PHD2 Guide logs (from ASIAir Pro) - How to read them

[Jonny_H]

Hi all,

Hoping that someone can assist or point me in the right direction.

I would like to try to understand how to 'read' guide logs, namely PHD2 guide logs from an ASIAir Pro as i am curious to see PE and/or whether anything looks particularly untoward. 

Is there a piece of software or an application exist where you can upload the log to get a simplistic view as to what is happening?  Or can this be seen by manually looking at the log?  If the latter......how do you read this 'Matrix' data?

For reference I have uploaded a log file from earlier in the week.

Any assistance appreciated.

Thanks,

Jon

PHD2_GuideLog_2022-01-17_202253.txt

[scotty38]

I know this isn't exactly what you asked for but, you can use this to more easily view the log:

https://openphdguiding.org/phd2-log-viewer/

and then this might help with fault finding:

https://openphdguiding.org/man-dev/Trouble_shooting.htm

[Jonny_H]

Thanks @scotty38

Admittedly I haven't had a detailed look at the 2nd link yet - i did however upload my PHD2 guiding log to the application from the first link and i have attached a screen-dump from the longest of the 4 guiding sessions from the night of the 17th Jan this year.

As previously mentioned - analysing guide logs is new to me but i will learn   Nothing is screaming out at me other than the fact that i see the occasional 'blip' from both RA and Dec axis' which seem to be outside of dithering.  I just dont know how to read this data properly yet to work out what these blips are or even if it is possible to do so?  

[wimvb]

A bit late, but I had a look at your data.

Calibration is ok, but your mount has some DEC backlash. I don't know if the ASIAIR compensates for this. If you have another computer to which you can connect your gear, you can measure this with PHD2. Although it doesn't seem too severe and guiding handles it.

You dither only 2 pixels. you might want to increase that to 10 pixels or more. This will make it necessary to let the mount settle longer between exposures.

The periodic error curve of your mount is very jagged. Normally this curve is much smoother. In your guiding algorithm, use PPEC for RA guiding (Predictive Periodic Error Correction). Your mount (AZ-EQ6) does have the capabality to use Permanent Periodic Error Correction, stored in the mount. But this feature is not compatible with guiding; it's either or, not both.

A frequency analysis of RA guiding shows main peaks at 120 s and 240 s. 120 seconds coincides with the motor timing pulley doing one full revolution. The much smaller peak at 10 s is this pulley advancing exactly one section/tooth.

Imo, you should do a test with guiding for one hour with 1 s guiding exposures. This will resolve the 10 s peak better. My guess is that it will increase. The 10 s peak (and to a lesser extent the 120 s peak) is related to belt tension. If you tighten the belt, this peak will decrease. If you use 2 s or longer guide exposures, the 10 s peak will cause what looks like noisy guiding.

[michael8554]

16 hours ago, wimvb said:

You dither only 2 pixels. you might want to increase that to 10 pixels or more.

A  comprehensive analysis from wimbv of the problems.

But you can't specify the Dither based only on the GuideLog.

You need to know the image scale of the imaging setup, and Jon has three very different OTAs.

The guide image scale is 3.3arcsec/pixel.

If the imaging setup is 1arcsec/pixel, a typical Dither of 12 pixels requires the guidecam to be dithered 12 x 1 / 3.3 = 3.6 pixels.

For 2arcsec image scale it's 12 x 2 / 3.3 = 7.2 pixels

Etc

Michael

 

[wimvb]

2 hours ago, michael8554 said:

The guide image scale is 3.3arcsec/pixel.

If the imaging setup is 1arcsec/pixel, a typical Dither of 12 pixels requires the guidecam to be dithered 12 x 1 / 3.3 = 3.6 pixels.

For 2arcsec image scale it's 12 x 2 / 3.3 = 7.2 pixels

Yes, exactly so. I was referring to dithering 12 p in the imaging camera of course. For 2 pixels dithering @ 3.3”/p in the guide camera to be sufficient, the imaging scale would need to be 0.55”/p at most, which seems unlikely.

[Jonny_H]

20 hours ago, wimvb said:

A bit late, but I had a look at your data.

Calibration is ok, but your mount has some DEC backlash. I don't know if the ASIAIR compensates for this. If you have another computer to which you can connect your gear, you can measure this with PHD2. Although it doesn't seem too severe and guiding handles it.

You dither only 2 pixels. you might want to increase that to 10 pixels or more. This will make it necessary to let the mount settle longer between exposures.

The periodic error curve of your mount is very jagged. Normally this curve is much smoother. In your guiding algorithm, use PPEC for RA guiding (Predictive Periodic Error Correction). Your mount (AZ-EQ6) does have the capabality to use Permanent Periodic Error Correction, stored in the mount. But this feature is not compatible with guiding; it's either or, not both.

A frequency analysis of RA guiding shows main peaks at 120 s and 240 s. 120 seconds coincides with the motor timing pulley doing one full revolution. The much smaller peak at 10 s is this pulley advancing exactly one section/tooth.

Imo, you should do a test with guiding for one hour with 1 s guiding exposures. This will resolve the 10 s peak better. My guess is that it will increase. The 10 s peak (and to a lesser extent the 120 s peak) is related to belt tension. If you tighten the belt, this peak will decrease. If you use 2 s or longer guide exposures, the 10 s peak will cause what looks like noisy guiding.

Thanks @wimvb & @michael8554

I really need to learn this stuff!

Thank you both for your comments. So in essence, i need to check my belt tension and tighten. Change my guiding contributes including looking at dithering settings.

I have recently changed my kit so that i can utilize NINA, PHD2 etc... mainly to free me from the ZWO bubble that the ASIAir restricts you to,  but also to open software capabilities. So next time I get out (whenever this weather decides to give it a rest) and have a play.

if I have said anything untoward above please feel free to put me straight!

regards,

Jon

[wimvb]

4 hours ago, Jonny_H said:

I have recently changed my kit so that i can utilize NINA, PHD2 etc... mainly to free me from the ZWO bubble that the ASIAir restricts you to,

Great. If you want to get the most out of phd, start with calibrating near the meridian, followed by at least 10 minutes (but preferrably 20 minutes) guiding assistant. This will give you a good baseline for your guiding setup. PHD will also determine your DEC backlash. Accept the settings suggested by the guiding assistant.

To analyze the mount, let PHD guide for an hour or so with 1 s exposures. (You can image during that time, clear skies are all too precious.) Using a short exposure time will give you better values for the short time periods in the belt drive. Normally you wouldn't use such short exposures, but rather 2-3 s.

If the 10 s peak is more pronounced for short exposure guiding, you will probably need to tighten the RA belt. This is a fairly simple procedure. 

[kens]

4 hours ago, Jonny_H said:

I have recently changed my kit so that i can utilize NINA, PHD2 etc... mainly to free me from the ZWO bubble that the ASIAir restricts you to,  but also to open software capabilities. So next time I get out (whenever this weather decides to give it a rest) and have a play.

Just FYI, the ASIAir is basically a Raspberry Pi and you can install Astroberry, StellarMate or a DIY INDI/Ekos/KStars onto it. These are open source are not restricted to ZWO hardware and there is even an INDI driver that lets you continue to use the ASIAir power ports. Plus you get the fulll blown versions if these and PHD2 rather than the de-featured versions implemented by ZWO.