Hi Everyone,

I am working with incremental encoder and presently am facing a problem with my incremental encoder. Attached are the output signals I measured using oscilloscope. I have also attached the circuit used to measure the signals.
As indicated in the output signals (in red) edges in both the phases coincide( with maximum zoom). I am using a mechanical encoder with 12 pulses per rotation with detent positions. I tested 3 encoders, all are behaving the same.
I am not sure why I am getting such a output, I was expecting 2 signals with quadrature phase.
Has anyone faced such a problem with encoder ?




Thanks in advance
Pradeep

 

Defective encoder.

 

If they are true quadrature encoders then there is something wrong, either with the nature of the encoder design or the method of monitoring?
Are you obtaining constant rpm of the encoder when monitoring? as you mention a detent.
Max.

 

Defective encoder.

I would have considered it defective encoder.. but I tested 3 samples all with same output. :-(

 

If they are true quadrature encoders then there is something wrong, either with the nature of the encoder design or the method of monitoring?
Are you obtaining constant rpm of the encoder when monitoring? as you mention a detent.
Max.

Max,
Encoder type : It is mechanical type, i opened up one and it has metallic strips inside with the metallic contact sliding across when rotated.
RPM is depending on how fast/slow I rotate, it is not constant..
I am not sure if there can be any problem with the monitoring method.

Pradeep.

 

If it is mechanical type, contact closure, there is probably going to be error, especially with any high rate of RPM.
What is the application?
Max.

 

Make sure that you are using sync on only one channel of your scope. If set for ALT sync, the sweep will sync on each waveform.

 

If it is mechanical type, contact closure, there is probably going to be error, especially with any high rate of RPM.
What is the application?
Max.

I think I would not need high RPM rates, application is for human interface something similar to volume control.
So max RPM would be the maximum speed at which any person can rotate it and I don't mind if I miss some pulse counts due to high speed of rotation.

Pradeep

 

Make sure that you are using sync on only one channel of your scope. If set for ALT sync, the sweep will sync on each waveform.

Please could you more details about ALT sync.
I don't have any such setting on my oscilloscope or may be it is named differently on my oscilloscope.
Pradeep.

 

The type of sync may have different names on different scopes. The one I'm talking about is to sync only on one channel that you select. My scope has selective sync between channels, ie. 1, 2 or both. Not being familiar with your scope, it is hard to know what it is called.

If you have an external sync, use a third probe and connect it to either the A or the B output of the encoder.

 

How about adding channel A and Channel B. This should show the contribution of each channel, and since A and B should never change at the same time, you should always see four distinct levels in the sum...

 

How about adding channel A and Channel B. This should show the contribution of each channel, and since A and B should never change at the same time, you should always see four distinct levels in the sum...

That would be three distinct levels, no?

 

If you set the vertical scale factors to different values, you could see four.

 

The fact it has indents indicates it is meant to be used no higher than can be achieved turning by hand, not high rpm.
Just place a 360° protractor scale on the shaft and mark off the rise and fall edges indicated by a ohm meter.
Max.

 

That would be three distinct levels, no?

Yep!

If you set the vertical scale factors to different values, you could see four.

and Yep!

 

I cross checked the channels in the oscilloscope and it doesn't seems to be the problem.
The encoder is generating such signals. Need to figure out the problem.

 

Can you give a link to the encoder's datasheet?

John

 

http://www.alps.com/prod/info/E/HTML/Encoder/Incremental/EC10E/EC10E1260502.html

Pradeep

 

Still looks like bad manufacturing, and you have units that do not meet their spec...

The output bit pattern is supposed to be a Gray code, therefore under no circumstances should two outputs be changing simultaneously...

 

I have also sent a query to them. Let's see if we get an explanation about the behaviour or the encoders are faulty.