PHD Autoguiding problem

[michael8554]

Good stuff from Malcolm.

Best to let PHD2 Autoselect a star - or stars if you're using Multi Star Guiding, recommended.

Pulse north until you see the guidestar move, to take up Dec Backlash, then Calibrate

Shift-Click on the Guide button to force a new Calibration.

A 30second test exposure at Max ISO on a DSLR will show your framing on most galaxies.

Not allowing a minute or two for PHD2 to settle is a common mistake. I see loads of GuideLogs where guiding is stopped after a minute and Calibration repeated for no reason.

Guide Assistant run will assess the Seeing conditions and suggest settings.

Michael

[Wael Hassan]

This is an example of the doubled stars I have now.

[teoria_del_big_bang]

Can you attach the image (the fits file) to a post ?

The stars look elongated as well as doubled up.

Steve

[teoria_del_big_bang]

Silly question maybe but how are you initiating the images - via laptop and usb cable or other method ?

[malc-c]

In that image of the camera screen it seems that all stars are elongated, but the fainter stars haven't left traces as the scope moved, which would suggest that the movement between the start of the exposure and the end may not have been linier.  It's like the exposure started, tracking started, and then tracking advanced slightly at a faster rate on one axis and then resumed at the end of the exposure.  The fact the fainter stars didn't register the short distance gives the impression the exposure is of double stars.  I've seen similar when the backlash in the mount hasn't been taken up after a slew.

Can you confirm your scope is fully balanced?   In this old but useful video explains the issue of balance and CogG

I'll also try and remote into the observatory PC and take some screenshots of my set up which may help

 

[michael8554]

A GuideLog would show what is going on, but a couple of guesses:

If your RA and Dec Errors are markedly different, then RA or Dec is significantly shifting during exposures.

Similar RA and Dec errors means guiding was good, but on a guidestar that was shifting due to a loose guidescope.

Or guiding was just rubbish.

Michael

 

Edited April 8, 2021 by michael8554

[Wael Hassan]

On 08/04/2021 at 12:15, teoria_del_big_bang said:

Silly question maybe but how are you initiating the images - via laptop and usb cable or other method ?

I made a profile on phd with mount HEQ5 and camera zwo 120 mms ....i use usb cable with my Windows tablet 

[Wael Hassan]

On 08/04/2021 at 13:17, malc-c said:

In that image of the camera screen it seems that all stars are elongated, but the fainter stars haven't left traces as the scope moved, which would suggest that the movement between the start of the exposure and the end may not have been linier.  It's like the exposure started, tracking started, and then tracking advanced slightly at a faster rate on one axis and then resumed at the end of the exposure.  The fact the fainter stars didn't register the short distance gives the impression the exposure is of double stars.  I've seen similar when the backlash in the mount hasn't been taken up after a slew.

Can you confirm your scope is fully balanced?   In this old but useful video explains the issue of balance and CogG

I'll also try and remote into the observatory PC and take some screenshots of my set up which may help

 

I balance my mount and scope always .... i use bubble level to balance the RA aziz at 90 angle then return it to home position at 0 angle. Also the same for DEC axis .

I was thinking in the RA and DEC aggression values at my PHD profile

[Wael Hassan]

On 08/04/2021 at 14:50, michael8554 said:

A GuideLog would show what is going on, but a couple of guesses:

If your RA and Dec Errors are markedly different, then RA or Dec is significantly shifting during exposures.

Similar RA and Dec errors means guiding was good, but on a guidestar that was shifting due to a loose guidescope.

Or guiding was just rubbish.

Michael

 

This is my guidelog : 

https://openphdguiding.org/logs/dl/PHD2_logs_fFq6.zip

[teoria_del_big_bang]

30 minutes ago, Wael Hassan said:

I made a profile on phd with mount HEQ5 and camera zwo 120 mms ....i use usb cable with my Windows tablet 

I really meant how do you start the imaging, do you have a usb cable to the DSLR camera and do it from a laptop, I really meant yu are not doing it manually by ressing the camera shutter button ?
Like I said a bit if a silly question.

Steve

[malc-c]

1 hour ago, Wael Hassan said:

I balance my mount and scope always .... i use bubble level to balance the RA aziz at 90 angle then return it to home position at 0 angle. Also the same for DEC axis .

I was thinking in the RA and DEC aggression values at my PHD profile

 

But was the centre of gravity checked as shown in that video... 

The thing is that looking at that image, after the initial jumping around the mount was being guided at the end. - Seeing that the zig-zag traces at the start are first in DEC and then in RA it would suggest that is some part of the calibration run, or similar before PHD finally gets round to guiding.... - It may well be that you are looking at the graphs whilst it does the configuring and thinking there is an issue, when in fact there is nothing really going on? - I must admit however that I don't see that behaviour with my guiding, so maybe there is still more to this than we know.

Can you describe exactly how you have the computer connected to the mount, if the ST4 cable is removed etc... just for clarification.

[teoria_del_big_bang]

27 minutes ago, malc-c said:

I must admit however that I don't see that behaviour with my guiding, so maybe there is still more to this than we know.

me neither, never seen anything like that. There is no scaling shown but assume it is not a super low scaling (not sure how low it can go).

But it does sort of seem something like calibration where something is moving Dec then moving Ra and eventually guiding (although not perfect does eventually settle down a lot and attempt to guide.

Steve

[teoria_del_big_bang]

A quick look at the Guidelog does show quite a bit of calibration going on, but certainly it does not calibrate every time guiding is enabled but there are a number of calibrations.
The calibration seems to last for about 90 seconds, which is pretty normal, and without doing it for myself I cannot remember seeing anything like that when it calibrates - I am not even sure the graph starts recording until calibration is complete and guiding begins (but not sure - I don't look that closely anymore now guiding just works).

There is a lot of info in the log but I am not the best person to interpret it all - but there are plenty on SGL that are.

Steve

 

[malc-c]

I'm no expert when it comes to dissecting PHD logs, but this isn't right

 

 

The axis should be at right angles to each other....

This calibration is better, but then you have all these south steps where the mount isn't moving

Looking at the calibration settings it shows the "Assume orthogonal axes = no" - basically from a bit of googling this is to do with the association of the RA/DEC axis and camera axis - 

Reading up on this error this is what the PHD2 website has to say

Quote

Orthogonality error - the camera axes are normally computed independently even though they should be perpendicular.  The angle calculations do not require great precision, but if they are significantly non-orthogonal, you should repeat the calibration.  If you see repetitive alerts of this type and the axes are significantly non-orthogonal, you'll need to identify the problem and fix it.  Common causes are bad polar alignment, large declination backlash, or large periodic error in RA.  Any of these problems can cause the guide star to move significantly on one axis while PHD2 is trying to measure its motion on the other axis. If you suspect these problems, go ahead and accept the calibration, then run the Guiding Assistant to measure your polar alignment error, declination backlash, and RA tracking error.  In other cases, the mount may not be moving at all, and the measured displacements of the star are just caused by seeing effects.  This sort of problem should be obvious in the calibration graph at the left of the dialog.  If the axis error is relatively small and you are convinced the hardware is working properly, you can avoid further alerts of this type by setting the option to 'Assume Dec orthogonal to RA' in the 'Guiding' tab of the Advanced Settings dialog. But you should do this only if the error is fairly small - otherwise, you are simply ignoring a serious problem.  

 

I think you need to possibly go back to basics and disconnect all the equipment.  Check the balance and centre of gravity of the set up.  Ensure the mount moves smoothly in all directions with no binding or has any backlash (if you lock the axis does the scope still move slightly). 

Ensure your guide scope is lined up with the same optical axis of the main scope.  Point the scope at a bright star and centre it in the field of view of the main camera - have the mount tracking at sidereal rate.  Then use sharpcap or similar application that may have came with the guide camera  to centre the same star in the centre of the  image the guide camera is seeing.  Now use the directional buttons on the handset or EQMOD to move the mount north and watch the motion of the star in both cameras.  If the guide scope image moves at a different angle to the main camera, then rotate the camera until the images move in the same direction, ie if the star moved directly up in the main camera image, then it should do the same in the guide scope image.   

Now polar align the mount.  Sharpcap has an excellent utility for doing this using the guidescope or main scope if the camera has live views. - Its worth the £12 annual cost.

Having confirmed that the mount has no major backlash issues, the cameras are aligned and the scope is perfectly balanced including the centre of gravity then load up just PHD2 .  Connect it to the mount using ASCOM, and the camera.  So now all you have running is EQMOD (ASCOM) and PHD2.  Set the slew rate to 4 in EQMOD and then use the NSWE buttons to locate a suitable start, ideally on the meridian and near the celestial equator - IE south.  

Clear any previous profiles in PHD2, have logging enabled and enable the calibration.  Hopefully you should see a nice set of steps at 90 degrees to each other and guiding will start.  Let it guide for  30 - 40 minutes.   If you still have issues then post up the log files again 

Now like I stated, I'm no expert, and certainly not in PHD2, but the above would be how I would approach the problem.  Hopefully some of the more experienced imagers might jump in and offer their assistance.

 

 

Edited April 10, 2021 by malc-c
typo

[malc-c]

17 minutes ago, teoria_del_big_bang said:

 

There is a lot of info in the log but I am not the best person to interpret it all - but there are plenty on SGL that are.

Steve

 

Steve - I had a bash

 

[malc-c]

On 31/03/2021 at 19:31, Wael Hassan said:

Iam using skywatcher eq 35 ....my main scope is Redcat 51

Heck, I've just looked that up.... the scope is tiny..... and only weights 1.5kg.   I think the issue you have is that you can't balance that small mass on the EQ35-M.  The counterweight may be right to the top of the weight bar and it still have issue as each one weighs 3.4kg.

Skywatcher do a 1.8kg weight - £42 form FLO (but out of stock)

Edited April 10, 2021 by malc-c

[teoria_del_big_bang]

One thing I see in the log is that 

13 minutes ago, malc-c said:

Steve - I had a bash

 

A dam good bash I would say. 🙂 

I was still looking and couldn't see anything jumping out at me.

[teoria_del_big_bang]

5 minutes ago, malc-c said:

Heck, I've just looked that up.... the scope is tiny..... and only weights 1.5kg.   I think the issue you have is that you can't balance that small mass on the EQ35-M.  The counterweight may be right to the top of the weight bar and it still have issue as each one weighs 3.4kg.

A beautiful scope for widefield  but yes I see what you mean about getting balance. 

[malc-c]

2 minutes ago, teoria_del_big_bang said:

A beautiful scope for widefield  but yes I see what you mean about getting balance. 

It wouldn't surprise me if the counterweight bar alone weighed around the same as the scope .... 

But seeing what scope it is, how is the guide scope and camera fitted ????

[malc-c]

On 31/03/2021 at 19:31, Wael Hassan said:

Iam using skywatcher eq 35 ....my main scope is Redcat 51

Can you please post up a photo of how you have set up your scope and mount - nice wide shot of the tripod, mount and scope so we can see what you are using 

[davies07]

Hello,

I'm rather late to this topic so forgive me if I ask some dumb questions.

Wael, I can see from your EQMOD screen that you are located somewhere in the Middle East (30 deg N, 31 deg E)  and your scope is pointed somewhere towards Polaris. 0 deg AZ, 30 deg alt, I hope you are not trying to calibrate your guiding with the scope pointing in that direction.

Assuming your scope is polar aligned -  Scopefocus has a great utility for polar aligning your mount - you should calibrate your mount with the scope pointing south and upwards towards the celestial equator, Hydra, for example. 

 

David

[Wael Hassan]

4 hours ago, malc-c said:

Can you please post up a photo of how you have set up your scope and mount - nice wide shot of the tripod, mount and scope so we can see what you are using 

Sure here are the photos....by the way i just remove the center plate that support the three legs because i want to transfer the mouny when i took the photos....also i dont use the st4 cable during my last imaging ...i disconnect it from the mount.

[Wael Hassan]

1 hour ago, davies07 said:

Hello,

I'm rather late to this topic so forgive me if I ask some dumb questions.

Wael, I can see from your EQMOD screen that you are located somewhere in the Middle East (30 deg N, 31 deg E)  and your scope is pointed somewhere towards Polaris. 0 deg AZ, 30 deg alt, I hope you are not trying to calibrate your guiding with the scope pointing in that direction.

Assuming your scope is polar aligned -  Scopefocus has a great utility for polar aligning your mount - you should calibrate your mount with the scope pointing south and upwards towards the celestial equator, Hydra, for example. 

 

David

Yes i wws doing it near polaris and i discovered it was wrong but i still have problem with guiding ....the RA curve has zigzag rough curve as in this picture 

[Wael Hassan]

4 hours ago, malc-c said:

Heck, I've just looked that up.... the scope is tiny..... and only weights 1.5kg.   I think the issue you have is that you can't balance that small mass on the EQ35-M.  The counterweight may be right to the top of the weight bar and it still have issue as each one weighs 3.4kg.

Skywatcher do a 1.8kg weight - £42 form FLO (but out of stock)

But i use only one and i could balance it by getting to counterweight to the top of the bar.

[Wael Hassan]

4 hours ago, malc-c said:

I'm no expert when it comes to dissecting PHD logs, but this isn't right

 

 

The axis should be at right angles to each other....

This calibration is better, but then you have all these south steps where the mount isn't moving

Looking at the calibration settings it shows the "Assume orthogonal axes = no" - basically from a bit of googling this is to do with the association of the RA/DEC axis and camera axis - 

Reading up on this error this is what the PHD2 website has to say

 

I think you need to possibly go back to basics and disconnect all the equipment.  Check the balance and centre of gravity of the set up.  Ensure the mount moves smoothly in all directions with no binding or has any backlash (if you lock the axis does the scope still move slightly). 

Ensure your guide scope is lined up with the same optical axis of the main scope.  Point the scope at a bright star and centre it in the field of view of the main camera - have the mount tracking at sidereal rate.  Then use sharpcap or similar application that may have came with the guide camera  to centre the same star in the centre of the  image the guide camera is seeing.  Now use the directional buttons on the handset or EQMOD to move the mount north and watch the motion of the star in both cameras.  If the guide scope image moves at a different angle to the main camera, then rotate the camera until the images move in the same direction, ie if the star moved directly up in the main camera image, then it should do the same in the guide scope image.   

Now polar align the mount.  Sharpcap has an excellent utility for doing this using the guidescope or main scope if the camera has live views. - Its worth the £12 annual cost.

Having confirmed that the mount has no major backlash issues, the cameras are aligned and the scope is perfectly balanced including the centre of gravity then load up just PHD2 .  Connect it to the mount using ASCOM, and the camera.  So now all you have running is EQMOD (ASCOM) and PHD2.  Set the slew rate to 4 in EQMOD and then use the NSWE buttons to locate a suitable start, ideally on the meridian and near the celestial equator - IE south.  

Clear any previous profiles in PHD2, have logging enabled and enable the calibration.  Hopefully you should see a nice set of steps at 90 degrees to each other and guiding will start.  Let it guide for  30 - 40 minutes.   If you still have issues then post up the log files again 

Now like I stated, I'm no expert, and certainly not in PHD2, but the above would be how I would approach the problem.  Hopefully some of the more experienced imagers might jump in and offer their assistance.

 

 

How could i solve the backlash or the periodic error ? Forgive me if i asked silly questions as i am still beginner and i have no one here either astro friend or astroshop in my country to give me advice about these issues 😊