Topics created by HebigT

@HebigT Which Duet 3 are you using?

I had a Google around, but there's really very little info on this spindle as far as I can see. The best info seems to be in the snapmaker forums, ie https://forum.snapmaker.com/search?q=cnc %23snapmaker-20
I found this thread where someone had take the module apart, and has some information: https://forum.snapmaker.com/t/cnc-spindle-not-running/24696
Someone replied:

The FG pin is the frequency generator internal to the motor that tells the machine how fast the spindle is actually going and is irrelevant to what you need.

So the FG pin is the tacho, but you'll need to put an oscilloscope on it to check the voltage and signal, I would guess, once you get the spindle turning. If it's within specs, you could set up a dummy fan and connect this to a the dummy fan's tacho input. While the Duet can accept a tacho input, it would need to conform to the Intel PC fan specification, which expects 2 Pulses Per Revolution (PPR) from the tacho, and the RPM reading calculated assumes that.

The PWM pin should set the spindle speed. The thread above talks a lot about 5V and 3.3V traces on the board; try 3.3V PWM, then 5V if it doesn't work. Try using the Laser/Servo/VFD pin, which provides a 5V PWM signal. All other PWM signals switch on the ground side, so you need a circuit to pull them up to whatever voltage you need to drive the PWM signal. Also try different PWM frequencies (Q parameter in M950).

Again, you'll have to play around with the Direction pin. I guess it's either high or low. So leaving it disconnected should get it spinning one way. Applying a voltage would, I imagine, make it spin the other way. Though it's also possible it is an enable pin, so needs a voltage to work.

For testing, you could set up a PWM pin to drive the motor only. Use M950 to create an output (use the Laser/Servo/VFD pin) and control it with M42. See https://docs.duet3d.com/User_manual/Connecting_hardware/Motors_servos#connecting-dc-motors

For actual configuration, see M950 and M453. There's also this page: https://docs.duet3d.com/User_manual/Machine_configuration/Configuration_CNC

Ian

@HebigT
I watched it earlier.
Cool project and kudos to the developer both for doing it and sharing it.
I can see OEM's going for something like that to make calibration something that users need not know anything about. But it seems a lot of complexity to essentially pick a value when you an do it by eye with no added cost.

@droftarts your not so easily replaced!

Where is heater and thermistor?

No more oozing is quite nice. You close the nozzle instead of retracting filament. That allows using remote Bowden tube extruders.

I only hope it works smooth at different nozzle temps and in heated chambers...so many different material expansion-rates to deal with.

@dc42 Thank you for the detailed explanation!

@Sindarius @chrishamm Thank you for the feedback on this!

@chrishamm

Thanks for that, this is helpful!

@hebigt
Cheers dude. Sorry I miss interpreted your question. Due to losing micro stepping I think you would need a really high resolution encoder to make a closed loop stepper worth it.

I think your right on the filament grinding side though!

@hebigt I think it is because you are using G91, ie relative coordinates, for moves. X is increasing each loop, so it moves the distance described by var.x not to that actual position. The first time it moves a distance of 0.05mm, next it moves 0.1mm (ie to X0.15), then 0.15mm (ie to X0.3) etc. My mental maths isn't good enough to know what will happen to var.y if you switch to absolute coordinates (G90), but I'm sure there's a simple way around that, too.

Ian

@hebigt said in How well does RTV silicone stick to a silicone heater?:

@mrehorstdmd Interesting, thank you for including the photo. I never would have guessed that would be the outcome of a bubble. Those traces must get hot!

It's fairly easy to trap a bubble of air under such a heater during application. It has to be done slowly and carefully. Using a seam roller to try and insure the best possible adhesion can also helpful.

Frederick

@martin7404 said in Bambu Lab Printer (for discussion, not promotion):

I Got one .
It is fast.
Also very close to experince like geting HP laser jet and runing it. Espesialy with PLA.
This thing will take away the good sleep of Prusa, ultimaker, Raise 3d etc

I got mine last week, Have to agree. Only going to get better as well.

@tenaja I perused a few of the reviews on all3dp.com and it was helpful, thank you!

@chrishamm Interesting, thank you for the starting points, I will give this a try.

I think I can close out this thread now

Both the Automatic Z and XY offset are working.

For XY offsets (using the optical endstops) I had to probe each side of the syringe tip - i.e. probe the tip on both sides in the x direction, probe both sides in the y-direction... I think this is how some dual extruder FDM printers set the XY offsets.

I used gcode meta commands to query current machine positions and then referenced those values later on. Every now and then I used an IF statement to check whether a variable already exists.

I used a USB microscope to check the syringe tip alignment across tools and it looks good.

For anyone coming across this in the future, I used Fisnar tapered tips and blunt-end steel tips. I found that some plastic tapered tips didn't work well with the optical endstops. (Blue, pink, orange colored tips might be more transparent to gallium-arsenide LED?). The steel, blunt-end tips work great.

Just to put the icing on the cake, if you have G28 to home automatically at power up, what happens when you have a power blip mid print? When the power returns and the board starts up, it will home the printer and crash into whatever print is left on the bed.

@phaedrux Thanks!

@dc42
@OwenD
@theruttmeister

Thank you! This gives me a good starting point.

The motor current must be very high, but it's quite impressive.