ABSOLUTE ENCODER TESTING

[spicychic]

Hi
I was just wondering whether anyone has developed any successful test procedures to test an absolute rotary encoder. I am looking for something basic, just to ensure that the configuration is correct.
Thanks,
Monica

 

[IRstuff]

You stick it on something that turns and compare its outputs to something that you know works or you use physical landmarks to check the performance.

TTFN

 

[spicychic]

um. . .that's not very helpful, can you be a little more specific please?

 

[IRstuff]

Are you having problems reading out the encoder?

In the simplest case, you could simply get a protractor and find the absolute zero of the encoder and then compare the rotation angle to the readout.

If you're concerned about accuracy, you'll need either a higher resolution, more accurate encoder to compare against or a theodolite to measure the angle change.

Any developed test procedures are specific to the encoder and the manufacturer. Very few users test the encoders stand-alone.

TTFN

 

[skills]

What exactly is it that you want to test of this encoder? Is is speed related or merely that it lacks missing codes or what. Set encoder to its zero and turn it four and a half times and back calculate distance/rev and see if your data matches up. If you provide more specifics on exactly what it is you have in your setup, we'll be able to answer your question in more detail.

 

[spicychic]

I need to make sure that it is all configured properly:
„h That an expected change is achieved
„h That the SSI interface has the ability to communicate and transmit data to the Wincon V
„h That the specified number of turns (4096) and the number of resolutions per turn (4096) is correct
„h That the specified supply voltage (11 - 27V) is sufficient

 

[skills]

I know this sounds trite; but I don't mean it that way. If you want to know if it has the right count/rev then why do you not simply revolve it one time and check the output? If it performs as indicated, then it is configured correctly, if not, then clues will be available. Perhaps, if you give us the encoder manufacturer and part number, a specific test can be described; but, the best way to test something is in the 'use' mode and compare result to specifications.

 

[Thorn3]

Hello Spicychic,
I won't ask why you chose that name..LOL. What I would do to test any rotary encoder, is to have it attached to a machined disk. On the outer edge of the disk, have a tongue of given dimensions. The dimensions really don't matter, as they will be subtracted from the total circumfrence of the disk. When the disk is being rotated, the tongue can come into contact with a premounted dial indicator, such as those used on a lathe. Run the disk CW and CCW for many cycles. Note the over or under run on the dial indicator with each rotation of the disk. This will tell you without a doubt, if the encoder is reading properly. It will also allow you an easy way to determine to what tolerance you wish the encoder to read. This kind of test is most certainly application free. Who cares what it is going to be measuring? The needed data is, does it count X amount of times, per degree of rotation and does it do it the same every time? This setup will tell you that.