Extruder EEPROM settings not being effective

[timskloss]

I'm trying to get a Zesty extruder up and running and it won't extrude faster than 7 mm/sec without stalling.

I'm repeating my problem here in the hopes that someone [expert at firmware sees it, because I think this is a general repetier issue. The EEPROM values for extruder 1 steps/mm and maximum extrusion speed are WORKING and I see the effects in the printer. But any change I make to start feedrate or acceleration are NOT having any effect on the extruder performance. This is weird and frustrating.

I have followed multiple setup guides posted here and elsewhere, and even chatted with Brian at Zesty today, without any luck.

I followed the setup exactly this way: http://forum.seemecnc.com/viewtopic.php?f=36&t=12016

Thanks, Ramai, for the excellent tutorial. I just wished it worked for me.

I am using a stock Rambo on a Rostock Max V3 from November 2016.
I have downloaded the latest firmware from GitHub pointed to by SeeMeCNC's documentation. It is dated August 22 2017.
I erased the EEPROM before downloading the firmware so the Rambo software should be clean.

I have tried varying the acceleration and start speeds over a wide range but no change is ever seen in the printer.

Any help would be greatly appreciated!

/Tim

[geneb]

Stepper motors will /only/ skip if the force they're trying to turn against is too high for them to overcome. There is literally nothing you can do via EEPROM to solve this.

First, check to make sure that the extrusion drive system will operate without any filament installed in it at all. Bring the hot end up to temp, pull the filament out of it and then try extruding via the control panel in either MC or Repetier Host(*) at least 10mm/sec. See if it'll do 100mm/sec. An unloaded extruder should do this easily. Even with a Zesty installed, there shouldn't be enough mechanical resistance in the system to prevent that kind of speed. The stepper MAY skip at 100mm/sec due to the higher speed required of the geared system in the Zesty, however if it skips at 10mm/sec, I'd disconnect the drive bowden from the stepper and see how much resistance you get turning by hand. Contact the Zesty guys to see how much resistance you should be getting from an empty extruder.

If there's no mechanical resistance problems with the drive system itself, then the other two things you have to look at are extrusion temperature and extruder stepper drive current. If the temps you're using worked fine for the exact same material under direct drive, the issue is likely related to the drive current - this can only be tweaked in the firmware - look for where it sets up the drive current values. I don't know exactly where, sorry. (I don't have the code handy to look at.)

* Repetier Host has much more reliable manual extrusion than MC does. No idea why.

g.

[timskloss]

Thanks, G.

With the motor disconnected mechanically I get the same results. So this isn't a mechanical resistance problem. When I switch from the Zesty-supplied motor to the motor on the stock EZstruder I also get the same results: stalling when the speed is around 7-9 mm/sec. So the motor is not to blame.

In firmware I swapped the control pins for extruder 0 and 1 and confirmed that it is not a chip problem on the RAMBO: I still got the same stalling behavior. So I put the pin settings back to the stock settings.

The stock motor current setting for the EZstruder is 140. I have tried this value, 100 from Ramai's post, and 48 by a separate calculation and came up again with similar results.

If I turn the steps per mm down from 730 for the Nimble back to the original 92.4 for the EZstruder, I can finally see the effects of a very low acceleration and start speed settings. But these effects disappear when I turn the steps per mm back up to 730 for the Nimble. I don't know if the RAMBO can work with this high of a steps per mm.

So at this point I am not sure what to do. I know some have made it work with other controller boards, but if anyone has made the Nimble work with a RAMBO board up to extrusion speeds of 30 mm/sec, please let me know so we can compare notes.

/Tim

[timskloss]

The problem seems to be tied to the steps per mm setting. If I use a setting under a 100, like the default setting for the SeeMe EZstruder, then I can see a difference in motor behavior by adjusting the max start feed rate and acceleration. But when I increase the steps per mm by about double, then the max start feed rate (jerk) and acceleration settings don't have any effect.

With the steps per mm set to 730, which is the correct calibration for 4 micro steps on the Zesty Nimble, the max start federate (jerk) and acceleration settings don't have any effect. I have tried numbers in the range from 10 to 0.1 for jerk and 120 to 0.001 for acceleration. In all cases, the motor starts and stops the same way--without any discernible ramp up or down in speed, just an instant on and off.

I think this is a limitation in the Rambo. I have a version 1.3L and running SeeMe firmware 0.92.2 20170822.

Would it matter what version of Arduino I used to upload the firmware? I am using the latest version 1.8.5 for the MacOs and the AVR ISP programmer.

/Tim

[timskloss]

The problem seems to be tied to the steps per mm setting. If I use a setting under a 100, like the default setting for the SeeMe EZstruder, then I can see a difference in motor behavior by adjusting the max start feed rate and acceleration. But when I increase the steps per mm by about double, then the max start feed rate (jerk) and acceleration settings don't have any effect.

With the steps per mm set to 730, which is the correct calibration for 4 micro steps on the Zesty Nimble, the max start federate (jerk) and acceleration settings don't have any effect. I have tried numbers in the range from 10 to 0.1 for jerk and 120 to 0.001 for acceleration. In all cases, the motor starts and stops the same way--without any discernible ramp up or down in speed, just an instant on and off.

I think this is a limitation in the Rambo. I have a version 1.3L and running SeeMe firmware 0.92.2 20170822.

Would it matter what version of Arduino I used to upload the firmware? I am using the latest version 1.8.5 for the MacOs and the AVR ISP programmer.

/Tim

[pouncingiguana]

Wow, 730 steps/mm. Didn't know it was that crazy high on a zesty nimble. That's going to severely limit the max speed when you're running on a Rambo. I don't know off the top of my head what it's max steps/sec is, but I'm guessing you're not going to be able to go very fast on that. Out of curiosity, you might try dropping to 2x microstepping (which would halve your steps/mm) and see if you can go twice the speed.

I don't think it's causing this issue for you, but the version of arduino does matter. As is says in the instructions, you must use 1.6.0 for everything to work properly. It'd be worth trying the correct version, but I'm still betting that you're hitting the room on max steps/sec

[Mac The Knife]

Stepper motors have great torque,,,,, when a pair of coils are energized. When you start microstepping a motor, the torque rating is going to drop off since to get fractional steps, the other set if coils are also getting energized. To add to the problem, the torque curve for a stepper motor fall rather fast the faster a stepper motor spins versus a brushed, or brushless motor.
GeckoDrive used to have some pretty good information on how steppers work, might be worth your while to dig around their website.

[timskloss]

Thanks, Ryan and Mac, and I agree with you both.

Math says that extruder retraction at 15 mm/sec with 730 steps/mm (4 microsteps) generates about 10,000 counts per second which is near the limit for a Rambo according to what I could find on the internet. I may try no micro steps which would reduce the steps per mm down another factor of 4, but not sure if that would result in a faster retraction speed or not.

So that is my limit for now unless I bite the bullet and switch to a faster board--which I probably won't because I think the Rambo is a perfect fit for the RMv3. Maybe in my next delta.....?

I did have to turn up the current to 180 in firmware to get 15 mm/sec to retract reliably. Right now the extruder has a thermocouple taped to the body of the motor and it doesn't get much above 100F. So I think this is a workable solution.

Thanks again to everyone who helped me get through this little transition. I feel the upgrade was worth it because I print a lot of PETG and it is much more compressible than PLA which gives me fits sometimes with oozing and stringing because of the long bowden tube.

Ryan, your team makes a great product! I can't wait to see the next generation.

/Tim