I have had this issue a few times where at the end of a carve it does not fully finish the G-code. The M5 command gets sent, however, the machine does not move to the normal position after a carve and the M9 command is not displayed in the serial display, however the coolant does shut off. Each time it has ended in the same spot and hasn't stopped mid-carve. Following this issue, the stop and pause buttons are still available, but do not do anything. The only way i have found to resolve this is to unplug the USB, plug it back in and re-connect the software to the controller. It is not specific to the file as i have a file multiple times and have had mixed results. Also, carve time doesn't seem to be the issue as this has occurred on 2 hour carves as well as 2 minute carves. Is this a USB issue or something going on with the CONTROL software?
You could give an alternative control application a go to help narrow it down between potential bug vs usb / emi: See docs:software:overview [OpenBuilds Documentation] for some alternative options What is connected to the Coolant port? (as this sounds like turning off that device, creates an EMI event)
I copied it to a text file but I seem to have lost that. The last item it shows is the M5 command with no other errors shown. Just the normal warnings after I unplug the USB then re connect. I will try to copy it again the next time I have the issue. I will try giving UGS a try for a bit to see if it repeats there to point us towards USB issues.
I use the coolant commands to control the relay in the controller which I have hooked up to a 240v pond pump for my water cooled spindle as well as the LED light for the spindle (24V OB design) plugged into the controller with the jumper configured appropriately. This way I know if the lights are on, the spindle has coolant.
OK that should be fine But do a test with the pump disconnected / relay jumper removed, etc too, just to help narrow things down
Peter, sorry for the delay in getting to this. I did some tests tonight with both OpenBuilds Control as well as UGS. I was able to narrow it down to something with turning off the pump based on your previous comments. I simply cycled M8 and M9 until it would freeze up. I then repeated with pump and jumper unplugged, followed by jumper installed, pump not installed. Only with the pump plugged into the relay was i able to get "results". All of these tests were run with the spindle off and the machine stationary so there should have been no other EMI. I have attached the CMD prompt from UGS from when this occurred following unplugging the USB cable. I am not sure what the best solution to this, or what is causing it. I would imagine a large in-rush current at start up of the pump, not sure what the physics is behind a disconnect that could cause this. I am running a 240V 75watt pump through the relay, not sure if that is an issue with the proximity of the relay to other components on the board as it appears the relay is rated for up to 250V.
Couple options: Use another IoT relay to switch the pump NB The onboard relay is not intended for mains usage, nor is it deemed safe. The PCB layout is not designed with mains isolation intended. it is intended for plasma trigger signals, misting solenoids, etc (DC loads). Or option 2, Control pump manually
Thanks for your help on this! I will likely just remove the relay from the equation and run the pump any time my system is powered up. I'm not a fan of manual control since forgetting could lead to an expensive mistake!
Agreed, manual is definately the 2nd option. But another IoT is cheap expansion (; so do consider that (or wait are you in a 220v country? No IoT option...) Depending on your skill/confort level with mains, consider a inline RF/EMI filter/supressor as well?
I am not in a 220V country, i ordered a 220V spindle just to have lower current as well as well as future expansion possibilities. They then sent a 220V pump to go along with it. I would have preferred 120V but have been able to make everything work up to this point. Having the pump running the whole time might actually be better as letting it run after a carve will help take all the heat out of the spindle rather than letting it heat-soak.
Ahhh good. So if you decide to automate after all, an IoT relay connected to the Coolant port, and a 120v pump from the pond section of your big box store, will be perfect. I assume your pump is one of those pond pumps with the brass fitting for the smaller hose? Or .. G4 P120 ; wait 120 seconds or whatever you think you need M9 ; Coolant off RE Heatsoak, the water cooling should prevent heat from building up in the first place - if a bucket+pump doesn't do it on long carves, investigate a "CW-5000" chiller
Correct, i highly doubt i am getting the rated flow from that. Not sure on how little flow is needed for the spindle, i don't think it takes much, just more temp increase as the water goes through. Maybe another add to the Fusion PP ? If other tool movements are needed following a carve (I move my Y 200mm out of the way after a carve is complete to remove clamps, etc) you could just stop the file, and turn coolant on manually.
