Got the CNC machine running and forming the PETG sheets (finally) with some hiccups along the way. We have raised a total of $3,336 between the Go Fund Me and other donations. A huge shout-out to the employees over at AngioDynamics. They had a fundraiser and donated $631 toward making shields. I cant thank you enough for the contribution. We will likely be able to reach 1000 shields total at the current funding. The Ender 5 Plus (new large printer with personal funds) has been pushed back 3-5 weeks unfortunately.
We have had a slow week on distribution because I have spent so much time with the CNC and because the elastic bands were delayed in shipping. We have made an additional 150 shields, 50 of went to Warren County EMS this past Friday and 20 to Warren Washington Albany ARC today. The remaining 80 will be deployed shortly. This brings the total in the wild to 365 and 445 made.
- 30: Warren Center
- 30: Washington Center
- 30: The Pines
- 30: Glensfalls Center
- 30: Granville Center
- 30: Fort Hudson Center
- 10: Slate Valley Center
- 10: Holdbrook Adult Home
- 25: High Peaks Hospice
- 54: Warren County EMS
- 60: Washington County EMS
- 20: Warren Washington Albany ARC
- 6: Others
Ender 5 Plus Delay
Contacted the vendor as to why I have not received a shipping notice for the Ender 5 plus. I was hoping this would be able to assist with getting prints done, but it looks like it will be another 3-5 weeks before they have stock again. Not a huge loss production-wise, and at least it will be one less thing to work on right now.
CNC Up and Running
Okay, so soldering up the wiring turned out to be worse than I was dreading. I do regret not just doing a simple box with good cable management, but its a nice control box now that its finished. The one thing I have not done yet is get a dust shoe designed which has been a bit of a problem so far.
It took hours to solder joint and crimp wires for the connections internal to the box and also to terminate the aviation plugs. The result is that I can remove the control box from the CNC when I want to move the thing. They are also pretty cool once wired up and functional.
There were issues with the extension cable for the spindle controller interface. Apparently something about having an extension cable of any length causes the controller to freeze. Based on my testing this has to be due to noise in the wire caused by how fast the controller is trying to send information. The longer the extension, the lower the RPM it would freeze at.
The solution was to ditch the cable extension entirely and leave the interface screen connected to the controller. I cant reach this while the box is closed, so I had to set up the Arduino running GRBL to set the speed. The PWM signal from the Arduino is pretty well tuned to within 100rpm of the resulting spindle speed so I don't really need to see the number on the front of the controller anyway.
There are some other pitfalls with this Chinese spindle controller too. Who the hell makes a spindle enable pin active low?... This means that if the Arduino turns off, or the system is starting up; the spindle will turn on. I decided to put a switch in place of the controller mount on the front of the case so that I can disable power to the spindle controller. This allows me to swap bits without risk of having my fingers ripped off if the Arduino turns off or if there is a power bump.
I also fried a cheapo power cable that I had initially used for testing, and had 100% intended to replace before actually using the machine. Except I didn't.
The current draw of the CNC combined with the vacuum relay heated the cable and turned it into a melted noodle that shorted internally and made it spark and die spectacularly. I lucked out, but that could have gone sideways fast. I nearly had a heart attack when it went off. Its replacement I believe is two gauge standards lower, which is a thicker wire (stupid American logic). I haven't had an issue since the cable was replaced and the new one.
I also had this weird issue where my spindle would randomly slow down and sometimes stop entirely. I am running the newest version of GRBL, which has the required options to control the spindle properly. I started off by using fusion360 to generate the gcode to run things once I set up the machine system variables properly.
After a bit of testing, and a fair amount of E-stop presses, I figured out that "GRBL Controller V3.6" was causing the issue and the hardware was fine. Apparently this was never updated to 100% support the newer GRBL V1.1 I swapped over to Universal G-Code Sender and things have been working without issue since.
I spent the afternoon increasing my feeds to match where the spindle rpm (12000). Runtime decreased from 32 minutes to 16 minutes and it seems like the machine is able to run at these speeds well and generates plastic chips better than before.