Ideas for cooling Nema 17 steppers?

[Kyle Kish]

The Z and X motors aren't to bad but the Y axis Nema 17 is very hot. but Xmachines left us about 5mm of space to work with so the typical 40mm fan and heatsink stuff people do wont fit.

Any ideas what we can do to cool the puppy? 

[Nic Sievers]

Somebody else mentioned them getting hot.  I wonder what the voltage is on these?  Perhaps they are getting too much voltage and have to adjust the pots down?   

[Robert Keifer]

I emailed them about the temperature and they said it was fine and that they have these running 24/7 and not a single failure yet.  Mine runs at about 50C.

I found this on the motors they use:

http://eng.bohongmotor.com/product/42BYGHB_4018.html

Says temperature rise is 80C.  I'm not crazy about them getting so hot either, but it seems well within operating specs.

[Kyle Kish]

Ok well thats good to know. I might still try and find some like of ...sloping fan shroud and a 40mm fan. maybe a project for another time

[Bob Kolbasowski]

This is what I have also seen on the extruder motors - these small motors are not the same as a regular 17. Given that the holding torque much lower, to get performance out of it, it must be worked hard and heat is the byproduct. 

FWIW I have changed out my extruders to real 17s and they run much cooler. I will also be changing the Y to a normal 17 - that is one place where you really need the most power.

[Kyle Kish]

Ah I didn't realize we got smaller versions. something to think about for sure.

[Jeremy Campbell]

Bob ,

Do you have any recommendations for the motors?

Thanks!

[Bob Kolbasowski]

Jeremy,

The included motors (not including Z) have approximately 33 oz.in hold torque. 

Something like this one has 83 oz.in holding torque and shouldn't be difficult for the driver to push.

[Sandro]

They are all NEMA 17 motors.  The NEMA 17 spec is for mounting purposes, which is to say NEMA 17, should all be 1.7" x 1.7" on the mounted side.  They can be any depth.  Similarly NEMA 23 motors are 2.3" per side on the mounting face.

Whether you change the motors or not, you should probably be adjusting the voltages.  I don't know if XMachines tested the output voltages for each motor before they sent the boards out, but it's worth doing again.  Since testing the voltages isn't too hard, I'd check them before I started changing things around.

I know from personal experience that the motors in the Lorei can do the job.  I'm not sure why we would need more holding torque on the X or Y axis, since there aren't really many forces acting on them during printing. (Z and extruders are another story).

[Bob Kolbasowski]

Sandro,

By stating 'not regular 17' I meant that they are a bit stubby - and it clearly shows on the specs. I apologize for not being specific.

They do the job but they do get worked hard, and heat is a clear indication of that. Of the other printer I have and have used, none have gotten so hot so quick. 

Also since they have limited torque compared to the normal, 'run of the mill' nema-17s that are used on most printers, I believe deceleration artifacts become more apparent. Once a large print is on the plate, Y-axis will have an increasing hard time as the print progresses and unless print speeds are slowed way down you will see this.

I will be upgrading the Y-axis extrusion with C-Beam within the next two weeks (whenever that parts arrive) and I will be using a standard 17, I'll keep all updated on the outcome.

Also, I have adjusted the potentiometers to feel but do you know what driver is used? 

[Sandro]

If by "stubby" you mean they are not very long and as a result they have lower torque ratings than longer NEMA 17 motors -- sure that's definitely true.

So the heat is an indication of lots of power being dissipated.  One reason could be lots of load on the motors, the other could be too much voltage. Easy test, run a bunch of moves on the y-axis *without* the bed attached.  If the motors still get hot, it's more likely over-voltage than over-loaded.

I have not experienced any problem with the y-axis bogging down, but I'm not sure what you mean by a "large" print.  I've printed several items, some as large as 8"x8", but nothing terribly larger.

The drivers are Allegro A4988 (or clones).

[Nic Sievers]

Yeah that is exactly what I was thinking. I was wondering if xmachines ever tested the voltage and the board is running too much.  My printer is built, will see if I can get it running this weekend.  If I have a chance will test mine.  

[Bob Kolbasowski]

I never said bogging down - actually referred to their ability to slow down / decelerate. 

[Bob Kolbasowski]

Actually, if or when or have time - print any cube test object, such as 40mm cube, and let me know if you see decel artifacts as I have. When you look at the corners, each will go slightly beyond where it was supposed to stop and you will be able to see a ripple effect pattern before that corner. 

Granted, from what I have seen, it's not awful or what I would call gross. And it can be easily rectified few ways, either mechanically or digitally. 

Maybe it's just my printer, but I think not.

[Sandro]

Printing one now.

It sounds like you're experiencing ringing.  But I don't think it's necessarily an artifact of the stepper torque.  I've had a ringing problem with my Prusa mk2 since the beginning.  A couple of things to note, the Prusa firmware supports two modes for stepper power "normal" and "silent".  Silent is much quieter at the cost of significant amount of torque.  I've printed calibration cubes in both modes, with no difference in the output.  The ringing seems to vary depending on my belt tightness (somewhat looser than as tight as I can go seems best).  According to:

https://www.simplify3d.com/support/print-quality-troubleshooting/#vibrations-and-ringing

the culprits are likely printing too fast, firmware acceleration, or mechanical.  Since on the Prusa I'm using the factory setup for slicing and firmware, I'm left with my problem must be mechanical.  I suspect the crappy IKEA tables I print on, and possibly something in the build since mine was a DIY kit.  

I'll edit this post later with the output from my Lorei of a 40mm test cube.

[Bob Kolbasowski]

Yeah, it's kind of like ringing but seems like the Y-axis cant seem to stop in time so each of the opposing corners on that axis have little humps on them. I did tweak settings / speeds and got it down to about .4mm but I could only get rid of it if I print super slow - like 1300.

Thanks for checking on your printer, tell me how yours turns out.

[Sandro]

Here's the results of the print (front, left, back, right):

I don't know if it's visible, but there's ringing on the left side of each face.  This is about comparable to what I'm getting on my Prusa.  The light and dark vertical bands are the infill showing through the faces.  And clearly, I had some lifting along one side (left).  Frankly, I think the biggest issue I'm having is come periodic oscillations on the Z-axis.

I checked the motors as it printed, and none became even warm, certainly not hot.

Voltages: I set the axis motors to on, and checked the voltage between the potentiometers and ground.  In each case I got around .2V.  I also checked the voltage while moving the x-axis, and got the same voltage.

[Bob Kolbasowski]

Sandro,

From looking at the top, how to the corners look?

[Sandro]

Here's a photo from the top.  The arrow points towards "home".  It looks like there's a tiny bit of over-run in the x and y directions. 

[Bob Kolbasowski]

Yeah, that is tiny - mine was slightly more evident until I slowed it down to about 2500.

[Sandro]

I printed the test cube at the default 3600mm/min.

[Bob Kolbasowski]

Interesting. Here are my current temps, been running for about 2 hours.

[Bob Kolbasowski]

Changed out to Nema-17 I referenced above, print quality is hard to judge - maybe slightly better but so hard to tell, it's prob the same. Motor does run a bit cooler it seems.

[Bob Kolbasowski]

The print looks pretty good, on the side profile you can still notice a little bit of an artifact - but nothing to be upset about. The print was also done in the blue, but with the new extruders, I don't think it poses any issue.

Thanks for the time assisting Sandro.

[arg2761]

Bob:

Did you figure out the motor drivers? I know Sandro says Allegro A4988. However, in the Pololu web site there is a page on this driver and it states that the Rcs are 0.05 Ohms; changed to 0.068 Ohms in Jan 2017. The drivers on the Lorei board have Rcs 0.1 Ohms.

Also did you ever installed the C-Beam? Any chance of showing a pic? Open builders is having a sale on it. 500mm for $14. I was thinking to buy enough to upgrade the Y axis to 12" inches.

[Sandro]

I'm pretty sure the driver chips are A4988, but it's entirely possible that instead of using the same design values as the Pololu drivers this is a modified design.  

I also just remembered that there are actually two different sizes of Nema17 steppers used in the printer.  I believe the extruder motors are a thinner design, so they probably have a different operating current.

[DroneOn]

Correct me if i'm wrong, but isn't the A4988 limited to 1/16 microsteps? If I'm correct then we have DRV8825 drivers which means that a Vref of 0.2 would give us the 400mA max rated current of the steppers.

[arg2761]

Oops! I think I installed an incorrect extruder2 stepper motor.
I have a 42HB40F04AB, not a "34".
Should the 40 be for the X axis or the Z axis?

[arg2761]

Should be Y axis.
Interesting I don't see the specs in the BoHo g site.

[Nic Sievers]

Yes that is the one I have on the y-axis.  It is on the site.  First one on the list...this page

http://eng.bohongmotor.com/product/42BYGHB_4018.html

[Nic Sievers]

Do you have the number for the extruders?  I haven't taken mine off yet so I can't see it

[arg2761]

The extruder motors are 42HB34F04AB

The one for the Y-Axis is 42HB40F04AB; this one is not listed in the site.

Regards,

Al

[Bob Kolbasowski]

Al,

I did figure out the driver, but tweaked to pot to feel.

The C-Beam is my project this weekend, I'll post pics. To note, I am going to keep it 500mm, and will be replacing x to 500mm.

[arg2761]

What will you use for the X axis? I don't think the C-Beam will fit without some "gentle persuasion".

Regards,

Al

[Nic Sievers]

Oh wow you are correct.  That is interesting. the '34' is what I have on the y axis as well.   I didn't look at the numbers because each motor was in a specifically labeled bag for that section of the instructions.  

I am looking to take it apart this weekend to do a full test on the motors and make sure they are in the right locations.  Perhaps mine are all 34's?  If they were 40's I would think they would be too tight to fit in that extruder alignment. 

[arg2761]

It fits.

[arg2761]

I went ahead and installed some heatsinks on the X, Y and extruder motors.

I was able to do a couple of prints; about two hours each. Finally!

I have to do some optimization; but, I can print now. No more stopping after an hour of printing, no more ticking and failure to feed filament, and no more THERMAL RUNAWAY ERROR.

Bob K.: Thanks for your info and the picture in discord

Here are a couple of pics (heatsinks):

I also installed silicon socks for the heating block and put some heat shrinking tubing on the thermistor brass fitting:

[Scott Roberts]

Where'd you get those neat little heat sinks for the extruder shaft? Also what sort of adhesive did you use for affixing the heat sinks? 

[arg2761]

Scott:

Heatsinks on Mk8 Gear: I had a 40mm x 40mm x 15mm heatsink. I just cut three row by three row pieces. Drilled them in the back to make space for the shaft (it protrudes a little). Glued the mini heatsinks to the gear only; so, I can remove from the motor if needed. I'm sure they can be snapped off with some authority.

Thermal Adhesive (Amazon):

https://www.amazon.com/gp/product/B0009IQ1BU/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

regards,

Al

[Daniel Schmiedel]

@arg
How do you fix the heatsinks ?

Do that work ( mounting heatsinks ) ?

Ty

[arg2761]

Daniel:

To attach the heatsinks I used this Thermal Adhesive from Amazon:

https://www.amazon.com/gp/product/B0009IQ1BU/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

I was having issues where the filament would stop feeding at around an hour of printing. There were clicking noises as the extruder gear slipped. If I stopped for a few minutes and restarted, it would print for a while and then the same problem would return. The motors got somewhat hot to the touch.

There are some hypotheses in the forum, low quality XMachines filament and/or extruders motors overheating transferring the heat to the shaft and gear causing the filament to heat, soften, and possible slip.

I had heatsinks available to me; so, I decided to install them and test (placement is critical; they may interfere with motor movements).

The problem went away. I have done prints longer than two hours without the issue recurring. So, it worked for me.

The XMachines filament may still be of low quality; but at the moment, it is all I have.

Regards,

Al