IONI SinCos encoder index channel

Questions
1

I have a question regarding IONI HC in IONICUBE regarding connecting a Hengstler SinCos encoder.

My hardware configuration:

I’m still making electrical connections, so I haven’t got Granity connected yet and don’t have firmware revs, etc.

Problems arise when:
I try to look up how to connect the index channel for the encoder.

How it behaves:
The Wiki (https://granitedevices.com/wiki/IONI_%26_IONICUBE_user_guide/Wiring_overview#X6_and_X8_pinout) shows both pins 14 and 15 as Z+ in SinCos mode. Is this a documentation error?
Furthermore, I can’t find any documentation of the expected electrical specification of the signal(s) in this mode.

How I would expect it to behave:
I’d expect one to be Z+ and the other to be Z- (as they are in other modes).
The A and B channels of the encoder are 1Vp-p centred on 2.5V (2048cycles/rev).
The Z channel of the encoder is also 1Vp-p centred on 2.5V, with a single peak per rev.
I need to know: Will the drive accept these signals?
My theory is that the A & B channels will be OK for direct connection. Is that correct?
I’m most concerned about the operation of the Z channel. What is the electrical spec for the Z channel?

Many thanks for your excellent products,

Leo

2

I would guess that it is a typo on the wiki page. I’ll let our hw engineer to verify it. From what I know of the software, index point signal is handled the same way in both quadrature and sincos encoder modes.

3

Thank you.

So the index channel remains digital in SinCos, as it is in Quadrature.

I’m pretty sure that means that I’ll need a bit of external logic to get the index pulse working (a comparator would probably be enough).

Is there any constraint on where the index pulse edge(s) need to be relative to the A&B sinusoidal channels?

Cheers,

Leo

4

At least the Sincos encoder datasheet says that " The zero signal appears once per revolution and reaches its maximum value at the angle where the amplitudes of A and B Signals are equal." and that the signal is differential, so it should be good to connect to the drive.

5

However, it’s an analog signal 2V < z < 3V (I scoped it), not a 5V digital signal.

It also seems to have a few funky small peaks before and after the portion where the differential pair actually cross.

Would you expect that to be OK for direct connection?

I was planning to run it though an MPC6542 (https://docs.rs-online.com/cd29/0900766b813814c5.pdf) to convert it to a clean digital signal.

6

I’m not sure how the servo drive would interpret it.

Does your application really need index point signal?

7

I think the index signal is nice-to-have rather than drop-dead-without-it.

I think I’ll see how it goes and use the index if it’s reasonably easy to get working.

Thank you for your help. I’ll report back if I discover anything surprising.