Motor and driver question

[Blaze_1]

Hello all!

So i’ve got a newb type question to ask.

When purchasing motors I ended up getting a good price on the .9deg stepper motors, so I got 4 of them. In my haste, I ordered 2x DRV8824 and 2x DRV8825, as the BOM lists.

Do I need 2 more high current Pololu’s to run the .9 deg motors I have on the z-axis and extruder, or should the low current ones work fine?

The motors are rated at 3V and 1.7A/Phase with a Holding Torque of 48N.cm(68 oz.in).

I seem to be having issues getting the z-axis to work properly, and I wanted to make sure the motors are working right instead of spending a ton of money on replacing the leadscrew. (BTW for now I’m just using some of the threaded rod I had leftover from the build)

I’m sooooooooooooooo close to having my Morgan working 100%, and it’s driving me batty that I have this fantastic machine that isn’t doing anything right now.
(Incidentally this is also what’s kept me from posting pictures. I don’t feel right about posting pics of a finished build that isn’t quite finished.)

[Quentin Harley]

The low current motors would likely be fine, especially since we only need the 32uStepping on the Theta and Psi axis.

Threaded rod is ok (but slow). I only replaced it because of the low quality rods I could lay my hands on – I had z-ringing artefacts all over my prints. What is the matter with Z at the moment?

[Blaze_1]

I’m having issues getting the first layer to print properly. Almost as if the z-axis isn’t replicating the movement accurately after calibration. Half of the print sticks (The half closest to the drive rods), and the other half doesn’t. So far i’ve only tried printing the calibration cross-thingie.

Although your comment about only Theta and Psi needing 32uStepping has me thinking………

I’ve got both the extruder and Z-axis going at 32uStepping. I wonder if that’s my issue. How many jumpers would I need to remove from RAMPS to get the .9deg motors to act like the 1.8deg motors that the directions call for? I tried removing one (and adjusting the steps in Pronterface), thinking that it would go to 16uStepping, but I couldn’t get the motors to turn.

I’ve got the z-axis mostly disassembled right now, but i’ll try to get it all back together and run thru calibration again so that I can get some video of it printing. (One of my first thoughts was that I somehow didn’t have the z-axis mounted properly. I’m not so sure that’s the issue now.)

I’m at work right now, so i’ll try to give some more detail when I get home (including the current steps I have set for the z-axis and extruder)

Other than the small issues i’ve faced, I love this machine! I can’t wait to build another!

[Blaze_1]

Ah HAH!

Using a DRV8824 when all the generic RAMPS stuff references the A4988 has given me some issues before, and it seems to have struck again. Silly me for not paying more attention.

After taking a look at the RAMPS 1.4 page again, the Pololu DRV8824 page, and the A4988 page, I noticed that the jumper options would be different.

Taking that third one dropped it to 1/8uStep. I need to remove the first 2 and just jump the last pins for 1/16, which should give me the same number of steps as a 1.8deg motor.

Welp, I need to do the Drv8825 mod too, so this seems the perfect time to fix those jumpers.

Thanks for helping me get the grey matter box jump started, Q. Hopefully i’ll have pics and videos up soon.

[ZS6HDV]

I’m still familiarising myself with the Marlin code (so double check), but I believe that you can also work around the problem in software. In the file Configuration_adv.h, line 192, the microstepping modes are set for each of the five stepper drivers.

The 8825 I believe is configured to automatically go to 1/32 microstepping when the software (written for the 4988) selects 1/16, all the other modes work identical to the 4988. I.e. no 1/16 is available on the 8825 drivers.

[Blaze_1]

I hadn’t even dug into the Config adv file. I’ll take a look and make sure I have 1/16 available.

Thanks!

[ZS6HDV]

I had another look at the DRV8824 Pololu board, and my comments above only apply when using the 8824/5 board as a drop-in replacement for the 4988. 1/16th is available but not when microstepping is selected under software control with software for the 4988.

Hein

[Blaze_1]

It does make sense. Thanks for helping me think through this.

The jumpers I was referring to are the jumpers on the RAMPS board that selects the microstep.

Both the DRV8825 that are running my Psi and Theta, as well as the DRV8824 that is running the Z axis and Extruder are at 1/32 microstep. So in order for them to run the same as a 1.8 deg motor they would need to run at 1/16 ustep.

I think I’m just going to have to take my Wife’s advice and spend the money for the right motors. I found a pair for $30 so hopefully this fixes the issue.

And as a bonus I’ll have an extra set of motors for my next machine!

[Quentin Harley]

Before you spend the money, you can just double the steps/mm on those axis and you should be good to go…

M92 Z(double the current) E(double the current) as a starting point.

[Blaze_1]

It’s not that the motors don’t work. I think there are too many microsteps.

I think I’ve explained it wrong.

My issue is that the z axis won’t stay level to get the first layer to stick. I’ve attempted to adjust the bed and calibrate it several times. On every print attempt the side closest to you when facing the machine from the front consistently gets too close to the head, restricting the plastic and preventing that half of the first layer to stick.

[Quentin Harley]

It sounds like the automatic leveling is not working.
Did you do the M370, M371 M372 M373 Adjustment then?

Remember to save the setting afterwards with M500

I have a machine that runs the Z axis on 32uStepping with a 0.9deg motor. No problems after calibration.

[Blaze_1]

Yup that’s after calibration. and a save with m500.

That’s why i’m scratching my head as to where the problem could lie. I even cleared the settings and re flashed the latest marlin (morgan branch) with the same end result.

I’m going to try and get some video to see if that helps illustrate my issue.

[Quentin Harley]

That will help. Also, post the output of M503 and M375

[Blaze_1]

Thanks for the info. It’s helped me figure out why it’s not working, but I’m still clueless on how to fix it.

I tried with just the third jumper enabled, and it seems to still be at 1/32 ustep. I couldn’t even find an option in Marlin for the third mode pin(which is M3 on A4988/ M2 on my DRV8824).

As you said, it doesn’t seem like 1/16 ustep is an available option.

I was hoping to get this issue resolved without having to make another purchase, but it’s starting to seem like I just need to purchase 1.8deg motors and save those 2 extra 0.9deg motors for my next Morgan.

Any suggestions?

[ZS6HDV]

My Morgan build is still in progress so I’m working off theory here…

The Morgan Marlin code does not make provision for 32 microstepping selection in software. The 32 ustep board automatically goes to 32 usteps when 16 usteps are selected (as they are).

The BOM shows 2 x 0.9 deg and 2 x 1.8 deg steppers, and 2 x 32 ustep and 2 x 16 ustep drivers. To me this means the z axis and extruder runs at one quarter of the resolution of the arms.

If you run 0.9 degree motors all round, my logic says the z and e drivers must run at 1/8 usteps to be the same step size as 1.8 at 1/16. This option (1/8) is selectable in software.

I’m not sure of your stepper drivers (the circuits for the ones I see online do not have hardware jumper options), but it SOUNDS like you have jumpers on the board?

If I wanted to select 1/16 on one of those (directly connected 32 ustep) driver boards I would de-solder the mode select pins (on the edge connector – just heat the pad and pull out the pin), and physically connect the pad (now not connected to the PC) to power or ground according to the selection table in the data sheet.

But, if you select 1/16 manually as above, you are running a 1/16 and 0.9 combination, whereas the design calls for a 1/16 1.8 degree combination, still double the resolution. I would first try 1/8 0.9.

Hope this makes sense…
Hein

[Author]

Posts