Encoder woes

PLC Pie Guy

Member
Join Date
Jun 2013
Location
Halifax
Posts
1,144
Hey all.
I have been troubled by an issue for some time and I thought I would ask here for general advice or somebody who might say its happened to them before.
We purchased an oven un-loader for our new plant. It has a gantry that moves back and fourth being pushed by a 2 Hp motor and a 525 drive. On the end of the motor is an incremental encoder. (SEW Euro Drive)
Over the past few weeks we have been getting f91 - encoder loss. I usually am able to reset the fault and keep running. Might see it once or twice a day. During downtime I have been going over everything looking for a loose wire or something that causes this. I have done the shielding on both ends of the encoder cable. This morning I have actually replaced the encoder cable and the termination cap. The final thing I did was to swap the encoder card from the drive in question to make sure the fault did not follow the card.
I am now simply left with the encoder itself. Do these things intermittently fail? I have never had one go bad on me yet.
One question I have is about the dip switch on the drives encoder card. It has a selection of 5V or 12V. Currently set to 5V. I would think this better at 12V. Any comments on this situation?

Thanks All
 
the selection switch should be set to the encoder supply voltage
if the encoder is powered with a 12VDC supply and the switch is set for 5V that could lead to your problem.
yes encoders do go bad in fact all electronic equipment can go bad.
you said that this is an oven un loader what is the temperature at the motor ?
high temperature can cause the encoder to go bad.
if you are just using the encoder for positioning then you may want to consider a laser measuring system.
 
Hey Gary.
Thanks for the reply.
In response to your question. The temp at the motor is low. The room is maintained at 65 degrees F. As well the motor is located about 16 feet from the oven and does not see any heat. To top it off it has an auxiliary fan to keep the motor cool. The motor, when it runs only runs for about 10 seconds, then its cycle is complete. twenty to thirty seconds later it will move again. The motor doesn't get a chance to heat up.

As for the voltage selection, I am getting conflicting info. The OEM of the machine says that they set all their encoder cards to run at 5 VDC. He says the 12 VDC could damage the encoder card. I don't think this is the case. As well the encoder is rated for 4.75 to 30VDC. The SWE Eurodrive rep says 12VDC all the time. For now I think I should stick with the advice of the OEM as to not void a warranty or something like this.

The comment about any electronics can go bad. I do realize this. I just have never had an encoder fail before and I am surprised to see it work for a period of time then fail. Then work for a bit and again fail. I am wondering what the typical symptoms of a failed encoder may be.

Thanks for your time and reply.
 
If it's a fairly basic encoder, then many factors could lead to a generic 'encoder loss'.

A quadrature encoder works by having 'A' and 'B' channels that are out of phase so the electronics can tell if it's going clockwise or counterclockwise. If A leads B then it's going one direction and the other direction if the opposite is true. The 'Z' channel is a marker pulse, going high (or low) once per revolution.

Missing a pulse or not properly seeing the regular progression would likely cause the electronics to fail the process, without explaining the exact error.
 
Some drives I've worked with give the generic encoder loss fault when any encoder problem arises. One thing that has bitten me more times than I care to admit in a public forum is to make sure that the coupling to the motor shaft is tight. I'm not familiar with the 525, so I don't know if that would result in a speed deviation type fault. You said it's new, so I don't know if you have a baseline for the type of current it should draw under normal conditions, but monitor the current and if you see abnormally high current, I would go ahead and replace the encoder after I made sure everything was tightened down.
 
I am indeed waiting for a new encoder now. Best case scenario is that there is one available in Can or the US. Right now SEW is saying that may not be the case.
Then, it would be coming from Germany.
Not good.

Thanks for the replies.
 
I am indeed waiting for a new encoder now. Best case scenario is that there is one available in Can or the US. Right now SEW is saying that may not be the case.
Then, it would be coming from Germany.
Not good.

Thanks for the replies.
 
You really need to get an oscilloscope on the leads and have a look at the waveform. You might find that it isn't to do with the encoder or drive but is to do with interference. One thing that did concern me is that you mentioned
I have done the shielding on both ends of the encoder cable
, normally you would earth one end of the screen and not the other, otherwise you can cause a current to flow through the earth screen which can induce voltage in the wires inside the screen.
Are the cables for the drive and encoder well separated, drives can create a lot of electrical noise.
Is the cable from the drive to the motor correctly screened.
Is the system installed according to the drive manufacturers instructions. Should there be a line choke, or output choke.
 
"I just have never had an encoder fail before and I am surprised to see it work for a period of time then fail. Then work for a bit and again fail".


Sometimes, the spring style encoder shaft couplers can break in their flexible spring area. They may still work at times, but at other times provide
intermittent or unreliable feedback as the broken spring catches and releases or changes direction.
 
One thing that did concern me is that you mentioned , normally you would earth one end of the screen and not the other, otherwise you can cause a current to flow through the earth screen which can induce voltage in the wires inside the screen.
Are the cables for the drive and encoder well separated, drives can create a lot of electrical noise.
Is the cable from the drive to the motor correctly screened.
Is the system installed according to the drive manufacturers instructions. Should there be a line choke, or output choke.
As Bryan said only one end of the shield on most devices get grounded and that is mostly at the supply end. But VFD cables get grounded on both ends.
 
I agree about the shielding and only bonding one end. This was something the oem asked me to do during the troubleshooting process.
I am fairly certain it isn't noise as when I ran a new cable this morning I ran it far from the tray and the original routing. I did specifically to avoid or eliminate potential noise.
I did manage to find and borrow a new encoder to try in the morning.
I wil keep posting my findings.
Thanks to all participants.
 
I have been watching this thread with interest and here is another thought on the problem.
Are you saying that your encoder is powered with a 12VDC supply but the selector switch is set for 5 VDC.
If that is the case then that is your problem
The voltage selector switch selects the threshold for the logic
With the selector set for 5v The low threshold would be about 2v or below would be considered low logic and 3v or higher would be considered high logic
With the selector set for 12v the low threshold would be about 5v or less would be considered to be low logic and about 9v or above would be considered high logic.
The key is the encoder on a 12v supply for low logic will only pull the signal down to about 5 volts. That is not low enough for input to read it as a low logic level.
Depending on the encoder output driver, some will pull the signal lower than others
Open collector, differential line driver, MOSFET totem pole are the most popular but there are others. Without knowing the exact encoder driver or the date sheet I can't tell what it is.

I don't know why the OEM doesn’t want you to change the switch without having a copy of the manual for the encoder input cared I wouldn't venture a guess.

Most of the encoder cards have 2 encoder supply voltages selected by the terminal you connect to 5V and 12V
Some encoders are resolver output but that takes a different type of input card
Gary
 
Encoder Issue

Encoders can be touchy. The drive output,input, and control leads should be in separate ferrous based conduit. The capacitive coupling is dependent on the carrier frequency setting in the VFD . The higher the carrier frequency , the higher the capacitive coupling. Mutual inductance can be occurring if the conduit is not ferrous based and properly grounded/ bonded. The larger the motor load and or current draw the larger the potential mutual inductance. a Fluke 123 will do wonders for troubleshooting your problem. There are "VFD" cables that have a shield and a ground conductor , and should be grounded at the drive and motor ends. However , Controls with standard 4 or 6 conductor wires should be shielded and grounded at the VFD end ONLY. Most manufactures provide a "screen" or ground terminal for the control inputs, and or options. If you are checking the supply voltage with a lower end multi-meter it may not have a fast enough acquisition time to catch an transient. VFD's are sensitive to a half cycle event. a complete sign wave is 16.4 ms, half cycle being 8.2 ms . Lower end meters have a peak , or min/max function limit of greater than 100 ms. Your drive can be dropping the power supply to the encoder , or a transient can be overpowering the pulses / index pulses and the lower end meters will not catch it. Fluke 87 IV / 189 /187 /287 / 289.. Agilent / Keysight --> U1272A / U1271A / U1253A / U1273AX are all capable of capturing these types of events. Even better would be a Fluke 123 / 124 / 97 /98 /99 /105 scopemeters. The 120 Series has trend Plotting that makes it easier to catch. Encoders are also sensitive to vibration. If your machine has a bearing or track that has become extremely worn or defective, it could produce enough vibration to vibrate the clear disc inside of the encoder and when the reflective marks goes by, it will no be smooth and causing a pulse to deviate in duration or get skipped. The higher the PPR of the Encoder, the more susceptible it is to vibration. There also could be debris in the encoder if the case was compromised. Hope this give you a good place to start.
 
Morning.
So to add to this story....
After much head scratching, talks to Rockwell. And many communications(emails and remote online support) between myself and the OEM they finally came to look at this F91 issue.
Before they got here I did figure out that by setting P535 to quad only and not quadracheck that I could actually run without the fault. We ran an entire day of production in this state with no issues. The OEM said this was a bandaid and we could not continue to run this way. Why I am not sure?
So when the OEM came we went over everything that I did. Then we used an oscilloscope to see if there was any noise and to make sure our pulses were looking good. The general consensus from the OEM and their suppliers was that I had a severe noise issue. Just as others here had mentioned. The scope revealed all was well. The OEM got me to put the quadrature back on the drive and after about 12 cycles of the machine the F91 returned.
After I left that day he was able to get this working with quadracheck enabled. He said he had to tune the drive with the encoder. I was not able to meet with him to discuss what he did and I am a little anxious to find out, and I will once I can get in touch with him, but for now this is all I can tell you about what he did to fix the problem.
I will post his explanation when I receive it.

Thanks
 
I got an explanation from the OEM tech on what he did to "fix" the problem with my F91.

"I did a static tune on the motor. This tunes the speed loop. The I set up the auto run, that new but I added, then I played with parameters 565 Pos Reg Filter and 566 Pos Reg Gain until the fault went away. "

To bad the fault returned after about 12 hours of running. I had to put the drive p535 back in to quadrature in order to run fault free.

Still no resolve.
 

Similar Topics

Hi everyone, This is my first time posting, so please forgive any omissions or mistakes. I am attempting to control the velocity of a stepper...
Replies
18
Views
741
Dears, i am trying to change the series of encoder from A to B, but the program do not has this option (Rslogix5000, 20.06 the old encoder was...
Replies
2
Views
137
Hi all, I am working with an incremental encoder (ABZ signals, 360 ppr (so 1440 counts per rev)) to replace the existing "manual" encoder wheel I...
Replies
25
Views
806
Hi all, I am implementing an incremental encoder sensor (ABZ) to replace the existing "manual" encoder wheel I have in my device. This is a 360...
Replies
0
Views
108
Hi Friends, I am facing a problem with encoder adjustment in Sew Electric Cylinder CMS71L. encoder using is Sick Hiperface Drive is MOVIDRIVE...
Replies
16
Views
628
Back
Top Bottom