Spikes in Dec while guiding with PHD2

[Chris M48]

In recent weeks I have been struggling, while guiding with PHD2, with the selected star suddenly lurching northward.

The occurrence is regular but not evenly spread. The display shows the star has jumped northward out of its box and the graph shows the dec trace having fallen vertically off the graph altogether.

The equipment I'm using is: SW ED100 on AVX with SW 50x9 Finder-Guider and SPC900NC. Software: StellariumScope, Stellarium, BackyardEOS, PHD2 and ASCOM.

Mount control is from Nexstar handset via serial to USB cables. Although the AVX mount is only six months old, I have used the remaining equipment successfully for a number of years. After a period of learning to use the AVX, I can now achieve a good alignment with the PHD2 Drift Align feature.

I have checked for cable snags and mount obstructions but these seem free and ok. I am at the earliest opportunity going to try different cabling to hopefully eliminate that as a possible cause.

I will be very greatfull for any suggestions, help or advice in solving this problem. It really would be nice to get back to some decent imaging.

Thanks.

 

[PeterCPC]

Are you dithering in BYE? If so you may need to alter the settings. Try turning off dithering to see if it improves it.

Peter

[Tiny Small]

It could also be stiction in the gears. A strip, regrease and careful re-assembly can work wonders.

[Thalestris24]

Hi

How does your mount perform with guiding off - have you run the guiding assistant?

Louise

[cfpendock]

That’s very interesting, Chris M48.  Last night I had a similar problem for the first time ever.  Unfortunately I had already set up and was imaging automatically, so I didn’t discover the problem until a couple of hours had passed because guiding during the first two subs was fine.  Then random oscillations started to appear in dec - sometimes up to 4 seconds of arc, whereas RA was fine. 

 

Like yourself, I checked for mount obstructions and cable snags.

 

Often this wandering in dec can be due PHD chasing backlash in the gearing.  You need to have a certain amount of backlash, or things will be too sticky - perhaps as Tiny Small above has indicated.  But I think the best way to cope with this is to ensure the motor is only driving in one direction.  For myself, I am going to try this by using a less accurate polar alignment, using either a north or a south only setting in PHD.

 

Otherwise I don’t know what to suggest, certainly running with the dec turned off got rid of my spikes, but sadly approaching clouds signalled an end to activity.

 

Hope you get it sorted.  Me too!

 

Chris

[rubecula]

Yes very interesting.  I've encountered this on a couple of occaissions when imaging almost vertically overhead.  If I choose a subject lower down all is well, so until I find a cure I'm avoiding overhead objects.  I feel sure it's a backlash issue but I'm nervous about stripping the mount down and third party hypertuning seems quite expensive.

[Joel Shepherd]

Hi @Chris M48: when you say " the dec trace having fallen vertically off the graph altogether" do you mean literally a near-instantaneous straight up-and-down (or very nearly so) trace, or a more gradual but persistent slope?

If the former, I've been wrestling with this for close to a year and unfortunately haven't found the solution. The fact that you're using a different guide camera and the same mount increases my suspicions that it's an issue with the mount. (I control the mount through the guide cam: also different from your set up and again pointing to the mount as the thing in common.) Things I've tried have included new cables, tying down cables to prevent laptop-side connections from moving, a high-quality powered USB hub, disabling dithering and checking fittings/clamps to eliminate sources of flexure. All to no avail. My next step, once I work up the courage, will be to flash the handset controller and mount and hope that Celestron's current firmware has addressed the issue.

If this is the same issue that you're describing, it's not backlash. My mount has plenty of that as well, but it manifests itself as PHD2 taking a long time to correct movements in DEC ... but that's a different problem from large and very abrupt changes in DEC position. For backlash I've found that setting PHD2 to correct DEC only against the natural drift of the mount in DEC (and let that drift correct for DEC errors in the other direction) works well to prevent large but gradual swings back and forth on the DEC axis.

Anyway: have the same issue from the sounds of it, regrettably can't report a solution.

[Chris M48]

Thank you all for your helpful replies. It seems that this is a common problem but with variations.

I have only used BYEOS and dither after obtaining stable guiding.

Yes Joel, vertical straight off the chart. Later, I increased the chart size to discover it did eventually stop and then slowly recovered, then repeated the cycle several times before my patience ran out with the test.

I've resumed imaging, with some success, over the last couple of evenings that allowed a few hours of clear skies. The first evening, with all new cables, the spikes were worse than ever so I turned off guide direction (usually set to Auto) altogether. Unsurprisingly, no spikes at all. Then I chose South and lastly North.

Off produced a distinct drift, South had a couple of spike instances but then settled down to none. North was best of all and allowed  5 minute exposures. I had reduced aggression and max move settings to 50% and 150 respectively. Not perfect, but then it never was.

I Need to play with settings to see if further improvements are possible.

[Thalestris24]

That sort of behaviour could be indicative of a comms problem.

Louise

[Chris M48]

That's interesting Louise. My logic can't quite work out that with minimum agression and movement settings why would PHD2 go so far outside of the parameters without error?

l'm currently using my Home PC set up for Astro in the conservatory but still got the spikes with my laptop. Could be the AVX hardware or firmware or maybe PHD2 or even ASCOM or Stellarium. Who knows? I certainly don't.

Nevertheless, we carry on.

[Thalestris24]

1 hour ago, Chris M48 said:

That's interesting Louise. My logic can't quite work out that with minimum agression and movement settings why would PHD2 go so far outside of the parameters without error?

l'm currently using my Home PC set up for Astro in the conservatory but still got the spikes with my laptop. Could be the AVX hardware or firmware or maybe PHD2 or even ASCOM or Stellarium. Who knows? I certainly don't.

Nevertheless, we carry on.

Hi

Well it sounds like it's not PHD2 actually moving the mount but if you lose comms then it could simply appear to make sudden movements because it's suddenly not getting the positional info back from the mount. How the mount behaves with guiding switched off in PHD2 e.g. by running the Guiding Assistant tool, should tell you something.

Louise

[Thalestris24]

ps set your PHD2 graph y-scale to +/-16 arc secs or to +/-16 pixels. That might give you a better idea on how far the graph appears to deflect.

Louise

[Joel Shepherd]

In my experience, PHD2 is trying to correct the mount and the mount is responding ... but it is the sudden, violent veer off axis that is the problem. It's way too much to correct in time to save the frame, and means stopping, recentering the image, etc. Very disruptive. If others have experienced this with other mounts I'd be interested to hear.

[Chris M48]

Sooo disappointing, after many hours of trying different settings it's no better. Worse in fact, it now happens when calibrating as well.

I can get rid of this when guiding by selecting anything but Auto from the drop-down list but this leaves me with unacceptable results.

I've attached Guiding Assistant page and drift alignment charts to try and illustrate what I mean.

I have concluded that it's the huge Dec Backlash that is the main cause and that I'm just looking for a work around.

Really fed up with this AVX mount now. Bought it from FLO so I think I might give them a word while it's still under guarantee.

UPDATE:

After several email discussions with FLO they sent me a replacement mount and handset. Guiding within 30 min of setup with no issues. A big thank you to SteveB and the team at FLO. I bought an HEQ5 at the height of my frustrations with the AVX. So now I need to let one of them go. After I've returned the deficient AVX to FLO of course.

Happy New Year!