Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

[Eric Watson]

Hello, 

I am still learning to use the software and look to be having some pretty bad DEC backlash if I am understanding the views from the history, I hope I don't have to send out my AVX mount for service. But here is my setup:

Celestron 8SE 

Celestron AVX mount

ZWO 60mm f/4 guidescope

ZWO ASI224MC USB 3.0 camera connected to guidescope with a Celestron Omni 2X barlow

First question is regarding the advanced settings guiding tab, the guide scope focal length is 280mm, but I'm wondering to I need to take into account the barlow lens when entering that information into the calibration and if so I think my focal length is 812mm taking into account the camera pixel size (3.75)and pixel resolution of 1304 x 976. Is it the 280mm or 812mm?

I have tried making the mount slightly east side heavy for RA and camera heavy or even balance on the DEC and I use a Polemaster for my alignment process.  I have done a manual guide to send pulses and I get movement of the star after I send a pulse in each direction. Any suggestions on correcting or am I actually OK and looking to much into it?

Thanks

Eric

https://openphdguiding.org/logs/dl/PHD2_logs_vstQ.zip

https://openphdguiding.org/logs/dl/PHD2_logs_ljGb.zip

[bw_msgboard]

Hi Eric.  I haven’t looked at your logs yet but maybe there’s no need.  You *do* have to enter the effective focal length of the guide telescope – that’s what determines the image scale and what PHD2 sees in the guide frames.  So you should re-run the new-profile-wizard and input 812mm if that’s what result the Barlow produces.  How are you determining the amplification factor for your Barlow – have you measured it to be the 2.9X you mention here?   This also means the numbers you’re looking at currently are “too big” by a factor of nearly 3X, so I can imagine they might look pretty bad. J  

 

For anyone else reading this thread, the same principle applies if you’re using a focal reducer with the guide camera – the guiding image scale is determined by all of the optical elements sitting between the camera sensor and the sky.

 

Regards,

Bruce

 

 

---

--
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.
For more options, visit https://groups.google.com/d/optout.

[bw_msgboard]

Hi Eric.  With the bad value for the focal length, it’s too hard to disentangle the results.  I see you did a GA run and tried to measure backlash but the test didn’t complete.  You should upgrade to the 2.6.5dev3 release which will make sure you always get a result.  It *looks* like there’s a huge amount of backlash but I’d like to see another test with the latest software and the correct focal length.

 

I’m not sure how your spreadsheet would calculate the correct focal length in this way.  The thing you’re trying to calculate is the actual amplification factor you’re getting with the Barlow.  It is nominally a 2x Barlow, but that number is probably for use with an eyepiece and will change as the sensor is moved further away from the optical element in the Barlow.  So you would basically need to measure the field of view on the guide frame and figure it out from there.  I’d guess your spreadsheet it trying to calculate the image scale for you, but that’s not what you need to know.

 

Cheers,

Bruce

 

---

From: Eric Watson [mailto:eric.w...@comcast.net]
Sent: Saturday, August 11, 2018 12:55 PM
To: bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Hi Bruce, I found an excel spreadsheet that calculates the new focal length when I entered the scope’s FL, Barlow type, and the pixel size and resolution. 

 

When you get a chance to look at my logs let me know your thoughts on the DEC backlash. I tried uni-directional and if the logs don’t show that I’ll send the other logs from last night. 

 

Thanks for your help, 

Eric

 

 

 

 

 

 

 

[bw_msgboard]

Ok, I’ve thought about this some more and I was able to use the PHD2 backlash test as another way to compute your effective focal length.  That gave me a result of 900mm so your spreadsheet result of 812mm wasn’t all that different.  Either value should be close enough for government work, it doesn’t need to be really precise at this point.  Why don’t you send me the spreadsheet so I can be aware of it.

 

So for the next test:

1. Upgrade to PHD2 2.6.5dev3
2. Increase the mount guide speed settings – via the hand-controller – up as close to 1x sidereal as it will allow
3. Create a new PHD2 profile with the correct focal length and mount guide speed setting
4. Re-do the calibration
5. Let the PHD2 GA try to measure the backlash again

 

Huge backlash is normally caused by a lot of slack in the Dec gear train.  Adjusting that yourself is usually not real hard but it depends on the mount.

 

Bruce

 

---

From: Eric Watson [mailto:eric.w...@comcast.net]
Sent: Saturday, August 11, 2018 1:42 PM
To: bw_m...@earthlink.net
Subject: Re: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

I just found the spreadsheet online and just plugged in my scope, Barlow, and CCD info. If you’d like I can send it to you to see if this is a good tool, but if you know of another tool or means that can help me calculate what the focal length should be I’m up for trying it. I’ll download the new version and run a new GA test and send it on to you. I’m really curious on what path I need to take to get this backlash issue fixed. 

 

Thanks

[Eric Watson]

Hi I’m looking for help from anyone with experience dealing with the DEC backlash issues on the Celestron AVX mount. Bruce has been kind enough to assist but if anyone else is familiar with this mount and can help. Please contact me and I’ll forward any logs I have.

 

Thanks

Eric

 

From: bw_msgboard <bw_m...@earthlink.net>
Sent: Monday, August 13, 2018 10:26 AM
To: 'Eric Watson' <eric.w...@comcast.net>
Subject: RE: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Hi Eric.  Yes, the Dec backlash is really severe, bad enough that we’re still not displaying it correctly in the GA tool.  Here’s what it should have shown you:

 

 

The red points are the north moves, and the green mounts are south.  You can see the south moves never really got going at a reasonable rate, not even after 11+ seconds of time!  This doesn’t look like typical backlash, I think there is some binding occurring in the Dec gear system that keeps the mount from reversing direction correctly.  Somewhat paradoxically, that can sometimes happen when the gear mesh is over-tightened in an effort to reduce the backlash.  Unless you can get this fixed, you will definitely need to guide in only one direction.  I see you tried that but you guiding in the wrong direction.  When you ran the GA test at 22:30 on 8/12, you were getting a very slow Dec drift to the south – which means you needed to guide in the *north* direction.

 

More importantly, I don’t think the RA tracking is working well either.  Look at the behavior during the GA run when guiding was disabled:

 

 

The red line is RA and you can that the tracking wandered off the mark by nearly 60 arc-sec.  This is very large, probably too large for astro-imaging.  I think this also explains why all the oscillations occur in RA when guiding is active.

 

Unfortunately, I think there are significant problems with your mount, so I think you’ll need to work with Celestron or other Celestron-knowledgeable people to improve its performance.

 

I’m going to be tied up for the next day or so at my observatory, so I won’t be able to do much more with this.  Whether intentionally or not, you’ve taken this thread off the PHD2 forum because you’ve been posting only to me.  That’s ok, but it means no one else is aware of what you’re doing or will offer any help while I’m away.  You could forward this message to the forum if you’d like to ask for help from others though.

 

Sorry you’re having all this trouble.

 

Bruce

 

---

From: Eric Watson [mailto:eric.w...@comcast.net]
Sent: Sunday, August 12, 2018 10:55 PM
To: bw_m...@earthlink.net
Subject: RE: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Hi Bruce,

 

A couple things, I was able to get my guide scope focused without the barlow lens. So I updated my profile to 280mm for the scope settings. I then ran several GA test one was a bit longer that I planned, but both are at or over 10 minutes. And I did get some results and recommendations. Still pretty poor on the DEC, lots of correction attempts and quite a bit on RA as well. I’m going to try the uni directional guiding and see how that goes and let you know but here are the logs after my changes.

 

Thanks,

Eric

 

 

https://openphdguiding.org/logs/dl/PHD2_logs_MAnn.zip

 

 

 

From: bw_msgboard [mailto:bw_m...@earthlink.net]
Sent: Sunday, August 12, 2018 8:57 PM
To: 'Eric Watson' <eric.w...@comcast.net>
Subject: RE: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Well, you should do what makes sense to you.  But the empirical measurement I gave you – 900mm – was based on an actual measurement of how far the mount moved in response to specific guide pulses.  I think that’s likely to be a better estimate than something you get by guessing at an optical focal length.  That aside, why are you using a Barlow anyway?  That’s something we rarely see done and probably isn’t necessary for a guide camera with such small pixels.

 

Bruce

 

---

From: Eric Watson [mailto:eric.w...@comcast.net]
Sent: Sunday, August 12, 2018 5:49 PM
To: bw_m...@earthlink.net
Subject: RE: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Hi Bruce,

 

In regards to increasing the guide speed, I have an AVX mount and Starsense autoalign camera. I was looking at the manual and the HC and the settings are between 50 % - 99% of the sidereal speed. I set it to 99% thinking that is as close to 1x sidereal as I can get.

 

I ran the spreadsheet again and measured from the CCD sensor to the focusing element of the Omni 2x Barlow, about 3 inches (76mm) but I haven’t been able to locate the focal length of the barlow but from what I did gather is 2x barlow’s are usually 75mm so I’m going to try that. That but me at 564mm so I’ll give that a try.

 

Thanks

Eric

 

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of bw_msgboard
Sent: Saturday, August 11, 2018 4:03 PM
To: 'Eric Watson' <eric.w...@comcast.net>; 'Open PHD Guiding' <open-phd...@googlegroups.com>
Subject: RE: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Ok, I’ve thought about this some more and I was able to use the PHD2 backlash test as another way to compute your effective focal length.  That gave me a result of 900mm so your spreadsheet result of 812mm wasn’t all that different.  Either value should be close enough for government work, it doesn’t need to be really precise at this point.  Why don’t you send me the spreadsheet so I can be aware of it.

 

So for the next test:

1.     Upgrade to PHD2 2.6.5dev3

2.     Increase the mount guide speed settings – via the hand-controller – up as close to 1x sidereal as it will allow

3.     Create a new PHD2 profile with the correct focal length and mount guide speed setting

4.     Re-do the calibration

5.     Let the PHD2 GA try to measure the backlash again

[peter wolsley]

Eric,
I am a Celestron CGEM owner...there are several AVX owners on this forum that can also give some advice.
I have been studying your guide logs and a few things come to mind. Specifically the last logs you put up from Aug 12th.  You had some sections where RA oscillated badly and other sections where the oscillations were not present.  The big difference I see that explains this is your choice of exposure time.  For some sections you had chosen exposure times as short as 50mS which is way to small for guiding.  These very fast (500mS and shorter) exposures are great for studying very fast oscillations or for users with adaptive optics.  The sections that showed wild oscillations were using exposure times of greater than 2000mS.  I would suggest you set your exposure time to 1000mS and maybe consider shortening this time to no less than 500mS. Your RA oscillations should go away and your RMS guiding error might get a little closer to 1 arc-second. I am not seeing much of a residual periodic error in your RA guiding which means either you are one of the lucky ones or you have done a great job training your mounts periodic error logic.  What also concerns me is that when you are guiding your RA axis is having to move a lot in one direction.  This could be a tracking problem as Bruce has alluded to.  It also could be a poor polar alignment which is what I am leaning towards.

It's difficult to assess your DEC guiding because you only used it once and it was configured for one direction only (South).  Virtually all of the calibrations show terrible values for DEC which Bruce has demonstrated with his graph.  I think these poor calibrations are caused by poor polar alignment.  I have seen in the past that if the polar alignment is poor then it can be almost impossible to get a good calibration.  The very last section of your 234120.txt guiding log is where you used DEC guiding...it shows three things. 1)RA oscillations which I have already discussed 2)Poor DEC guiding that drifts to the North and never manages to get back to zero deviation. 3)The overall movements of your DEC axis during this time are flying away towards the South which is typically a big indicator of a very poor polar alignment.  What I believe is happening during your calibrations is that when it comes time for PHD2 to issue South pulses the net result is that the star barely moves because the South pulses are all that it preventing your guide star from drifting to the North.  Instead of moving the star to the South PHD2 is just able to stop the star from drifting even more to the North.

I know you said that you use a Polemaster for polar alignment.  All the reports I have seen from others are that the Polemaster is a great device.  I think you should start afresh and polar align your mount using the polemaster, align/calibrate your mount, calibrate PHD2 and then perform a drift alignment using PHD2.  If the PHD2 drift alignment proves you polar alignment is great...then I'm a fool.  Then go ahead and guide using an exposure of 1 second.  Run the guiding assistant and then guide for 1/2 hour minimum being sure to avoid saturating the signal.  Send us the guide and debug logs and we will have another look.

Peter

[Eric Watson]

Hi Peter,

 

Thank you very much for looking hat my logs and your suggestions. I will try a fresh PA and then run the DA tool again. I do a polemaster PA every time I get started and it’s not that hard so I don’t think I’m screwing that up, that’s why I bought it for the simplicity.  It’s making me wonder about this mount and the DEC issues. My routine has been after I do a PA with the polemaster I’ll start PHD2 and Stellariumscope and Stellarium. I’ll slew to the meridian and equator and look for a star, do a manual guide and then calibrate.  Then I’ll test with the DA tool, usually just running the AZ adjustment.  I let it run for some times for usually 5+ minutes while at the azimuth and I’m looking at it and it will get down between 0 and 0.5 arc sec but then it will begin to fluctuate and the DEC line may move up or down and not stay with the middle line. Shouldn’t it remain somewhat consistent? It makes me concerned with the issues I’m having with the DEC backlash that it may be causing issues with doing the drift align and knocking things off.  Also when doing the DA what exposure speed is best? Also when I do the DA does that data show in the logs? Maybe I can run that for a while and sent that if it does to get your thoughts. It’s going to be cloudy for the next week so I’ll send new logs when it clears up.

 

Thanks

Eric

--

[peter wolsley]

Eric,
I appreciate that you have confidence with your polar alignment method.  I am playing a hunch based upon the last guiding log you submitted.  Your method for getting started every evening sounds fine. I would assume that you use the "Quick Align" method on your AVX to get it tracking.  Polemaster polar aligning is pretty bullet proof but there is a "gremlin in the machine" that I want to flush out.  My procedure for starting up is to Polar Align using my own custom method.  Then I perform the full 2+4 align/calibration on the CGEM.  This is primarily because I like using the Precise Goto feature in the CGEM...it saves me alot of time composing my astrophotos.  I am a computer programmer and I have written a suite of tools that allow me to quickly polar align, align/calibrate the CGEM...right now they are for my own use.

During the PHD2 drift alignment the DEC axis is not moved but the RA axis is being guided.  The mechanical backlash in the DEC axis shouldn't matter but there could be some differential flexure that could affect both RA and DEC over a longer timeframe.  Bruce had mentioned previously that your unguided GA run showed a huge RA drift.  Significant unguided RA drift can be an indicator of differential flexure.

I would suggest that you use a 1000mS exposure time for all of your next testing.  In the mean time you should take a good look at your mount.  Try to identify any looseness in RA or DEC.  Pay special attention to how your guider scope/camera are affixed to your telescope/mount.  You might stumble on something that could explain what's happening.  When you do a guiding run please configure your DEC guiding using the resist switch method and with bi-directional guiding so we can better see any backlash effects and guiding actions.  Set the DEC aggression to 80%.

In the meantime I will take a look at all of the logs you have submitted and see if anything else comes to mind.

Within the last week I had another user having RA guiding issues which turned out to be a corrupted copy of the motor control firmware in the mount.  There are two separate copies of the firmware running in your mount...one for RA and one for DEC.  This user's RA controller software was corrupted.  It may be possible that your DEC controller software is corrupted.  I am really going out on a limb on this one but you may want to do some research on how to update or reload your mount's motor control firmware.

Peter

[Eric Watson]

HI Peter,

 

I have a Starsense camera (not sure if I mentioned that or if it makes a difference) , I usually do a Starsense auto align and let the camera pick up the stars and do a plate solve. Then I calibrate to center the star. I’m going to try the quick align (never tried it before) and add a few reference point stars and see if that helps with the tracking. I’ll set the DEC as you suggested and do some guiding runs to gather some data. I’ll do some investigating on the mount and check for the latest firmware also.

 

When you say looseness in the RA and DEC, I think I follow what you mean, but to be clear this is more towards listening for something odd in the gears, not when I loosen the clutches when doing a balance? Is there a certain speed I should set the motors when listening for issues?

 

Thanks

Eric

[peter wolsley]

Eric,
I am not familiar with the Starsense camera so I do things based upon the previous generation of mounts.  I don't do any plate solving with my set-up.  Now that I understand more about the Starsense I doubt that a quick align will do anything of benefit for you.  I would assume that when you use a Starsense you need to follow the Starsense methodology.  Mixing old and new methods will most likely confuse the issue.  But you know more about Starsense...maybe there is some merit.  I suspect that the quick align and the 2+4 align/calibrate methods are available on your AVX so that the firmware can be standardized and is compatible with a host of different set-ups.

Because I have fiddled with my CGEM a fair bit I look for looseness by engaging the clutches and manually pulling on the DEC and RA to see if the discernable backlash has changed.  I also look for nuts and bolts that may have loosen up with use.  Listening to the mount while it rotates is based upon using the same speed each time and listening for changes in loudness.  This varies a lot with how I have balanced the mount.  I tend to balance the DEC axis slightly camera heavy and balance the RA perfectly.  I feel that the more loading on the RA axis due to unbalance, the more gearbox oscillations will influence my guiding.  The gears in my CGEM are not metal so I never want to put much load on them. Of course, the problem with doing this is that the speed that really counts is the guiding speed which is incredible slow.

Carefully look at your guide scope/camera mounting.  It needs to be fairly rigid.  I don't know if your system suffers from differential flexure but it worth thinking about it while you look at your system. 

The topic where a user had problems with his firmware can be found via this link https://groups.google.com/forum/?fromgroups=#!topic/open-phd-guiding/qKoOz4o_RuU  It's called "Guiding Issues only in one directions".  He invented some methods for diagnosing his mount's faulty RA guiding during daylight hours.  You could use his methods to confirm your pulse guiding is functioning correctly.

Peter

[Eric Watson]

Peter,

 

I checked my mount and there seems to be odd noises when its going in a clockwise rotation on the DEC axis when using the HC,  almost like a skip while the gears are going. Wonder if it would make sense to record it. But it doesn’t seem very “smooth”. It also seems that my RA balance is a bit too heavy on the OTA side, even when I have the weight at the bottom of the rod, so I’m going to order a 7lb weight to add. I’m also going to try using my 71mm refractor and test that out with the DEC when the skies clear. I’m curious to know the variance between my SCT and the retractor on the DEC axis and see if that affects the backlash on that axis from the weight difference. The GS/camera is locked down pretty good.

 

I’m on the fence if I should get the AVX hyper-tuned. I’m kind of nervous to open it up myself, not sure I’ll know what to exactly look for or correct.  Maybe a bit more nervous to reassemble.

 

Thanks for your help,

Eric

 

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of peter wolsley
Sent: Wednesday, August 15, 2018 8:32 PM
To: Open PHD Guiding <open-phd...@googlegroups.com>
Subject: Re: FW: [open-phd-guiding] Newbie: I think I have bad DEC errors and question regarding the calibration/guiding tab

 

Eric,

--

[peter wolsley]

Eric,
My CGEM does make some intermittant noises while slewing but my guiding is fine.  On my CGEM there are two inspection holes that are accessible via two flat-head screws.  When these screws are removed I can see the gearmesh between the brass gears that connect the wormgear to the gearbox.  If you have some kind of simular inspection hole you can watch to see if the gears move smoothly.  There should also be very little gap between the gearteeth of the two brass gears.  There needs to be a tiny gap but thats all.  You can also just gently rest your hand on the end of your telescope as the DEC axis rotates.  All your feeling for is some roughness in the movement.  The movement should be smooth but this comparison is not definitive because the roughness while rotating quickly (2 degrees/sec = 7,200 arc-seconds/sec) can be completely different when pulse guiding (~15 arc-seconds/second).

If you can't full balance your mount then you should not operate your mount.  The motor/gearbox in your AVX (and my CGEM) is made with plastic gears which are intended to be used as a precision clockworks...it's not designed for heavy lifting.  I would suggest that you perfectly balance both RA and DEC for now.  Once we get to the bottom of your issue we can work on optimizing.  I don't think the weight difference between your SCT and refractor should affect you DEC guiding because your mount should be balanced to start with and then off-balanced by a small amount to optimize guiding.  That small amount should be the same regardless of what telescope you are using.  I base the off-balance on how much effort I need to lightly exert with one finger to keep the axis stationary...it's not very much.

I am worried about your inability to balance your RA axis while using the SCT. Don't use the SCT until you get another counterweight.  You don't want to screw up your RA gearbox.

While I have never had my CGEM hyper-tuned I can see the benefit of having someone with experience take a hands-on look at your mount.

You may want to try using your set-up during the day to view a far-off object.  You can use very slow slews to see if the motion is smooth.  You can even use PHD2 to manually guide and verify you have EAST/WEST/NORTH/SOUTH guiding.

Peter

[peter wolsley]

Eric,

To view an object during the daytime you will need to turn off tracking.

Peter

[Eric Watson]

I’ve decided to send the mount to Celestron to get the issues with the DEC and RA resolved. I really hope they clean up those gears because both the RA and the DEC barely make a single rotation without stopping. If they don’t fix it, then I’m sending it off to get hypertuned.

 

Thanks for your help. More to come when I get the mount back from California.

[bw_msgboard]

Hi Eric.  That’s the decision I would have made, these don’t sound like the kinds of problems you should have to deal with.  Good luck with the repair, let us know how it works out.

 

Regards,

Bruce

 

---