts about what you want to see come from us in the future, make sure to share it through our form.

FAQ

Here are a list of questions people have been asking us. Please note that there are more FAQs on the main AltMill landing page.

Will the Spindle Kit be available for sale on its own?

A lot of people, especially LongMill owners, have been asking about if the spindle kit for the AltMill will be available as a separate option. The short answer is, yes, because we won’t stop people from buying stuff from us if they really want to. However there are a few things to consider.

First is that the 80mm spindle is pretty big and heavy. There is some testing and validation we’d want to do before making it an “official” LongMill spindle option because of this.

Second is that we’re also working on a “Sienci Router”, which we plan to launch in the next couple months, which will be a drop in replacement for the Makita router, with speed control, and a brushless motor that allows for around 800 to 1000 watts of usable power, or around double the power output of the standard Makita router, which will be cheaper, lighter, and more suitable for the LongMill.

I do recognize that at around $500USD, this spindle kit still offers a lot of value, and is priced competitively compared to other, plug-and-play kits. I think there will be a number of people outside of the LongMill and AltMill ecosystem that we can serve with this kit as well, so we will explore that option in the future.

Why offer a 1.5KW spindle when the AltMill can handle a 2.2KW or larger spindle?

Based on our testing between a number of 1.5KW and 2.2KW spindles, we felt like the power difference between these two options were not large enough to justify the extra complexity of having the extra 220VAC wiring. We are continuing to do both real-life testing and establishing a bench test to measure the true output of the spindle.

Based on our estimates, to get the full potential from the AltMill, a 3-5KW spindle would be best suited for the machine. However, most households would need to invest in having an electrician set up the appropriate power outlets installed to allow this use.

This comes to another question, which is “how powerful do we really need to make the AltMill”? The other bottleneck that we need to address is the strength of the end mills. Based on our real-life testing, another reoccurring issue is that if we run the job too fast, the end mills break. For us to cut even faster, we’d need to use larger-diameter end mills. In some cases, this may be useful, such as in projects like surfacing and boring lots of material, but this is a small subset, and may not make sense to offer a spindle designed to do that all-day, every-day, if the alternative is to slow down the machine during these types of jobs.

Will you be making a larger/smaller AltMill?

Very likely yes, since the AltMill offers a platform that can be made larger or smaller by changing the size of the rails and linear motion parts, as well as mixing and matching some of the components. Making a smaller machine is probably the easiest, since it doesn’t have the same problems as making a larger machine, as we will discuss next, and we have some ideas for making a machine more dedicated to milling smaller projects or metals, similar to the Langmuir MR1.

There are two main challenges to making a larger machine, which in this case is probably going to be a machine that is for cutting 4×8′ sheets. The first is in the power transmission system for the Y axis. The longer the ball screws get, the more prone to “whipping” there is, especially at higher speeds. To avoid whipping, we would need to consider other options such as a rack and pinion system, which isn’t something we have a design for. The second challenge is in shipping. The AltMill comes in three large boxes which can be shipped with a courier like UPS. A 4×8′ machine would need to have parts that are around 10′ long to make up the Y-axis, which would be heavier and harder to transport. This adds extra complexity in shipping and packing these parts safely.

That being said, we definitely see a strong interest for a 4×8′ machine. If you’re interested in one, please share your feedback in our Product or Feature Request Form.

Will there be a toolchanger available for the AltMill?

We do have a lot of people asking about this feature. With the SLB’s input and output suite and stronger Z-axis capabilities, the AltMill will be able to support a toolchanger in the future.

As in typical Sienci Labs fashion, we want to do a lot of testing and see if we can build something high-quality, reliable, and affordable, or work with another company to add this feature.

At this time, users will need to invest in retrofitting their own aftermarket toolchanger to the AltMill.

Can I buy the AltMill without the table legs?

Initially we were planning to have this a separate option, we’ve made the final decision to have the table legs a default part of the AltMill kit. There are two reasons for this.

Survey data about the table legs

First is that based on survey data from over 400 respondents, the majority said yes to wanting to order them. To save money overall, we decided that it would be most effective to design the packaging with the legs included.

This in theory leaves a small but not insignificant group of people who don’t want the table legs (24%), which is still a lot of people.

Back when we didn’t have a default router mount size (65mm) for the LongMill, a lot of people ordered the 80mm mount thinking that they were going to toss on a spindle instead. What happened in reality was that most of these people decided to not go down that route and just get the Makita router we’ve been recommending. Because these customers would need to order the 65mm after getting their machine with the 80mm mount, we found that it was just more cost-effective as a whole to make the 65mm the only option, and have the other mount sizes at an additional cost, especially if you took into account the extra customer service, shipping, packaging complexity, returns processing, and time wasted.

I feel like the table legs may follow a similar story because designing and making an AltMill bench will probably cost a lot more time and money than slapping on the ones we’ve made. I believe if we make the legs optional, there will be a lot of people who don’t order the legs at the beginning, realize this fact, and then have to spend the extra shipping cost to get a set of legs sent to them.

For someone who truly doesn’t need the legs, yes they may go to waste. However, in the grand scheme of things, making the legs defacto makes the overall cost to the customer lower.

AltMill Launch and Production Schedule

Hey everyone, we’re excited to share our launch date for the AltMill.

The AltMill will launch on Wednesday, March 27, 2024 at noon, EST. You can access the order page at https://sienci.com/product/altmill/ when the page goes live.

Our livestream will be happening on the same day at 1PM EST. Please join us at https://www.youtube.com/watch?v=QufxkgPRxCU

If you’d like to learn about the AltMill itself and the engineering behind it, please read our Everything You Need to Know about the AltMill article.

For more information about the AltMill project, please see https://sienci.com/altmill/. If you have any questions about the AltMill, please see the FAQ.

Pricing

The AltMill will come at a base price of $2950USD/$3990CAD, which includes the table legs.

Users can also purchase the Spindle and Dust Shoe Kit for an additional $515USD/$690CAD.

The first 50 machines

As noted in past updates, we’ve jumpstarted the process by starting production on the first 50 AltMills in December 2023. This allowed us to tackle some of the major unknowns/questions, such as:

  • What will it cost for us to make the AltMill?
  • How difficult will it be to manufacture certain critical parts, such as the rails, linear motion, and table that we were most concerned about?
  • What will our QA and assembly process look like?
  • What sort of performance and reliability should we expect from the AltMill.

As of the time of writing, the plan is to offer the first 50 machines directly to select users and for internal use before our “main batch”. The first batch of AltMills represents our trial-run for production and comes with a couple of you-should-knows, especially if you’re planning to be one of the users in this batch.

We also plan to collect comments and feedback from our first batch of AltMill users to improve the user experience and tackle any initial quirks and issues in the first part of the product launch.

Some parts are still in shipping and manufacturing, and we expect the first 50 machines to start shipping in May 2024.

The “main” batch

This is what we expect most users will be part of. We will begin taking pre-orders at the end of March. Please check www.sienci.com/altmill for more information and a link to the order page.

The goal for our first main batch is to build enough units to leverage economies of scale to make our relatively low price for the AltMill viable. This not only involves the unit cost of the machine, but the work and labour needed to build each batch of machines, which might include work done to set up tooling, packing stations, and the ordering of parts.

Please note that to place your order for the AltMill, the total amount must be paid to hold your place in our queue.* You may cancel your order at any time before your order ships for a full refund. Once your order is in the possession by the courier or arrives at your door, our standard store policies apply.

The number of machines we’ll make in the first batch is still undetermined and will be based on the number of orders we get at the beginning of the launch. 

We expect the “main” batch to start shipping in July 2024. However, we will ship orders based on when they were placed, which means that if your machine is in the later part of the batch, you will receive your order accordingly after July 2024.

Future production

If you feel that pre-ordering the AltMill now isn’t right for you, you will eventually be able to order and have an AltMill ship to you in a shorter amount of time, just like the LongMill. However, when this will happen is dependent on when our production capacity can meet the demand for the product, which is unknown at this point.

The goals for the future production of the AltMill is as follows:

  • Have a reasonable lead time for us to build and ship AltMills. For us, two weeks or less from when we recieve an order to when it gets shipped is a pretty good number to hit, but the lower the lead time the better.
  • Produce larger numbers of machines to leverage economies of scale and either reduce the price of the AltMill or invest our increased profits into additional resource development and R&D that benefits the CNC industry
  • Take our learnings from this new product, especially in the production and QA side to create variations to the AltMill, such as a smaller, stouter, more rigid machine focused more on metal milling, or a larger 4×8 machine.

The size of future batches will be adjusted based on demand once our main batch has completed.

How to get updates

We will continue to share and provide updates in our Production Updates which are released at the start of each month at www.sienci.com/blog

Additionally, we will write order updates as we currently do with our other products at www.sienci.com/order-status

For the most reliable way to get news and updates, please sign up for our email mailing list.

March 2024 Production Updates

Hey guys, it’s Andy again with March 2024 production updates. I am currently writing this in China, where I am taking a bit of a “work-acation” but also to visit some suppliers and manufacturers that we work with. 

This also means we’ll film the typical production update video a bit later, probably on the week of March 11th when I get back.

March is expected to be a busy month, especially as we continue to make progress in our projects like the CO2 laser and Sienci Router, as well as prepare for shipping and launch of the SLB and AltMill.

LongMill

Last month we paused shipping for LongMills as we waited for more controllers to arrive. We expect around 100 controllers to arrive in the next week or so (shipped on Monday). Once these parts arrive we will continue to ship machines and clear the backlog. More controllers are expected to finish the first week of March.

Additional production is underway for the LongMill, with motors, power supplies, and fasteners in production now.

LaserBeam and Vortex

LaserBeam and Vortex are shipping as usual. Ikenna and Abeku have developed a riser mount for the LaserBeam which allows for easier use in combining LaserBeam and Vortex to do engravings with the Vortex.

They are also working on some different magnetic mounting designs for the LaserBeam to make removing and attaching the LaserBeam faster and easier, and should have more stuff to share in the coming weeks.

AltMill

This month we have finally put the machine together and started running it through the paces. Check out Daniel’s video on some more updates. I would have been there for the video, but I am currently away.

For more info and FAQ, please check the AltMill landing page.

Testing is showing some promising results. Here’s an excerpt from Daniel’s notes about the rigidity of the machine.

Also just finished doing some preliminary deflection testing of the machine with some pretty good results. This was done using the standard Sienci testing parameters/setup for the most part.

  • In the Y-axis, we have 0.003” of deflection at the tool with 80N applied 
    • This is 1.05 N/μm rigidity
  • In the X-axis we have 0.0025” of deflection at the tool with 80N applied
    • This is 1.26 N/μm rigidity

For comparison sake, here are some misc numbers of other machine’s rigidity:

  • 0.1515 N/μm in the Y-direction of the Shapeoko 3 XXL
  • LongMill MK2 48” Y-axis rigidity sits around 0.13 N/μm
  • LangMuir MR1 2.9188 N/μm in the X-direction, 4.3782 N/μm in the Y-direction
  • Onefinity (with added ‘stiffy’ rail) estimated to be 0.5 N/μm (realistically much less) based on one user’s measurement of ~1 N/μm at the bottom of the Z20 plate.
    • This pretty much only accounts for beam bending in the Y-direction, and not much torsion for which is the Onefinity’s achilles heel. It wouldn’t surprise me if this was even as bad as 0.3 N/μm.
    • This is mostly speculative, so not a fair comparison but worth mentioning.

I also checked the X-axis rail’s isolated deflection contribution. The rigidity of the X-axis rail assembly is ~3.75 N/μm. This is pretty good considering the rail was sized to be 4.9 N/μm and this is real life with extrusion and alloy defects and the like. 

  • For comparison sake, AvidCNC’s 8016 extrusion was estimated to be 3.8047 N/μm. Considering it weighs (I think) like 4 times more than ours, this is amazing.

In other news, we are continuing to put together the online ordering infrastructure to prepare the AltMill for launch at the end of March.

Sienci Router

At the start of the month, we received the sample motor we’ve been waiting on to do another round of development and testing. If you’re not up to date on the development here, make sure to check out the last post.

The new motor is much more powerful, and showing promising results. However, we are waiting on some improved motor tuning to happen as we have found some issues with the speed control to achieve a full 1KW of mechanical output. We are waiting on an updated control board expected to arrive in the next week or so.

Additional to this is that we’ve started exploring more spindle options for applications needing higher power past the 1KW the Sienci spindle can put out. If you saw Daniel’s update on the AltMill, the new machine is so powerful, that even the 2.2KW spindle ends up being the bottleneck in our ability to remove more material.

Eventually, we hope to provide several options, the standard Makita as a simple, powerful, and inexpensive option for routing, the Sienci Router as a step above with more features and power to run the LongMill at its full potential, and spindle options to maximize the AltMill’s performance.

Spring Loaded Anti-Backlash Nuts

I’m excited to say that the first set of the injection molded nuts has arrived. To learn more about this project, please see the long post about them here (put link here). While the T8s overall look good and function properly, unfortunately, we are still experiencing some warping and inconsistent threading on the T12 nuts. Since not all the nuts are affected, we’ve put on the store all of the nuts that are currently ok. We will work with our manufacturers to iron out the issues with the T12 nuts.

Demand for the new nuts has been super high, with all of the T8s already sold out, and with T12s expected to be close to selling out by the time this post goes out. Not to worry, however, we are working on making another batch of a few hundred sets and make sure we don’t run out.

It should be noted that existing LongMill kits will continue to ship out with the original style of nut. Once we catch up on orders sold for replacement, we will start moving to making them a default option for new machines. We currently don’t have a specific timeline for it, but likely in about 2 to 3 months, since production and assembly of the nuts can take a long time.

SuperLongBoard

We’re excited to share that the new SLBs have started production and should be ready to ship in the next few weeks. We are also waiting on parts for the controller and estops to arrive in the next few weeks.

Work currently being done with SLB primarily revolve around checking for reliability and making bug fixes. We’ve also sent the SLB for testing to key grblHAL community members for feedback.

gSender has now been updated to natively support SLB and it’s features. You may have seen a toggle when connecting your machine to allow for GRBL and grblHAL available.

Additionally work on building controllers for the AltMill to provide external driver support, higher voltage, while sharing the same features is also underway, with first versions of the design expected to be ready in the coming weeks. However for the full development cycle, we expect it to take till end of April to have production-ready designs and firmware ready.

In addition to this, we have continued to work on the computer side of the SLB at a bit of a slow pace. However, we have put together this proof of concept where we have attached a VESA mount arm to the threaded holes at the front of the machine to allow for use with a touchscreen, as well as a mount for the computer. This design was created by one of our engineering students working at the company this term.

Feb-ulous SLB news

Greetings all, Chris here again to emerge from my SuperLongBoard development cave and provide you with my news!

For anyone who still has yet to hear about our new and most ambitious Sienci Labs electronics projects to-date, our arguably aptly named SUPERLongBoard, is the graduated version of our current LongBoard and has been designed to be even more SUPER at all things hobby CNC. 

This is a serious upgrade for a serious board, even if we had some fun with the name: culminating in what we hope to be an all-in-one solution and in many cases an upgrade to anything currently in its pricepoint on the market. The result is a 32-bit processing, on-board motor driving, Auxiliary output supporting, 4th axis controlling, laser, TLS, macro button -having board.

With that summary out of the way, you can catch up on past progress by looking at my update last month which also has links to past videos and past updates before it: https://sienci.com/2024/01/25/slb-january-updates/ 

Now let’s roll along to what new stuff I have to update you on this month 🙂

New Stuff

We’ve now pre-sold nearly 450 boards now…. and wow I’m so excited to see that everyone else is just as excited about this board as I am. To put into context: the current batch of boards we’ll be manufacturing was intended to be 500 but had to be reduced to 470 (I’ll mention the reason shortly) which means that we’ve nearly already sold out the whole batch before starting shipping! Needless to say, Andy and I already began discussions about a month ago on what our next steps are going to be if the reception to the SLB ends up being as good as we hope it will be to ensure we can have more on-hand in the coming months if possible.

Now the reason why we had to reduce to 470 is basically a math error, but in short:

  • Right on time to the schedule I posted in my last update – a couple weeks ago – we finally ordered the final production run of SLB boards! After the last run of prototype boards arrived we were able to very quickly redistribute them to Beta testers and vet any last changes we felt needed to be made so we could turn around and start prompting for production.
  • Already knowing that Lunar New Year would interrupt our delivery progress, we’d already ordered every other overseas part that we needed to produce the SLB (except the boards) at the start of January to ensure things would arrive on time by boat.
  • Knowing that we couldn’t do this for the boards since we needed more time to test them, we instead pre-ordered lots of the parts used for the board circuitry in advance as a way to help speed up production after Lunar New year finished. This included important stuff like the STM32F412 ‘brain’, TMC2660C motor drivers, and some other rarer items. We pre-ordered around 525 of these parts, but shortly after realized that we needed to do one final set of prototype boards and this used up 40 of the 525 parts.
  • After we add a small buffer for potential board failure, this left us with about 470 left for production SLBs that we can make available and would ship by air to catch up with everything else so they would all arrive around the same time.

Speaking of ordering parts, let’s take a look at what our full part table looks like right now as we start preparing to intake SLB parts and set up packing stations to check and ship them out:

SubsectionNameAmountShipping Status
Board PartsSLB important PCB components1Pre-ordered and now being used for SLB board production
SLB PCB assembly1Underway, scheduled to arrive by air on March 18th
USB-C Cable1To be ordered shortly after some more testing/validation
E-stopE-stop Button1Ordered, currently on boat and scheduled to arrive March 5th
E-stop PCB1Underway, scheduled to arrive by air mid-March
E-stop Cable1Completed, will be shipped alongside E-stop PCBs
Injection Moulded Case1Completed, currently on boat and scheduled to arrive March 15th
#4 screws3In stock
Enclosure PartsAluminum Extrusion exterior1Completed, currently on boat and scheduled to arrive March 15th
Front Panel1
Rear Panel1
Acrylic Cover1Arrived, to begin in-house production mid-March
Steel Mounting Bracket1Underway, should be made and plated in a couple weeks by our local steel manufacturer
M5-10mm SEMS3In stock
M5 T-nut3In stock
M4 Thumbscrew1Arrived
#4 screw6In stock
PackagingCardboard Box and inserts1Still being designed, should have a 1 week turnaround time by our local manufacturer

We’ve tried to time things out using a combination of boat and air freight for international parts and the shorter timelines of our local producers. Hopefully the information isn’t too much to look at, but the long and short of it is that everything looks like it’ll all show up around the same time!

Some of the great-looking production samples!

In the meantime, we’ve begun working on making space for packing and testing stations, working on designing our quality assurance procedures to test boards before they go out the door, all the while internal testing and Beta testing still continues forward as small tweaks to gSender support and to the Firmware are still being made to get everything working how we like it.

Please still bear in mind though, due to there being things that could still be outside our control, I would still conservatively estimate that SLBs begin shipping out the door last week of March or the first or second week of April. This would cover instances where delivery by boat or air has a slowdown, or we find something with the final batch of boards that need our attention. For example:

  • We’ve now had problems with a couple RGB LEDs out of the 50 prototypes we’ve made where they light up random colours rather than turning the colour they’re told to be. This makes us think that our manufacturer is giving us slightly water-damaged components, so we spoke to them about baking the LEDs before installation and are also working on a way for them to validate the boards before shipping so they don’t show up broken. This is an example of nothing we’ve done wrong, but could pop up unexpectedly on a board as complex as the SLB.

Will there be more Videos?

Yes of course! To be honest, after the inrush of pre-orders after the SuperLongBoard launch I’ve had more of a sense of duty to test-test-test the board as much as I possibly can as a higher priority than filming the boards capabilities. This is because, though I know there’s lots that the board can do, clearly everyone who’s already pre-ordered is already on ‘board’ with the SLB so I don’t think it would be fair for me to spend my time building more excitement for the board when I can instead focus my time on making it more bulletproof 🙂

And I think the nose-to-the-grindstone work has been paying off! There’s been tons of bug squashing and pushing machines to their limits by us and our trusty team of Beta testers:

  • We’ve now had over 15 firmware iterations since the start of the project
  • We’re continuing to work closely with the grblHAL firmware creator to ensure great compatibility between the board and its functionality, we actually sent him a pre-production SLB of his own that he’ll be receiving shortly
  • Major work thus far has included implementing major features like reliable USB and Ethernet, standby current reduction and individual axis holding, 4th axis control, TLS support, action buttons and ensuring their behaviour is predictable in different situations when using the CNC, controlling the custom outputs, honing in on the new LongMill default settings
  • Lots of tweaks have been made to gSender to make it know all the things it needs to know to effectively communicate with a faster and more feature-diverse controller. There have been more delays than originally thought due to continuing to find unexpected edge-cases in how the SLB behaves differently from the LongBoard, but we’ll keep trucking along and expect to be done in time
  • Ikenna has now thoroughly vetted the SLB to work with our LaserBeam laser diode and is very happy with how it’s working, even now using SLB in his LaserBeam livestreams

When it comes to roadblocks we’ve hit, the last month has actually been very good. Beta tester feedback has been a split of 40% gSender compatibility problems, 30% improvements to board documentation, then about 15% tweaks to firmware and the last 15% changes we had to make to the board design before we began production. This is good news because software and firmware are things that we can -and have- continued to work on while things are being manufactured and shipped over.  This means that the gSender and Firmware teams still have most of March to finish polishing things up which seems very doable. The last remaining hurdles are to get things like 4th axis cutting and switching between spindle and laser working as smoothly as we can in gSender, and then completing the remaining documentation tweaks based on Beta testers continued feedback, then writing up a guide on how to do a full board swap for existing MK1 and MK2 owners. I’ll also be setting aside some time mid-March to establish final movement speeds and motor noise that can be expected.

With all that said, videos will definitely still be making more of an appearance as we begin to near the delivery date since I want you guys to see what you’re going to be getting before shipment begins 👍 Alongside this will of course be: continued updates to FAQs, starting to release parts of the SLB manual, starting to release a list of recommended hardware to use alongside your SLB for 4th axis, lights, and more. I might even do some Livestreams so you guys can see more of the board in action and have any of your other questions answered! Some of my current video/livestream ideas are:

  • Speed and noise comparison to the LongBoard
  • Trying and SLB retrofit onto a Mill One
  • I’ve been messing around with a RapidChange unit and am hoping to see if I can get it fully set up and working to answer any questions about the SLB supporting an ATC. If you haven’t seen this unit yet I’m very excited about what it might do to shake up the hobby CNC space for more affordable automatic tool changing, and Don who I’ve been speaking with over the last year is a really nice guy who’s direction I really love so far

Conclusions

If you’ve made it this far, here’s your award 🏅. This is to recognize that you once again lasted through another one of my treacherous, Engineering-writing posts, complete with its poor sentence structure and also a lack of pictures on this post especially.

In all seriousness, thanks for everyone’s continued support for this project and the other projects we’ve continuously strived to do with Sienci Labs over the years. It’s never been easy for us but the kind words and reassurance that the work we’re doing matters really helps to keep us going. Thanks for anyone who’s been helping give me and the gSender team the needed feedback to keep making more cool features and fix iterations that sometimes break things and thanks for posting all the cool stuff you do and helping each other out too.

If you have any other ideas for SLB content you’d like to see, please leave them in the comments of wherever you were when you saw the link to this post whether it was on Facebook, our User Forum, or elsewhere and I’ll try to find them and write them down.

Until next time!

-Chris

AltMill 48×48

Experience next-level power and precision with the AltMill.

You will need to provide your own*:

  • Computer
  • End Mills
  • Set of metric Allen keys, standard bits, and drill for assembly
  • MDF for the wasteboard (4ftx4ft sheet)

*Items such as the spindle, t-tracks, and additional shelving as shown in the photos do not come by default with your kit. There may be some cosmetic differences between the machine in the photo and the production version of the AltMill.

Specifications

  • Rigid frame and linear motion design using HGR15 linear guides and ball screws on all axis mounted to a custom-designed, machined aluminum extrusion
  • Closed loop stepper motors for over 600IPM
  • Support for 80mm spindles and routers
  • Approximately 49″ by 49″ inch working area with a max z-height of 5.5″
  • Footprint of 59″ wide and 62″ deep
  • Inductive sensors for homing included
  • Sturdy metal legs with the ability to add 4ft wide shelving underneath

Pre-order information

Please note that this is a pre-order. Machines are currently in production and some aspects are still under development. For more information, please see the AltMill Launch and Production Schedule article.

AltMills from this pre-order are expected to start shipping July 2024.

February 2024 Production Updates

Hey everyone, welcome to our February 2024 Production Updates.

Media Room and Workshops

Since we’ve moved into our new space, we’ve dedicated an area as a “media room”. The idea is to build a space that allows us to make content more quickly with dedicated space, lights, and machines for filming and education. Additionally, we’ve gotten a lot of interest in doing workshops, and so we’re now looking into planning workshops in the space as well.

If you’d like to provide some feedback and let us know what sort of workshops and content you’d like to see, please check out https://sienci.com/2024/01/10/fill-out-our-cnc-workshop-survey/

LongMill MK2s

Production for the LongMill continues to move smoothly. Orders are shipping out within one week, however we are running low on controller boards. Lead times may get longer this month.

Check out this new racking we got for all of the rails! It looks very visually satisfying.

Injection-molded middle feet that are used for supporting the rails have finally completed production and are on the way to us. We expect these feet to arrive in early Feburary. For those who haven’t been following along on this change, we decided to start injection molding these parts since we make a lot of them using the print farm and we crossed the point where it would be faster and more economical to injection mold them. It should be noted that this change is to improve production efficiency and reduce costs, but won’t make a difference to the LongMill’s performance.

Injection molded feet

The bristles that we use for the LongMill dust shoes have come in earlier this year but we have been dealing with quality issues. We have been able to use some of the good bristles, but we’re also working on sourcing a new manufacturer to improve the quality.

#image_title

We are now starting production on Batch 9 LongMill MK2s. We currently have around around 750 LongMills in stock, and expect to start shipping Batch 9 machines in the spring of 2024.

Spring Loaded Anti-Backlash Nut

Second batch of prototypes

I know a lot of people have been anxiously waiting for the spring-loaded anti-backlash nuts. While they seem simple, these have been a really fascinating but challenging project as we needed to make changes and considerations to the design and manufacturing process of the nut.

For more details about the process of design and making the nuts, I wrote another blog article. The first 200 sets of T12 and T8 nuts are expected to arrive in the first week of February. Please note that the blog article will include more updates once the first batch of prototype nuts arrives.

Vortex Rotary Axis and LaserBeam

Parts for the Rotary Axis have arrived and are being packed and assembled. We have another 300 units in stock now.

Ikenna and Abeiku are also working on a new magnetic mount design and also a riser mount to be used with the Vortex to allow for easier laser engraving on round objects soon, so make sure to keep an eye peeled for that.

AltMill

We have our major components arrived here and are working on putting together and testing the first prototype. Based on looking at the linear motion and extrusions, everything looks great and we’re excited to get everything in to start building the first batch.

AltMill table, Daniel for scale

If you’re interested in ordering an AltMill, make sure to fill out our form.

Here are some other updates:

  • While we have received one set of extrusions, the full batch of 50 sets have had some QC issues and are being worked on now. We expect them to be finished in the next 2 weeks and get prepped for shipping
  • We have received a few additional closed-loop stepper motors for testing and will be working on having them set up for testing
  • We are working with Andrew at Expatria to figure out what modifications we need to make for the SLB to allow for use with AltMill.

Also, check out this new logo that Leandro made for the AltMill.

We are tentatively looking at a launch date for the end of March. We’ll keep people updates so make sure to follow along on the development through the blog and such.

CO2 Laser

Ikenna and his team have been continuing to work on the CO2 Laser. Here’s a photo of the mockup in progress.

I probably won’t be continuing to put updates for this project on the production updates here because Ikenna will make a separate post as updates come. Make sure to sign up for the CO2 mailing list for all updates as they come.

Sienci Router

Testing with the 400 watt motor looks to show that using BLDC is a promising technology and shows that power output even at 400 watts is comparible to the Makita router. However, we feel that to bring the most value to users, having a bit more power will be beneficial since:

  • Cutting using larger bits, such as the surfacing bit causes the Makita router to bog down
  • Additional headroom allows us to run the LongMill faster alongside other future improvements to speed and rigidity
  • Potential to be a viable option for higher-end machines and the AltMill.
  • Creates a differentiation between our router and the Makita router

Having a larger motor is more expensive, but still within our budget. Pricing is still yet to be determined, but we believe that if we have an option around the $250 mark will allow us to provide a tool that sits somewhere between a traditional router like the Makita RT0701 and a 3 phase spindle.

A second batch of motor samples are expected to ship in the first week of Feburary. We are also in the design and sourcing stage for the motor body and bearings.

A section view of one of the router designs

SuperLongBoard

Development continues for the SLB and third version prototype is currently in testing. Here’s some news:

  • SLB resources continue to be developed ahead of shipping
  • E-stop injection molded case, buttons, and circuitry have arrived for testing, and have started on full scale production
  • Enclosure parts are getting prepared for shipping

Otherwise we are just working through general bug fixes and testing as usual.

Demand for the SLB has been strong, and we are expecting to sell out of the first batch before we start shipping, so we are working on

If you haven’t checked out Chris’ last update, make sure to read it here.

AltMill

The AltMill represents the pinnacle of hobby CNCing, bringing industrial level technologies to a price point and ecosystem for both beginners and advanced users.

General Specifications

  • Rigid frame and linear motion design using HGR15 linear guides and ball screws on all axis mounted to a custom-designed, machined aluminum extrusion
  • Closed loop stepper motors for over 600IPM
  • Support for 65mm and 80mm spindles and routers
  • Approximately 49″ by 49″ inch working area with a max z-height of 5.5″
  • Footprint of 59″ wide and 62″ deep
  • Inductive sensors for homing included
  • Sturdy metal legs with the ability to add 4ft wide shelving underneath

News and Updates

  • Time to grow the team at Sienci Labs

    As we talked about in the blog post “Everything You Need to Know About the AltMill”, the launch of this new machine represents a pretty big shift in our company,…

  • April 2023 Production Updates

    April fools LongMill MK2 LongMills continue to ship out as usual. We received another batch of controllers after being out for another few days. Batch 9 production continues and we…

  • The AltMill CNC Is Open for Preorders!

    We’re thrilled to announce that the AltMill CNC is now officially available for preorders! Don’t miss your chance to secure yours by visiting our website!

  • Everything you need to know about the AltMill

    Hey guys, it’s Andy here. We’re excited to be launching a new CNC machine, the AltMill! This article is designed to tell you everything about the AltMill that we possibly…

  • AltMill Launch and Production Schedule

    Hey everyone, we’re excited to share our launch date for the AltMill. The AltMill will launch on Wednesday, March 27, 2024 at noon, EST. You can access the order page…

  • March 2024 Production Updates

    Hey guys, it’s Andy again with March 2024 production updates. I am currently writing this in China, where I am taking a bit of a “work-acation” but also to visit…

FAQ

Q: Will the AltMill come with the SuperLongBoard?

The AltMill will come with a specialized variant of the SLB which allows for use with a higher 48V power supply and with external driver support. Features that the SLB comes with, such as independent 4th axis control, RS485 protocol, and more, will be shared between the AltMill and LongMill.

Q: What’s the difference between the AltMill and the LongMill

The AltMill was developed to be a higher-performing machine compared to the LongMill. Our main design goal was to create a machine that had the best possible value for dollar at a higher level and pricepoint.

  1. The AltMill is larger, with a working area of 48×48 inches, whereas the LongMill comes in 12×30, 30×30, and 48×30 working areas.
  2. The AltMill is more powerful, allowing higher rapid speeds and deeper cuts. For context, the AltMill will allow for speeds over 600 inches per minute versus 160 inches per minute.

Q: What software does the AltMill come with?

The AltMill works with our free gSender software for controlling your machine. Any software that works for CAD/CAM with the LongMill also works with the AltMill.

Q: What else do I need to provide with the AltMill?

Besides the standard items we sell (machine, legs, spindle) , users will need to provide their own wasteboard (we recommend 3/4″ MDF), computer, and end mills.

Q: What accessories does the AltMill work with?

The AltMill will support accessories such as the LaserBeam, Vortex, and both the Autozero and regular touch plates.

Q: Will the AltMill come in any other sizes besides the 48×48 inch working area?

For the time being, we are focusing on developing and maturing the design of the 48×48 size. However, a successful launch and release of the AltMill 48×48 does pave the way for larger machines that could cut full sheets of wood, or smaller machines dedicated to cutting metals. We believe that the AltMill isn’t just a machine, but a new platform we can build upon for new form factors.

Q: What is the footprint of the AltMill?

The width of a fully assembled AltMill with our table legs is approximately 59″ wide and 62″ deep. Users should allow for a few inches of extra space on each side.

Q: What are the power requirements for the AltMill?

The AltMill comes with a 48V10A power supply, which can supply 500 watts of power. Users should dedicate additional power based on the spindle they choose to use, which can range from 1.5KW and above. This may require having two or three available breakers, or 220V available for your spindle.

Q: Can I add a toolchanger to my AltMill?

At this time, we do not have development or plans to release a toolchanger, but the AltMill and it’s hardware and electronics offer inputs and outputs to allow aftermarket toolchangers to be added by the user.

Q: Can I modify, update, trade-in, my LongMill to a AltMill?

No, because of the significant differences between the LongMill and AltMill, there is no path for upgrade or modification. Additionally, we currently do not have the infrastructure to support trade ins for the LongMill.

SLB January Updates

Hey all! Chris here again with some more juicy SLB updates

There’s been a lot of great news to come since you last saw me with my major SLB update post at the start of November, and then our very successful SLB launch on December 4th. If you missed them feel free to look back and get caught up:

What an Exciting Start

Firstly, thank you to all of you who have pre-ordered! We sold almost half of our first batch of 500 in the first 3 days, and are now at 357 total pre-sales, we might be sold out before we’ve even shipped the first board! Me and the whole team behind this effort really appreciate the trust and excitement we’ve seen for the SuperLongBoard and will keep working diligently to come out with something that we’re proud with and in a timely manner.

Thumbs up from one of our test setups at the new Sienci HQ!

Just a reminder that we’ll still be sticking to our word and offering anyone who orders an SLB before Jan 31, 2024 a reduced price, this means that if you’d like to show us some early support in this project and save some bucks while you’re at it, you’ve only got less than a week left to join the pre-orders 🙂

Some Meat and Potatoes

SLB Beta testing and board development rolls on 🚂. Since you last heard from me late December, nothing notable was hinted to since there’s A LOT of complexity to timing a product like this that has been going on behind the scenes. The one fun you might’ve seen was an SLB-powered Christmas Tree display which was a great break for me to enjoy a hands-on project, especially after the many countless hours I’ve spent recently behind a computer on documentation, emails, bug testing: https://www.facebook.com/groups/mill.one/posts/1811830652621591/

Happy SLB-mas!

Getting along to the juicy stuff, the SLB production progress has been continuing to pass more milestones:

  • SLB Enclosure 3D design finalized
  • E-stop PCB finally designed after the hurdle of the E-stop connector not being able to be tracked down so we got around it by attaching a connector to the other side
  • Enclosure samples arrived, tweaks made, production started
  • Major slowdown in receiving new SLB prototypes because of shipping confusion from UPS since we moved offices
  • Now at 5 Beta testers who are all talking on an the expanding private SLB Forum (will become public at launch)
  • E-stop samples and sample 2.5m E-stop cable arrived (longer length allows for more options on where to place the E-stop)
  • Found we won’t be hitting any major roadblocks when it comes to certification
  • Manufacturing of E-stop wires and injection molded E-stop enclosure complete
  • Manual & documentation now reaching reasonable progression in outlining all board features and testers now having hands on with the majority of options the new board makes available
Most-finalized SLB design, basically ready to go!

I hope you’re all as excited as I am about all those major steps that have been taken. We’ve also been iterating very quickly on improvements to the SLB’s firmware and bringing gSender up-to-speed to properly support the new board. These we know we have more time to complete since physical production is the MOST important to move along ASAP, but software and Firmware is still great to put our minds at ease to ensure everything works as expected as soon as possible:

  • 6 new SLB firmware versions since my last big update, now at 5.0.1 which fixes and improves some great stuff
  • Status light overrides
  • Improved PWM switching with separate inversion control
  • TLS inversion should now be fixed
  • SwitchBank control now happens over software
  • Laser will be changed back to not be default behaviour for safety reasons, but gSender is being worked on to support easier changing between Spindle and Laser
  • 3 more gSender Edge builds and now the new 1.4.0 Main version to better support SLB (see more about it here: https://sienci.com/2024/01/23/gsender-1-4-0-release-rotary-grblhal-jobs-stats-and-more/)
  • Controller jogging finally smoothed out, improved time estimation, more progress to resolve HAL Rotary behaviour, among other things
  • Whole ordeal now almost straightening out where we had massive delays due to software certification regulatory changes, requiring us to reapply and have our costs doubled to $800/yr just to be a trusted distributor

Here’s some more detail you can see on some of those points. Please read to the end if you want to see information about the remaining steps we have ahead of us and where my thoughts currently lay for the delivery timeline ↓

Finalized SLB Enclosure and E-stop

Sooo much iteration we went through to try to make a from-factor we were happy with

To give some context, we started ideating the enclosure the SLB would fit inside before the board even existed! We wanted to make sure it would accomplish a range of things to accomodate all the new plugs, keep wiring clean, look nice but also robust, not break the bank since the value should be coming from the board itself, and much more. A while ago I even started a thread on our forum when I thought I’d finally reached the end of our design vision (https://forum.sienci.com/t/official-call-for-feedback-on-new-board-in-development/8519/)….. aaand it was met with mixed reviews. In light of this we went back to the drawing board and I hope what we came up with will be able to meet everyones needs.

Finalized aluminum SLB enclosure and cable, E-stop to be injection moulded

This new design should give the best of both worlds! Universal flat mounting via flanges to accommodate a range of setups or machines inside enclosures, with a subtle bracket that will allow for Y-axis rail mounting for any LongMill MK2 owner if they want to keep all their wiring tidy and together. The front is easy to open and allows some wire management inside the enclosure itself, with most wires going out the rear, and I’m sure you’ll be happy to see how the status lights turned when we do a final reveal. You can also see the detached E-stop with it’s longer cable and 3 customizable Action buttons! These will all come standard with each SLB kit

Beta Testing

We’re in our final push on Beta testing to see if we can find any last issues with the board. This process has been ongoing for several months now with us finding little quirks here and there – but luckily most of these have been easy fixes that weren’t due to the boards themselves and instead the software or firmware. Below you can see a new setup for one of the testers, and if you pay close attention you’ll see some of them posting updates on their testing experiences on our Facebook group or Forum (like https://www.facebook.com/groups/mill.one/posts/1829844527486870/).

I rotated this picture to fit better onto the page, sorry Ian 😐

Documentation

A labour of love to keep our level of product support to the standard you’ve come to expect from us, we’ve now passed 50 pages of documentation covering all aspects of mounting, features, configuration and more for the SLBs. Another thing I wanted to let everyone know of is that one of my goals for this year is to transition all of Sienci’s documentation to be publicly contributable so that anyone can submit improvements or entirely new write-ups on how to use our products and just to share CNC resources in general. I’m very excited about this prospect and hope that I can do it in a way that best honours our ever-growing community.

Ongoing work on over 50 pages of documentation on all aspects of the SuperLongBoard

What this all Means

It means we’ve been working hard and seeing results! There’s still many more steps to go though, and with now everything in production except the boards themselves, I have to admit that the March delivery date might push into the end of March or start of April depending on our luck with slowdowns overseas from Lunar New Year. The board design as of tonight is fully complete with any other small tweaks we felt it needed to be ready for production, so the last thing holding us back is any remaining feedback from Beta testers that might tell us that there’s anything else we might’ve missed on the boards – otherwise if we have the confidence then that would be the last big piece of the puzzle completed. We want to try the best we can to not miss anything and have been trying to run Beta testing and in-house tests for over 5 months now to gain the confidence that we hope to see play out once everyone else gets their hands on the SLB.

Most recent round of prototypes being sent out to Testers to collect as much feedback as we can

Any other remaining steps and timeline as we start to see the light at the end of the tunnel:

  • Ongoing Internal validation on the new batch of boards, between Johann on General checks and Ikenna on LaserBeam compatibility alongside Expatria on their own setups we’re doing one last big effort to ensure boards can begin production and Firmware is as progressed as possible 
  • As of the start of this week and until the end of next week, the -hopefully- last version of the board prototypes arrived in the hands of Beta testers alongside the mostly finalized E-stop, enclosure, and new firmware. We’ll be looking for any remaining feedback to hopefully conclude any remaining findings they can provide before we begin production: using the E-stop and Action buttons, the form-factor and fitment of the enclosure, anything else they’d still like to see improved or fixed with their SLB experience
  • Continuing to order full 500 quantities of anything we’re sure about like the thumbscrews that just arrived today and will soon be starting to vet USB-C cables
  • Board PCB production should begin the last week of January or first week of February, and in the meantime we’ll be pre-ordering all the board components to hopefully reduce production time
  • Design box packaging
  • Start to prepare jigs and processes for QA
  • Throughout February and until we begin shipping, continue to make updates to FAQs, short videos to show board capabilities, SLB Manual, gSender HAL support, and SLB Firmware

Me and the SLB and gSender teams have been, and will continue to be, working diligently as the March delivery window continues to approach. This has been an especially big project for us to undertake so we want to do whatever we can to set up the SLB to succeed. As 2024 starts to pick up steam, I’m really excited for the impact that the SuperLongBoard might have on the future of hobby CNCs and hope to be proud of the 1.5 years spent to bring it to life.

Watch out for more videos I’ll put out to show what the board can do and feel free to keep asking questions 👍

gS release schedule

Please let us know what you think of the new 1.4.0 (https://sienci.com/gsender/)! It’s an accumulation of over a years worth of work in a package that we called “Fundamentals”. We wanted to bring new things to the table but also really take the time to go back into what fundamental things we could improve about gSender to make the day-to-day CNC experience more pleasant and reliable. This included new joystick controls, touch plate probing on all corners, much more accurate time estimation, faster file loading, endeavouring into maintenance reminders, and more. We know it may have a couple bugs in it here and there but we’re committed to fix those in the coming weeks and make 1.4.0 the new, best version of gSender for everyone to use!

Our next big push we’ve decided to call “New U”, where we’ll be aiming to take everything we’ve learned about the CNC control experience and put it to work in doing a complete overhaul on gSender’s UI. I know this may be a polarizing topic but we do feel like we can take something that we already felt we’ve done so well on and push it even further. We hope to take plenty opportunities to get everyone’s feedback as we begin this process but we feel hopeful that once we come through on the other side this will certainly be for the better and a better gSender overall.

Thanks yall for your time and hope to see you more soon!

-Chris signing off