Long settling time after dither

[A hwang]

https://openphdguiding.org/logs/dl/PHD2_logs_w9hF.zip

I'm trying to learn NINA advanced sequencer and adding a dither after a sequence of exposures with different filters and then loop back to the beginning of the sequence of exposures with different filters after the dither is finished.  I'm experiencing time out warnings from NINA.  I have read the NINA advanced topic dithering and the PHD2 documentation.  I tried increasing the pixel tolerance, shortening and lengthening the settle time and increasing the settling time out parameters.  What I see on the PHD2 graph is a large change in RA and DEC when the dither occurs and then big movements up and down before it finally settles down and resumes guiding.  It seems to take about 3 minutes to settle down. 

I have an astrophysics mount with encoders and followed their YouTube video on how to set it up for use with PHD2 as they instructed.  I noticed in the Target window of PHD2 that when the dither occurs the red X seems to make a spiral pattern around the center. I have the dither setting on Random in the Advanced Settings of PHD2.  The Algorithms settings are on Lowpass2 as recommended by AstroPhysics and the minimal moves were set by the Guiding Assistant.

I think that I understand the guiding parameters in NINA after reading the NINA documentation and PHD2 documentation about pixel tolerance, settle time, settle time out.  Am I missing a setting somewhere or is PHD2 fighting the encoders?

thank you

Allen

[Brian Valente]

Hi Allen

i'm familiar with that video ;)

To me it looked like your dithers are exceedingly large, i'm not sure of your image scale but about 1 pixel should be fine for your dithers (within NINA). PHD dither scale should not be changed from default of 1.0

I think the rest of your issues are likely within your NINA settings. I don't know what those are, but they control the definition of settled. You might check with NINA, or if you want to post your values here we could venture a guess

--
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/9fda20b9-ca34-4074-a0ca-4f27fbf263c7n%40googlegroups.com.

--

Brian 

Brian Valente

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

portfolio brianvalentephotography.com 

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

[A hwang]

I followed the NINA dither instructions (snippet below) to shift the main imaging camera 10 pixels so I set NINA to move the guide camera 12 pixels (0.617 arcsec/pixel main camera and 0.52 arcsec/pixel on the off axis guider camera) if I did the calculations correctly.  I would be happy if I didn't need to dither 10 pixels on the main imaging camera. 

An explanation of the two most important dithering-related settings follows:

• PHD2 Dither Pixels: The amount of pixels (on the guiding camera) that the dithering action will shift. This value should take into account the guiding imaging scale and main camera imaging scale, in arcsec/pixels. To select the appropriate value you need to consider how many imaging camera pixels will shift between two exposures as a consequence of a dither move. Obviously, 2 pixels at one focal length and pixel size will cover a different amount of sky than another setup with a different focal length and pixel size. It is usually recommended to dither a number of guide camera pixels that will shift the main imaging camera of about 10 pixels.

my NINA settings right now are Dither pixels 12, Settle pixel tolerance 2 px, settle timeout 150 sec, Minimum settle time 15 sec.  It takes at least 150 seconds for the guiding to settle.  The guide star image in PHD2 crosshairs and the Target view seems to show the guide star making a loop around the center of the crosshairs with each successive exposure as it gets closer to the center and that takes about 3 minutes and then it is back to normal guiding.  If I don't dither it seems to guide normally.  I have PHD2 set to Random and not Spiral dither in the advanced settings.

[Brian Valente]

Hi Allen

I think the NINA recommendation of 10 pixels is far too high for your setup. For fixed pattern noise that dither addresses you do not need that much, and it's clearly impacting your ability to dither and recover. A much more reasonable few pixels would still achieve what you need. I have no idea what they recommend a dither that high. maybe for DSLRs and widefield setups? 

your other issue is the minimum settle time. this means your dither has to be under 2 pixels for 15 continuous seconds. If it goes outside of 2 pixels, the counter starts over. You might consider a lower minimum like 3-5 seconds. That should get you to a point where your dithers aren't failing, and you can then decide if and how you want to tune it from there

Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/445f432e-53d6-4c98-9b4e-a3d7541c1455n%40googlegroups.com.

[A hwang]

Thanks Brian,

I'll try smaller dithering. Maybe 3 pixels? My focal length is about 2300mm. Does that sound about right?  Unfortunately, the clouds have returned and I won't be able to test for a while.  Monsoon has been a long one this summer.

Allen

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i5aS%3Dc3COeQr48T7%2B597dLHWi%3DoJJNoYR9XoMzehbsbfw%40mail.gmail.com.

[ChrisR Oz]

Hi Allen,

I used to use the same guider imaging scale,  but now do much better with 2x2 binning, and more sensitivity to find suitable guide stars.

Cheers, Chris.

[mj.w...@gmail.com]

Hi Brian

With no particular expertise in the matter, it seems to me that CMOS sensors suffer from a many-pixel wide tartan/plaid block pattern over the whole sensor on long exposures.

That require random dithers of up to 12 pixels to correct.

Again, no idea how that commonly quoted figure was derived.

Different to single or groups of hot pixels.

Michael

Wiltshire