Hello,
I have my CEM40 for just over a month.
The mount is good but sure has some character.

I already reported on this strange guiding behavior.
I thought I fixed this by removing the USB3 cable which I fed through the through the mount channel.
But no... it keeps happening.
A few night ago I had a pretty bad case of it.
Some nights it doesn't happen.
Last night I had just two single spikes within one hour.
This night, I started the guiding about one hour before actual shooting and it was bad from the very beginning.

What happens: SNR momentarily goes to zero and than back up.
Or it stays at zero and only momentarily goes up.
This makes the guiding plot look horrendous but actually the image doesn't shift and the resulting sub still looks good.

Any idea on what's going on?

From earlier today:

Hi Max,
Have you tried guiding with PhD2 as if the behaviour is replicated it suggests a mount issue although unlikely.
I have not had a problem since changing to one pulse dithering and keeping aggression down in RA.
I have noticed recently that star detection generally (in focusing, solving and guiding) is significantly adversely affected by binning. I have never noticed this before. I tried 2X binning with the ASI120mm and a pop up said that this was not supported. I am pretty sure that this has not been the case previously.
Luckily removing binning in solving and focusing modules has restored normal star detection behaviour.

Mike

No I didn't try PHD2. Want to use the internal guider only.
I use 1x1 binning for guiding.

I understand wanting to use the internal guider, but trying PHD2 would be a great experiment. If the problem persists with PHD2 that would point in the direction of a mechanical/electrical connection issue. If PHD2 guides without the 'SNR goes to zero' phenomenon, that would suggest a bug in Ekos.

SNR zero would suggest that it got an empty frame, or otherwise cannot detect stars. Have you tried debug mode and store guide images to have a look at them?

Guider log/info didn't complain about anything.
Captured guider images looked fine.

I am not sure how to reproduce this...
Some intermittent issue...

I am shooting right now and the guiding is perfect from the beginning. I am 1 hour in.
Last session I did have this issue (which I reported in this thread) and guiding stop/restart fixed it.
Resulting subs were fine.

I didn't change anything between these two sessions.

Thanks!

So it's not a camera driver issue that delivers black frames....
If you have normal looking images with stars but the guider doesn't find them that would point at the SEP profile used there. Have you tried changing that?

No. The guider finds stars I don't get any messages about losing the guiding star.

Ah, my bad - so it's only about the SNR jumping up and down?
Do you use dark correction for the frames? Maybe if that one is old (and/or too large) you end up with a 0 background without noise so you cannot compute S/N....

No, I do not use the dark correction