Help identifying problem with Declination

[kostas stavropoulos]

I use an heq5 pro with an rc 8'' or an ed80 scope for my imaging with an atik 383l+ and a lodestar sx guide camera via an OAG. Recently i modded my mout with rowan belt mod in order to improve the guiding and reduce/remove a problem i have with my DEC axis. While guiding within 5 mins i get a spike at the DEC axis that leads to trail at the dec axis. The phd does its best to correct this but to no result. In the included log files you can see guide graphs form both the ed80 and the rc 8'' i use. Now it is possible that i chase seeing but i would like some help with identifying the problem. I have tried off-balancing the scope at the mount but i had the same result. Also i noticed the following behavior during calibration. Once the east - west axis is calibrated and phd goes for North - south the mount moves towards the north but when when it comes to the south i get a huge delay at the movement of the mount. The calibration is completed and then phd moves the star to the original position.

[Andy Galasso]

Hi Kostas,

It looks like the backlash problem is still present.  Have you tried using the manual guide tool to get a sense of how much backlash is there?

I also notice from your logs that your calibrations have only about 4-5 North steps.  I think you may get a more accurate calibration with about 10-12 steps.  I see that you have you calibration step size up at 5.7 seconds.  This indicates that you have your guide rate set too low in EQMOD.  Try increasing your guide rate to 0.5 to 1.0x sidereal rate.

Andy

[kostas stavropoulos]

By using phd2 i managed to correct the declination being to tight and the calibration can move the dec axis N-S with no problems. But some new problems have risen and i would like some help identifying them.

1. Last night i started making some tests with my rc 8'' scope and i had a serious problem with calibrating phd. After many attempts i went back to my polar alignment since the review calibration option showed me that my PA was off by 6.8'' at a time, needless to say guiding was terrible. Even though i drift align and i did not have any issues in the past i checked my polar alignment with alignmaster and corrected it under 1'' at both axes. When i redid the calibration phd showed me an error of 3.6''. I tried the phd polar alignment but i had no success since every time i adjusted the AZ phd would show me a terrible drift at the AZ. i would like some input on the subject.

2. I few days ago i was watching a youtube video of getting the most of phd. A question popped in mind. The devs mention that we need to have the graph as flat as possible, if we see a sawtooth graph then we need to adjust the RA aggression, hysteresis and exposure time. Given the fact that i use an oag for my scopes and based on my latest guide graph am i overcrrecting my guiding?  I do realize that my mount is not the ideal but i would like any input on the subject.

I include a log file with two calibration events, the first shows me bad PA while the second calibration is after i corrected with alignmaster. Also any observations at my guiding graph would be much appreciated.

[bw_msgboard]

I’m afraid I don’t quite follow what you’re saying here. The calibration results and calibration review dialog don’t have anything to say about polar alignment.  So I don’t know what you mean.   Are you talking about the Guiding Assistant?  That’s the only place we try to measure and report polar alignment error unless you’re using the drift alignment tool.  Perhaps you misinterpreted what you were seeing.  Did you get a calibration alert or are you just looking at the calibration results for yourself?  Those places talk about an orthogonality error, but that has nothing to do with polar alignment error.  In any case,  I don’t see anything very alarming about either of the calibrations.  The orthogonality is off a bit in the first calibration but that’s probably not a big deal.  Likewise, the guiding results aren’t terrible either – something below 1.3 a-s total RMS.  You can probably improve on this although you may be approaching the limit of your seeing conditions.  I don’t know where you’re located, but if it’s hot there you will need to pay attention to local seeing conditions right around your scope.  Thermal convection can wreak havoc with guiding, and the rapid sawtooth deflections you see could be caused by seeing.  Looking at the guiding graphs, your mount actually seems to be behaving pretty well.  I do see some evidence you may be under-correcting a bit in both Dec and RA.  You can experiment with that by reducing the min-move settings a bit.  You may also want to try increasing your RA aggressiveness a little at a time, but probably after you look at the min-move changes. You are getting down into a range now where you are fine-tuning rather than trying to eliminate gross problems, and seeing may be your limit.  In any case, you’ll need to make adjustments one-at-a-time, then measure the behavior for a good stretch of time (>10 min) before deciding whether it’s improved or not.  This stuff is not fun and it’s easy to be fooled by randomness.  <g>

 

I’d say whatever you did to the mount mechanics is looking fairly reasonable – nice job.

 

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.

[kostas stavropoulos]

Maybe i did not understand well, i assumed that by watching the orthogonality error after the calibration is complete, at the review calibration, that it showed the PA error. I will pay close attention at the guiding assistant next time and read all the documentation again. I did not get any calibration errors. If the orthogonality does not represent PA then what information can i learn form it and how can i improve it if it affects guiding? As for the environmental conditions the temperature at nights is around 20-25 C and this summer is unstable with windy afternoons so seeing maybe the issue as you mentioned. How can i understand that i undercorrect, if i recall i should be looking for some sort of drift at the graph? i was not able to find any examples for overcorrecting or undercorrentig other the youtube video getting the most out of phd2. Maybe at a future update of the software some sort of troubleshooting for guiding should be included at the help file. As for the mount i used the rowan astronomy belt mod.

here is a test pic of m17 (if i remember correctly), i took 6 pics with this being the best, i had to discard two pics that showed trail

[bw_msgboard]

No, the orthogonality error and polar alignment error are not related in any straightforward way.  Substantial polar misalignment can contribute to orthogonality error, but so can other things including seeing and declination backlash.  If you aren’t getting calibration alerts about orthogonality error, it’s probably safe to assume it isn’t causing you any significant guiding problems.  I would ignore it based on what I see in your log file.

 

The best tool for looking at details of your guiding performance is the PHDLogView app.   Look at the graph it shows and zoom in to look at the details over a typical sequence.  Be sure the Corrections checkbox is checked to see when guide commands were issued and for how long.  The first thing you’ll see is that the oscillations are quite random and aren’t being caused by guide commands – the guide commands are reacting to them.  That is suggestive of seeing issues.  It’s often helpful to look at one axis at a time to better see what’s happening.  In your case, there are fairly long periods where no guide commands are issued at all – that’s the result of the min-move settings you’re using.  Then look where commands are issued and see what results they achieved.  If it often takes multiple commands back-to-back, all in the same direction, you are under-correcting a bit.  That’s not a big problem, just something to consider. What you don’t want to see are situations where most of the guide commands push the mount too far and you then have to reverse direction right away.  These behaviors are mostly affected by aggression and related guiding parameters. 

 

If you’re working in conditions that hot, it’s probably going to be hard to fine-tune guiding.  Your scope may take a long time to reach thermal equilibrium, and the surrounding terrain may be generating heat convection until well after midnight.  And if the winds you’re getting don’t create a laminar air flow, they can also cause problems.  I’d be cautious about making a lot of changes until conditions improve.

 

We’ve been asked before to provide some kind of help for analyzing guiding performance and interpreting the graphs.  It’s definitely on the list of things to do, but it will take some time because the interpretation can be fairly tricky depending on the physical capabilities of the mount.

 

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of kostas stavropoulos
Sent: Friday, July 10, 2015 1:41 PM
To: open-phd...@googlegroups.com
Subject: Re: Help identifying problem with Declination

 

Maybe i did not understand well, i assumed that by watching the orthogonality error after the calibration is complete, at the review calibration, that it showed the PA error. I will pay close attention at the guiding assistant next time and read all the documentation again. I did not get any calibration errors. If the orthogonality does not represent PA then what information can i learn form it and how can i improve it if it affects guiding? As for the environmental conditions the temperature at nights is around 20-25 C and this summer is unstable with windy afternoons so seeing maybe the issue as you mentioned. How can i understand that i undercorrect, if i recall i should be looking for some sort of drift at the graph? i was not able to find any examples for overcorrecting or undercorrentig other the youtube video getting the most out of phd2. Maybe at a future update of the software some sort of troubleshooting for guiding should be included at the help file. As for the mount i used the rowan astronomy belt mod.

here is a test pic of m17 (if i remember correctly), i took 6 pics with this being the best, i had to discard two pics that showed trail

--

[kostas stavropoulos]

i attach some guide logs and i would like some insight.

At 12/7 i polar aligned my mount via the poalrscope, did a calibration, guided and took some really nice shots of m20. The guide assistant showed me that the PA was by 1.7'' off (if i remember correctly) - logs 2015-07-12_215146 and 2015-07-13_003132 (i did another calibration after the meridian flip after restaring phd since i had my mount disconect)

At 13/7 i used alignmaster for polar alignment and had it under 1'' but phd showed it was by 6''ish, the guiding though was better than 12/7 and the images were much better. Logs 2015-07-13_220753 and 2015-07-14_010312 (i did another calibration after the meridian flip after restaring phd since SGPro crashed and i had to finsh imaging at maximdl )

Things start getting weird at 15/7. I polar aligned via alignmaster and i have my PA under 1'' at each axis again. Phd guide assistant shows it is off by 6''ish and after the calibration is done i get a message that the rates are different at the axes (can't remmber the whole message), after a few calibration attempts the calibration was ok with no errors. Guiding was fine up to the point when i was near the meridian, then guiding started to get weird which i am guessing was due to the seeing deteriorating rapidly. The same situation continued after i did a meridian flip but the images were acceptable. At the time of the meridian flip there was no wind and temperatures were near 20C. Log 2015-07-15_215500

At 17/7 things continued the same way as 15/7. This time thoug i drift aligned but phd after calibration showed me the same message that the axe's rates were different and guiding was really bad and the images were bad. This time there was a light wind and i could see that the seeing was bad so i had to stop my imaging session. Log 2015-07-17_214609.

One thing i noticed at 15/7 was that after each calibration the error between the axes and therefore the PA was reduced until it was acceptable excpet from 17/7. Can seeing cause havoc to the calibration proccess or do i need to check again the mesh of my gears? The scope had reached thermal equlibrium before calibration

[Andy Galasso]

Hi Kostas,

Here's a plot of your declination guiding from the 2015-07-15_2155 log:

That looks like good guiding to me .. no significant drift, no signs of backlash or over-correction.  You are only limited by your seeing conditions.

Your RA is showing some over-correction/oscillation -- alternating E/W corrections.

and at a larger scale some E/W back and forth drift.

I think you could improve your RA guiding a bit by:

  1. change your guide rate in EQMOD (both axes) to something larger, like 50% to 90% sidereal rate.  It looks like you are using 15% sidereal rate which is too low. At that low guide rate your corrections are taking up to 2000ms which is introducing excessive latency into the guiding and perhaps contributing to the oscillation.

  2. Try using the default values for hysteresis (10%) and aggression (70%).   Your values of 0% hys and 100% aggr are contributing to the oscillation (though your min motion value is mitigating it.)   Slowly increase hysteresis and/or decrease aggression until you see the oscillation (alternation of E/W corrections at every cycle) go away.

Andy