Moving X & Y together produces loud grinding noise

[vigilantesculpting]

I've successfully flashed a Cricut Maker 1 (with serial starting Q) using the cutcutgo firmware, using a pickit3, on MacOS.

I can connect to the device using picocom, a commandline serial monitor, and successfully send it gcode commands.
Sending a command like G01X50F1 works very well, the unit moves the tool head smoothly in the X direction.
Similarly, a command like G01Y50F1 works well too, the unit moves the bed smoothly in the Y direction.

However, a command like G01X50Y50F1 also works, but the unit makes a grinding/rattling noise while moving the bed and toolhead together.

At this point, I presume that the motor-control (which controls the DC motors while monitoring their rotary detectors) is responsible.
Do you have any insight into this?

[virtualabs]

Yes, the way CutcutGo drives Cricut's motors is sketchy and especially when X and Y moves simultaneously. I've tried a lot of different ways to drive these motors but I've always avoided going into some PID stuff, in the end I got something that kinda works, but a bit noisy. This is due to the fact CutcutGo is driving these DC motors like steppers, kicking motors with high duty-cycle PWM signal for a very short time and keeping track of any drift (compensated when necessary). When a single axis is moved, there is absolutely no issue as the motor is continuously driven. When two axis are involved, it alternatively moves both of them during a very short time, resulting in a grinding/rattling noise (motors are basically vibrating). As far as I've tested and despite this noise, I almost get clean cuts 😅...

I've ordered some extra motors with the goal to make a testing rig and design some good PID-based code to drive such motors, but the motor I received had its coding wheel damaged... I was thinking about buying another Cricut just to get some motors and design a correct algorithm and code to drive it smoothly (but also to measure the inertia and tune PID default values for this specific motor).

The problem is I've lost motivation after I've published the first (kind of) working version of CutcutGo, but it may change in a near future. Contributions to this project are also open, so if you find a solution to this and have some working code to drive this DC motors, that would be awesome !

[vigilantesculpting]

Interesting. I was thinking that maybe the DC motors + encoders would be the problem. There have been requests for GRBL to add support for these, but AFAIK it has never been done. Either because everybody uses steppers, or those that don't use servo motors, which present like steppers (so they have control circuitry inside to drive the dc motor / servo like a stepper).

It caught me off-guard that cutcutgo manages to drive one axis at a time very well, but struggles with two simultaneously. From what you describe however, this does sound difficult, and I think you've done an amazing job.

I have a couple of dc motors with servos lying around from inkjet printer salvages, and I have some h-bridge drivers. I can see if I can test some code out on these, and see if I can make something work. I can't promise anything though, at least not soon!

BTW it is very normal to lose motivation like this. Again, congratulations on getting this far!

[audryhome]

Hi Damien, I have 3 Cricut maker i bought as spare parts. Can send you a pair if you like. Kind Regards Jean Le mar. 2 sept. 2025, 08:08, Damien Cauquil ***@***.***> a écrit :

*virtualabs* left a comment (virtualabs/cutcutgo#22) <#22 (comment)> Yes, the way CutcutGo drives Cricut's motors is sketchy and especially when X and Y moves simultaneously. I've tried a lot of different ways to drive these motors but I've always avoided going into some PID stuff, in the end I got something that kinda works, but a bit noisy. This is due to the fact CutcutGo is driving these DC motors like steppers, kicking motors with high duty-cycle PWM signal for a very short time and keeping track of any drift (compensated when necessary). When a single axis is moved, there is absolutely no issue as the motor is continuously driven. When two axis are involved, it alternatively moves both of them during a very short time, resulting in a grinding/rattling noise (motors are basically vibrating). As far as I've tested and despite this noise, I almost get clean cuts 😅... I've ordered some extra motors with the goal to make a testing rig and design some good PID-based code to drive such motors, but the motor I received had its coding wheel damaged... I was thinking about buying another Cricut just to get some motors and design a correct algorithm and code to drive it smoothly (but also to measure the inertia and tune PID default values for this specific motor). The problem is I've lost motivation after I've published the first (kind of) working version of CutcutGo, but it may change in a near future. Contributions to this project are also open, so if you find a solution to this and have some working code to drive this DC motors, that would be awesome ! — Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABTQJV2MN2MAPDKECJDDRYD3QUX7RAVCNFSM6AAAAACETSTBOKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENBTHEYDMNZRGQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

*Jean VISCONTE*

[virtualabs]

Thank you @vigilantesculpting for the kind words, it is great to see some people use this project (even if not perfect) 😄 ! Motivation is a strange thing though, seeing other persons working on it may make me get back on track on this project. In the end, I may take some time to help fixing the code driving these DC motors (I have a good understanding of how GRBL works and how my sketchy code works too), having other people wanting to contribute and solve this long-lasting problem I had with this Cricut Maker is just awesome !

@audryhome That would help a lot setting up a testing rig for these motors, indeed. I can cover the shipping cost if needed, and will send them back to you once these beasts can be driven to make smooth and silent moves. We can get in touch by mail if you want.