Interesting fault on machine.

[Chris1234]

Had an interesting fault today. Don’t know if any of you like working out problems, but anyway.

A traveling saw, Lenze servo controlled travel, Servo speed is controlled by encoder. Saw blade is controlled by VFD.

Problem:
Servo travels to over travel limit very jerkily when turned on & saw won’t cut. Current length not displayed as read by encoder.

 

[westward10]

Have you ruled out a mechanical issue.

 

[Chris1234]

It’s fixed now, I just wondered if anyone liked guessing games

Yes, first thing was lubricate linear slides and ballscrew. It wasn’t anything mechanical.

 

[Lucien Nunes]

Hmm, so many variables. Could be anything from sawdust in the encoder, fatigue fault in the cable chain, faulty PSU in the servo drive (that's presumably AC rather than DC?). I'm supposing that the spindle motor won't run because the VFD is locked out by servo not ready error, or is there an electrical fault common to both the spindle and servo drives? Anybody been fiddling with the wiring?

 

[Chris1234]

It’s none of the faults you mention.

I think your right about the VFD not running because the Servo wasn’t happy though. That’s what I’d assumed.

 

[static zap]

a) Any optics in the system ..seeing a flickering light ?
b)(or extra interference -- Poor earthing.. like Lucines"Wiring-fault")?

 

[rapparee]

Was it an absolute or incremental encoder?

 

[PEG]

Any limit switches?

 

[Chris1234]

Incremental encoder.
No flickering lights.
Three proximity sensors:
Start
Another one, possibly to tell the saw to start.
Overtravel

Faulty earth? On encoder cable?

 

[rapparee]

Incremental encoder.
No flickering lights.
Three proximity sensors:
Start
Another one, possibly to tell the saw to start.
Overtravel

Faulty earth? On encoder cable?

So the proximity switches would be two limit switches and the dog switch.

Was the encoder wired in SY, was the sheath correctly earthed like you said.

 

[Darkwood]

Limited info' here, usually the saw starts when certain conditions are met, the main one is safety, it needs to know it is housed safely and ok to start, if it doesn't then often this is down to the saws 'home' position is open exposing the blade (usually a canopy lid so the blade can be changed) this also could be its seated position, if it needs to raise first and hasn't acknowledged it. This would initially lead me to check the limits for the saw blade home enclosure.

You then go on to express it does move the saw albeit in a jerky fashion and runs into the cutout limit (over-travel)... there can be various reasons here but another possible answer given the saw hasn't started is hydraulic/pneumatic pressure low (if used), the saw if it needs to raise may be low, this means sensors are not triggered but actions may still go through the process dependent on programming quality of software running it.

A machine should not move a saw that hasn't fired up as that would simply run it into a job and could mechanically damage the machine, I believe the problem was a simple one but I question the programming of this saw to give these results, if your explanation is correct.

 

[marconi]

Problem:
Servo travels to over travel limit very jerkily when turned on & saw won’t cut. Current length not displayed as read by encoder.

I begin my guess from the bold text in your problem statement. Since an incremental encoder only measures displacement and not absolute position, the position sensing system needs to sense and record a pre-defined reference position. Some incremental encoders output a Reference/Index signal when they are in a particular angular position/displacement. Those without would rely on a separate signal when the sensed part is at the reference position.

I think/guess that since the current length is not displayed the problem is caused by the failure of something in the absolute referencing sub-system eg: loss of reference mark (internal or external to the incremental transducer), broken R/I wire, defective reference transducer, defective incremental encoder interface circuit.

On first switch on I assume the the travelling saw must null the index error to then be able to track absolute position and know when it is in the home position and approaching or at end of travel - the servo then knows the extent of travel by counting up or down using the A and B quadrature output signals : One would not want the travelling arm to have to stop by running into the safety end stops and shutting off power totally - this would slow down the rate of sawing operations.

So something wrong with the index referencing system.

If allowed my second guess is the loss of an A or a B quadrature signal (but not both) so neither the servo nor the displacement counter receive any increments of displacement and their direction. As the control logic to sense the index position attempts to drive the actuator the A and B signals to the servo follow an incorrect pattern - one square wave and the other constant. It is not possible to sense direction from this faulty signal pair and the increment signal derived from them is jerky ie: on/off/on/off from one and constant on or off from the other and unsuitable for speed control because they don't form an analogue of rotation using the orthogonality of A and B.

 

[Chris1234]

Sorry for delay, it’s been a busy week.

To answer questions:

Encoder didn’t use SY, it was screened cable however.

There are no hydraulics or pneumatics on the machine.

It also isn’t a saw, I said this as I assumed people would understand what it was, rather than saying ‘travelling flying knife’ which is what it is. So the saw blade/motor doesn’t need to rise up. Just the knife turns a single rotation.

I may be wrong about it being an incremental encoder, they are not my strongest subject. The machine counts pulses until it reaches its set cut length, then whine it cuts, its resets the count. I think it was a 1000ppm encoder.

Yes, it shouldn’t hit the end limit during its cycle, some of our saw really don’t like this!

These machines usually have a maximum line speed programmed in, if the line speed is above this they know that they won’t be able to travel and make the cut in time.

I believe the carriage return speed is fixed on this and not dependent on encoder pulses telling what speed to travel.

Some interesting theories, so convinced was I that it was the encoder, I forgot the basics.
All voltages between phases 415v ish.
All voltages to neutral 241v ish

Then back back to encoder and servo drive connections!

Got shock of 38v AC between encoder and some steelwork.

Checked voltages to earth:
L1 - 193v
L2 - 283v
L3 - 193v