servo motor run away

[HANSJUERGENKARL]

hi everybody
this is probably something that everybody wants to avoid
i have a multitek box(tek10 amp) with the deskcnc controller mounted to one of my mills (it is running quite good for the last 2 years, maybe 3)
now suddenly the x axis motor takes off at random(i push the fast feed button) but stoppage only via kill switch,
on top there is a random x-axis creeping(control on, but in standby)and no change in the position display, until it hits the solid endstop.
i redid the wiring,replaced the amp with a spare i have,replaced the wiring to the motor/encoder, but no change
i had at one point a repeatable run away (6 times) of the x-axis and when i jogged the y/z axis (no problem there) the run away situation evaporated until next time.
can anybody shed some light on it, i am at a loss.
cant use the machine,do not know where it will go next

[Al_The_Man]

Assuming there is no tach on the motor, I would check the encoder signal output.
Al.

[HANSJUERGENKARL]

the funny thing is when i start the control, the run away condition does not appear always,if it does then only at the x-axis.this morning it did it again.
1. jog x-axis in rapid, x-axis started to run away, killed it
2. jog x-axis in rapid, run away again,killed it
3. jog y axis in slow, back and forth
4. jog x axis in rapid, servo moves at set jogging speed
5. run small test program with feed/rapid in x/y plane a fw repetitions, everything fine
6. let it sit for some time (1/2 hour) x-axis starts to creep in slow feed,no position change
7.scratch at the last few hairs on my head
hans

[Tartan5]

Not familiar with your hardware, but could your X-encoder be slipping? If it were loose, that might explain why at slow speeds it's fine, and acts up more at high speeds....

Can you swap the x and y servo/encoder signals?

[HANSJUERGENKARL]

saga continues
started back after coffeebreak , jogged x in rapid and again run away of the axis
so i removed the x axis servo motor,checked wiring trough the housing connector
(why they route the encoder cable and powercable through the same conduit is something i do not know)
but checked the encoder, nothing lose,cable seams to be alright
it is an accu-coder model 775a -1000 encoder
not knowing if there is cable disconnect or not so i swapped the whole motor/encoder
against a known working one.(motor is on bench)
joggingin slow/rapid seams to work and my test program is running.
my problem might point to a faulty wiring at the encoder/motor end
so far test program still running
hans

[Al_The_Man]

Originally Posted by HANSJUERGENKARL (why they route the encoder cable and powercable through the same conduit is something i do not know)

This is often done, especially when using differential encoder with correct cable and shielding etc & the motor conductors are twisted together.
Al.

[HANSJUERGENKARL]

the sparemotor/encoder is doing okay,
it looks that either the encoder/or the big plug at the servo has a loose wire(encoder)
i will reconnect back to the machine axis reroute the encoder wires seperately and see what i get
thanks
hans