S4 and MAE3 encoders..,. What?

[matthewdenny]

I've been looking at the S4 and the MAE3 encoders from US digital and I have no idea how that thing should be placed in the robot. How does it physically connect to anything? Where do you put them? Am I being dumb?

[Joe G.]

Originally Posted by matthewdenny

I've been looking at the S4 and the MAE3 encoders from US digital and I have no idea how that thing should be placed in the robot. How does it physically connect to anything? Where do you put them? Am I being dumb?

The S4 is panel mounted, meaning that you insert the threaded boss through a hole, and secure with the included nut. From there, you can link the shaft to your robot through any number of means, including a shaft collar, surgical tubing link, or simply pressing or gluing it in a hole in your shaft. If you've been using the E4P encoders before, you won't be able to directly drop in an S4, but rigging something up shouldn't be too tricky.

[matthewdenny]

so does it mount such that the end of the encoder is in line with the axle, or do you connect it with a belt?

What is a surgical tubing link?

Does anyone have pictures of these on their 'Bots?

[mman1506]

Originally Posted by matthewdenny

so does it mount such that the end of the encoder is in line with the axle, or do you connect it with a belt?

What is a surgical tubing link?

Does anyone have pictures of these on their 'Bots?

[billbo911]

Originally Posted by matthewdenny

so does it mount such that the end of the encoder is in line with the axle, or do you connect it with a belt?

Simple answer, yes! Either way works. It will depend on your system's design. It could be directly in line, or parallel to the shaft. Belt linkages work as well as chain. The are things to consider with either approach. Putting too much tension on the encoder's shaft could be a problem and damage it. Not enough tension could cause slop and produce inaccurate measurements.

Originally Posted by matthewdenny

What is a surgical tubing link?

Does anyone have pictures of these on their 'Bots?

Sorry I don't have an actual picture, but with a bit of imagination you can see where a piece of surgical tubing stretched over both shafts here would make a fairly decent coupling. It would allow for a small amount of misalignment between the shafts as well.


The encoder in this picture has ball bearing shaft supports and practically zero rotational friction.

[matthewdenny]

In the second picture how is the encoder connected to the shaft?

Also doesn't the surgical tubing cause cause the encoder to not reliably show the angle of the shaft?

[asid61]

Originally Posted by matthewdenny

In the second picture how is the encoder connected to the shaft?

Also doesn't the surgical tubing cause cause the encoder to not reliably show the angle of the shaft?

Generally you use the surgical tubing method for incremental encoders unless I am mistaken, so the error is minimal.
For the MA3, you stick it into a 3/8" hole and screw the nut onto teh threads to keep it in place. Then you stick into a hole at the end of a shaft and set screw it tight.

[Ether]

Originally Posted by asid61

Generally you use the surgical tubing method for incremental encoders unless I am mistaken, so the error is minimal.

Why is the error minimal with incremental encoders but not absolute ones?

[Joe G.]

Originally Posted by matthewdenny

In the second picture how is the encoder connected to the shaft?

Also doesn't the surgical tubing cause cause the encoder to not reliably show the angle of the shaft?

There's practically zero resistance from the encoder, so the surgical tubing doesn't twist any appreciable amount.

[asid61]

Originally Posted by Ether

Why is the error minimal with incremental encoders but not absolute ones?

Becuase in the drivetrains above, the encoder is spinning at a rapid speed.
For the purposes of a swerve drive or arm, an absolute encoder would be spinning very slowly by comparison. Would that be correct?

[Ether]

Originally Posted by asid61

an absolute encoder would be spinning very slowly by comparison.

1) There's no reason why an incremental encoder cannot be used in a slow-spinning application.

2) Properly designed and implemented, surgical-tubing shaft couplers can successfully be used in some high-speed applications.

[asid61]

Originally Posted by Ether

1) There's no reason why an incremental encoder cannot be used in a slow-spinning application.

2) Properly designed and implemented, surgical-tubing shaft couplers can successfully be used in some high-speed applications.

True. I'm sorry, I phrased my statement wrong, because I am used to fast-spinning incrementals on drivetrains.

I totally agree with #2. I'm saying that's a better application for it, although apparently that's not the case due to the negligible resistance of the encoder.