Ho ho hold your horses. Today I just assembled my Maslow with 3 encoders and it wont work 
At the moment i just pushed the test motor button and didnāt get any readout. Tomorrow i will test one encoder on osciloscope to see what is the problem.
1 Like
Ok. I tested one of them on my old maslow shield at homeā¦ I swapped maslow shields yesterday then i tested encoders 
Pfff.
You know what? I spend 2h to swap the shields and my encoders works only on old shield
Ok. I need to check what is wrong with my maslow BIG shield
did the motors move at all?
as long as they are wired to the h-bridge you should have them move, even if the
ātest motorsā doesnāt see the encoders and fails.
Yes. Motors did move
I prabbly messed up the power trail for encoders
1 Like
U know what? Iām stupid. I just connected left motor to right encoder
pff.
UPS. Wrong Pic
Ok. Now the maslow with optical encoders finally responds.
Yup. Its working already
All three motors responds to orders. No modification done :P.
Ok. Now its time for benchmark
2 Likes
Story of this work in 6 parts
6th part will be about how it looks as finished project. I need to put it better order
Yey. Automatic subtitles works as well xD
Itās a playlist so watch all of them. Probably 4 available atm because Iām still learning how to upload things to my channel
1 Like
ok, if the motors move, either the controller isnāt seeing the pulses from the
encoder, or itās seeing them backwards, try switching either the power wires to
the motor or the encoder A/B pins (either will work)
note, the test motors routine should look for the motors moving a given distance
IN EITHER DIRECTION and then detect the direction and set a motor direction
variable
This is far from the first time that someone was using non-standard motors and
got bit from the assumption about which direction the motor turns when power is
applied
I had passed motor test maybe few times. I calibrated and now I cant pass test on top motor and Z motor
During calibration evertything was working great. But when it comes to Z axisā¦
Right motor signal interfere with Z axis encoder
What a fail. No wai. Z axis build is a failure
Robber band as pulley works as buffer and causes to run Z motor in very bad manner.
At the moment only top motors works correctly.
One step from trowing this thing to trash 
Top motors act like with magnetic encoders :P. Maybe shielded cable would help preventing interference.
Funny/bizzare observations while moving motors individually:
When i move Z axis ā both chains starts shaking
When i move Right motor ā Z axis starts moving randomly and tries to adjust itself.
Only left motor act like it should because it got grounded shield on whole length.
Experience Iāve get from this project is priceless
@dlang Iām using only my DIY shields (I already went with 3 versions of them). I can prepare one shield in 1h so its convenient for me. I tried to make it ass foolproof as possible. I probably had all possible issues with them. So Iām making all tinkering in real time.
I canāt mount Z axis encoder directly to TR screw so i should try GT2 timing belt rather than rubber band.
Lets Cry a bit at my Maslow
I think this is a good hunch. How many ppr are your encoders? I found that witch encoders higher than the stock ones I was losing pulses in the cables.
2400 PPR and it extends chains to perfect length. 2000mm set --> 2000mm I get
1 Like
Ok. I will try to mount Z axis motor on GT2 timing belt. Then i will have free shaft where i will mount encoder to it.
I will change to shielded cables. Normal 2x0,5 to every motor and 4x0,5 + shield to encoders.
1 Like
Ok. Iām going back to business.
Half of Monday i will spend on remaking Z axis to look same as my top motor mounts
Optical encoder mounted directly to motor shaft. Then transmission via belt to TR screw.
Then i will change wires for my encoder to shielded ones.
1 Like
Ok. I rebuild the Z axis
And still. Even on shielded cables when i move Z axis iām getting movement on both top motors.
I will share video in a moment.
Video need to be processed.
Last idea 
I can multiply resolution by adding transmission for encoders
Two optionsā¦ I can add this encoder to the end of motor shaft. 3mm w/o any key doesnāt look good. 2400*192 = 460800 PPR xD
Mounting it to output shaft via GT2 belt is possible but there are only fev combination. Minimal GT2 wheel got 20T. Maximal Iāve got on market is 40T.
Is the chatter on the gears from non optimized pid loops or from much lower resolution of encoders?
1 Like
This are 2400 PPR Optical encoder 9x lower resolution :P.
I donāt know how to fix it now
PID settings are default becouse I donāt know what will be result of changing this values.
I think your PID settings are the issue and the ācontrol system loopā with the encoder thinks the shaft is never reaching its setpoint so it keeps moving back and forth to find the final position. Maybe someone else on here knows the technique to tune the pid so you can get the pieces to stop moving. Typically when PID tuning starts with the P value, but Iāve never done it with Maslow. I would look to adjusting those, but Iām not sure the direction to change it or the amount you can change it without it not working at all.
1 Like
I think that the Maslow firmware PID loop has no deadzone, and it calculates in
floating point numbers, so it is never going to settle, itās always goin to
bounce back and forth across the setpoint.
I donāt know enough about PID setups to even start on fixing this though.
David Lang
Tomorrow morning i will try to lower floating point spaces to 2 places.
MarcinanBarbarzynca,
From a newbie perspective who was just really introduced to Maslow recently. Your project is insanely fascinating. Being able to read all these discussions as we can clearly watch your progress is fantastic and makes me want to start ordering parts of amazon and try it myself. (A lot easier to get wife to stomach $25 dollars worth of parts a week and work on a project than a chunk of like $500.00) Even though it will eventually be thereā¦ bite size bytes I guess. Also since the whole COVID 19 China scare I am stuck at home for months from work. So I need a new hobby and this looks like a great one to help pass the time and learn some more about coding.
2 Likes
Let us know if you get stuck anywhere, weāre here to help!
1 Like