Guiding Assistant and Dec backlash - EQ8

[Dave Newbury]

Guiding on my EQ8 has been degrading for past few weeks/months. Was getting RMS in the 0.4 to 0.8 range. Now lucky if I can get it down to 1.0.  Tried running the Guiding Assistant for the first time and am getting a rather large backlash measurement (58,000ms !).

I presume this means I will need to dig into mechanical adjustment of the mounts Dec gear meshing.  However, before I start on that, is there anything else I should consider?

Any suggestions appreciated.

DaveNL

[peter wolsley]

Dave,

58 seconds to cater for an 85 arc-second backlash implies an autoguiding rate of only 10%.  Check your autoguiding rates...they should be a minimum of 50%...preferably 75%.  Also please supply your guide and debug logs for this guiding assistant run.

Peter

[peter wolsley]

Also your RA and DEC aggression settings are too low.  We will know more once we see your guiding and debug logs

Peter

[donwat...@outlook.com]

I am new at this. What is range of values I should expect for high frequency variations? What is a good value? Is this affected by mount at all or just seeing?
Thanks
Don Waters

[oliv deso]

Hello,

The dec backlash measurement is strange : the backlash seems to have 2 steps,. At the begining it is OK then there is another step at 58s. Why? Something has moved? the camera or its cable? or else?

Anyway, maybe a false measurement of the mount backlash itsef...maybe the mount backlash is not the only reason.

Now concerning the EQ8, (and other worm geared mounts) I am also guiding in one direction in Dec with a low agressiveness, (30 like you) which gives best results IMHO.

I mean, this is the only way I found to catch lower FWHMs than 2". In that case, the DEC guiding must be extremly smooth and damped. Typically  I get 0,3" RMS for RA and 0,2" on DEC with EQ8. In case of very good seeing, this could be lower, at 0,25 and 0,15.

With the AP900 and 1100, I use the same process, the Dec is the same of course, the RA is just a little bit smoother, which helps at 1,5". The only concern is the point where/when the DEC drift changes of direction. It occurs once a night...

If you have a fixed setup, you will learn, where is that point and in what direction. you should guide.

Of course the PHD utility is ver useful' for that also.

Of course it depends a lot on the seeing...when the seeing is average or poor, guiding in both direction is sometimes necessary...especially if there is some play somewhere in the DEC assembly or telescope assembly.

When I saw your DEC graph, poor seeing is the first thing that I suspected...

Still thinking...

Olivier

--
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.

[oliv deso]

Hi again Dave

Another thing that may explain that : tje calibration data. If the calibration is not perfect, there will be a coupling between RA and DEC movement, which may explain the uncontroled DEC.

- enable the "suppose Ra and DEC orthogonal" option

- perform a "good" calibration on the equator, near the meridian, and re-use it forever (as long as you don't modify your guide cam orientation)

Olivier

[Dave Newbury]

Peter,

Thanks for the response.  Just realizing that I had switched to pulse guiding a ways back and yes, the guide rate is set to 0.10 - doh!   The aggressiveness settings had been higher, but I started to play with them recently when trying to address the guiding issues. Attached are extracts from the guide and debug logs. The debug log was huge (app been running for a few days - disconnected - and not clear to me what the associated dates are). Hopefully it is useful.  If not, I will try again next clear night (with higher guide rate!) and start new log files.

Olivier:

Thanks for the input - some good suggestions for me to check out.  I agree the backlash graph is strange - I don't think there was a cable or other snag/interference. Another time I tried running guide assistant it seemed to give up trying to measure the backlash after a long period of time.  I did try one-direction only guiding, as suggested by the GA and this helped but only lasted about an hour before guiding went wonky.

My seeing is not typically good which is why I was surprised when GA suggested lowering min mo.

DaveNL

[Andy Galasso]

Hi Don,

Welcome to PHD2.

The high frequency value is a measurement of high frequency variations in the star position that cannot be guided out. This generally means fluctuations from atmospheric seeing, but it could include other sources of high frequency fluctuation like vibration of the mount for example.  The purpose of the high frequency variation measurement is to establish a good baseline value for the Min Move setting. Since the the high frequency motion (seeing) is random it cannot be guided out and we would like phd2 to ignore the high frequency movements. The min-move setting provides the threshold for what to ignore.

A "good" value for high frequency variations would be 0, but unless you're imaging from space you won't get that!

Andy

[Andy Galasso]

Dave,

That backlash measurement graph could be indicating slippage in the dec gear train. We see a narrow plateau at the top where the direction first reversed from N to S which probably represents your actual backlash.  The wider plateau is indicating that while PHD2 was sending S guide pulses the dec axis stopped moving, failed to move for a very long time, then eventually it started to move again.

I think you should take a close look at you dec axis drive and make sure there is nothing that is slipping (a loose coupling somewhere? loose clutch and/or large weight imbalance?) and that there is no end-play in the worm (worm assembly moves laterally while worm wheel remains stationary), or any other mechanical issue that could result in the axis not moving while the dec motor was moving.

Andy

[bw_msgboard]

Hi guys.  I think it could also be a symptom of a dec gear system that is over-tightened.  The initial, correct response could be allowed by the very small residual relaxation in the gear system.  But at that point, things lock up a bit and the axis stops moving.  This static resistance continues in the flat area but is eventually overcome and the axis begins moving again.  We have seen this pattern before, sometimes on premium mounts that have been poorly adjusted.  This can also be exacerbated by using such a low guide speed, which can make it harder to push through any resistance.  Of course, it’s almost impossible to diagnose complex mechanical behavior by just looking at such a simple software test.  Whatever the reason, I think it needs to be explored and eliminated if possible – this doesn’t look like simple backlash to me either.

 

Bruce

 

---

--

[bw_msgboard]

Hi Dave.  I thought I’d come back to the original question now that we have a guide log.  Your last 2.5 hour session showed a total rms of 1.15 a-s but the Dec was somewhat better than the RA (0.75” vs. 0.87”).  So whatever mechanical problems you have in Dec don’t seem to be the limiting condition here.  The combination of the extremely slow mount guide speed and the low RA aggressiveness made the mount very slow to respond to RA excursions. Beyond that,  I think much of your problem here was probably seeing.  The GA run shows a lot of rapid fluctuation in Dec along the lines of 3 arc-sec peak-to-peak:

 

 

Unless something is really loose or you were operating in windy conditions, this is probably an implicit measure of seeing.  It’s not likely you’re going to be able to guide this out.  Have you checked with local seeing forecasts and jet stream maps to see if poor seeing conditions were expected?  In many locations, winter and early spring often produce poor seeing.

 

Regards,

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Dave Newbury

Sent: Wednesday, April 18, 2018 5:53 AM
To: Open PHD Guiding

--

[Dave Newbury]

OK - lots of great feedback - thanks to all !  I had a little time and small window of clearing last night, so tried again. Only two changes were: setting guide rate to 0.7 and doing a fresh calibration. It did not make the significant impact I was hoping for :=)    Attached is guide log. Here is screen shot of Guide Assistant result. Backlash measurement is much better compared to previous.

After I completed the GA, I changed to ST4 guiding - seemed to be a little better, but not a lot. In the past, I have generally had better luck guiding with ST4 instead of pulse guiding. But that is probably a user limitation :=)

Additional info:

• This is an EQ8 in a dome
  
• EQ8 encoders are disabled
  
• Scope/Mount has been balanced as best I can possibly do (EQ8 is prone to off-balance issues).
  
• Using EQMOD
  
• Using ASCOM driver for camera (Lodestar X2) instead of direct driver.
  
• Dec  (and RA) clutches are "firm" but "crazy tight" (I've seen reference elsewhere that over-tightening can warp plate and impact guiding)
  
• My seeing is often not good but didn't think it would be the biggest factor (especially for backlash measurement issue)
  

Questions:

• Where does one change the guide rate when using ST4 guiding?
  
• I presume select "on-camera" in PHD2 sets the guiding via ST4 regardless of how EQMOD is configured?
  
• If seeing conditions are bad, wouldn't the Guiding Assistant suggest larger value for Min Motion?

Next steps:

a) Try again but with "Assume Dec orthogonal to RA" checked.

b) Any other "software" changes/tweaks that are suggested (maybe increase aggressiveness some more).

c) Try playing with Dec clutch tightness.

d) Deeper dive into mount/gear/mesh mechanics

Thanks to everyone again for you help.

Dave

[bw_msgboard]

Hi Dave.  This looks like the same log file you already sent us so there’s not much more we can say.  But 8 sec of Dec backlash is really pretty bad, I would say you should dig into the mechanical issues if you want to improve this.  Other answers to your comments/questions are shown below.

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Dave Newbury

Sent: Friday, April 20, 2018 9:26 AM
To: Open PHD Guiding

Subject: Re: [open-phd-guiding] Re: Guiding Assistant and Dec backlash - EQ8

OK - lots of great feedback - thanks to all !  I had a little time and small window of clearing last night, so tried again. Only two changes were: setting guide rate to 0.7 and doing a fresh calibration. It did not make the significant impact I was hoping for :=)    Attached is guide log. Here is screen shot of Guide Assistant result. Backlash measurement is much better compared to previous.

 

 

 

 

 

 

 

After I completed the GA, I changed to ST4 guiding - seemed to be a little better, but not a lot. In the past, I have generally had better luck guiding with ST4 instead of pulse guiding. But that is probably a user limitation :=)

 

Not likely.  The only way to know is to run multiple sequences of back-to-back guiding sequences of relatively long duration, all on the same night.  I think it’s pretty unlikely you’ll do better with ST-4. and there’s really no reason you should unless EQMOD is getting in the way.  If you’re getting 1.1 arc-sec total RMS as you did in the 4/10 log, you may have reached the limits of the mount.  You should definitely be getting nice round stars based on the 01:30 guide session on 4/10.

 

 

Additional info:

·         This is an EQ8 in a dome

·         EQ8 encoders are disabled

·         Scope/Mount has been balanced as best I can possibly do (EQ8 is prone to off-balance issues).

·         Using EQMOD

·         Using ASCOM driver for camera (Lodestar X2) instead of direct driver.

·         Dec  (and RA) clutches are "firm" but "crazy tight" (I've seen reference elsewhere that over-tightening can warp plate and impact guiding)

·         My seeing is often not good but didn't think it would be the biggest factor (especially for backlash measurement issue)

 

Questions:

·         Where does one change the guide rate when using ST4 guiding?

This has to be set in the mount controller somehow, presumably through the hand controller interface.

·         I presume select "on-camera" in PHD2 sets the guiding via ST4 regardless of how EQMOD is configured?

Yes, guide commands will be routed through the camera and then on to the mount controller.  EQMOD won’t ever receive any PHD2 guide commands in this scenario.

·         If seeing conditions are bad, wouldn't the Guiding Assistant suggest larger value for Min Motion?

Yes, but only if you’re not greatly under-sampled on the guiding assembly.  Unless something has changed, it looks like your guiding image scale was over 7 arc-sec/px so that is seriously under-sampled.  In that case, you are mostly impervious to seeing,  The GA uses a “floor” value for min-move and that’s what you’re seeing – it represents a rational estimate of the accuracy in computing the guide star location.

 

 

Next steps:

a) Try again but with "Assume Dec orthogonal to RA" checked.

 

It’s ok to try this, but it essentially just paves over whatever pot-holes you have with the mount.  I don’t think it has much bearing on your current problems.

 

b) Any other "software" changes/tweaks that are suggested (maybe increase aggressiveness some more).

c) Try playing with Dec clutch tightness.

d) Deeper dive into mount/gear/mesh mechanics

 

It would be good if you could upload a guide log with your updated guide speed settings and one that contains a fresh calibration.  In the last log I saw, your Dec guiding wasn’t really the limiting condition.

 

Hope this helps,

Bruce

 

Thanks to everyone again for you help.

 

 

Dave

On Wednesday, 18 April 2018 14:14:26 UTC-2:30, Andy Galasso wrote:

Dave,

 

That backlash measurement graph could be indicating slippage in the dec gear train. We see a narrow plateau at the top where the direction first reversed from N to S which probably represents your actual backlash.  The wider plateau is indicating that while PHD2 was sending S guide pulses the dec axis stopped moving, failed to move for a very long time, then eventually it started to move again.

 

I think you should take a close look at you dec axis drive and make sure there is nothing that is slipping (a loose coupling somewhere? loose clutch and/or large weight imbalance?) and that there is no end-play in the worm (worm assembly moves laterally while worm wheel remains stationary), or any other mechanical issue that could result in the axis not moving while the dec motor was moving.

 

Andy

 

--

[Dave Newbury]

Thanks Bruce for all the input.

Although the guide log filename has the same date, it does contain the newer guide data (look at the dates inside the file). Its become a rather large file so before uploading, I removed all the old data. And yes, I should restart PHD2 to force a new log file :=)

Dave

[Daniel Lorenzo]

Hi, I'm reading this in 2020... I also have a crazy DEC backlash on my EQ8.

Did you end up finding a solution or workaround?