connect scope to encoder for troubleshooting ?

milmat1

Member
Join Date
Aug 2005
Location
North Carolina
Posts
209
Trying to troubleshoot a new installation using a AB PF525 vfd with 25-enc-1 encoder card. Encoder is a DYNAPAR H20 series, This is one of several axis' on this machine, others are working correctly.

counts are acting strange when viewed in RS LOGIX 5000. In fwd direction it seems to gain counts, in reverse it seems to loose counts. And never returns to 0 when back at home position. Voltage on "A" and "B" are at 4.75vdc. (Opposite polarity ). Encoder is powered with 24vdc.

I want to look at the signals with a scope, I have a 2 channel old B&K CRT scope. Im not sure how to connect the probes ?
should the probe be connected to "A" and "A _" for ex. Or to "A" and ground ?
and of course the same for "B" channel....
the cabling is an igus shielded twisted pairs cable and the A and B channels are on pairs.
shield grounded at the vfd and cut / stored on encoder end....as usual...
 
how many pins is the encoder?
https://www.dynapar.com/uploadedFiles/_Site_Root/Products_and_Solutions/H20_701879-0002.pdf

there is a data sheet. You are going to want to connect your "common" probe from your encoder to the common pin and then put your signal probe on to the signal wire. Spin the encoder and zoom in on the peaks. They should be nice and clean.

In my experience, when you are looking at an encoder count and it seems to jump or go all over the place, there has been an issue with the encoder.

Make sure the reset switch (in the case of our machines and proximity switch and bolt or something like such on the shaft to trigger a reset every full rotation) is properly lined up.

Check your encoder and make sure the electrical connections are clean and tight.

Check whatever form of lovejoy or coupling you have and make sure that they are all tight and lined up.

Replace the encoder.

We don't have an oscope for troubleshooting, but most of the time it has just been a loose coupling or the encoder mounting bolts have came loose.
 
Last edited:
Thank You,
It is a 10 pin encoder. However we are connecting to an encoder card (25-ENC-1) in a AB PF525 VFD.
we are using external power from 24VDC PWR Supply. And the signal connections to the VFD/card, A, A_, B, B_, The card is configured as quadrature. The programmer is trying to tell me to switch the encoder to a different voltage model. (12V rather than 5V). But my card has a dip switch setting for 5V so that's what i'm using.
I am going to scope these signals today and see what they look like...

I'll report back what I find...

THANKS !!
 
Actual specs for the encoder(s)
Dynapar Part No. HS20 2048 834A7
2048 PPR Differential Bi-Directional with Index (A,A_,B,B_, Z, Z_,)
5 - 26 VDC Input 5VDC Line Driver Output...
Max RPM = 6000
2 ch Quadrature operation "A" leads "B" for CCW rotation.
Phasing 90 degs. +/- 22.5*
Square Wave with rise and fall of 1 micro-S (@1000pf load)
Electrically isolated,
Noise Immunity to EN50082-2
 
Per the Dynapar Encoder Installation Manual :
https://www.dynapar.com/uploadedFiles/ST67_Man_703048_1_.pdf


if you have problems:

help is at your fingertips! Simply pick up
the phone and dial our Applications Engineering
Department at 1-800-234-8731 (US & Canada) or
847-662-2666 from 8:00 AM to 4:45 PM (Central
time) Monday - Friday. One of our engineers will

gladly help you solve the problem.

 
Per the Dynapar Encoder Installation Manual :
https://www.dynapar.com/uploadedFiles/ST67_Man_703048_1_.pdf


if you have problems:


Thanks !
I have sent emails to tech support with no answer, I'll give them a call in the morning.

At least with this antique scope I have here (B&K 15MHZ) The encoder signal looks awful. I cannot even get it to see a square wave, it's just fuzzy and like each cycle runs into the next,.... Very Strange !
 
1) You are using DC coupling for the scope, right?
2) Set the scope for differential mode, and look at one channel at a time (One probe to A, one to !A).
3) Look here.
4) How is the encoder coupled? Hollowshaft !

5) - This isn't cheap, but does a fantastic job for testing encoders... http://www.dynapar.com/Products_and_Solutions/RIM_M100_Encoder_Tester/

Yes on DC Coupling,
Diff mode,
I have only one probe so I can look at one signal at a time.
? When looking at "A" Should the probe ground be connected to Ground?

Something strange I just found:
When I set my Fluke 87 to Frequency (Hertz) And I place the probes between "A" and "A_" I read 60 hz EXACTLY.
Doesn't this indicate I have a 60Hz signal infiltrating the encoder line ?

That would explain the "Fuzzy" unstable traces..
 
Last edited:
Changing the encoders power source seems to have helped the problem.
The Power was from a dedicated SOLA SDN10 24vdc pwr supply isolated from everything except the encoders, 2 Encoders we actually had working that way. But even they seemed to loose counts but being very short moves it wasn't an issue.
This Axis moves several feet so it really showed up badly...
Now I am getting 12VDC directly from the Encoder Card/VFD And the Encoder seems to like it. (Encoder will operate with a supply of 5-26VDC)...My electricians are doing some more testing right now but I think we will be good....
 
As a general rule of thumb, and this is in my experience, you don't necessarily want to mix DC power from different supplies. There is a reason most cards / vfds/ electronics / etc will have an output source to feed back to the input. It is to avoid issues like the one you are having. With AC distribution, generally one area of the plant is all going to be fed from one or two main transformers that are in phase with each other. This isn't necessarily the case with the electronics in power supplies.
 
It appears this issue has risen again. Except now the machine is in the customers facility in OK. So I am traveling there next week to take a look at the system again.
Something I have discovered recently is that during the install none of the 15 + VFD cables are properly grounded at the motor end. I used Belden 295___ Cables and the installers did not connect the shield at the motor end. I feel like that is likely trashing the entire ground system with PWM harmonics...
 
Often the VFD has a +24 vdc supply you could use for the encoder power.
The 5 volt signals are referenced to the 0 volts (gnd) of the card and VFD.
The A and A_ are inverse of each other as are the B and B_ signals.
The A and B over lap by 50% (the A is 50% through its high when the B transitions).
The A and B are sufficient by themselves, but the not signals help to suppress interference in the signals.
If these lose position the problem can be loose coupling in the encoders etc giving false counts where the encoder edge triggers and the system does not clean it up in logic.
That would be several false up counts etc with no corresponding down counts int some mechanical vibrations. Using the Z or turn signal helps keep things straight.
 
This same issue has gone on for the life of this machine. Chasing noise / positioning errors etc. Once again it is presenting severe inaccuracy. The onsite tech has moved some grounds around, reconnected some shield grounds etc. And gotten it to work "Fair" but still not 100%.

When I visited there over a year ago the encoders were completely trashed with noise. So much so that I seriously thought my scope was defective. Because I could not even see the encoder signals.
I had them install a ground rod at the main panel and bond several parts of the machine, trying to keep all parts of the machine at the same potential. "0"...

Just for illustration I'll include the picture of the final result, the scope image here is the best we could get. And it was working well. But as you can see it's still Really Horrible !..
Before we could not even distinguish the signals...Even with a separate cable ran across the floor, it was not much better..
The PF525 drive / 25-ENC-1 closed loop drive setup, just isn't working well at all..

..
20160127_140911.png
 

Similar Topics

Hello, friends, I am trying to upgrade a system that uses an Onrom incremental encoder (E6B2-CWZ6C) connected to a Danfoss VFD (FC360), but now...
Replies
4
Views
245
I will find cable series but I don't know for fs260 model, kindly suggestions and help to me for connect cable
Replies
1
Views
95
Hi all, I'm having difficulties trying to connect FactoryTalk View SE Local Station (V13.00) to MS SQL Server Express. I state that they are...
Replies
2
Views
109
I am using Mitsubishi PLC :FX3g and downloaded the program in it. When I connect Mitsubihi FX2N-4DA expansion module, it lost its communication...
Replies
3
Views
242
Hi, The VersaMax PLC suddenly stop running (no LED on run function) so I'm trying to upload the program. The HMI connected to it says "Host...
Replies
0
Views
117
Back
Top Bottom