Guider Settling, Sequence Won't Initiate

Hi. Last night was my first attempt with SGP and was hoping to run a simple sequence. I completed most of my setup as I always have (Target centering, CCD setup/cooldown, focusing, guider calibration and initiation, etc) before even opening SGP, My main reason for switching to SGP was for the dithering capabilities with PHD2 and so I just wanted to run a sequence with dithering. All the devices seemed to connect fine. I chose dithering parameters based on others suggestions with a FL 2300 and imaging scale of 0.65". I couldn’t get the sequence to initiate as initial “guider settling” would not complete. Of note, it was a particularly bad “seeing” night so the PHD graph which usually shows RMS error in the 0.3-0.5" range was about 0.7" and the “distance” reported by SGP was often more than 1 (pixel?). When I increased the value from 0.3 to 2, it seemed to complete settling, but I still had issues initiating a sequence. The log file is here:

One more question if I may. I am perfectly happy with my temperature compensating focuser (TCF-SI) and entered 83 steps/degree which I had previously determined and checked the temp compensate box in SGP focuser tab. I don’t intend to use any other autofocus routine. Do I need to check the “use autofocus” box or set any parameters within SGP. Or should I just try to run the Optec program independently of SGP (I seem to have difficulty doing this on a dry run).

Thanks so much.

BTW, my equipment are: QSI540wsg, Edge9.25, Lodestar x 2/PHD2, Optec TCF-SI, Avalon M Uno

Am I correct that you intend to focus at the start of the night and then use just temp comp? If I have this wrong, please correct me. If I have this right, it is a really bad idea and will not give decent results. You can use temp comp along with periodic auto focus but although I have a solid system and solid temp/focus data, I no longer even do that. I set SGP to autofocus every 1.5 degrees and that keeps it in the zone regardless.

Also, FYI, if you do use SGP temp comp, it will only change focus between exposures, not during. Most focuser programs do that continuously. Which is better is a matter one can debate but for short exposures, the SGP method is better IMHO.

Thanks CCDman for your reply and advice. But I have generally gotten excellent results with just temperature compensation focusing with the TCF-SI on the 9.25 Edge. Perhaps my standards are not as high but for the past year I have been quite happy with it. I was surprised, but after an initial compensation learning session, I have maintained excellent focus, visually and by FWHM, over unattended sessions as long as 7 hours and with temperature drops as much as 8-9C. And my understanding from reading these forums is that the SGP on-target autofocus doesn’t work as well with OTAs that have a central obstruction and at long FL. And I would rather not have to add the complexity of having to slew to a bright star for each focusing run. Regarding focusing only between exposures in SGP, that could be a problem, because I at times take subexposures as long as two hours. The Optec focuses continuously.


It is correct that tubes with obstructions have more of an issue - I understand that they are working on this, however. You do not have to slew to a bright star with SGP - that is pretty much a FocusMax thing and one reason I moved to SGP to begin with after many years of MaxIm and FocusMax and ACP.

I will be blunt. Two hours is much too long. Period (unless maybe you are cooling with liquid nitrogen). You will get better noise levels with not more than 45 minute subs and thereby having more to use for sigma reject. As proof I offer my Squid Nebula Narrowband image from this summer. The blue O3 is super faint and these were 30 minute subs with 3nm narrowband filters at f7.

Squid Narrow

Also be aware that just because you focus between subs does not mean you cannot also use the Optec compensation. If you are going to use the continuous Optec compensation, I would also do periodic autofocus - using offsets if you are imaging with narrowband filters. You can just do the focus less often if you are using temp comp.

Thanks again CCDman, bluntness notwithstanding :slight_smile: Wonderful squid by the way. Of course two hour subs are not my norm. Its typically 5-10 minutes for LRGB and 15-45 minutes for narrowband. But I have gone 1-2 hours when searching for super faint detail such as suspected PN halos. In that instance, so long as you have not buried your read noise in sky noise, ( a possibility even at two hours with 3nm filters) there is no reason why two hour or even longer subs are not possible (though perhaps impractical), The stars will likely be saturated, cores blown out etc., butt it depends on what you are trying to do. And again, I have gotten more than adequate results with nothing more than temperature compensation. Why are you so sure that is impossible? And are there folks here who truly are doing on-target focusing successfully with long FL SCTs?


Impossible, not really. But I think not practical or very helpful would be a better description. I really don’t think that, for example, that (3) 2 hour exposures would net you a better result than (8) 45 minute exposures. I suspect noise rejection becomes as important as the increased signal at that 6 hour total exposure.

The question of mine that you quoted above actually referred to the focusing issue, and that you don’t seem to believe it possible that you can get excellent results with just the continuous temperature compensation feature alone. We could argue all day about whether or not you can pull out the faintest details with the longest subs attainable (so long as you remain below sky noise limitations) that may not be possible no matter how may shorter subs are stacked. But that’s not why I started this thread. I’m just looking for help, as a beginner with SGP, regarding the guider settling question. Imagers get great results using different methods, some that may run counter to your beliefs.


Indeed. In fact I have been to imaging conferences where there have been very heated arguments by experts on both sides about subs lengths! :smiley:

So back to the main issue, maybe it is possible to get good results with temp comp alone but that has not been my experience even when I owned carbon fiber RCs (12 inch tube then 14 inch tube then 14 inch truss). I really do not believe that there is any substitute for occasional autofocus (or manual but not many do that these days). In my opinion actual always beats theoretical, which is what temp comp really is at the end of the day - an estimate. It may be a very scientific estimate, but I have not found it to be as accurate or reliable as an actual focus run.

This is not to say temp comp is worthless, far from it. It makes an excellent adjunct to focus runs and can mean having to do them less often with some level of confidence. But to depend solely on a single focus and then just temp comp all night? I do not know anyone at all that does that or would suggest it is a good idea. In fact, if it were possible to predict focus that well one could just focus once when you set up the system and then just use temp from then on!

The other thing to realize is that not just temp is involved. It is very possible that sky position might also cause shifts in equipment that might alter focus somewhat. Temp would, of course, do nothing for that.

Well, in fact that is exactly what I have done for some time now. In the end, it is whether or not you are satisfied with your images. I have been with mine. I want to dither because I know it will substantially reduce the amount of noise that I have to deal with. I don’t need to improve upon my focusing routine. I can show you examples of first and last subs over an evening where temperature dropped significantly and my FWHM was nearly identical using just temperature compensation. You are very dogmatic in your opinions. While I can appreciate that, I’ve done fine with a different approach. I need to prioritize, and focusing is just not at the top of the list. Thanks.


I’m not saying that the discussion so far hasn’t been relevant, but let’s get back to the original questions and try to stay on track.

Derek, if I understand you correctly, you had set the original dithering settling amount to 0.3 px and then later changed it to 2px. This is in the autoguide tab of the control panel. But the distance is only half of the equation. How long is your settle time?

Also, I haven’t combed through you log, but it would be helpful if you could describe more clearly what “issues initiating a sequence” means, or at least about what time you tried to start a sequence so we can find that area in the log.

If you want to use your TCF during exposures, your focuser software will need to do that. SGP’s temperature compensation only takes place in between exposures. I don’t use a TCF so take this with a grain of salt, but I don’t think that you can continuously compensate during exposures from within SGP. You’ll need to use the Optec software. If that is indeed the case, then I don’t know if you can have both SGP and Optec connected to your focuser at the same time. If you are committed to just using TCF, then I would suggest not connecting it in SGP and see if that makes a difference in getting the Optec software to work. These are just suggestions as I don’t really know what I’m talking about with this :grinning:.

I am using the Feathertouch focuser and the SW is basically the same as with the Optec (Optec makes it). In my experience you can indeed have both the SGP and Optec connected at the same time but would not want to have temp comp enabled in SGP if you want to temp comp continuously, just in the Optec . So bottom line is you can have both connected at the same time, but should only have temp comp active in one or the other. SGP if you want to correct only between exposures, Optec if you want to correct all the time.

Of course I still think a single focus per night is a bad idea but at the end of the day (OK, end of the night), it is not my setup and not my images so really not my business. :smile:

Thanks Joel. I am trying again tonight. I am able to run the Optec software temp compensation independently, without connecting it from SGP, so focusing is not an issue for now.

The pertinent times on the log are 10:51:08 when I tried at 0.3 pixels and 10:57:56 when I increased it to 2 pixels. I have tried to settle from 2 to 8 secs.

So tonight I tried 1.5 pixels for 2 seconds. When I try to “run sequence” the message will be something like “auto guider settling distance 0.5 for 2 secs”. Once it has settled for below the specified amount and above the specified time it will say " integrating NGC…event 1 frame 1 for 60 sec". But then it will either go right back to “auto guider settling…” Or it will begin exposing (green bar moving from left to right), but the exposure will never complete more than 30 seconds (I set it for 1 minute for the test) before aborting and going back to the “settling” message. I can post this log file tomorrow.

Any thoughts? Thanks,


Your settling values are fine (if you like them). Your issue is that you have told SGPro to restart the current exposure any time the PHD2 pixel error exceeds 1.0. From your settling issues, this parameter obviously needs some adjustment.

[10/10/2015 10:58:04 PM] [DEBUG] [Auto Guider Error Thread] Frame restart: distance was 1.0 pixels

It is in the CP->Guider tab.

Thanks for confirming that stuff CCDMan.

Just to be clear, the settle time means that the guiding error must stay
below your threshold for the whole duration you set. So an error of 0.3
and settle time of 2sec means that the error must stay below 0.3px for at
least 2sec. 8sec is a really long time for such a stringent pixel error.

It sounds like the guiding is settling ok with 1.5px error for 2sec and
then the exposure starts but something fouls it up. At this point I’m not
sure what to tell you and hopefully Ken or Jared will chime in.

Here is the log file from last night. The sequence starts around 8:33:23

Thanks. Hope someone can help me figure this out.


Derek, Ken found your issue, please see his post above.

Hi. I finally got around to trying this again, and by unchecking the parameter noted above by Ken I am finally dithering. Thanks Ken/Andy/Joel. However, I am still having an issue with premature sequence termination. Last night I found that SGP had terminated the sequence and parked my scope after completing only 9/20 subs. There was a message stating sequence completed. Can someone please look at this log file and help me figure out why this is happening?



A couple things…

  • The version of SGPro you are using is pretty dated (and no longer supported). Please upgrade to the latest release.
  • And why your sequence quit…

[1/3/2016 9:21:11 PM] [DEBUG] [Sequence Thread] Guide star lost! Aborting sequence.
[1/3/2016 9:21:11 PM] [DEBUG] [Sequence Thread] Run event requested sequence abort…

Using automatic sequence recovery mode can be helpful. You did not have it on so your sequence just quit.