Dither disruption in guiding

[mark T]

I recently acquired an AP1100 with absolute encoders (I have it setup in PHD2 with the encoder settings) and I have noticed that dithering guiding disruption in DEC and Az continues well after settling is "complete" and is resulting in increased eccentricity for the first sub after the dither.  I have uploaded the most recent PHD2 log and a sample screen shot is below.  

https://openphdguiding.org/logs/dl/PHD2_logs_4xuo.zip

 I am guiding with an OAG and the OTA is a SVX130T.  I did not use a point model last night but this behavior occurs regardless.

Notice the red lines in the attached CCDI table - all of which have elevated Aspect.  Dithers caused a great majority of them, and autofocus did the remaining. You can match up a couple of the dithers in the PHD2 log with the lines in the CCDI table below.  Autofocus is also not settling out quickly enough.   

See the NINA guiding settings below .  My guiding camera pixel scale is 1.29 and the main camera is 0.84, so there is a "dither scale" of 1.5 camera pixels per dither pixel.  I reduced the number of Dither Pixels in NINA down to 3 in hopes of getting better dither settling, but that did not help.    I previously had Minimum Settling set at 15s but then the settling did not complete so I reduced it to 10s.  

Also, with the encoder settings in PHD2, guide pictures are taken 10s apart, so that seems to complicate the time it takes to settle.  Maybe if dithers could be set to use the "short" exposure of 1s, the settling process could occur smoother.

Any suggestions how to solve this would be appreciated

Mark

[Brian Valente]

Hi Mark

Just a few clarifications. Bruce and/or i should have a chance to review your logs later

1. I assume you watched and followed this tutorial: https://youtu.be/wCu8PKjDZ20

2. The short exposure time is used during the dither settle time. it seems like your issues after dither relate to lost star events rather than additional settling required. A dither should not result in a lost star event.

3. 3 pixel dither still seems like a lot for that setup. Have you tried a lower value?

--
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/8aadc10b-5995-4b6e-a990-cc660932d650n%40googlegroups.com.

--

Brian 

Brian Valente

astro portfolio https://www.brianvalentephotography.com/astrophotography/

portfolio brianvalentephotography.com 

astrobin https://www.astrobin.com/users/bvalente/

[mark tenboer]

Brian - 

1.  yes I followed the Astro-Physics encoder guiding instructions on YouTube and I am using the "encoder default" settings.  I will note that the current PHD2 default "long" exposure is 10s where the one in the YouTube appears to be 5s.  

2.  I have also noticed the lost star problem - forgot to mention it in my message.  I noticed last night when running GA.  I have the Minimum Star HFD set at the default of 1.5 pixels, so increasing that would help.  What do you suggest ?  

3.  A 3 px dither = 5px on the main camera = 4 arc sec at 0.84 pixel scale.   I didn't realize that is considered too large an amount.  How many arc seconds is considered necessary to get the benefit of dithering?

Mark

  

You received this message because you are subscribed to a topic in the Google Groups "Open PHD Guiding" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-phd-guiding/zSKgELLECGg/unsubscribe.
To unsubscribe from this group and all its topics, 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/CAJa45i7FK%2BjqHznZCw_tM74cjHtKk0tMpGYkZG85e3YApuQqhg%40mail.gmail.com.

[Bruce Waddington]

I have some suggestions to make:

 

1.       Increase your guide cam exposure time to 2 seconds or use auto-exposure.

2.      Set the Min-HFD value to 1.2

3.      Increase the “short delay” to 2 seconds

4.      Decrease the size of the dithers as Brian suggested unless your main camera has a large amount of fixed pattern noise you’re fighting. 

 

I don’t think it will be difficult to get things smoothed out – I think you’re being too aggressive with the guiding and perhaps the dithering.

 

Bruce

--

[mark tenboer]

Thanks Bruce -  I will give these a shot.

Mark

You received this message because you are subscribed to a topic in the Google Groups "Open PHD Guiding" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-phd-guiding/zSKgELLECGg/unsubscribe.
To unsubscribe from this group and all its topics, 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/003b01d8de9b%24a109c9a0%24e31d5ce0%24%40earthlink.net.

[mark T]

Bruce - I adjusted the guide settings per your suggestions - much improvement but I think some more adjustment might be needed.  Last night's log is linked below.  Consider that the seeing was not very good.    Also recall that my mount is a new AP1100 with encoders.

https://openphdguiding.org/logs/dl/PHD2_logs_f2sU.zip

I am no longer losing stars, and the dithers are "settling".  However, subsequent to a dither successfully settling, the guiding is still sometimes very volatile for over a minute.  Some dithers go smoothly, so do not.

After an autofocus ends there is often volatile guiding for a while. See the start of log 7.   

I have NINA set to stop guiding for AF runs.  AF is always done with the Lum filter.   So if I'm imaging Ha, a filter change back to Ha occurs after the AF, but NINA is also set to stop guiding when there is a filter change.

It seems like restarting guiding, whether due to a dither or AF, is frequently a long and rough patch.

I continue to see way too many large, seemingly random, excursions for an encoder mount.   Dec and RA can often move from -1" to +1" in one or two consecutive steps   Example:  See log 9 at 00:48.  They are easy to see all over the logs.

Do they look like sticktion to you?   My cables are all snug on top of the OTA (I have a NUC up there with an USB hub), with the exception of two cables that run through the mount - a power cable and a USB cable from the CP4 to the NUC.  I am wondering one of those is snagging inside as RA tracks.  This cable setup is exactly the same as I used with my prior mount, with the exception of the two cables through the mount with the new mount.  The old mount did not show any signs of sticktion.

 Could the gear mesh be too tight?  This is a brand new mount so maybe shipping had a detrimental effect on the factory settings.

I greatly appreciate your suggestions.

Mark Tenboer

[Brian Valente]

Hi Mark

The thing that jumps out at me is not your guided results, but your unguided run. without any guiding, the results look really choppy with some notable excursions.

How good is your seeing there?  You might consider increasing your exposure time to 4-5 seconds. I know it seems counterintuitive, but guiding with 

This run averages about 0.55" total RMS

 

your longest run averaged 0.37" total RMS. well below that

 

so aside from your specific questions around dithering, it seems like generally things are working reasonably well, given what seems to be somewhat choppy conditions

For encoder mounts, remember you aren't using guiding in a traditional sense. Encoders are doing most of the work, guiding is only supposed to kick in gently to keep things on track

It may help a longer exposure will smooth things out. 

You might also consider reducing the aggressiveness on each axis. It's at 0.8 right now, i think i'm at 0.6 with my 1600ae

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/02d40e09-4e8a-44b9-b432-17e30fb877fdn%40googlegroups.com.

[Brian Valente]

Mark

I was just thinking about this, do you have any sort of sky modeling active?

On Fri, Oct 14, 2022 at 10:13 AM mark T <mten...@gmail.com> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/02d40e09-4e8a-44b9-b432-17e30fb877fdn%40googlegroups.com.

[Bruce Waddington]

I agree with Brian, I think you need to wait for a night of good seeing in order to get a better assessment of the results.  1 arc-sec deflections are hardly rare when the seeing isn’t good.  The rate of Dec reversals seems too high, which is often the case when the seeing is poor and the Min-Move is too low.  I know you ran the GA but that was several hours earlier and necessarily in a different part of the sky from where you spent most of the night imaging.  I generally try to get a Min-Move value that results in Dec corrections being issued only 15-20% of the time.  In your worst guiding session at 23:30, the guide pulse frequency was much higher than that.  I think you were chasing the seeing here.

 

A good way to judge seeing is to take a few 10-sec exposures, unguided, and measure the FWHM of the stars in the image.  That will give you a working estimate for the seeing conditions.  I do this as a side-effect of doing automated focus runs.  It’s not uncommon in a single night to get results that might range from 1.5 arc-sec to 2.5 arc-sec and the guiding results usually reflect that.  Of course, 2.4 arc-sec is hardly the worst it can be, some locations and conditions are much worse.  Also, since you were imaging at a high Dec value, you need to be sure the OTA is well-balanced and you might, as you suggested, want to take a close look at the cable routing.  Through-the-mount cabling doesn’t give anyone a free pass on these kinds of problems, it just make it harder to find them. <lol>

 

Regards,

Bruce

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i6avCA1MM-eL1dkzJ96%2BsNV%2B4wDpiU1%3DNMrxKZ1_5e-8w%40mail.gmail.com.

[mark T]

Yes I was created a dec-arc model last night that covered both targets I imaged.  I am new at that, but it went smoothly and all the points were "OK" in APPM.  The final corrections in the APCC table looked fairly small,  so it seemed ok.  This week I have guided with and without a model and the guiding has been choppy regardless.  Seeing has not been very good.

[Brian Valente]

If you do experience more of this, you can also try disabling the model just to see how it may change things. 

I'm not saying that's the culprit, but it's one more thing you can try to eliminate variables.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/c3e0e809-8674-44b1-90ed-747f924d91c3n%40googlegroups.com.

[mark T]

Yes - I was using a dec-arc sky model that I created last night that covered both targets I imaged.  In APPM all the points were "ok".   Over the past week, I have guided both with and without point models and the guiding has been very choppy regardless.  Seeing has apparently been pretty poor. 

Bruce  - thanks for the method of measuring seeing and your insights on this.   I have been using the "high frequency star movement" in GA as a way to determine that, but I have seen nights where the RMS was low but guiding was very choppy anyway.  I didn't know if that is its purpose.

Regarding balance, I have triple checked it, and both Dec and RA are very well balanced.  I can nudge either of them in any direction and they slowly stop and stay at any position.  There is no tendency to move one way or another.  If there are some ways to check it that are unique to AP mounts, I would love to know. 

Regarding through the mount cables - I plan to take them out of the mount next week and see if the spikes are reduced.  They seem to have a nice amount of slack, but there is no way to tell if they are just rubbing on areas inside the mount.

I will make the guiding adjustments you recommend above, and I accept that the results may not improve much until the seeing conditions improve.  

Thanks guys   - your suggestions and insights - they are immensely helpful.

Mark

[mark T]

OK - I will disable the model too.

[mark T]

Bruce - you commented that the GA was in a different part of the sky than where I imaged.  I'm not sure if that was just an observation, or it implied that I should do GA's in the area of the target.   I always do GA's near DEC 0 and the meridian (same for calibrations).  If you recommend that sometimes GA's should be done in the target area, please expand on this.

[Brian Valente]

>>>  I always do GA's near DEC 0 and the meridian (same for calibrations).

I always do calibrations and GA at Dec 0 as well

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/f07d80ce-7494-4393-9ec5-ae2c711e48efn%40googlegroups.com.

[Bruce Waddington]

Actually, what I said was “necessarily in a different part of the sky from where you spent most of the night imaging.”  <g>  I said that in order to avoid this confusion.  The most accurate GA results are gotten around Dec = 0 but you have to remember that seeing can depend on the pointing position for various reasons.  What you did was fine so long as you remember that the MinMove recommendations relate to a time and place and may need adjustment when conditions change.  That’s why it’s good to keep an eye on the Dec guiding activity and perhaps increase the min-move value if the frequency of corrections is high.  I think that’s particularly true with configurations like you have.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/f07d80ce-7494-4393-9ec5-ae2c711e48efn%40googlegroups.com.

[mark tenboer]

Understood Bruce - thanks

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/006301d8e01f%24621e71c0%24265b5540%24%40earthlink.net.