Cross Slide Rotary Encoder

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Speedy Builder5 on 26/11/2022 10:30:45:

[…]

I had a quick scout around to see what the angular error might be, but didn't find anything on the net.

.

The joint is not exactly the same, but for a reasonable approximation have a look at the velocity analysis for a Hooke joint … Loads of examples available, in writing or video, because it’s classic text-book stuff.

If you read between the lines, it’s obvious [because time remains constant] that the error is angular.

MichaelG.

[blowlampParticipant @blowlamp]

If a coupling ever does gives up the ghost, then you'd probably get away with a piece of decent rubber/plastic hose, for the amount of torque needed to move the encoder.

Martin.

[blowlampParticipant @blowlamp]

Posted by Michael Gilligan on 26/11/2022 12:25:00:

Posted by Speedy Builder5 on 26/11/2022 10:30:45:

[…]

I had a quick scout around to see what the angular error might be, but didn't find anything on the net.

.

… but for a reasonable approximation have a look at the velocity analysis for a Hooke joint … Loads of examples available, in writing or video, because it’s classic text-book stuff…

MichaelG.

I don't think the two couplings are comparable with regard to their angular velocity characteristics. The in-line couplings are much more of a constant velocity arrangement than a Hooke's joint, The in-line drive coupling basically being a torsionally stiff coil spring.

Martin..

[Michael GilliganParticipant @michaelgilligan61133]

Perhaps we could have a forum-poll on that, Martin

… I remain convinced that it’s a ‘reasonable approximation’

Let’s see if anyone other than we two is prepared to opine

MichaelG.

[mgnbukParticipant @mgnbuk]

At a previous employment, we used Huco couplings for connecting encoders to servomotors & ballscrews on CNC machines – more reliable than the type shown, which always broke due to metal fatigue unless the shafts were in very close alignment .

Very low backlash at the low torques required for encoders at "normal" line counts, I only had issues on a high line count direct mount rotary table encoder, which showed in the reqion of 0.009 degrees wind-up / backlash – not good on a 0.001 degree resolution RT & cured by fitting a special Heidenhain coupling that cost 10 x as much as the Huco.

Nigel B.

[John HaineParticipant @johnhaine32865]

I was going to mention here that years back one could buy a little board from the US with a 400 steps/rev stepper motor that would act as an encoder – at the time a cheaper solution than a proper encoder. Now I see that you can get these optical encoders from Aliexpress at about £15 delivered so no need!

**LINK**

Or of course use one of the nice magnetic ones that Joe Noci recently described here.

[Speedy Builder5Participant @speedybuilder5]

Michael, I did buy a pair of Hooke joints, but there was so much backlash in them that it wasn't worth installing them. Hooke joints have poor angular velocity variations. Its interesting that the MINI used to have 1 x Hooke joint and 1 x CV joint – I wonder how that worked out !

I did have 1 x beam joint of a far superior quality, but I needed 2! The ones I have are several magnitudes stiffer !

This looks like the Rolls Royce Beam couplings

https://www.ruland.com/technical-resources/application-spotlights/encoder-applications/beam-couplings-for-encoder-feedback-systems

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Speedy Builder5 on 26/11/2022 15:45:29:

[…]

Hooke joints have poor angular velocity variations.

[…]

.

and as previously stated, I believe that the couplings that you have used will be seen to have what is essentially the same geometry. [backlash being a separate issue, down to quality of manufacture]

… Martin disagrees with my ‘analysis’

MichaelG.

.

Edit: __ I may have to admit defeat !

These little beauties specifically claim to be Constant Velocity:

https://www.abssac.co.uk/uploads/docs/gdxfl6e8_Make_your_part_flexible_Catalogue.pdf

See the text on page_5

Edited By Michael Gilligan on 26/11/2022 16:13:13

[Ian PParticipant @ianp]

I worked for a company (45+ years ago) who were the UK agents for some USA manufactured welded bellows. We used their bellows shaft coupling on one of our products but changed to Abssac helical beam for cost reasons. We never had any failures and I have every confidence in their products however this quote is from the Abssac datasheet Michael linked to (on page 5) which I do not fully understand.

'Torsional variations, which can induce differences in hub-to-hub velocity when subjected to dynamic loading, are minimal in steady-state applications of the HELICAL Flexure'

It mentions torsional variations but then refers to steady state. If its steady state then where do the dynamic loadings come from?

Just curious

Ian P

[Ian PParticipant @ianp]

I worked for a company (45+ years ago) who were the UK agents for some USA manufactured welded bellows. We used their bellows shaft coupling on one of our products but changed to Abssac helical beam for cost reasons. We never had any failures and I have every confidence in their products however this quote is from the Abssac datasheet Michael linked to (on page 5) which I do not fully understand.

'Torsional variations, which can induce differences in hub-to-hub velocity when subjected to dynamic loading, are minimal in steady-state applications of the HELICAL Flexure'

It mentions torsional variations but then refers to steady state. If its steady state then where do the dynamic loadings come from?

Just curious

Ian P

[Ian PParticipant @ianp]

My first (I think) double post on here.

I did nothing different to what I normally do!

Ian P

[blowlampParticipant @blowlamp]

Posted by Speedy Builder5 on 26/11/2022 15:45:29:

Michael, I did buy a pair of Hooke joints, but there was so much backlash in them that it wasn't worth installing them. Hooke joints have poor angular velocity variations. Its interesting that the MINI used to have 1 x Hooke joint and 1 x CV joint – I wonder how that worked out !

I did have 1 x beam joint of a far superior quality, but I needed 2! The ones I have are several magnitudes stiffer !

This looks like the Rolls Royce Beam couplings

https://www.ruland.com/technical-resources/application-spotlights/encoder-applications/beam-couplings-for-encoder-feedback-systems

Bob.

It was a (very stiff) rubber/metal bonded kind of U/J that was located next to the gearbox and because of that, it didn't need to flex very far with suspension & steering movement, when compared to the C/V joint, which was mounted on the streered wheel hub.

Martin.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ian P on 26/11/2022 16:43:50:

[…]

It mentions torsional variations but then refers to steady state. If its steady state then where do the dynamic loadings come from?

Just curious

Ian P

.

I think they are admitting that the spiral will wind-up [or unwind] with dynamic changes … and their ‘steady state’ reference is to the shaft running at constant speed.

MichaelG.

[Author]

Posts