Hi all, I built this Workbee a couple of years ago, and never got it fully working, so it was collecting dust for a long while, now recently have dusted it off and started trying to get it working right. The main problem always was the Y-axis; I had to limit the speeds to 1000mm/min in the firmware in order to mitigate the Y-axis problems a little bit, but it still is problematic. Eventually the Y-axes get out of sync, and that's when it at latest stalls. Sometimes it gives an exclamation mark on the black box, usually on the Y2, when this happens: And I need to unplug the Y-axis motors and roll the axes by hand so that they would be aligned again, and then reconnect the motors. Sometimes I've had it working for a while if I loosen the backlash nuts, but in this case at least one of the axis will be very noisy: I tried reading through some threads with this same problem, and in some cases it had helped to get a black box, or a meanwell PSU, but I've had both of those since start, along with this problem. Also someone had wrote that they needed to loosen the rolls on Y-axis carriage, but I have those also as loose as they go, and none of them really spin freely, but I can rotate them with my fingers anyway. I wouldn't imagine them being the problem as they are equally tight as on my 3D printer's carriage. Any pointers to how I could finally get this working and operating in full speed would be welcome.
See docs:blackbox-4x:currentadjustment [OpenBuilds Documentation] - fault indicators are documented in the 2nd half of the page, though the first part usually relates Sounds like docs:blackbox-4x:faq-identify-motor-coils [OpenBuilds Documentation] though
Thank you Peter. I do recall measuring them when installing, but I will do that over again next week when I have a moment. I'll report back then. Just to make sure; if the steppers were wired correctly, it should be able to move the y-axis at full speed even if the backlash nut blocks are tight enough not to make noise?
The noise in the video sounds like the first section of Pay close attention to the fact that page covers both initial wiring, but also, more appropriately, situation where wiring break / loose terminals etc causes one coilpair to lose connection either permanently or intermittently (yours)
Ok, took a while, but I went through the wiring and then reinstalled everything. Then I proceeded to align the y-axes again and started jogging the machine back and forth to adjust the backlash nuts until it no longer makes a noise when jogging at 1000mm/min. Feeling victorious, I turned the max speed to 2000mm/min in the firmware. As soon as I try to jog it, it gets misaligned and starts screaming like this: No doubt the screaming is because the axes get misaligned and it no longer can move. I guess the reason for it getting misaligned instantly is that one of the axes is too heavy for it to move now that the backlash nuts are adjusted to silence? Is there something about the leadscrews I could doublecheck to determine if there's something I could do to get it to love more freely?
The nuts should just take out backlash, not excert pressure The high pitched noise is called stalling, could be caused by binding (could you spin leadscrew by hand when you were adjusting nutblock?) but could also be Acceleration or Max Rate set too high Grbl v1.1 Configuration and Grbl v1.1 Configuration Are the motors from OpenBuilds?
Thanks Peter. I didn't adjust the acceleration, that remained the same value it was when I first adjusted the nut blocks at 1000mm/min max speed, before upping that to 2000mm/min when I had the Y-axes aligned and nut blocks adjusted and this stalling immediately started happening and the y-axes got misaligned again. I didn't try spinning the leadscrew by hand when I adjusted the nutblock, I went by whether the leadscrew was making noise or not when adjusting it. Am I approaching this completely in the wrong way to going by the noise heard in the first video? I mean, when I carefully adjust the backlash nuts bit by bit, it eventually gets noise free, but maybe it starts binding then? Should I just detach the motor couplers and rotate the leadscrews by hand and adjust the backlash nuts until there is no "gap" to be felt at all, and just ignore the noise heard on the first video if I can get to the full 2000mm/min this way, without y-axes getting misaligned? The motors are chinese Nema 23's bought elsewhere.
Ours are low inductance and low voltage specifically to get good speed out of it. Nema-xx is but the size. What matters is the electrical Specs
Thanks Peter. I sent a message to the seller if I could get the electrical specs of them. But I started wondering whether this problem could have something to do with the leadscrew tension. Mine have the flexible couplers, and they don't have the tensioner hardware depicted in these instructions at all: CNC Commissioning Guide – Maker Hardware Mine has the bearing and the lock collar on the "inside" of the profile, and on the outside the other end has the flex coupler and nema23, and the other end has nothing.
We've seen enough chinese motors with high inductance to hold off on that till you know the specs Consider this, why was this one of my questions right away? The symptoms revealed something we see all the time (; Leadscrews are only tensioned in our 1500mm actuators and LEAD1515 kit. For the 1010 and other machines we don't tension either.
I now started installing the beefier motors and tested how the y-axis turna by hand. It turns quite well but I still keep wondering where this noise comes from: Would be great if someone has pointers of what to do to remedy this while the motors are off.
Sounds a bit like one or more V wheels are binding. I would slacken off all the V wheel adjusters and check everyone of the wheels spin freely on their bearings while they are not in contact with the C beam. If the V wheels are all good then I would suspect the nut blocks. Also check the leadscrew bearings run freely.
Thanks Christian, I slacked all of the V wheels and moved it by hand, seemed to move okay. But then it started stalling again, and it feels like in one part of the Y-axis C-beam they get a bit tighter. I can still move the wheels by hand though. I checked the C-beam with a straight edge but couldn't find any bends. Anyway, even though I can move the axis back and forth by hand, it still stalls even with the beefier motors. It stalls pretty much instantly if I try an y-axis speed of more than 1000mm, but it stalls at 1000mm too. I tried adjusting the current on the y and y2 all the way between min and max in many steps but it doesnt seem to do much, maybe stalls a bit easier on high current. I am a bit at loss with this, doesn't seem to get any better no matter what I do.
You mentioned in response to one of Peter's questions that you had not changed the acceleration values, but what acceleration values do you have set for the Y axis? (i.e. the $121 value should be your Y acceleration). Try reducing the acceleration level down (e.g. try a 1/4 of the current value) and then seeing how fast you can move over the 1000mm/min before it stalls. If the acceleration is set too high for the motor/drivetrain combination, then it can cause stall problems like you are seeing, especially when trying to go for a higher G0 speed. If you need very fast motion with a stepper-driven axis, there becomes a point where the acceleration has to be wound down to allow it. I have two different configuration saved for my machine, one uses a low acceleration but a high top speed for jobs that need fast but long travels (such as cutting profile paths in large sheets), and a second configuration for 3D work where I have very high accelerations, but also appropriately low top speeds.
Thank you Evan. The acceleration values can be seen on the latest video, they are 100mm/sec2 for all axis. I haven't had a problem with other axis on speed or acceleration, x seemed to be 2000mm/min and 100mm/sec2 and it has never stalled on x. I did test the y at 1200mm/min and 75mm/sec2 but it stalled on that too.
Keep the guy reading it in mind post a Grbl Settings backup (more values than just those onscreen worth reviewing, and shaky video is hard to read compared to a nice backup
Thanks, yes I agree that is the right way to do it, gotta figure out how to do that when I am back at the machine on monday.
CONTROL > Grbl Settings tab (where you were in the video) > Backup Settings button on the toolbar there, right next to the Save button
Ah, yes I did not spot the acceleration in the video. The motion of the Y sounded ok, up until the motor stall. My hearing is not great, but I could not hear any 'thuds' that can occur when a motor misses steps. Yes the rates you have are not unusual. Although the X runs ok at the faster speed, the Y motors are pushing a larger mass, which will change the maximum acceleration that can be achieved on the Y. That is, even though Y has 2 motors, it only doubles the force, and if the X gantry metalwork is heavier than the Z-assembly and router, the mass difference between X-pushing-Z and Y-pushing-X-and-Z can be a bigger factor to overcome. I am not sure of what the mass of the Workbee components are though, so you may be in a situation where Y could potentially be faster than X if the gantry is not too heavy. A couple of questions: is it always the same side of Y on the machine that stalls, even if you swap over the Y1 and Y2 connections at the controller? I doubt the problem is at the controller end, but the test would help rule out if one of the drivers is not acting as anticipated, and also then confirm the problem is mostly related to one side of the machine. I know you have changed the motors and probably would have spotted this one, but are the grub-screws on the couplers between the motors and the drive screws all fully tight onto the shafts, just in case one side is slipping and causing the rack to eventually jam? I had a cheap coupler a few years ago that it turned out had a big chunk of swarf at the bottom of the threads. When I tightened up the grub screw, it did not actually clamp onto the shaft properly. When I turned the shaft by hand, it all felt tight, but under acceleration, there was a fraction of slipage. Unfortunately it was on X so it never stopped the machine running, it just drove me nuts as to why sometimes I appeared to slowly lose position during a print! The last question is when the machine is powered up but idle so that the steppers are engaged in holding the machine in its current position, is it possible to rotate the motor-to-shaft couplers by hand (just by one step to get an idea of the holding torque)? They should both be holding pretty solid and if they can be rotated by a step, they should take about the same force to do it.
The machine was making the noise with the motor removed, as per the video, so can't see how the motor or settings can cause it. Got to be something in that Y axis binding.
Not likely the cause of the stalling, but either failing bearings (this isn't an OpenBuilds machine correct?) or the Lock Collars aren't tight up against the bearings correctly. Sounds like a little axial free play in the leadscrew, it moves along its length and slams an endplate. Hard to locate source of noise on video though so try to use stereoscopic hearing, closer inspection and visual checking to confirm. Or rebuild the axis per the original assembly video - if not the lock collars something else is a bit loose
