Phd2 problem

[Crispy5573]

Hi I'm using Phd2 with NINA for guiding. I am using an ASI174 MM with a Celestron OAG for guiding and a QHY 268m to image.. The cameras are connected through USB3 ports on a Pegasus pocket power box advance to a NUC running windows 10. 

The problem I'm having is Phd2 cuts out every other exposure and says star lost low SNR then goes back to a normal exposure.. My telescope is 2850 fl  f/9 so I use 7 second exposures in Phd2.  The image I get in Phd2 has never been great but guiding was (in ok seeing) always between 0.3 and 0.6 (at the worst) guiding error. Now it is never better than 0.6 and often goes to 0.9. I am using a JTW Trident P75 mount. I always keep it well balanced and polar aligned. 

I have uninstalled and reinstalled NINA, Phd2 and ASCOM and installed ASI camera drivers and have tried using both the ASI and ASCOM drivers. 

Any ideas anyone PLEASE it's driving me mad. 

[TiffsAndAstro]

It doesn't sound like a usb cable issue, but it's still worth checking.

[Crispy5573]

Checked and changed just in case. 

[Budgie1]

What setting do you have for the low SNR in PHD2, I think the default is 6, and what happens when you increase the gain on the ASI174MM?

Are you using narrow or broadband filters, if both then does the issue persist no matter which set you use or is there a difference. Have you changed the filters recently, maybe gone from 7nm down to 3nm?

[Crispy5573]

1 minute ago, Budgie1 said:

What setting do you have for the low SNR in PHD2, I think the default is 6, and what happens when you increase the gain on the ASI174MM?

Are you using narrow or broadband filters, if both then does the issue persist no matter which set you use or is there a difference. Have you changed the filters recently, maybe gone from 7nm down to 3nm?

Hi the SNR is at 6 and I have the gain at 100 (max in Phd2) on the 174. The filters will make no difference as they are after the OAG in the image train.

[Crispy5573]

I see plenty of stars in Phd2 for one exposure then it decides there aren't any for the next then back to plenty.  The only thing I think changed ( please bear in mind the UK weather and days or weeks between sessions) was a windows update but I can't remember if it started playing up immediately after that or not.

[markalot]

1 hour ago, Crispy5573 said:

I see plenty of stars in Phd2 for one exposure then it decides there aren't any for the next then back to plenty.  The only thing I think changed ( please bear in mind the UK weather and days or weeks between sessions) was a windows update but I can't remember if it started playing up immediately after that or not.

Have you tried just looping only to see what you can see?   Have you turned off star mass detection (this caused all kinds of issues for me when on)?  Have you bumped up the minimum star SNR for autofind?  I had all kinds of issues when I started using a 2600 MC duo that I solved by increasing minimum DNR and turning off the star mass detection (I think it defaults to off).  If the stars are actually vanishing (during looping) then I suspect the camera or the cable.  I'd aim at a tree and loop during the day to see if its the camera or cable.  

Edited August 25 by markalot

[Stuart1971]

3 hours ago, Crispy5573 said:

Hi I'm using Phd2 with NINA for guiding. I am using an ASI174 MM with a Celestron OAG for guiding and a QHY 268m to image.. The cameras are connected through USB3 ports on a Pegasus pocket power box advance to a NUC running windows 10. 

The problem I'm having is Phd2 cuts out every other exposure and says star lost low SNR then goes back to a normal exposure.. My telescope is 2850 fl  f/9 so I use 7 second exposures in Phd2.  The image I get in Phd2 has never been great but guiding was (in ok seeing) always between 0.3 and 0.6 (at the worst) guiding error. Now it is never better than 0.6 and often goes to 0.9. I am using a JTW Trident P75 mount. I always keep it well balanced and polar aligned. 

I have uninstalled and reinstalled NINA, Phd2 and ASCOM and installed ASI camera drivers and have tried using both the ASI and ASCOM drivers. 

Any ideas anyone PLEASE it's driving me mad. 

The P75 mount, is a friction drive mount, and so should not suffer any backlash but does has a large periodic error, so is the main error on the RA axis by any chance…? also is there by stiction on either of the drives, as I know they can be quite hard to get the brake or the clutch, (not sure what it’s called on that mount) set to the right torque..

[Crispy5573]

42 minutes ago, markalot said:

Have you tried just looping only to see what you can see?   Have you turned off star mass detection (this caused all kinds of issues for me when on)?  Have you bumped up the minimum star SNR for autofind?  I had all kinds of issues when I started using a 2600 MC duo that I solved by increasing minimum DNR and turning off the star mass detection (I think it defaults to off).  If the stars are actually vanishing (during looping) then I suspect the camera or the cable.  I'd aim at a tree and loop during the day to see if its the camera or cable.  

Hi I tried the camera during the day but couldn't get a short enough exposure time with Phd2. I did manage it by setting it as the main camera in NINA and everything looked very good ( I put it on a guide scope to get focus) . I have tried increasing the minimum SNR to 10 which I think is the max but it makes no difference. As for the star mass detection it seems a bit better with that switched on. Not sure what the DNR is ?? 

[Crispy5573]

39 minutes ago, Stuart1971 said:

The P75 mount, is a friction drive mount, and so should not suffer any backlash but does has a large periodic error, so is the main error on the RA axis by any chance…? also is there by stiction on either of the drives, as I know they can be quite hard to get the brake or the clutch, (not sure what it’s called on that mount) set to the right torque..

The mount works fine as I was saying I was getting only 0.3 and sometimes below total error at times. There is no clutch as such the two drive pins (they are about 2mm pins that drive the wheels) are spring loaded and driven direct by the stepper motors. I'm pretty sure it is a software problem and not mechanical. My next thought is to connect the guide cam direct to the NUC and run Phd2 on it's own and not through NINA in case there is some sort of conflict with anything else going through the PPBA but I can't see it doing anything. 

[Crispy5573]

53 minutes ago, Stuart1971 said:

The P75 mount, is a friction drive mount, and so should not suffer any backlash but does has a large periodic error, so is the main error on the RA axis by any chance…? also is there by stiction on either of the drives, as I know they can be quite hard to get the brake or the clutch, (not sure what it’s called on that mount) set to the right torque..

Just checked to be sure the spec on the mount states 8" peak to peak periodic error over 24 hours. 

[markalot]

40 minutes ago, Crispy5573 said:

Hi I tried the camera during the day but couldn't get a short enough exposure time with Phd2. I did manage it by setting it as the main camera in NINA and everything looked very good ( I put it on a guide scope to get focus) . I have tried increasing the minimum SNR to 10 which I think is the max but it makes no difference. As for the star mass detection it seems a bit better with that switched on. Not sure what the DNR is ?? 

Sorry, SNR.  Look at the Guiding tab screenshot on this page:  https://openphdguiding.org/man-dev/Advanced_settings.htm

 

I have mass detection off as it kept losing the star when on, especially during poor seeing or transparency.  

[Crispy5573]

5 minutes ago, markalot said:

Sorry, SNR.  Look at the Guiding tab screenshot on this page:  https://openphdguiding.org/man-dev/Advanced_settings.htm

 

I have mass detection off as it kept losing the star when on, especially during poor seeing or transparency.  

Oh ok lol. I have that set to 10 and mas detection is on and seems to work better but the problem happens even in good seeing. 

[PhilB61]

When you say 7 sec exposure, do you mean 7 sec guide interval? From my understanding the actual individual exposure length is set in the camera tab and PHD stacks these exposures "frames" at the frequency set by the guide interval, so for my typical 2 to 3 sec guide interval I can see that PDH is stacking around 15 or more frames (indicated in bottom left), if I increase the exposure length this figure reduces correspondingly. I understand the exposure length must be less than the guide interval to allow for processing of the data and the sending of guide pulses etc. How much less I am not sure but for me PHD seems to default to 100ms, I usually increase this to 500ms to reduce the workload on the data cables and processor by having to deal with less amount of data transfer and processing. Could it be that your exposure length is too long for PHD to process it sufficiently quickly for every guide interval or that the exposure is too short and PHD is trying to process too many frames, might be worth experimenting with these settings. 

I have also experienced dropped frames using USB3 cables with both active powered cables and powered USB3 hubs, these went completely when I reverted to USB2 cables. 

[Crispy5573]

8 minutes ago, PhilB61 said:

When you say 7 sec exposure, do you mean 7 sec guide interval? From my understanding the actual individual exposure length is set in the camera tab and PHD stacks these exposures "frames" at the frequency set by the guide interval, so for my typical 2 to 3 sec guide interval I can see that PDH is stacking around 15 or more frames (indicated in bottom left), if I increase the exposure length this figure reduces correspondingly. I understand the exposure length must be less than the guide interval to allow for processing of the data and the sending of guide pulses etc. How much less I am not sure but for me PHD seems to default to 100ms, I usually increase this to 500ms to reduce the workload on the data cables and processor by having to deal with less amount of data transfer and processing. Could it be that your exposure length is too long for PHD to process it sufficiently quickly for every guide interval or that the exposure is too short and PHD is trying to process too many frames, might be worth experimenting with these settings. 

I have also experienced dropped frames using USB3 cables with both active powered cables and powered USB3 hubs, these went completely when I reverted to USB2 cables. 

I had been wondering if the USB3 socket was giving enough power. I do have a powered (up to 3A) USB2 on the PPBA. I did mean to try that and will next clear night. 

As for the guide interval I was told when I bought the mount to use 7 or 8 secs exposures. Any less and apparently it gives the mount too many instructions and causes bad guiding. The drop down at the bottom says Camera exposure duration. I can't find anything to set the guide interval. Maybe if I set everything I can to auto and try that ? 

Thank you for the help all I can do is try all of the suggestions folks have made and hope something works. 

[PhilB61]

Sorry I don't think I explained very well. I would definitely go with the guide interval that you have been advised, this is what is set in the drop down box as camera exposure duration. However this total guide exposure duration is actually a stack of a number of shorter duration frames, the number of frames is very briefly displayed bottom left, same place as it displays looping etc, after each total guide exposure (with a fast computer it is almost a blink but my slower main imaging laptop is easier to see. With my ASI120 the duration of these "sub frames" can be set by clicking on the camera icon next to the brain icon, as can gain and USB speed etc. The actual settings available here seem to vary by camera/driver. Altering these setting will not affect your guide interval, but could impact the guide data/processing load which may be beneficial for the reasons I poorly explained in my previous post.

[Crispy5573]

Hi Phil thank you for that I didn't quite get your point the first time  . I have noticed the camera icon next to the brain but it's always been greyed out so not accessible not sure why as it's an ASI camera like yours. I was hoping to get a chance tonight again but as usual the forecast in the morning changed by evening. 

I have a few things to try now next time I can and I'll post back here if I find the solution. 

Thank you for your help. 

[michael8554]

1.  Please post a LINK to a GuideLog, make sure the date and time in the selected filename are relevant.

2. Are you using basic 7 second exposures, or Delayed 7 sec exposures ?

3. Describe your OAG setup.

Were you able to lower the prism enough to actually show a shadow on the Flats, before backing off ?

Is the prism adjacent the long side of the QHY 268m ?

Are both cameras focussed ?

Michael

 

Edited August 26 by michael8554

[Stuart1971]

16 hours ago, Crispy5573 said:

The mount works fine as I was saying I was getting only 0.3 and sometimes below total error at times. There is no clutch as such the two drive pins (they are about 2mm pins that drive the wheels) are spring loaded and driven direct by the stepper motors. I'm pretty sure it is a software problem and not mechanical. My next thought is to connect the guide cam direct to the NUC and run Phd2 on it's own and not through NINA in case there is some sort of conflict with anything else going through the PPBA but I can't see it doing anything. 

It’s the screw pressure on those pins that I was referring to, the pressure has to be just right…hence they now supply a torque driver with the latest scopes for this job.

[Crispy5573]

Hi Stuart I do remember seeing they supplied those now. It's not been a problem as far as I know and I doubt it is having this effect on Phd2. The screws are just finger tight. That was emphasized when I bougjt the mount.

Thanks for your help.

Ron

[Crispy5573]

1 hour ago, michael8554 said:

1.  Please post a LINK to a GuideLog, make sure the date and time in the selected filename are relevant.

2. Are you using basic 7 second exposures, or Delayed 7 sec exposures ?

3. Describe your OAG setup.

Were you able to lower the prism enough to actually show a shadow on the Flats, before backing off ?

Is the prism adjacent the long side of the QHY 268m ?

Are both cameras focussed ?

Michael

 

Hi Michael,

I will get the guide logs from my obsy NUC and post them here today. 

I'm not using delayed as my mount doesn't have the encoders.

I'm not sure what you mean by describe my OAG set up? It is in the imaging train after the field flattener/corrector and before the filter wheel and has the ASI174 screwed on top after a 15mm M42 extension. It's connected via USB3 to a Pegasus PPB advance and from there to my NUC.

I haven't noticed any shadows on my flats. I just lowered it enough to get a star image. Having said that I have been having a problem with flaring on very large stars something like two extra short thick diffraction spikes which I have been wondering if they are caused by the OAG prism. See attached image.

The celestron OAG is designed to rotate to pick up stars if not enough are visible in whichever orientation it is in but I generally have it adjacent to the long side of the 268 sensor.

Yes both cameras are in focus. 

Ron

[Crispy5573]

6 hours ago, Crispy5573 said:

Hi Michael,

I will get the guide logs from my obsy NUC and post them here today. 

I'm not using delayed as my mount doesn't have the encoders.

I'm not sure what you mean by describe my OAG set up? It is in the imaging train after the field flattener/corrector and before the filter wheel and has the ASI174 screwed on top after a 15mm M42 extension. It's connected via USB3 to a Pegasus PPB advance and from there to my NUC.

I haven't noticed any shadows on my flats. I just lowered it enough to get a star image. Having said that I have been having a problem with flaring on very large stars something like two extra short thick diffraction spikes which I have been wondering if they are caused by the OAG prism. See attached image.

The celestron OAG is designed to rotate to pick up stars if not enough are visible in whichever orientation it is in but I generally have it adjacent to the long side of the 268 sensor.

Yes both cameras are in focus. 

Ron

Sorry been out all day. Here is the guide log from last time I used it 24/8/2024 attachedPHD2_GuideLog_2024-08-24_212903.txtPHD2_GuideLog_2024-08-24_212903.txtPHD2_GuideLog_2024-08-24_212903.txtPHD2_GuideLog_2024-08-24_212903.txtPHD2_GuideLog_2024-08-24_212903.txt

PHD2_GuideLog_2024-08-24_212903.txt

[michael8554]

Your problems are nothing to do with faulty software.

1. Poor Calibration with large Orthogonality error "ortho.err. = 26.0 deg"

The Cal was riddled with "Star Lost" messages.

Do you have to use 8000ms exposures to get guide stars, because the prism isn't low enough (see point 2 below) ?

You can safely Bin the guidecam as the pixel scale is only "Pixel scale = 0.42 arc-sec/px,"

Which will increase sensitivity, allowing faster exposures.

Best to Cal pointing near south and around Dec = 0, not  "Dec = 57.8 deg,  Az = 49.7 deg".

The PHD2 Cal Assistant will point the mount correctly, and clear Dec Backlash, before starting the Cal.

2.  The ZWO ASI174MM has a large sensor and big pixels, ideal for OAG use with a large prism.

But I suspect that " just lowered it enough to get a star image." is not low enough.

The prism must be just clear of shadowing the imaging camera, where the brightest and least distorted part of the available light cone is.

If you can't achieve that you have the wrong OAG.

Note how poor the focus is "HFD = 7.38 px".

PHD2 will guide with the distorted stars that OAG usually has, but you've got to offer the best possible to start with.

3.  Periodic Error is about 10arcsec peak to peak, the Trident spec is :

     3" Peak-to-peak periodic error (16 minute period)

     8" Peak-to-peak periodic error (24 hour period)

4. Imaging at 0.29arcsec/pixel, you will need to guide at well below that figure.

But RMS figures are RA = 0.65 , Dec = 0.33.

And Peaks are RA = 8.06, Dec = 2.61.

So stars in images will be elongated and bloated.

I would Bin both cameras and seriously speed up the exposures, to give PHD2 a chance to do it's stuff.

Give the PHD2 PPEC RA Algorithm a try for the PE. 

Michael

 

 

Edited August 27 by michael8554

[Crispy5573]

Thank you Michael I will take that on board and try to correct these things. The reason I thought it was software related was calibration and guiding were fine until a windows update followed shortly after (the same night I think) a power outage which caused chaos with everything. 

I do bin my main camera 3x3. My stars are usually fine, no trails and not badly bloated. 

I've attached an image from two or three nights before my last session when things weren't so good just to show the stars. They aren't over processed because I'm not good enough at that lol.

[Crispy5573]

That's not a very good example to judge by really I don't think. That was taken with 6 minute subs btw. Here is a jpeg of a single 10 min sub binned 2x2 from April this year. I was still trying to get the scope set up properly but there's not a lot of star movement for 10 mins @2850fl. By the way if you're ever in Nottinghamshire I would happily pay you to use your knowledge and set it up for me

Edited August 27 by Crispy5573
Added comment