Bad DEC response after dither

[Tom Palmer]

I'm dealing with a problem with DEC response.  It happens almost exclusively after a dither but I've seen it at other times too.  A representative example was at 4:51:39am, frame 1450 in the attached log: it hammered away at DEC for almost a minute before there was appreciable response.  And the most recent DEC move prior to this (frame 1432) was in the same direction and produced an immediate response so it doesn't seem like backlash should be a factor.

Is this stiction?

Mount: EQ6R-Pro with EQMOD

Guide scope/camera: Evoguide 50 / ZWO ASI290mm Mini (2.47"/px)

Imaging control: NINA

I've recently done the following (prior to this session):
- Implemented PPEC in the mount
- Changed the mount speed to 0.5 pulse rate for both RA and DEC (had been 0.8 but I've seen this problem before that too)
- Run the Guiding Assistant a couple of times (for long periods) and accepted all recommendations.  GA said that DEC backlash was around 900ms.  I made that change but then you can see it's adapted down to the 116-155 range for this session and is currently 128 in the DEC configuration.

Log attached - thanks!
- Tom

[Bruce Waddington]

Hi Tom.  Thanks for providing the specifics about the issue you’re dealing with.  I agree with you that this doesn’t look like simple backlash and it’s definitely correlated with large guiding displacements resulting from dithering.  It happens in both Dec guiding directions but not on a consistent basis.  I can see 3 or 4 consecutive large dithers in both Dec directions that recover very quickly – and then there will be one or two that take a minute to recover.  Here’s just one example:

 

 

Dithering is happening in both directions and only the single event in the center results in the long recovery time.  I really think this is a mechanical problem but I can’t guess at a cause.  Since these things happen sequentially at the same pointing position, the Dec axis isn’t moving much so the gear orientation should stay pretty consistent.  I do think you’d be better off running with higher guide speeds on both axes because those improve the ability to push through small amounts of static resistance on the axis and to more quickly clear any Dec backlash.  Obviously, you have to be sure the Dec axis is well balanced and there is no possibility that the guiding assembly can move around or be affected by dragging cables.  Have you looked at the various backlash test graphs to see if they provide any insight into how the Dec axis responds to aggressive guide pulses?

 

For what it’s worth, your overall guiding is quite good and these slow recovery times shouldn’t cause you much trouble if you relax your settling requirements a bit.  As it is, these events result in settling timeouts and it’s likely your imaging app is starting the next exposure before things have stabilized.  If you see that these events are causing you to discard images, I think it would be easiest to just revise the settling requirements.  But you’ll have to work through the trade-offs between lost images and more time spent dithering and settling during the imaging session.  You could also think about whether you need to dither so often – unless you’re using an imaging camera with substantial fixed pattern noise, it may not be necessary to dither so often.

 

Sorry I can’t be of more help on this one,

Bruce

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/a642ba75-a1b1-4e1a-85ba-f89078024e94n%40googlegroups.com.

[Tom Palmer]

Thanks Bruce - it does smell mechanical.  I gather many EQ6R-Pro mounts are stiff out of the box and mine is no exception.  I've had it a couple of years but it's still stiff.  For now I'll try to reduce dithering and also speed up the EQMOD pulse rate a bit and watch for this.  I might also boost the settle time.  We'll see how it goes.  Not too enthusiastic about any hypertuning but who knows?

Thanks -

- Tom