Ioptron CEM70 RA oscillation and spikes with OAG and a bit longer focal lengths

[Mike Ales]

  Team,

    I've been struggling with two new rigs both with Ioptron CEM70, OAG's and a bit longer longer focal lengths.
    Both of these Ioptron are new in the last 3-6 months depending on the rig.

   I was convinced for awhile that it might have been related to shorter exposures and mono with frequent filter changes and focuser shifts with filter offsets.

  While I have guiding kind of tuned in I really wished I could run with longer exposure times and stop chasing so much of the seeing but when I do go for longer 

   PHD2 exposures I do see degrading image sharpness.

  Last night I finally did a full Polar Alignment, Calibration, and Guiding Assistant runs.
  I did not apply the minmov for the RA axis but did apply the backlash and most of the minmov on the dec axis since polar alignment was good.

  Rig for this set of logs is  Ioptron CEM70 on a tripier, with CarbonStar 200 Newt including mirror mask, QHY OAG and QHYMiniCam with QHY's 7nm SHO set.
   Exposure times on this rig are pretty short 30 or 60 seconds, 120seconds for SHO.
   I have added some additional waits after Autofocus/filter changes and that has not really addressed much.

  I've gone through balance a couple times and also checked for cable drags without any big improvements.

  While the RMS values look good with the 0.5 second exposures I am a little concerned with the spikes.

  Two rigs are doing very similar behaviour the other rig as a Touptek ATR585MM on it also with OAG but with longer exposure times.
    I would just like to focus on the one rig for now.
   I have pauses for both autofocus and filter changes turned in NINA.

   This particular rig is running NINA 3.3 Nightlies but the other rig is running NINA 3.2 with nearly identical results.
  

https://openphdguiding.org/logs/dl/PHD2_logs_aHaQ.zip

If I could resolve the non dither spikes and be able to stretch out the PHD2 exposure time to avoid chasing seeing I would be overwhelmed.
Clouds did roll in this morning around 3:45 so the rest of the log is not very useful.
I did change the PPEC Prediction weight about 10:40pm but behaviour seems similar before and after.

RMS I am happy with but concerned about the Peaks and would like to be able to run longer exposures

  I know you will take a bit to go through the logs in detail and your efforts and time is greatly appreciated once I get this rig cleaned up a bit.
  I will leverage whatever your advise on the other rig before going into the forum with logs and all.

  

I have a much shorter refractor (WO81) also on an Ioptron CEM70 and that has been rock solid for 2-3 years but that is using a dedicated guide scope and camera.
So no confusing being created by stop guiding/start guiding commands, focuser shifts, or risk of reflections with filter swaps.

While I am writing just would like to confirm a couple questions on PHD in my mind.
1: For OAG with frequent interruptions do you still recommend the PPEC algorithm for RA
2: If the frequency plots are showing RA harmonics at 1/2 and 1/3 should the period be reduced to try and catch these also.



Thanks a bunch

Mike Ales

[Bruce Waddington]

The guiding actually looks pretty good so I'm not entirely clear on the problem.  You say that longer guide camera exposure times result in images that are less sharp but I don't know what kind of numbers you're talking about or whether you've tested carefully enough to develop statistics that support that conclusion.  I assume you're familiar enough with atmospheric seeing to know that longer exposures with the main camera are usually less sharp than very short exposures with the same camera - that's the whole premise of "lucky imaging".  Seeing conditions are highly variable from hour to hour and day to day, and with hotter atmospheric conditions they generally get worse.  So I don't know know what your expectations are for your setup, 1.5 - 2 arc-sec guide star excursions are common in most locations.  There are some things about your configuration that strike me as being off.  You've set the RA min-move to 0.05, a value I don't think I've ever seen before.  That effectively means you will issue a guide command with every exposure and you're probably approaching the limits of the centroid calculations.  I think that's too low, there is no way to mask seeing effects with this approach and from looking at the data, I don't think your mount requires it.  Next, it looks like you've mis-configured the PPEC parameters.  Here's what your FFT looks like:

So the period you want to lock on is around 118 sec, not the 348 seconds you're using.  This FFT analysis is coming only from the 18 min GA run you did, there aren't any long guiding sessions in your log.  Once you've identified a target period like this, try changing the predictive weight of PPEC up to 90 while leaving the reactive weight alone.

Good luck,

Bruce

[Bruce Waddington]

Sorry, I forgot the two specific questions you asked.  PPEC retains its model across interruptions and small slews so those are not a problem.  And as I said, you should change the period length to go after the frequency component we see here. I don't know to what extent "harmonics" apply to these strain-wave mounts where there isn't the classic worm-gear/worm-wheel interface that operates in a highly predictable way.

Bruce

[Mike Ales]

Thanks Bruce.

Just to clarify the Ioptron CEM70 is not harmonice.

The older worm and gear standard.

I'll update the ppec period per your suggestion.

I set it originally based on the quoted web period for the worm which does seem right based on the Raw graph.

Also once i do that trial I'll do a run with the longer guide exposures and redo all the normal steps before posting again.

Thanks as always for the clarity and time.

Sent from my iPhone

On Mar 23, 2026, at 10:47 AM, Bruce Waddington <bw_m...@earthlink.net> wrote:



--
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.
To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/717ddd04-14b3-4b62-ab54-f08487cb8712n%40googlegroups.com.

[Bruce Waddington]

Ok, I guess I made the bad assumption that the short exposure times and hyper-aggressive guiding were being applied because of using a strainwave mount.  So I would emphasize my previous remarks even more... :-)  As I said, I don't really see a guiding problem here so if you're unhappy with your final images, that's probably an entirely different matter.

Bruce