by Been a little over a month since I put one of these out. There is a lot going on, and this is a quick update on things at home and in the shop.
During this holiday season, I hope that everyone is home safe and healthy. I hope that the holidays brings some happiness to everyone, it did for me and my family. Between Thanksgiving and Christmas; Lisa and I spend time with our boys, the boys’ girlfriends, and all the dogs. Obviously with everything going on in the world right now, we did not venture outside of our small circle. Lots of cooking at home, video calls with Lisa’s parents, and staying away from any sort of crowds. A good time was had by all and we are all hoping for a quiet and back to normal 2021.
Finally after three years, my heavily modified workhorse Anet A8’s mainboard failed. This was not unexpected. It has been having trouble for about the last year handling temperature management of the hotend. No amount of PID tuning would get it to work properly. I would get a constant 10 degree fluctuation in the hotend temperature. For the most part it worked OK, but recently if I made a “sliced” temperature change, say going from layer 1 to 2; it would get an heat fault and quit. The replacement Bigtreetech SKR 1.4 Turbo has been on the shelf waiting for this moment. I also purchased the TMC 2209 drivers, and TFT 3.0 screen to replace all of the electronics. After several hours of work, including changing the end stop pinouts around, up and working. Special thanks to Caggius for a post on the updated pinouts. Saved me a lot of time messing with the schematics. I was able to get everything working in Visual Studio Code and merged the Anet sample and the one that I was using with Marlin 1.9. I am now on Marlin 2.0.6. The first shot at this was a disaster. Turns out that the bugfix github branch for 2.0.8 had some sort of issue with the location of the X and Y axis after homing. No matter what I did everything was offset wrong. I gather that the version that I pulled had some sort of bug. This seems to me a common thread for me. Remember the printing parts issue with the MPCNC. I downloaded the 2.0.6 version and everything work properly. I had to order some memory cards, I guess that the SKR boards have a 2 gig limit. All I had was 16 gig.
The SKR board and Marlin 2.0 is a game changer. The hardware has never worked this well, and this quiet. The speeds that I am getting are impressive and I have little to no issues with the final product. I have to make some small slicer changes, but looks great and is a terrific upgrade. The TMC2209 drivers a so quiet, I did not expect how quiet. In lieu of the noise from the 4988 drivers, I have to physically go check the printer to see if it is still working.
I did start down the path of upgrading the ATXPiHat plugin from Python 2 to Python 3. This turned into a lot of work. With me discontinuing selling the hats, it no longer makes sense to continue this development. I have two customers, that have everything working with the latest Octoprint and the original plugin. So that is what I am doing for the moment. Unless we get a serious need for the upgrade to take place, I am not going to do it. There are some features that I would like to change, but not enough for the effort. I have not done a full upgrade on my copy of Octoprint running on the Anet, on the list of things to do. I have purchased a Pi4 to do this. There is also an issue of amperage requirement for the Pi4, it is a bit higher than the Pi3 and I believe is might be more than the 5VSB pin the ATX supply puts out. That might kill the whole thing. We will have to see.
As for the MPCNC project, I have redone the electronic in the housing three times. I just do not like how it was coming out. I have completed all of the printed parts and have installed most of the electronics. The first one was a big mess of cables and hard to deal with. Additionally, the first two times I did this, I followed good industrial control design and put the electrical in the top left corner. However, based on the way that the wiring was running, and the proximity to the back panel, it could/would introduce electrical noise interference with the steppers motors. So I moved the power related items to the top right. This seemed to be better. During this part of the project, is when my Anet died; so the whole thing got put on hold. I still have parts for the MPCNC that need to be completed and some that need to be reprinted.
 |  |
| If you look at this photo, the power cables run right above the motion controller. | In the revised configuration, the power cables all run on the other side of the case |
I also come to find out that Bart Dring, the maker of the MPCNC GRBL ESP32 board, is no longer going to be making the MPCNC specific version. He is currently out of stock and is not going to restock them. This is truly disappointing. I have been recommending this board to anyone who will listen. In talking with Bart, he is spending all of his efforts on the modular ESP32 GRBL board. Nice product, not as easy to put together as the MPCNC version. I looked high and wide on Discord for the conversation and I cannot find it. I am using the browser version which is not very good. As a side note, there is a bug in the MPCNC board, supposedly this is a known issue. I could not find it anywhere. The cooling fan ground pin is actually not wired to anything. A jumper wire and some solder took care of it. The lack of long term support will most like accelerate my move to Mach 3/4. No reason to invest in the latest board from Bart if I was wanting to move to something else (Mach3) long term.
Over the holiday, I did work on the bill of materials (BOM) for my MPCNC build. This will be posted as a part of this BLOG entry. The pricing is not what I actually paid for things; it is the piece prices. So, if I had to buy a bag of parts to get the two I needed, say 10 screws come in the pack; I only show the cost of the 2 I needed, not the cost of the entire bag.
I have also had a lot of issues with the plastic parts on the MPCNC breaking without much of a reason. I have already broken and end stop clip and one of the truck clamps. I am at a loss on why. The machine was not even being used when this happened. One day, I found the end stop clip just snapped. As for the truck clamp, I barely tightened the screw on the clamp and I heard a snap and the part broken in a spot that I did not expect. I am just frustrated with this project. Been a great learning experience, but it is a lot of work and repairs. I have not even cut anything yet. Maybe I am just to much of a perfectionist.
Organizing my shop and “stuff” has been a focus of recent. Turns out that I am a hoarder. Not as bad as the television shows, but I never throw anything out. Several attempts at organizing things into bin’s, boxes, etc are just not working. It is going to take a much larger effort on my part. My shop is really small, 12 x 12 (maybe). I also have things in my office and in the finished section of our basement. A few years ago, I moved most everything out of the office. I could not find anything anymore. With the pandemic and me working out of the house, I am glad that I did this a few years ago. I would have had a heck of a time. I went through all of my 3D printer bits and have been selling off some parts on Ebay. So far it has worked out ok, I am not making any profit, but I am not putting things into the garbage, which is always a good thing. On the list for the winter is installing some real shelves in the garage for things that I very rarely go after. This should clean out the basement a bit more so that I can repurpose the walls for better hanging storage.
Thank you to all of my family, friends, customers, and support from individuals across the net. I appreciate all of you.
–BA
