SSI encoder problem

Hello to all

I have a problem with encoder from bei sensors.

encoder is dual with ttl and ssi channels.

When i turn off and turn on my system it gives me different position than it was before power off.

We have custom control unit that evaluate ssi and ttl signals. We tried different ssi modul from Lenze and same thing happens.

Encoder is on motor with electomagnetic brakes. Manufacturer told us that magnetics is the problem and sent differet type of encoder but same thing happens.

Cable is twisted pairs and shileded. Length is 15m max.

Everything works normal bud after power off and on we haave different position value and if drive went positive before power off new value is smaller than what was before power off

Exmple:

Drive moves from 10.456 to 12.665, off/on, new position is 10.354

Drive moves from 13.445 to 11.344, off/on, new position is 13.221

Any suggestions?

Thanks in advance

What are the position units you are quoting - motor Revs or application units
If application units convert to motor revolutions and try again

Couple of things to check to assist what is going on

Power on the drive, lift the brake, do NOT turn the motor, Power Off / On
Monitor the motor position the whole time - preferably trend it

Does the position change every time - If so then something is causing the motor position to change {eg brakes, surge from the drive power on, Drive performing a motor resistance check and the wires too close}

If you are using application units then is the position change a multiple of the motor revs to position units?

We use aplication units [meters]. Position is changed only when we move and then off/on.

If its not moved we get same position after power off/on. Even when i activate brakes and not moving it stay same.

I would be curious to see what happens if you return to the same spot then cycle power.

i.e. Move from 10.234 to 13.456, stop, move back from 13.456 to 10.234, off/on.

This could tell you if the problem is related to the distance being moved or if it is related to running the motor.

This is also a good time to check to ensure that the encoder is not mechanically slipping by making sure it does, in fact, return to the same point.

nikoldije said:

...
We have custom control unit that evaluate ssi and ttl signals...

Click to expand...

Problem is with the controller software. It's not storing the last position before power off.
You say it won't do this if you don't move before power off. What do you mean?
What do you do different after making a move or not making a move?

Edit:
Okay, I see it here: "Everything works normal bud after power off and on we have different position value and if drive went positive before power off new value is smaller than what was before power off".
Is there play in the mechanics?

I did not have access to machine for few days I hope i will be able to play with it tomorrow.

We also thought that the problem is with software but then we put a basic ssi modul from lenze and connect encoder to it and same happens.

I am looking to get encoder from different manufacturer and try. The system and hardware that we implemented works 100% because it was implemented before but with different encoder (different manufacturer)

possibly your encoder is a singleturn absolute, or a multi-turn and you are "rolling over" from absolute max to the absolute min capability of the encoder. On power up, the majority of encoders out there report their absolute position. Most controllers out there just look for the absolute position from the encoder at startup, then keep tract of the encoder counts "incrementally" from there. If it is a single turn absolute encoder, then each time you power up you should always see a low value. If it's multi turn absolute, then there are only so many turns the encoder can report as a positive value before it "rolls over" and reports its absolute min value. Again, if the control only used the absolute value on startup, then it keeps tract so this is not a problem during running, only an issue when you power up. Many encoders can only report up to 4096 turns, which is really -2048 to +2048 turns, plus it's position within 1 turn. Its not the encoder "lying" to you. It's telling you where it is when it's powered up.

to rule out noise or electrical connection issues, can you verify that when you move from 10.456 to 12.665 that it is moving accurately? It is possible that the TTL encoder signals are noisy and counts are lost/gained. You should be able to do a repeatability test to see if returns to the same position every time. If not, then you are losing/gaining counts. If you move from 10.456 to 12.665 and the axis movement is long, then you are losing counts likely due to TTL levels which could be power supply problem, 0V of power supply not common to controller interface, poor connection. If your move comes up short then you are gaining false counts, which is likely noise on the TTL siganls. Really need an o-scope to see them. I worked for a linear encoder company for a very long time and saw errors like this daily. I'd be very interested to hear your results.

We did some testing yesterday. It appears that we have problem with magnets from electromagnetic brakes!

Our encoder is mounted on motor that has two electromagnetic brakes and is just above them.

We moved encoder further away from brakes and problem is gone. I am very happy but it is strange that it works fine until power off/on when it makes mistake.

It is common to put encoder on that spot but it seems that not all encoders are immune to magnets.

Having a quick look at BEI's products, it appears they have models of both optical and magnetic encoders. I'm guessing you have the latter.

The catch is that optical also uses magnetic reading on power up (at least that is how they explained)