Core One back to head banging with firmware 6.4.0 - any tips?
Hi
My C1 is back to doing a lot of head banging when it's homing the head prior printing. It does 4 slower, then 10 rapid knocks. After an earlier firmware update it got down to just doing a few. The belts are in their happy range for tension and I'm not seeing any print defects or hearing any noises that otherwise have me concerned.
If anything though, 14 knocks is far more head banging than I remember previously.
Aside from the belt tension what other adjustments/settings might be making the homing sequence do this again?
Thanks
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Before we start exploring other potential causes -- are the belt tension and gantry alignment actually ok? Using the new LED-based belt tuning wizard, are the two belt frequencies in the right range? And is the gantry squared at the same time?
When you pull the gantry forward against the end stops (motor power off), it must touch both stops. Try to gently push it against the stop on the left, then on the right, so see whether there is still wiggle room, i.e. a small gap between gantry and end stop. If you don't get "zero wiggle", you will have to straighten the gantry. On the new belt tuning instructions page, follow the Quick gantry alignment guide in the yellow box.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Before we start exploring other potential causes -- are the belt tension and gantry alignment actually ok? Using the new LED-based belt tuning wizard, are the two belt frequencies in the right range? And is the gantry squared at the same time?
Yes perfectly square, touches both sides with no play, and solidly in the frequency range. No adjustments are recommended by the tuner.
Belt tuning was my go-to but there's nothing to report there. Do we know of anything else causes longer homing?
RE: Core One back to head banging with firmware 6.4.0 - any tips?
For more info I ran the test just now and upper belt is 91.5 and lower is 92Hz
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Ok, looks like the gantry and belts are in good shape!
You could try re-running the Homing Calibration (Control > Calibrations & Tests > 5 Homing Calibration). That one makes a real racket bumping into the stops, but I understand it determines how an end stop hit looks in the Stallguard readouts of the specific printer and stores those values. Maybe that helps to make the routine homing quicker again?
A couple of users have reported that the heater (bed and/or nozzle?) seems to interfere with the XY homing on their Core Ones. To my knowledge, it is not clear where that interference occurs -- maybe via the power supply? But the affected users report that homing with the heaters turned off works more reliably. Unfortunately firmware 6.4.0 no longer lets you do a precise homing without heaters from the menu -- the Control > Auto Homing will only do the quick, less accurate homing. Sending a G28 command while the heaters are off is probably the quickest test?
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Thanks - I forgot there was a calibration for homing to try. It has printing to take care of for the next 6 hours so will jump on that tomorrow and see if it helps.
I also bought a lube kit for the linear rail in case there's a possibility that's adding drag that's upsetting it, but there's no noise and it doesn't feel tight (as much as you'd remember) when moving it with the motors off. Regardless will try that too. It's a shame there's no feedback from the machine as to why it needs to try more knocks when homing.
Truly appreciate your thoughts on this so far. It's always good to get a fresh perspective.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
You could try re-running the Homing Calibration (Control > Calibrations & Tests > 5 Homing Calibration).
That has reduced it by around a half! Also I found in the settings menu something about calibration that was set to ASK that I've moved to AUTO. Haven't found any documentation on that yet, and you'd have thought it might have asked already if that was going to be helpful, but I've changed it anyway. I'm encouraged, but the quest continues...
RE: Core One back to head banging with firmware 6.4.0 - any tips?
My Core One upgrade from Mk 4S had an awful issue with incessant, loud head banging, even with 6.4.0-RC. I recently installed 6.4.0 and re-tuned the belts to see if there was any change in the final release that might have improved the belt tuning module. I tweaked my belts to the higher end of acceptable range and the head banging was markedly improved. In part, it appears that the final 6.4.0 release slowed the acceleration of the print head during homing. There is still head banging before each print, but it is nowhere near as loud. The number of bangs has also been reduced, but is still way more than what I had with my Mk4s. Prints are looking good. For whatever reason, the Core One belts seem to loosen more with time than the Mk 4S. Maybe it just takes repeated belt tunings to finally get all the stretch and slop out of the Core One belt system. The belts are certainly a lot longer than in the Mk 4S so maybe there's a stretch issue that just needs to work itself out.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
For more info I ran the test just now and upper belt is 91.5 and lower is 92Hz
The upper belt shouldn't have a smaller frequency than the lower belt. It should be 4-6 hz higher.
Just as a note, when checking for gantry squareness, it must be done with the belts slack. It's possible, for example, that the gantry didn't start off square but inadvertently got square after tensioning the belts. Just after I built my kit back in late June I noticed a similar issue (incessant head banging). I noticed that with the belts under tension the gantry was "square" but then when I released tension there was a 0.8mm skew.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
As above, upper belt should be higher by more than a couple Hz. Remember that you can't adjust both belt frequencies. You can adjust one belt frequency and the gantry alignment. I use the upper frequency and let the lower fall where it may, following the belt tuner instructions for how much to turn both, then tweaking one for alignment. Rinse and repeat until both are in the acceptance window and the gantry is perfectly aligned.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
OK, so if there is skew with zero belt tension, how do you get rid of it. The guide page says to put the gantry half way along the Y axis then pull the side with the slack and push the other side to presumably "bend" something. what is the next step if this doesn't work?
RE: Core One back to head banging with firmware 6.4.0 - any tips?
what is the next step if this doesn't work?
Bend some more. 🙂 Be sure to do that with the belts fully loose.
RE:
OK, so if there is skew with zero belt tension, how do you get rid of it. The guide page says to put the gantry half way along the Y axis then pull the side with the slack and push the other side to presumably "bend" something. what is the next step if this doesn't work?
In my case, I placed the flat side of the universal wrench on the side without the skew to use it as a lever and bent the skewed side by hand until it was square. I didn't have to go caveman on it, a medium amount of pressure was all that was needed to bend it. What you are bending is the bracket that attaches the gantry to the linear bearing holder on the Y axis rod. You are trying to get it as close to 90° as possible
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Interesting. One of my brackets in the upgrade kit came bent. I thought I did a pretty good job bending it back to spec. I'l try bending again after my current print finishes. Thanks.
RE:
You might want to check out this post. Your alignment may be good and it could be this issue.
https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/4596
I had the head banging issue and also the z axis issue where it would detect a crash sometimes, but not others, and then continue on. It would take 15 mins or so to start a print. Bed probing was also an issue at times. Now it is around 5 mins. Home works fine now. It seems the heater is causing multiple issues with homing, z axis, and bed probing. I used ferrite cores that I had laying around and put them on the heater wires and fixed my issue. It looks like Prusa has a fix with a new twisted cable that they are testing.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Is there a LOG file? There is an option to turn on the log files, so if those are activated, perhaps they record the steps that the printer is performing.
Otherwise, watch the screen display. It will flash the activities of the printer. So I briefly saw the message 'failed X homing' once.
Interesting. One of my brackets in the upgrade kit came bent. I thought I did a pretty good job bending it back to spec. I'l try bending again after my current print finishes. Thanks.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
Best way to mitigate this is to periodically check belt tension and gantry alignment.
RE:
I contacted tech support and they ran me through the head alignment procedure from scratch. I loosened the belt tensioning screws to the point of release, then deactivated motors and pushed/pulled the nextruder carriage until it touched both left and right sides with zero gap. I then initiated the manual belt tuning routine under settings and set the frequency for the upper belt to 98Hz. I then tightened both upper nd lower belts the exact same number of turns until the upper belt started to oscillate in the belt tuning module. I then clicked through the module to the lower belt tension to see what its oscillating frequency was. It read slightly lower at about 96Hz. I called it good and tried a print. First, I confirmed the Nextruder had no gaps after the belt re-tensioning. At the initiation of the print it homed with two bangs to the right (X axis), two bangs to the front (Y axis) then about ten more, softer bangs in the X axis again.
So the upshot is the banging is somewhat less, and a lot quieter, but still I think is excessive. The print quality is excellent, however.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
That also might be a 'one time' calibration.
I think everytime it is two bangs on X and two on Y. It seems that sometimes but not always that is followed by a diagonal XY movement for a number of times that results in a softer bang.
At the initiation of the print it homed with two bangs to the right (X axis), two bangs to the front (Y axis) then about ten more, softer bangs in the X axis again.
RE: Core One back to head banging with firmware 6.4.0 - any tips?
FWIW, mine will do two bangs to the side and two bangs to the front (can't remember which is first), followed by four diagonal bangs. That's if everything is perfect with regards to alignment and belt tension. When things drift out a bit, it does six diagonal bangs. It seems one should do the belts quite often, but I don't know if you should go into the menu and do the homing there as often. That always seems to do an excessive amount of banging.