05-23-2022 06:52 AM - edited 05-23-2022 06:55 AM
N-743, I went through that thread and still ask myself what you want to measure:
Amplitude of a single sine with offset or the angle (over time) of a shaft with a mounted encoder (giving sine&cosine and probably Z)?
Can you read both signals (sine & cosine) at the same time ?
Can or have you establish a differential measurement if your encoder provide differential outputs? (that helps agains noise 😉 )
If you want shaft/encoder angle over time and sample sine and cosine things get easy 🙂
Put it to the edge: For the offsets and sine and cosine amplitude errors/differences there is a nice method called Heydemann correction
P. L. M. Heydemann, Determination and correction of
quadrature fringe measurement errors in interferometers,
Applied Optics 20, 3382 (1981)
(should have some vis for that .. somewhere ...)
05-23-2022 10:42 AM - edited 05-23-2022 10:45 AM
Interesting abstract Henrik, can you link the full paper?
Unfortunately, this actually appears to be a mechanical allingment problem. Hence the approach to moving the signal into Polar coordinates.
There will be at several sources of error;
Imagine riding a bicycle with no springs, no rear tire, rusty chain, with broken sprockets on a rear wheel out of true and missing spokes over railroad ties.
You have to pay attention to the hardware!
05-27-2022 06:18 AM
Hi Jpb,
You have to pay attention to the hardware!
I agree with you. Only issue there is, in my situation, is difficulty in getting responsible person to admit it. There are issues with grounding, physical mounting etc. But I guess I am only supposed to call out if there is "spike/noise" or not.
I will mark your post as solution since there is not enough one can do without resolving hardware issues. For now I will just raise a flag if voltage values go out of bound.
Thanking all of you for your input on this.
05-27-2022 07:48 AM - edited 05-27-2022 07:50 AM
@N_743 wrote:
Hi Jpb,
You have to pay attention to the hardware!I agree with you. Only issue there is, in my situation, is difficulty in getting responsible person to admit it. There are issues with grounding, physical mounting etc. But I guess I am only supposed to call out if there is "spike/noise" or not.
I will mark your post as solution since there is not enough one can do without resolving hardware issues. For now I will just raise a flag if voltage values go out of bound.
Thanking all of you for your input on this.
Show them the Polar plot. If faced with empirical evidence they can try to explain its causes themselves. A responsible person would own up to the mistakes or oversights.
You are welcome.