placing rotary encoder in turntable

[ilirvg]

Hi all,

I have a motorised turntable, which is driven by a synchronised AC Motor, and I need to get angular position of the rotation. Because it was hard to hack electronic structure of the table I have decided to use rotary encoder.
But now I am facing another problem as you can see on the photo I cannot put my rotary encoder in the middle of turning plate because I have a screw there(fig 1). I was thinking to put the rotary encoder in the side of the plate (fig 2) and to wear both the encoder and the plate with kid of a rubber or something else that would make the friction higher. Do you guys think that this would work if not dose anyone have another idea how to do it.

And if this is not the right place to ask this question where can I?

Best,
Ilir
Fig 1.

Fig 2.

 

[drawoh]

ilirvg,

We cannot see your pictures, or at least, I can't.

Is your encoder incremental or absolute? If it is incremental, how are you getting your zero position, and how accurate do you need it?

An encoder acting on the side of your turntable is almost certainly not being driven at a 1:1[ ]ratio. There is not much point running an absolute encoder at anything other than[ ]1:1. An incremental encoder's zero pulse only works at [ ]1:1.

You can attach an optical switch or a proximity sensor somewhere on your turntable. Probably, this is not extremely accurate.

--
JHG

 

[GregLocock]

Why not put a black and white sticky tape around the outside of your turntable and use and optical encoder? Color one of the white marks black for zero.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[MikeHalloran]

Friction drive of an encoder will not be repeatable.

Attaching a cross-striped tape to a diameter near the outside and using a retroreflective sensor as an incremental encoder will sort of work, except that you will get a long or a short bar where the ends come together. I was peripherally involved in a project where we used a multi-row pierced stainless strip wrapped around a ~12" diameter turntable as an absolute encoder, with a 8-row transmissive sensor, and the holes comprising an absolute encoder. It worked okay, once we were able to hold the diameter of the faying surface to 'tenths', to make the angles come out right and the ends to meet properly.

I.e., a sticky tape's angular resolution will be affected by the diameter to which it is adhered, the thickness of the tape, and the thickness of the adhesive. If you are trying for very fine angular resolution, you may need a lookup table somewhere to compensate.

If you could provide some photos or a model number, we can surely help more.

It would also help to know what angular resolution and accuracy you are trying to achieve, and what you are intending to do with the measured angle, e.g. feed it into a motion control to drive the table, or just provide an angle readout, or whatever.





Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

A better approach would be to use random dots on the tape and then indulge in signal processing.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[ilirvg]

Hi guys thank you for you're responses, and sorry about the photos.
I really like the idea of using sticky tape and I will have clockwise and counter clockwise rotation. So if I go this idea in what exactly can I print the white and black strips, and can I use LED as transmitter and receiver.

Thank you very much.

here are the pictures that I send in the beginning

 

[asimpson]

You could wrap magnetic encoder tape around table and use pickup like below:

https://www.google.ie/search?q=line...wBA&ved=0CAYQ_AUoAQ&biw=1258&bih=642#imgdii=_

Should get resolution of 0.01 mm on circumference with low end brand.

Pickup should have index channel to initialise every revolution

 

[MikeHalloran]

Can you trade it for a table with the encoder built in?
Homebrew will be more expensive when you are done.

It's possible to use an actual LED as a photosensor, but you need a lot of gain in the circuit because it's not supersensitive. It may also pick up random incandescent sources like your flashlight.

For reading a striped tape, Texas Instruments used to make a nice retroreflective scanner comprising an LED and a phototransistor in one package, prealigned so you didn't have to fuss much with optical issues. You still needed some analog circuitry to interface it.

I'm sure you can also find higher level black/white reflective sensors from, e.g. Banner, Balluff et.al., that interface direct to a PLC, if you don't like to solder.




Mike Halloran
Pembroke Pines, FL, USA

 

[dvd]

Curious what the motion is. Rotary indexers are cool in my book - a tribute to mechanical engineering. Today we can slop out a motion and let the electrical guys take over. Too bad mechanical motions are out of style.

 

[hendersdc]

As Mike mentioned it would help to know what you are trying to measure with the encoder output and to what accuracy.

Do you need to constantly monitor the angular position or do you really only care about a few specific orientations? If it is the latter you can get by with a few pins and photogates (depending on the number of specific positions) and save a lot of effort that you would have spent working with an encoder.

Also, I still cannot see your pictures, can you add them as attachments instead of embedding them?

Doug

 

[BrianE22]

Have you looked to see if you can access the shaft from the bottom? Also, you can get encoders with hefty enough bearings such that you can use a pulley and small belt.