Ahh I understand. 
Ok got it, that makes sense.
After the farmbot is done making that sound does it say that itâs in the 0,0,0 position?
or does it say that the movement was not complete?
Iâm not using encoders when doing this action. It happily reports 0,0,0.
Thatâs good to know! and just to make sure, its the Z axis thatâs making that noise, not the Y axis trying to push farther than the track allows?
It sounds like itâs not spazzing out but just not getting enough juice to keep going up.
1 Like
Itâs not Y.
1 Like
Then itâs the Z stepper driver that needs some tuning with the screw.
The bot sends itâs 10000 something pulses for 400 mm up so the bot should theoretically go all the way up. Without encoders it doesnât know of that worked or not, so it most assume it arrives at 0,0,0. If in the meantime the motor doesnât get enough current to stay in position you get this situation.
1 Like
This page could benefit greatly from a follow-up step where you tune the potentiometers so that newbies like me donât run into these problems. When I get back to work on my bot Iâll try to increase the voltage a bit for the motors and see if this still reproduces.
I wonder if you could reproduce this problem as well. My drivers are all measuring at 600 mV. If you follow the steps above, do you get the same?
1 Like
Could you elaborate on the tuning screw stuff? Iâd like to give that a shot but donât really know what to do.
Then itâs the Z stepper driver that needs some tuning with the screw.
For testing, I just turn the screw once clockwise if the motor canât move like it should, when the motor is not activated. Repeat until motor runs smooth enough. Thatâs not the correct way, so Iâll give you the official stuff below:
Motor specifications:
LDO-42STH47-1684BC (official motor).pdf (143.2 KB)
The exact procedure from the manufacturer:
1 Like
Would this help my X axis movement to not sound so loud and âgrindyâ?
maybe my X axis motors both need a little more juice.
Anyone attempting this really should measure the voltage on the tuning screw before activating their motors. As such, donât tune it on a live Z axis (unless you turned off its âalways onâ function)
1 Like
I think tweaking your top speed might help a lot. Could be that youâre hitting a resonance frequency and if you slightly increase or decrease steps per second your motors wonât vibrate at parametric resonance
Ordinarily, when running, your ponytail would move up and down opposite your body motion, and side to side a little bit. But, if the frequency of your run is twice the swing frequency of your ponytail, the ponytail swings wildly from side to side. The two frequencies resonate, and the running motion magnifies the motion of the âunstableâ ponytail. Thatâs parametric resonance.
You can really notice this when your motors are accelerating and decelerating. They will pass a few of these frequencies as they wind up. So make sure your top speed is not dead on one of these frequencies. They way your motor is mounted can alter its resonance frequencies. Re-tighten some screws or worst case, add some rubbers.
Hmmm very informative @mdingena
thanks for the cool read on parametric resonance! and actually, I did tweak the speed a bit. got it to be a lot less loud now!
I appreciate the resources! this has indeed helped.
-C
Iâve turned the screw for the z axis and it doesnât seem to be solving the problem. Unlike what @mdingena is seeing, my bot just suddenly stops and says it might be stuck. There is no clicking or anything. Just stops.
@Cjaramillogrows @mdingena are you able to get your bot to do much of anything consistently? Mine craps out after 1 or 2 moves rendering essentially useless. Trying to figure out if Iâve somehow assembled or installed something incorrectly or whether itâs just the current state of the software.
@detchi
indeed I have had some issues but in total, our farmbot has successfully executed about 13 out of 20 sequences of my âplant seedâ sequence.
Iâm going to throw a few quick questions at you so I can get an idea of where youâre at and what your bot might be doing.
Q1: So I see that your farmbot Z-0 position is all the way up, ok so if thatâs the case then the farmbot should make + movements just fine. The use of negative coordinates wouldnât even be needed here as you z-0 position is all the way to the top so your only move options are to go down.
Q2: are you entering specific move commands via the control tab or are these part of a sequence youâve set up.
I did read above in one of your earlier posts:
Did you try and add some wait commands of about 1 second to see if that has helped fix the issue?
Does the movement issue happen with both manual move commands and sequence move commands?
It has been a good while since Iâve had movement issues like that. Last Monday I was experiencing some of those issues but seem to be fine now.
And just to make sure, your Z axis 0 position is all the way to the edge of the track? that includes your Y axis being all the way to the left or right of the track and its been already established that your Z axis 0 is 100% retracted or at the top most position?
We noticed that when the bot is moving, you can still send a new command. That new command is ignored by the firmware because itâs already executing a command and the command times out. The RPI goes into panic mode because the command timed out and assumed lost communication. Restarting communication restarts the bot with immediately stopping all motors and sets the position to 0, 0, 0.
There is a (partial) fix for that in the latest update and weâre working on a permafix. It also helps by adding enough pause time between commands and not pushing buttons while a sequence is running.
1 Like
This is really a matter of personal taste. You could power the motor such that moving down is negative only. It depends how you plugged it into RAMPS and the way you think about position. Reason could be: itâs going down so numbers should go down too.
For making the bot fool-proof, it shouldnât require those waits. And since the frontend seems to know the bot is moving (the absolute axis input boxes update on the go) the web app should either ignore your additional clicks until itâs done, or put those in a queue.
Like I said:
Yep, I do have the z-axis all the way up and X & Y zero points in the corner. So Iâm using all positive coordinates.
You made a good point about the manual controls vs. sequences. I have noticed it is more reliable using manual commands although I did have 1 second waits between each command in my sequence. The bot usually worked with manual commands all though it did stop and reset once.
Iâve made modest improvements in reliability by doing the following:
- I increased all speeds from 400 to 800.
- I never move the z-axis when x and/or y movements are happening. Always move the Z axis by itself.
- I increased the waits between commands within a sequence from 1000 to 2000
While this seems to have provided some help, the bot still does randomly stop and then reset itself. Iâm definitely not at a place where I can leave the bot unattended. After every stop interruption, I have to manually move the bot back to 0,0,0 which is an especial pain for the Z axis, having to manually screw it back to the top.
@Tim @roryaronson @Gabriel @RickCarlino Any other ideas?
I too have begun to add wait commands between all chained sequences. this has indeed helped me a lot also.
Also, I never noticed that but you are right. Whenever I have a complicated combo of x,y and z-movements the farmbot usually has a bit of a delay when executing the sequences and sometimes will just stop mid-command. saying the movement was incomplete and then sets itself to 0,0,0.
@detchi have you ever used the calibration command? I personally havenât done it and wonder if maybe, by setting up the calibration the farmbot will have a better understanding of its space, movements and incomplete moves.
@Tim thatâs super exciting to know that a perma-fix will be implemented!! 