It’s been a heck of a ride with building LongMills and ironing out issues in production to get the best quality machine to our customers. With any new product, there are lots of issues to fix and resolve. We’ve kept detailed records for every customer issue we have to review and improve on our machines and customer service, as well as to place new methods to prevent issues from happening again. In total, we shipped approximately 380 units between October 2019 and Jan 2020.
This post will discuss some of our common problems and issues and delve a little deeper into what we did on our side. We hope that this helps provide some transparency on how we help our customers as well as give everyone a little insight into some of the day to day things we sort through for our customers.
Number of reported incidents: 31
We found that with so many different components that make up the LongMill and the variations for all of the kits, we found that it can be easy to miss a part.
A good portion of missing components were bags of fasteners missing. We have updated the packaging for batch 2 to:
- Be labeled with the fastener type to make it easier to visually identify different types of fasteners.
- Weigh checked at each step. All components must be in the bag to pass the correct weight.
- Labeled in different colors so that they are harder to mix up or miss.
One of the areas we found that were error-prone was where there are variations package to package. We’ve isolated more of the components to reduce variations and pre-pack them so that there is less chance for a mix-up.
Broken 3D printed parts
Broken during assembly: 6
One of the common reasons for breaking a part during assembly was from dropping the rail with the 3D printed parts attached to it. We provided new parts free of charge, and we have been using a stronger plastic to reduce the chance of parts breaking during assembly.
Broken during shipping: 14
One of the most common issues we’ve had with parts breaking was with the Z-axis motor mount holder. The rear mount for the drag chain was prone to breaking. To mitigate this issue, this part has been redesigned to be made from steel, eliminating the chance for that to break during transport.
Control box issues
Defective control board: 9
Several boards, especially at the beginning of the production run, did not work properly. Most problems came from individual motor drivers not working properly or switches not working properly. Customers received replacements control boxes and old ones were shipped back for testing and repair.
We found that several of the drivers had missing or damaged components, which we suspect were from shipping.
After finding these issues, every subsequent board was more thoroughly checked for defects and tested before assembly.
Batch 2 boards were improved in design for reliability, uses better quality components to reduce the chance for issues. Every driver will be tested before being soldered on the main control board.
With the new steel enclosure design, the LongMill’s control board is also expected to be better protected during transport.
Control boards also have higher quality motor connectors which means that there is less chance of misdiagnosis for improperly seated cables.
Manufacturing defects: 18
There were some manufacturing defects with some of the parts with the LongMill. This includes:
- Angle aluminum out of tolerance
- Defective power supplies
- Tapped holes are stripped on ACME locknuts
- Improperly manufactured Delrin ACME leadscrew nuts
Luckily we’ve had very few issues with most of the manufactured parts, with most of the issues coming from the power supplies. We’ve found a new supplier that is able to provide us with better quality, more robust power supplies.
Most components are now visually checked to reduce the chance of bad parts, but typically the error rate has been very low. We are usually able to get replacements parts within a few days to customers in North America.
With the exception of the power supplies, we usually only had one or two issues with each part, so the rate of defective parts is quite low (less than half a percent).
At Sienci Labs we are anal about making sure that the quality of our products are as high as possible. We want to make sure that the experience of receiving and assembling our machines are as smooth as possible, and that starts with how we pack and manufacture our products.
Tracking our issues and solving them in our process has been a lot of fun and we expect to continue to improve our machines. I hope that this post provides some insight into our process and how we made some of the decisions for our packaging methods.