RA and DEC Spikes

[Thomas Butts]

Hey everyone,

The past few months I've been trying to use Pulseguiding with ASCOM after about 2 years of guiding with my Shoestring GPUSB , with no luck. Luckily for me last night I finally figured out it was a firmware issue with my handset and a software glitch with EQMOD's "Mount Limit" that was giving me grief. 

Once I got my guiding working, it was going OK (not great, certainly not what I would call good..but it was working!), but I would have these very significant reoccurring spikes in RA and Dec. The mount I have is a HEQ5 Pro that has been belt modded. 

Now what I am trying to figure out is if my mount really requires a(nother) tune up, due to bad grease/grime in the worm, bad bearings, or would this be caused by the light cloud overhead? I didn't recall seeing any star mass warnings. I do see a little bit of a pattern in the graph, so if someone else could offer me their thoughts, that would be great. In between the spikes, I had guiding in the 0.65-0.8" RMS range, and overall my guiding after the update inclusive of the spikes was 1.35" RMS)

For those looking at my log, you can ignore the guiding prior to 23:06pm, as that was before I did the update.

Clear skies,

Tom

[Thomas Butts]

I should clarify, I had checked the balance of the mount and had no snagging cables. 

T

[Andy Galasso]

Hi Tom,

I notice a definite pattern: the RA spikes are happening whenever there is a large correction in Dec! (i.e. the large backlash comp Dec corrections)   It seems like there is some kind of mechanical (or electrical?) coupling between RA and Dec.   I have no idea what could cause this.

You may be able to diagnose the problem using the Manual Guide tool. Set your Dec Guide mode to None and RA Min Move to the largest allowed value (20 pixels) to effectively disable guiding, yet still allow the Manual Guide Tool to work.  Then you should be to trigger the RA jump by sending a large Dec pulse and seeing if anything happens in RA.   You can use the RA/Dec overlay (View menu) to see which way is which.

Good luck, I hope you are able to find the issue.

Andy

P.S. unrelated to your problem, but I see that you did a re-calibration when you meridian flipped.  In case you didn't know about it, it is not necessary to re-calibrate after the flip since EQMOD tells PHD2 about the flip so PHD2 will automatically adjust for the flip.

[Thomas Butts]

Hi Andy,

I'll give that a try!

For the recalibration, I did one as I updated the firmware, and was being a little extra cautious.

Cheers,

Tom

[Bruce Waddington]

Hi guys.  This sort of Dec/RA interaction, despite a clean calibration, can sometimes be caused by a problem with the Dec bearings.  A large torsional force on the Dec axis – like a large backlash comp pulse and a direction reversal - could cause a “rocking” motion in the RA direction if the bearings don’t hold the Dec axis firmly in position.  There’s absolutely no way for us to say this is the problem, it could probably be many other things -  I’m just mentioning it so you are aware of the possibility.

 

Hope you can track it down quickly,

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.

[Thomas Butts]

I'll be changing the bearings shortly to see if it helps. Could a faulty cable cause these interactions?

T

[Bruce Waddington]

Hi Tom. Since you're using an ASCOM connection for guiding, you can pretty much rule out a cable as the source of this type of problem. It's nothing like using an ST-4 guide cable. Any sort of electrical interference or cross-talk would have to be inside the mount controller itself which seems to me unlikely.

Good luck,
Bruce

-----Original Message-----
From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Thomas Butts
Sent: Thursday, May 31, 2018 3:29 AM
To: Open PHD Guiding <open-phd...@googlegroups.com>

[Thomas Butts]

Looking at the graph, I would think the issue is either with a bearing, or a damaged or dirty worm. I'll be inspecting it tomorrow. New bearings will be in next week...they probably need replacing anyways. 

[Thomas Butts]

A little bit of an embarrassing realization...

Those large Dec/RA spike combinations are dithers...

Still doesnt explain the individual RA spikes. I did have "disable guiding during image download" in SGP...perhaps the disabling of the guiding with the clouds was enough to cause it? I'm still going to inspect the gears, and replace the bearings though.

Tom

On Wednesday, 30 May 2018 16:22:08 UTC-4, Thomas Butts wrote:

[Bruce Waddington]

Hi Tom.  The correlated large RA/Dec excursions that Andy and I were talking about were not associated with dithers – or at least they didn’t come from dithers that were accomplished via PHD2.  Dither events are delineated in the graphs of PHDLogView and we routinely ignore those.  So I don’t know what you’re looking at to convince you they were dither-related.

 

Here’s one sequence:

 

 

RA is in red, Dec in green.   The dither events are shown by the silver/gray areas, but you can see four other times when a large Dec guide pulse (the thin green lines) resulted in an unwanted large RA excursion.

 

Cheers,

Bruce

 

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Thomas Butts

Sent: Friday, June 01, 2018 3:40 PM
To: Open PHD Guiding <open-phd...@googlegroups.com>

--

[Thomas Butts]

Things have taken an interesting/optimistic turn for the better. 

For some reason after deselecting Declination Backlash Compensation, these rouge Dec pulses have completely disappeared, as has my RA/Dec spikes. I went from getting them every few minutes, to nothing the rest of the night.

I know get significant dec backlash, but I think using the EQMOD dec backlash settings that can be reduced.

If anyone knows why that solved my problem, Id be more than happy to listen to why...

Tom

[Andy Galasso]

Thomas,

I think you are just avoiding the problem (not necessarily a bad thing!)   As we saw, there is some kind of mechanical coupling between RA and Dec so that a large Dec correction causes the RA axis to lurch.  By disabling the Dec Backlash Compensation in PHD2 you no longer have the large dec guide pulses, thus you don't see the problem.  If you enable backlash compensation in EQMOD (not recommended in general) you will get the problem back again, but worse because PHD2 won't be aware of the backlash compensation pulses and your dec guiding is likely to become unstable (oscillation.)

If you cannot find the underlying mechanical source of the problem, it's not a bad idea to keep BLC disabled, but you may need to rely on uni-directional dec guiding (dec guide mode North or South, not Auto, more info available in the user manual and in the best practices doc.)

Andy

[Thomas Butts]

Sorry for the late response. I took the entire mount apart and discovered the culprit....a very worn out dec bearing. 

I replaced it, and have seen a noticeable difference. Even with some good wind gusts and poor seeing later in the night I was still getting sub arcsec guiding.

Tom

On Wednesday, 30 May 2018 16:22:08 UTC-4, Thomas Butts wrote:

[bw_msgboard]

Hi Tom.  Thanks for reporting back, this is very helpful to us. And congratulations on nailing the problem!

 

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Thomas Butts
Sent: Friday, June 15, 2018 3:30 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re: RA and DEC Spikes

 

Sorry for the late response. I took the entire mount apart and discovered the culprit....a very worn out dec bearing. 

--