Aufschaukeln mit JMC iHSS60-36-30-21-38 in Verbindung mit Linear Encodern

More
11 months 1 week ago #338739 by Jensner
Hallo Leute,

Ich habe in den letzten 2 Wochen meine Portalfräse mit einer neuen, sauberen LinuxCNC-Installation incl. Probe Basic versehen.
Die 3 Achsen laufen hervorragend auf JMC Closed Loop iHSS60-36-30-21-38.
Als Steuerungskarte lief eine MESA 7i96s.

Nun wollte ich die Genauigkeit nochmals verbessern und habe meine Achsen mit Magnet-Linear Encodern versehen.
Zur Signalverarbeitung kam eine 7i85s dazu und die 7i96s wurde mit der entsprechenden Firmware geflasht.
Nach etwas testen und Jumper setzen, erhalte ich nun auf allen 3 Achsen saubere Signale über die ersten 3 Encoder-Eingänge der 7i85s.

Soweit so gut. 
Nun zum Problem.
Ich habe die HAL nun testweise nur mal für die X-Achse entsprechend angepasst.
Sobald ich nun die Maschine aktiviere beginnt die X-Achse selbstständig, sich ganz langsam sich aufzuschaukeln bis sie in einen Positionsfehler fällt. Nun habe gibt es 2 Optionen.

1. ich bin zu blöd die Pins in der HAL korrekt zu verbinden (was durchaus möglich ist)
2. Der Closed Loop versucht ständig zu korrigieren und der Messwert (PID) arbeitet dagegen.

Ich komme hier irgendwie nicht wirklich weiter.
Könnt Ihr mir hier bitte helfen.

Hier mal die HAL und INI (wie gesagt, ich habe erst mal nur X zum Test umgestellt)
 

Attachment not found


 

Attachment not found



Vielen Dank und viele Grüße
Jens
   

Please Log in to join the conversation.

More
More
11 months 1 week ago #338742 by PCW
My first guesses as to what is wrong are these:

1. P term too high (1000 is way too high for encoder feedback)

2. either stepgen scale and encoder scale (or both) are not correct
 

Please Log in to join the conversation.

More
11 months 1 week ago #338743 by Jensner

My first guesses as to what is wrong are these:

1. P term too high (1000 is way too high for encoder feedback)

2. either stepgen scale and encoder scale (or both) are not correct
 
 
HI,
I did some testing before i conected it in the HAL to make sure that die Signal of the Encoder incl. Scale is correct. These are Encoders with 0.005mm, so "ENCODER_SCALE = -200" are correct.
I tested it with HAL Show and it was identic with X shown in the GUI.

Further more I did some testing too with the PID-Config, but P=1000 and FF1=1 works the best. 
Every other (for example P=50-100 or FF1=0.1-0.5) leeds to errors. 
Best regards

Please Log in to join the conversation.

More
11 months 1 week ago #338744 by Jensner

Schau mal an das:
forum.linuxcnc.org/10-advanced-configura...glass-scale-feedback
I Think my config sadly is exactliy what you mentioned with "THIS IS NOT FOR CLOSED LOOP stepper systems with encoder wired to their drives."

What do you think?
 

Please Log in to join the conversation.

More
11 months 1 week ago #338745 by tommylight

Zur Signalverarbeitung kam eine 7i85s dazu und die 7i96s wurde mit der entsprechenden Firmware geflasht.
Nach etwas testen und Jumper setzen, erhalte ich nun auf allen 3 Achsen saubere Signale über die ersten 3 Encoder-Eingänge der 7i85s.
   
Are the encoders wired to Mesa 7i85S?
Or are the encoders wired to JMC ?
Or, i really hope not, did you remove the encoders from JMC and wired to Mesa? This will not work without changing the drive config so the JMC are set to work with no encoders.
I assume this is the case as you mention the motor moves slowly when enabled, and this is exactly what will happen when encoders are detached from JMC.

Please Log in to join the conversation.

More
11 months 1 week ago #338746 by Jensner

Zur Signalverarbeitung kam eine 7i85s dazu und die 7i96s wurde mit der entsprechenden Firmware geflasht.
Nach etwas testen und Jumper setzen, erhalte ich nun auf allen 3 Achsen saubere Signale über die ersten 3 Encoder-Eingänge der 7i85s.
   
Are the encoders wired to Mesa 7i85S?
Or are the encoders wired to JMC ?
Or, i really hope not, did you remove the encoders from JMC and wired to Mesa? This will not work without changing the drive config so the JMC are set to work with no encoders.
I assume this is the case as you mention the motor moves slowly when enabled, and this is exactly what will happen when encoders are detached from JMC.

Yes, the 3 Linear-encoders are wired to the Encoder-Ports at the MESA 7i95s.
But the iHSS60 itself are Closed Loop and i thought, that the internal controller maybe tries to correct the position.
Than the linear-Encoder gets the signal to the PID to correct and this leads to the growing shaking. 
Do you know what i mean?

Has someone maybe a similar config? Closed Loop Stepper together with Linear-Encoder on Mesa?

Please Log in to join the conversation.

More
11 months 1 week ago #338748 by Jensner

This will not work without changing the drive config so the JMC are set to work with no encoders.

Do you know how to config this in the JMC?
I did some investigation, but could not find how to do this...

Please Log in to join the conversation.

More
11 months 1 week ago #338751 by PCW
If the scaling is correct, FF1 must be 1.0

I would lower the P term (till it no longer oscillates), raise the ferror limits
 And the plot the following error (and commanded velocity) to get an idea of
what is happening.

Another guess is that the drive responds too slowly to step commands
(this may be a tunable feature)

Please Log in to join the conversation.

More
11 months 1 week ago #338760 by tommylight
The link i posted, explicitly mentions making the machine work without encoder feedback to LinuxCNC first, so enable the stepgen feedback in the hal file and omit the encoder feedback, test the machine, set the scaling correctly, and only after getting this done activate the encoder feedback and omit the stepgen feedback.
Closed loop stepers when loop is closed on the drives are viewed as normal open loop steppers from LinuxCNC.
So yes, it should work with your drives, just make sure they work properly in open loop mode to LinuxCNC.

Please Log in to join the conversation.

Time to create page: 0.580 seconds
Powered by Kunena Forum