Servo Motor Position Drift After Continuous Cycles on Pick and Place Machine

552 11

#1 21673404 13 Feb 2015 04:20

Ankim Tandon Ankim Tandon

Anonymous

Post #1
21673404 13 Feb 2015 04:20

Hello Friends,

I am facing problem on my servo position.

I have pick n place machine , where my pick n place position are servo motor position controlled.I am facing problem is after few continuous cycle the position is not the same.
For example my calibration position is 85mm but after continous cycle position is out to 95 mm.I have crossed check my PLC code there is no calculation mistake.

Once i have origin my servo position starts from 85mm but after few cycle it starts to drift again.

Is there a possiblity of Noise on Servo pulse train cable ?
Cable are twisted pair cables for servo to move ccw and cw direction.

Can someone help me how to reduce noise or what could be cause of noise.?

Appreciate some direction
ADVERTISEMENT
#2 21673405 18 Feb 2015 10:59

Rohit Dubla Rohit Dubla

Anonymous

Post #2
21673405 18 Feb 2015 10:59

I would look at what the feedback signal is when it starts to go out of specification. It could be temperature related...
#3 21673406 25 Feb 2015 03:12

Frank Bushnell Frank Bushnell

Anonymous

Post #3
21673406 25 Feb 2015 03:12

Possibility: loose or worn position sensor in the servo.

Do you have a spare servo to replace the part and test it again ?
ADVERTISEMENT
#4 21673407 27 Feb 2015 04:09

vikasbly44 vikasbly44

Anonymous

Post #4
21673407 27 Feb 2015 04:09

DEAR...

If you have a spare servo to replace.. if yes, then you can replace the part and test it ..again..

THANKX...
ADVERTISEMENT
#5 21673408 27 Feb 2015 10:07

Frank Bushnell Frank Bushnell

Anonymous

Post #5
21673408 27 Feb 2015 10:07

Hi there Vik,
Is it your intention to be sarcastic ? Most people here are trying to be helpful.
ADVERTISEMENT
#6 21673409 01 Mar 2015 00:54

Ankim Tandon Ankim Tandon

Anonymous

Post #6
21673409 01 Mar 2015 00:54

Mr Rohit ,

Temperature on monitoring showed it is very stable and machine is in clean room enviornment.

The command pulse and received pulse on the servo drivor has differs....Looks like servo driver receiving pulse Cable (CCW/CW) has noise picking up from panel wiring.

I have added some ferrites and Isolated AC/DC wiring from eachother.

I will try to monitor it now...

Thank you,
Ankim
#7 21673410 01 Mar 2015 00:57

Ankim Tandon Ankim Tandon

Anonymous

Post #7
21673410 01 Mar 2015 00:57

DEAR I am trying to find out root cause.....before i suspect its a servo driver or servo motor fault.
#8 21673411 01 Mar 2015 00:58

Ankim Tandon Ankim Tandon

Anonymous

Post #8
21673411 01 Mar 2015 00:58

Mr Frank,

My servo sequence is Position based and not sensor based.
#9 21673412 31 Dec 2015 09:51

Zachary Pick Zachary Pick

Anonymous

Post #9
21673412 31 Dec 2015 09:51

Ankim

Servos have position sensors (encoders) built in. What kind of actuator are you using? If the encoder of the servo is not damaged or loose, and if the actuator is a belt, is the belt slipping?
#10 21673413 02 Jan 2016 10:32

Justin Spencer Mamaradlo Justin Spencer Mama...

Anonymous

Post #10
21673413 02 Jan 2016 10:32

Hi there Ankim.

Though it has been 10 months, I would like to share my ideas on this problem.

I think you are having a problem with offset. Position controlled feedback systems are very vulnerable to this. I suggest using a PID controller. You just have to add a few resistors and capacitors here and there. The schem. can easily be googled.

If this doesn't work, then the problem may indeed lie in parts replacement.

Regards,
Justin
#11 21673414 03 Jan 2016 14:05

Sambath Kumar Sambath Kumar

Anonymous

Post #11
21673414 03 Jan 2016 14:05

Hi Ankim Tandon ,

I think this is a mechanical problem your arm mover will have a spring loaded
system which will help the arm to reach the zero level point after every forward
movement , may be the spring fixing is loose or tension of the spring is gone or
wrong mech fixing. the servo motor is driven usually by transistors some times
one of the power transistor or its driver transistor may be weak replace it.

pl review my rating.

V.Sambath kumar
#12 21673415 27 Jul 2020 01:48

Manoj Kumar Manoj Kumar

Anonymous

Post #12
21673415 27 Jul 2020 01:48

Dear.
I have also faced the same problem .Reason:1. Pulse train wiring interracted with ac power supply cables2. amplifier issue3. 99percent - noise from ac power suply in the panel Remedy : use noise filter between ac power supply.
Create an account, log in here. You will receive points by participating in discussions.
Join this discussion.

Install Elektroda application

Didn't find an answer? Ask Artificial Intelligence

*I agree to send the question to OpenAI, Anthropic PBC, Perplexity AI, Inc., Kagi Inc., Google LLC - owners of language models in order to prepare the best response. The companies may monitor and log information entered into the form.

*I agree to publicly display my question and answer. The question and answer will be publicly available to everyone. The process may take a few minutes. Upon completion, you will be redirected to the page with the answer.

Wait...(2min)

Topic summary

✨ A pick and place machine using servo motor position control experiences position drift after continuous cycles, with calibrated positions shifting from 85mm to 95mm. The PLC code was verified with no errors. Potential causes discussed include noise interference on the servo pulse train cables (twisted pair for CCW/CW control), temperature effects, loose or worn position sensors (encoders), mechanical issues such as belt slipping or spring tension loss, and possible faults in servo driver transistors. Noise coupling from AC power supply cables into pulse wiring was highlighted as a common issue, with recommendations to use ferrite beads, isolate AC/DC wiring, and install noise filters. Suggestions also included testing with a spare servo, monitoring feedback signals, and implementing PID control with additional resistors and capacitors to address offset errors in position feedback systems.
Generated by the language model.

FAQ

TL;DR: In one documented case, 99% of similar servo drift issues traced to AC power noise; “use noise filter.” [Elektroda, Manoj Kumar, post #21673415] Why it matters: This FAQ helps pick-and-place and motion-control engineers stop position drift that ruins placement accuracy and throughput.

Quick Facts

Reported symptom: calibration at 85 mm drifting to 95 mm after continuous cycles. [Elektroda, Ankim Tandon, post #21673404]
Environment steady: temperature stable in clean room; drift persisted. [Elektroda, Ankim Tandon, post #21673409]
Observed mismatch: command vs received pulses on driver differed; ferrites added and AC/DC wiring isolated. [Elektroda, Ankim Tandon, post #21673409]
Root-cause claim: “99percent – noise from AC power supply in the panel”; fix with an AC noise filter. [Elektroda, Manoj Kumar, post #21673415]
Alternative causes: encoder fault or belt slip on actuator mechanics. [Elektroda, Zachary Pick, post #21673412]

What causes servo position to drift after many cycles on a pick-and-place machine?

Common triggers include AC mains noise coupling into pulse-train wiring, amplifier faults, or mechanical slip. One contributor reported 99% cases tied to panel AC noise. Adding an AC line noise filter resolved it. Inspect feedback integrity and mechanics after power quality checks. [Elektroda, Manoj Kumar, post #21673415]

How do I quickly test if AC power noise is the culprit?

Do a controlled run with an AC noise filter added to the panel supply. Keep pulse-train and power cables separated. If drift disappears, noise was the source. This mirrors a reported 99% noise-caused drift outcome and recommended filter remedy. [Elektroda, Manoj Kumar, post #21673415]

Can temperature changes explain the drift?

Temperature can affect sensors and mechanics, but one case showed stable temperature in a clean room while drift persisted. That points away from thermal causes and back to electrical noise or mechanics. “It could be temperature related...” is worth checking, but verify power noise first. [Elektroda, Rohit Dubla, post #21673405]

What is a servo encoder and why does it matter here?

A servo’s built‑in encoder reports shaft position to the drive. If the encoder is damaged, loose, or its coupling slips, commanded and actual positions diverge. On belt or linear actuators, an encoder can read fine while the belt slips, causing drift at the tool point. [Elektroda, Zachary Pick, post #21673412]

My command and received pulses don’t match—what should I check?

Treat it as signal integrity noise. Separate AC and DC runs, twist CW/CCW pairs, add ferrites, and verify shielding and grounding. One user noted mismatch and reduced interference by isolating AC/DC wiring and adding ferrites before continued monitoring. [Elektroda, Ankim Tandon, post #21673409]

Could the amplifier/drive be faulty?

Yes. A weak or failing amplifier stage can miscount or distort pulses, leading to drift. One report listed the amplifier among likely causes alongside AC noise, with the latter being far more common. Swap with a known-good drive to A/B test. [Elektroda, Manoj Kumar, post #21673415]

Is this a purely position-command issue or sensor-based control?

Even in position-command sequences, the servo relies on encoder feedback. Confusion arises when sequences are “position based,” yet actual position still comes from the encoder. Validate encoder health and mounting regardless of command mode. [Elektroda, Ankim Tandon, post #21673411]

What mechanical issues can mimic electrical drift?

Look for belt slip, loose couplings, worn pulleys, or weak springs in return mechanisms. If mechanics relax under load, the system returns to a shifted zero. Check and retension or replace suspect parts before retuning controls. [Elektroda, Zachary Pick, post #21673412]

How do I reduce pulse-train noise on CW/CCW lines?

Use twisted pairs, short cable runs, proper shields grounded at one end, and physical separation from AC power. Add ferrite beads and install an AC mains noise filter in the control panel. Re‑test placement repeatability. [Elektroda, Ankim Tandon, post #21673409]

Should I try PID control or tuning to fight drift?

If offset accumulates due to plant dynamics, a tuned PID can help. However, if noise or slip causes missed counts, tuning will not cure root causes. Fix power noise and mechanics first, then tune. [Elektroda, Justin Spencer Mamaradlo, post #21673413]

Edge case: can a weak power transistor cause drift?

Yes. A degrading power stage transistor or its driver can reduce torque or distort drive signals, allowing position error under load. Replace the weak device or the drive module and verify repeatability after repair. [Elektroda, Sambath Kumar, post #21673414]

What’s a fast 3‑step diagnostic to reproduce and isolate the drift?

Log command vs received pulses over 500 cycles at nominal speed.
Add AC mains noise filter and separate AC/DC cable trays; repeat the run.
Swap in a spare drive or motor; retest to isolate electronics vs mechanics. [Elektroda, Ankim Tandon, post #21673409]

How big can the drift get during production?

One case documented a calibration at 85 mm shifting to 95 mm after continuous cycling. Ten millimeters is enough to ruin placements, so treat early warnings seriously and act before yield falls. [Elektroda, Ankim Tandon, post #21673404]

If temperature is stable and drift remains, what’s next?

Proceed to electrical and mechanical checks. Prioritize AC noise mitigation, verify pulse integrity, then inspect encoders and belts. This sequence matched a clean-room case where temperature was steady but pulses differed at the drive. [Elektroda, Ankim Tandon, post #21673409]

Generated by the language model.