CNC V3 + TMC2009 on GSO DOB 12" #tmc2209 #cnc #dob


Davide Preziusi
 

Good morning, I'm Davide and i'm writing from Italy
I'm here to thank Howard Dutton and everyone who contributed to the wiki.
I'am new to the Arduino world, and, also an absolute beginner to the astronomy, but i managed to build a CNC Shield with two DRV8825 for my Dobson GSO 12". 
I used THIS project from Thingiverse for the mechanical part, with a bit of remix in AutoCAD to adapt it to a geared stepper and, in my opinion, improve it; i'll upload my remix to Thingiverse to help anyone who wants to build the same project



After some fine tuning, the project worked really well, but i noticed that the noise from stepper was VERY loud, i was afraid i might bother my neighbors, especially late at night; so i bough some TMC2209, thinking they were plug&play, but i was WRONG.
I didn't know there were multiple brands making the same microcontrollers and so i bought the BTT 1.2; but after a couple of day of testing, i managed to get them to work, with no additional soldering, here's how:


 
Simply insert jumpers in END/GND and MS1 - MS2, as shown in the picture, to get them working at 16microstep, i found that was perfect with resolution and torque. The only issue is that you can slew max at 0.5-0.8deg/s, even with STEP_WAVE_FORM PULSE otherwise the motors will stall, i still don't know why; i'm using "TMC2209" in config.h (without _QUIET or _VQUIET or other), the vRef is correct, I calculated it following this formula

Formula for calculating Vref for the TMC2208 or the TMC2209

which gave me 1.73v (the shield is powered at 12v); anyway:
 
For the altitude, i used 17HS19-1684S-PG100 - Nema 17 - 100:1 - 1.68A with a gear reduction of 8.75 (105/12)
For the latitude i used 17HS19-1684S-PG51 - Nema 17 - 51:1 - 1.68A with a gear reduction of 37.2 (558/15)

Hope to help someone with the same setup


Simply one question: what is the best way to find the gear ratio? Measure diameters or teeth?

Thank you to all, again

Davide



 


 

On Tue, Jan 11, 2022 at 11:39 AM, Davide Preziusi wrote:
which gave me 1.73v (the shield is powered at 12v); anyway:
That trimpot on TMC2209 should be set to give Vref not highger than 1.0V for 1.68A stepper (thats equal to 60% of Imax), regardless of decay mode. Too high Vref can cause motors to stall.


kevin
 

Nice build, 
just a few  thoughts,
Did you remove the resistor from the CNC shield,
By grounding the Enable pin you are shorting out the signal from the processor which is not good practice, maybe cut a PCB track somewhere.
You may  want to check the value of Rsense, some drivers i buy have 0.22 ohms some have 0.1 ohms.

Gear ratio is usually done by counting the teeth, calculating from a diameter will depend on measuring and manufacturing tolerances.


Davide Preziusi
 

I think that you were right, because my vRef should be 1.03, i think that is the problem!


George Cushing
 

Count teeth. I usually photo the gear and bring it up in a photo editor for the counts.
Nstr Gears.jpg


kevin
 

I usually use a felt tip pen to mark one tooth and work my round, often several time because i loose count :-)
I think he actually 3d printed his so should have the CAD model for it.


Frank Henriquez
 

Hi Davide,
The TMC 2209 is a fantastic driver.
 Fystec has a good page on how to set the motor current. Their mode select columns are incorrect, since the jumper settings for 1/32 are not listed. The Watterot page lists the proper settings.

Frank


Davide Preziusi
 


thank you all; I set the vref to 1.03v, but the motors continue to stall over 0.8deg / s. Not a big deal, in the end the goto and the tracking are OK, it's just a little slow but that's okay
 


Khalid Baheyeldin
 

On Tue, Jan 11, 2022 at 05:39 AM, Davide Preziusi wrote:
For the altitude, i used 17HS19-1684S-PG100 - Nema 17 - 100:1 - 1.68A with a gear reduction of 8.75 (105/12)
For the latitude i used 17HS19-1684S-PG51 - Nema 17 - 51:1 - 1.68A with a gear reduction of 37.2 (558/15)
You don't say what microstepping you are using.

Using your numbers, I put these in the spreadsheet:

31,111.111    = 400 step motor,    1/32,    8.75,    100.00
33,728.000    = 400 step motor,    1/16,    37.20,    51.00

And with 2.5 deg/sec, I got these RPM numbers:
364.6    RPM
790.5    RPM

Which should not stall, at least for axis1.


kevin
 

acording to the data sheets
17HS19-1684S-PG100 is actually a 200 step motor and the gearbox ratio although call 100 is
Gear Ratio: 99.05: 1


similarly 50.9: 1


just in case you wondered why your goto were out a  bit :-)


Khalid Baheyeldin
 

Based on Kevin's specs for the motor gear boxes, and steps, it is now:

361.1 RPM, 15,407.778 steps per degree = 200,    32,    8.75,    99.05
789.0 RPM, 16,830.933 steps per degree = 200,    16,    37.20,    50.90

Still, axis 1 at least should not stall at 2.5 deg/sec (SLEW RATE BASE DESIRED)


ghpicard
 

Hi Davide
    From the photo you published you left the "Spread" pin, which is equivalent to MS3 in other stepsticks, open. As the pin has a pulldown, the mode is set to be StealthChop, which is the most quiet but IIRC can cause stalling at high speeds.
If you put a bridge to VIO then the driver will work in SpreadCycle mode, which is supposedly less quiet but is more able to handle higher speeds.
Could you test setting the Spread/MS3 bridge to VIO?


Davide Preziusi
 

Thank you so much guys, you are really kind!
However I had studied the motor spreadsheet well, and I had correctly set the configurator



Here it is, in case it helps



On Tue, Jan 11, 2022 at 03:54 PM, ghpicard wrote:
Hi Davide
    From the photo you published you left the "Spread" pin, which is equivalent to MS3 in other stepsticks, open. As the pin has a pulldown, the mode is set to be StealthChop, which is the most quiet but IIRC can cause stalling at high speeds.
If you put a bridge to VIO then the driver will work in SpreadCycle mode, which is supposedly less quiet but is more able to handle higher speeds.
Could you test setting the Spread/MS3 bridge to VIO?

Great to know, I'll try it as soon as I get home, thank you!!!
 
 


Davide Preziusi
 

to supplement the previous post, here is the position of the jumpers for the various drivers of the TMC2xxx family to set stealthChop or spreadCycle


Davide Preziusi
 

Even having inserted the jumper in MS3, which should have enabled the spreadcycle mode, the motors stall over 0.8deg / s.
I don't know, then, what could be the cause, but I think I'll be satisfied even with this speed