Guiding/mount help needed

[Filroden]

Hi

Last night was not fun. There was a clear forecast for the entire night and it had been clear all day so I was really looking forward to capturing HaLRGB data on the Monkey Head Nebula before the moon rose. Now I put my hand up and admit that some of the issues I faced last night were entirely user created but I've had repeated issues for two or three sessions and I think it's time to ask for advice before the dent in the wall gets much bigger!

The kit: I'm guiding an AVX mount/Esprit 80 using OAG. I have a ZWO 1600 camera for the main imaging and guide with a ZWO 224. The optical train is Scope > FF > spacers > OAG/224 > FW > Camera/1600. Spacing is not quite spot on, Pixinsight suggests my stars a slightly more eccentric on the left of the image compared to the right (values of eccentricity range from 0.55 on the left down to 0.4 on the right).

Issue 1 - PHD2 calibration, non-orthogonality and poor RA tracking

These may or may not be related issues, but I've lumped them together. In the last few sessions I have calibrated using my target and PHD2 reports a non-orthogonal issue saying the expected RA rate is much higher than it is expecting. Last night I changed my calibration process, no longer calibrating for each target (a setting in SGPro) but finding a target on the close to the meridian at zero dec to calibrate at the start of the session and using that for the rest of the night. The issue has been the same using both methods.

Here's an image of last night's calibration (using the area around Uranus as it was just about in the right spot):

The RA line looks ok, but the dec looks all over the place. Is the second fainter dec line a result of backlash? What's causing the variations when moving north? I don't remember seeing it look like this last year.

I've previously achieved tracking errors of 1" or less over long periods. In the last few sessions the dec errors have been their usual 0.6" or so, but the RA errors have increased a lot, so the overall error rate is now approaching 2" and therefore noticeable at my imaging scale. Why would the RA error rate increase?

I've got the scope balanced in RA but it's not well balanced in dec. As I've been imaging close to the zenith I had hoped this wouldn't be an issue but my last two targets have been lower. Could dec balance be an issue? It's currently back heavy.

Issue 2: ZWO driver conflicts

This is really becoming a pain. I use ZWO cameras to image and to guide. Again, in the last few sessions PHD2 and/or SGPro have been having fights over cameras. I think SGPro accesses the ZWO 1600 via its ASCOM driver (native support is in the latest v3 beta which I'm not using yet). PHD2 is set up to access the ZWO 224 using the native driver. Both native and ASCOM drivers are up to date. I'm running the latest live versions of PHD2 and SGPro.

What happens is that I start a sequence and at some point PHD2 starts to report it has lost the star and at a similar point (I don't know if before or after) it swaps to the second camera (the ZWO 1600) which then causes SGPro to hang on image download. This is a real issues as I cannot leave the gear to run a sequence.

One thing I noticed which was a new issue last night was SGPro was taking a lot longer to download/analyse the image during framing/focusing. If I restarted SGPro it would speed up again and then slow down. I wondered if this was also a driver conflict issue?

I wondered if any of this was related to the recent Windows update to "fix" the recent processor security flaw as I'd read this broke ASCOM. However, these issues predate that fix and last night everything seemed to connect and work fine, i.e. SGPro and PHD2 both connected to the mount via ASCOM and guiding/dither commands worked from PHD2 and goto commands worked from SGPro. Likewise, SGPro could access and control the ASCOM focuser driver with no issues.

What am I doing wrong? (Apart from picking the wrong hobby!)

To end on a positive note: after 2 hours of "imaging" last night I achieved 5 x 240s images of NGC 2174! So here's a quick integration (no flats) with no processing other than an initial gradient division to remove vignette and a quick histogram stretch:

And I've realised that depending on orientation, there are two monkey heads. As presented above I can see a smiling lemar but if you rotate it 180 degrees I can see an orangutan.

[Filroden]

I've attached a couple of older logs which show the calibration issue on other occasions (plus some more normal looking calibrations).

Any advice would be greatly appreciated

PHD2_GuideLog_2017-12-28_160934.txt

PHD2_GuideLog_2017-11-18_175322.txt

[michael8554]

Before calibrating PHD2 you should first move the scope north to  clear the backlash, enough to see the guidestar actually move.

Calibrate once only pointing south at zero dec - PHD2 will compensate for other positions (unless you're ST-4 guiding which means no positional data fed to PHD2).

That ought to get rid of your wandering dec calbration and give similar steps in ra and dec.

Michael

[kens]

Can you also attach the guide log corresponding ti the screen shot in the first post?

[Filroden]

8 hours ago, michael8554 said:

Before calibrating PHD2 you should first move the scope north to  clear the backlash, enough to see the guidestar actually move.

Calibrate once only pointing south at zero dec - PHD2 will compensate for other positions (unless you're ST-4 guiding which means no positional data fed to PHD2).

That ought to get rid of your wandering dec calbration and give similar steps in ra and dec.

Michael

Thanks Michael, I'll do that on next set up. I had assumed that as I was letting SGPro slew and centre the mount to position, it would always end the movements in the same directions but I'm guessing the backlash still isn't clearing.

[Filroden]

1 hour ago, kens said:

Can you also attach the guide log corresponding ti the screen shot in the first post?

Here's the calibration log and the first 6m of guiding until the camera driver bug hit. The second log has a longer guide (16m I think) which covered three more of the subs. The RA is very spikey. In previous sessions, outside of dithers and settling, I don't see spikes larger than about 2.5" in either axis and I can average 1" RMS over 2 hours.

PHD2_GuideLog_2018-01-07_172014.txt

PHD2_GuideLog_2018-01-07_183907.txt

[Nigel G]

Ken I would love to be able to help but don't think I can, I guide using ST4 cable, so don't use EQmod or Zwo cameras. I never have any major issues but occasionally PHD2 will loose the guide star, probably to do with backlash as it will bring the star back after about 30s if I let it run on. A bit of a pain if your 10 minutes into an exposure but it is a rare thing, not bad for an EQ3 though.

I don't do a drift align, just polar align the mount by eye and polar finder, then 2 star align the mount, find target. manual guide a few steps N & W then calibrate. but I do need to re calibrate if I change targets or flip the mount. It would certainly improve if I took more time polar aligning. Need Polemaster really  

I have attached a screenshot of PHD2 guiding by ST4, this seems to be an average guide graph.

But I do like your Monkey head  

Nige.

[Filroden]

24 minutes ago, Nigel G said:

I never have any major issues but occasionally PHD2 will loose the guide star, probably to do with backlash as it will bring the star back after about 30s if I let it run on.

I'm thinking that each of my logs show PHD2 losing the guide star. I'm guessing for one of two reasons:

• the driver bug kicks in and swaps the camera, so PHD2 no longer sees the guide star, or
• it loses the star for whatever reasons - backlash, cloud, bird, plane, superman - and then decides to swap camera

I have a feeling there are a bunch of issues that have all happened at similar times that are making it hard for me to resolve each of them.

I think today I will uninstall all drivers and the ASCOM platform, and do clean installs. I'll also bite the bullet and upgrade SGPro to the latest v3 beta. That way I can use the native ZWO drivers in both PHD2 and SGPro rather than me using a mix of ASCOM and native.

[kens]

Heres what your calibration really looks like. It looks like the fast recenter option is on so its not fully informative but good enough to see that you have a lot of backlash.

The RA (West and East) segments look ok but then PHD2 tries to clear any Dec backlash (yellow) by moving the mount north till it gets a sensible amount of giode star movement. It looks like it never really succeeds then starts the norh clibration Looking at the steps it seems the backlash finally clears about halfway through. But then on the southbound leg it has to deal with the backlash again so the guide star hardly moves at all.

Try adjusting the worm mesh especially on the dec axis to reduce the amount of backlash

The West and North legs are also not exactly orthogonal. So its possible, if not likely, that your polar alignment error is quite highcausing the RA leg to drift north of where it should be. So check your polar alignment error

Other things I saw in your guide log:

• Your exposures are 1s - too short. Try 2s
• You are guiding in dec in both directions (Auto). With this amount of backlash guide in one direction only.
• You are dithering on both axes. Dither in RA only or not at all until you sort out your mount issues.

It is absolutely dritical to get a good calibration or PHD2 simply wont work as it is effetively flying blind

[Filroden]

22 minutes ago, kens said:

Heres what your calibration really looks like. It looks like the fast recenter option is on so its not fully informative but good enough to see that you have a lot of backlash.

The RA (West and East) segments look ok but then PHD2 tries to clear any Dec backlash (yellow) by moving the mount north till it gets a sensible amount of giode star movement. It looks like it never really succeeds then starts the norh clibration Looking at the steps it seems the backlash finally clears about halfway through. But then on the southbound leg it has to deal with the backlash again so the guide star hardly moves at all.

Try adjusting the worm mesh especially on the dec axis to reduce the amount of backlash

The West and North legs are also not exactly orthogonal. So its possible, if not likely, that your polar alignment error is quite highcausing the RA leg to drift north of where it should be. So check your polar alignment error

Other things I saw in your guide log:

• Your exposures are 1s - too short. Try 2s
• You are guiding in dec in both directions (Auto). With this amount of backlash guide in one direction only.
• You are dithering on both axes. Dither in RA only or not at all until you sort out your mount issues.

It is absolutely dritical to get a good calibration or PHD2 simply wont work as it is effetively flying blind

Blimey, that's some backlash! I feel sorry for PHD2 now.

I think any mechanical adjustments may be beyond me but I'll do some searching to see how I adjust the worm mesh. 

I can definitely do the other things. I used 2s for exposures until the other night when I dropped it to 1s. I think it was a result of a recommendation from the guiding assistant but I can't remember. I'll switch to guiding dec in one direction and stop dithering until I get better results.

On PA, I use SharpCap and it consistently reports PAs of under 1' each session. When I ran PHD2's guiding assistant, it reported PA to be under 10' (it didn't agree with SharpCap but was within tolerance so I accepted it). Would a PA up to 10' be enough to cause the RA leg to drift? I've never seen field rotation in subs over a 2 hour period.

[RayD]

12 minutes ago, Filroden said:

Blimey, that's some backlash! I feel sorry for PHD2 now.

I think any mechanical adjustments may be beyond me but I'll do some searching to see how I adjust the worm mesh. 

I can definitely do the other things. I used 2s for exposures until the other night when I dropped it to 1s. I think it was a result of a recommendation from the guiding assistant but I can't remember. I'll switch to guiding dec in one direction and stop dithering until I get better results.

On PA, I use SharpCap and it consistently reports PAs of under 1' each session. When I ran PHD2's guiding assistant, it reported PA to be under 10' (it didn't agree with SharpCap but was within tolerance so I accepted it). Would a PA up to 10' be enough to cause the RA leg to drift? I've never seen field rotation in subs over a 2 hour period.

Hope you get it sorted, I feel your frustration.

I adjusted my AVX (they aren't that good from the factory in many cases) and it was actually much easier than you may think, and well worth doing.

There is a pretty good video here showing how to do it.

[Filroden]

7 minutes ago, RayD said:

I adjusted my AVX (they aren't that good from the factory in many cases) and it was actually much easier than you may think, and well worth doing.

Actually, that looks about the same difficulty as fitting my auto-focuser motor so I might be able to do it. I notice in the video he sets anti-backlash settings. Will this be needed if I run PHD2's guiding assistant ant let PHD2 account for any remaining backlash?

[RayD]

36 minutes ago, Filroden said:

Actually, that looks about the same difficulty as fitting my auto-focuser motor so I might be able to do it. I notice in the video he sets anti-backlash settings. Will this be needed if I run PHD2's guiding assistant ant let PHD2 account for any remaining backlash?

No you don't need to touch them, PHD2 will work that out during calibration.

One thing worth checking is that the brass gears are secure on the shaft as they are only secured with a small grub screw, which can come loose.

[kens]

51 minutes ago, Filroden said:

Blimey, that's some backlash! I feel sorry for PHD2 now.

I think any mechanical adjustments may be beyond me but I'll do some searching to see how I adjust the worm mesh. 

I can definitely do the other things. I used 2s for exposures until the other night when I dropped it to 1s. I think it was a result of a recommendation from the guiding assistant but I can't remember. I'll switch to guiding dec in one direction and stop dithering until I get better results.

On PA, I use SharpCap and it consistently reports PAs of under 1' each session. When I ran PHD2's guiding assistant, it reported PA to be under 10' (it didn't agree with SharpCap but was within tolerance so I accepted it). Would a PA up to 10' be enough to cause the RA leg to drift? I've never seen field rotation in subs over a 2 hour period.

You shouldn't see much drift on the RA leg with 10' or less. More likely then that the dec axis was floating a bit. Another thing you can try next time you calibrate is to make sure that the last movement of the mount is north and west to remove any backlash before you start