Guiding spikes and a possible solution
Ed Wiley
I have encountered severe spikes in guiding (+/- 12") and spend last night trying to solve the puzzle. Reading several forums brought up suggestions to change parameters such as aggressiveness, hysteresis and Min move. Interestingly comments about Mx RA and Mx Dec seem rare. Here is my equipment and analysis.
AP900, C11, 66mm refractor hard-coupled to OTA with ZWO guider. The AP900 has outstanding rms values (like 0.07) and is well polar-aligned. PEC is used and excellent (don’t have to value in front of me, the observatory is remote.) So, I am guiding with PEC.
I run guiding assistant each night and also do a new calibration each night.
Observations: Spikes are almost certainly caused by seeing and refraction. It is not wind and appears not to be gears (although no doubt the gears are not perfect) or cables on this particular night (of course wind can be a factor as can the other parameters, but not last night).
Good seeing early in the night, no spikes (above +45 Alt) until (repeat until) I try following a target below +40 altitude; then the spiking starts to happen. On nights of relatively poor seeing I can get spike at higher altitudes. In severe cases the spikes set up an irregular oscillating pattern. No amount of reasonable changes in aggressiveness, hysteresis and Min moved seem to help. The only other parameters are Max RA duration and Max Dec duration, which I had set at the default of 2500.
What I fixed: Why not go whole hog? I set both Max RA duration and Max Dec duration at 50 (!). RA: Hyst = 10, Agression = 70, Min = 0.10. Dec: Agression = 50, min = 0.10, Auto. Image every 4 seconds.
Result: No spikes at lower altitudes. Question is: was I even guiding?
I will continue to play around with these parameters. But my question: why is there so little discussion of Max duration as a variable parameter? Or, have I simply missed the relevant discussions or instructions?
bval...@gmail.com
Hi Ed
Can you please upload your logs so we can see the details?
http://openphdguiding.org/getting-help/
Brian
--
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 Ed. We need to see the guide log to have anything sensible to say about this. By cranking down the max-move values, all you’re doing is not letting PHD2 correct for these big excursions. But they are still happening, right? So your images are going to show the effects of these large moves, and those frames are probably hosed. So what have you accomplished? I think that’s why nobody talks about max-move values – they’re not generally a useful tool for this sort of thing.
If you’re imaging at a long focal length, you are going to be strongly affected by seeing. Once you have reached an altitude of 30 degrees, you are imaging through 2x the air mass that you are at the zenith. It’s very common to see conditions deteriorate in this area, and many of us won’t image at all below 30 degrees elevation.
Cheers,
Bruce
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Ed Wiley
Sent: Wednesday, January 10, 2018
12:16 PM
To: Open PHD Guiding
Subject: [open-phd-guiding]
Guiding spikes and a possible solution
I have encountered severe spikes in guiding (+/- 12") and spend last night trying to solve the puzzle. Reading several forums brought up suggestions to change parameters such as aggressiveness, hysteresis and Min move. Interestingly comments about Mx RA and Mx Dec seem rare. Here is my equipment and analysis.
--
bval...@gmail.com
>> many of us won’t image at all below 30 degrees elevation.
Really? That is an interesting bit of knowledge
B
bw_msgboard
Hi Brian. That’s just my personal experience and the experience of others whom I know. Obviously, if you want to shoot a target that’s always low in your sky, you have to make exceptions. But when I’m working on a target that is reasonably positioned in Dec, I try to shoot my luminance frames in a time interval centered on the meridian transit. I shoot my RGB frames further from the meridian because I can tolerate poorer resolution for those. I typically start the session at 30 degrees elevation in the east and stop at 30 degrees elevation in the west. Again, there are exceptions, I always watch the seeing and guiding behavior but I can typically see a marked deterioration in the seeing as I get down below 30 degrees elevation. Also, the lower you get, the more you’re going to be affected by surrounding structures and local seeing effects, beyond just the increase in air mass. This is all just personal opinion, of course, and I’m imaging at 0.6 arc-sec/px so I have to pay a lot of attention to seeing.
Bruce
Ed Wiley
Brian Valente
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guiding+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
bw_msgboard
Hi Ed. You’ve got some interesting mechanical problems here in RA. Here’s a big-picture view of a section starting at 1:55 (RA only in red):
These huge excursions are not caused by guiding, they are external events of some kind. And of course they are image-killers. Here are some more after you started changing the RA Max-move value:
And here’s an even more interesting example when guiding was completely disabled:
So this is clearly some kind of mechanical problem that you’ll need to eliminate. What’s interesting about this one is that it seems to have self-corrected. If some of these are self-correcting, that would explain why reducing the max-move would seem to have an effect. If we had tried to correct for this excursion, we would probably have created an oscillation, and there is an example of that earlier in the session. But of course, not all of them are self-correcting:
This one took over 20 seconds to recover.
I think the main point is that something mechanical – perhaps wind gusts – is creating these big impulse moves of 10-15 arc-sec and there’s no amount of guiding trickery that will fix them. Those images are probably toast. I think you’ll have to find the root cause of the problem and eliminate it.
Hope this helps,
Bruce
Ed Wiley
bval...@gmail.com
I’m reminded again how a bad PEC curve can really mess things up.
B
From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Ed Wiley
Sent: Monday, January 15, 2018 5:57 PM
To: Open PHD Guiding <open-phd...@googlegroups.com>
--
bw_msgboard
Hi Ed, thanks for getting back to us with the resolution – we wish more people would do that. It’s a good lesson for me because I never think to suggest that people disable PEC when they see these inexplicable big moves in RA.
Happy guiding…
Bruce
From:
open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Ed Wiley
Sent: Monday, January 15, 2018
5:57 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re:
Guiding spikes and a possible solution
As it turns out I had no evident mechanical problems or PHD2 problems. I upgraded the AP CP3 control chip to the latest and re-did PEC with Pempro. The run last night was perfect, no guiding issues at all and with default Max settings (and other reasonable PhD2 setting, nothing exotic). Apparently it was a combination of my CP3 firmware and my PEC that was causing the problems. Probably a corrupted piece of firmware communicating with the PEC. I did not expect this because the spikes appeared only recently, but since I need to upgrade the firmware in any case (mine did not interact with APCC), I decided to run a new PEC analysis. Happy ending. PhD is not-guilty.
--