AO-X: Low Speed / Rate

Discussion in 'Guiding and Adaptive Optics - StarChaser and AO' started by sixburg, May 4, 2016.

  1. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    My AOX rarely achieves a rate about 2.5 Hz. I've seen it go higher, but that was some time ago with a different camera of the same model. What could be the causes of such a slow rate? It seems to be independent of these variables which is completely counterintuitive (i.e., the rate stays are ~2.5Hz no matter what). I understand that "slightly higher speeds" can be obtained by moving the star towards the upper extent of the chip. However, I understand that these variables are more impactful.
    1. Brightness (48K)
    2. Mag (4.2)
    3. Track Box (8x8)
    4. Rate = 2.5
    1. Brightness (5.7K)
    2. Mag (8.7)
    3. Track Box (8x8)
    4. Rate = 2.5

    Videos of each of the above tests cast can be viewed at the link below. Additionally, previously recorded sessions showing the AO achieving much higher speeds are at the same link. The only difference between the systems is the a different camera of the same model. The RGH is the same, all the cables are the same.

    There is nothing out of the ordinary running on the system. All software, firmware and drivers are up to date.

    https://onedrive.live.com/redir?res...43&authkey=!AMU3v_tMqwbm8Zg&ithint=folder,mov

    Thanks for any direction.

    -Lloyd
     
  2. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    Another experiment in tonights session shows speeds consistently above 6 and topping out at 7.5Hz with exposures around 0.06s. RMS wander is around 0.2 x 0.2. I increased the track box to 16x16 to make sure the star stayed in the box which should have made the rate slower. This was before I recalibrate the guider for ACP (which I had done improperly before).

    Some combination of things improved the rate 3x. I'm still curious of the SBIG experts can comment on the above results.

    Thanks,

    -Lloyd
     
  3. Bob Denny

    Bob Denny Cyanogen Customer

    Joined:
    Oct 12, 2014
    Messages:
    807
    Location:
    DC-3 Dreams, SP, Mesa, Arizona +1 480 396 9700
    Note that ACP eliminates the need for drive calibration on the AO. Only the tip-tilt needs calibration in MaxIm. That's the "calibrate AO". The "calibrate drive" is unnecessary as ACP knows the PA and sets the bump polarities and axes directly.

    I don't know what factors limit the tracking rate apart from the exposure interval. I suspect the CPU speed is important as it has to do some fast calculating between samples. But I am not the expert. 7.5 Hz is a decent speed.
     
  4. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    Hi Bob!
    Thanks for your input on this. I should clarify the workflow a bit better. I calibrated the AO - Locate, Calibrate Drive, Calibrate AO. I actually do all three steps (just one time). I also calibrate in Maxim for ACP. All this is done in the proper part of the sky. I understand the Calibrate Drive is not necessary because ACP , but I wonder if it is detrimental? Things seemed to have worked.

    One other workflow related issue to point out: ACP will automatically populate the Guide Sensor Rotation Angle which is great. However, I enter the angle manually which is very simple to do not because I want to but because with my particular camera there is no easy way to switch from AO to non-AO mode which is a requirement of ACP's automatic guide sensor rotation angle procedure. "No easy way" means when switching between SBIG w/ AO and SBIG Universal camera drivers the camera must be power cycled and sometimes that does not even work. It's far easier in my opinion to do the "regular" Maxim guider calibration and enter the guider sensor angle manually...plus, it should only have to be infrequently.

    Bob, this last paragraph is more of a question for SBIG than for you. I appreciate your help.

    Summary question
    1- Is is problematic to do Calibrate Drive as described in the first paragraph?
    2- More important: Is power cycling when switching drivers normal procedure (described in paragraph 2, 3rd line)?

    Thanks,

    -Lloyd
     
  5. Geoff Stone

    Geoff Stone Cyanogen Customer

    Joined:
    Mar 7, 2015
    Messages:
    143
    Location:
    Shaver Lake, CA
    For what its worth, the fastest speed I've ever seen is about 8 Hz. Presumably the limiting factor is the latency involved in acquiring a guide frame and getting it into memory in the computer. I doubt there is any pipelining going on in that cycle to bury the latencies. This could probably be seen by examining a driver log.
     
  6. Bob Denny

    Bob Denny Cyanogen Customer

    Joined:
    Oct 12, 2014
    Messages:
    807
    Location:
    DC-3 Dreams, SP, Mesa, Arizona +1 480 396 9700
    It won't hurt. It simply lets MaxIm determine the bumping polarities and axes for your mount and your current PA.

    When doing the guide sensor rotation angle (with respect to the main camera, for other readers), it's fine to do the manual conventional guide calibration to determine the angle. Getting your main camera to a 0 deg. PA manually can be tricky because there is always a difference between the mechanical rotator angle and the actual sky Position Angle. Follow the directions in ACP help where you run an observing plan for 0 PA then kill it. That will guarantee 0PA so you can read the guide sensor angle from MaxIm.

    For further info and help, please move this to our forum.
     
  7. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,175
    Geoff, pipelining will not work because the phase lag would induce control instability.

    I'm not sure of the cause of the slow guiding in this case, but it would have to be computer or software related. (Possibly USB bus but less likely.)
     
  8. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    Hi Bob...
    What I normally do is plate solve my way to 0PA. I can record mechanical angle in my telescope control system so getting to 0PA is pretty quick. I've done it your recommended way too.
    I would rather let ACP do the automated guider cal / sensor rotation, but the SBIG driver makes this very difficult. SBIG Universal as opposed to SBIG w/ AO must be used. Switching between those drivers is problematic. Sometimes the guider will not even function in this mode. Power cycling usually helps, but not always. Therefore, I just do it manually in MDL without changing the camera driver.

    I would love for this to driver behavior to be explained / fixed / and/or my procedure corrected. This has not been answered either way.

    -Lloyd
     
  9. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    The slow guiding is not an always situation. I reported on this yesterday (236AM post). Guiding routinely hit 7.5Hz. I don't know why it varies...that was the basis of my original question. I've witnessed this now on 2 different computers...sometimes fast guiding, sometimes slow. There is no other software on these systems except that which is necessary to run an imaging system.

    I'm open to any suggestions on this or any other question that I've posed.

    -Lloyd
     
  10. Bob Denny

    Bob Denny Cyanogen Customer

    Joined:
    Oct 12, 2014
    Messages:
    807
    Location:
    DC-3 Dreams, SP, Mesa, Arizona +1 480 396 9700
    Again, we should take this cal with ACP over to the Comm Center. I know it's cumbersome for you to switch between conventional and AO. But just do it, even if you have to shut down the system and restart. Then do the cal using the CalibrateGuider script. Now you have the guide sensor angle. Now shut down if needed to switch back to AO. You're done until you physically remove/replace the guide sensor in the pickoff or or the main camera... Anything that would change the RELATIVE angle between them. Normally that would be weeks or months...
     
  11. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    Bob, thank you for your help. I'm glad to move this to your forum. Just to be clear...I understand the ACP procedure completely. Also note there is an unaddressed issue of why the guider will not always function when switching to the non-AO driver (i.e., from SBIG w/ AO to SBIG Universal). That will not be solved on your site.

    My approach (the manual way) is a workaround to this apparent issue / limitation / operator error. The cause has never been made clear through information I have received here....I have not received any information on the topic.

    -Lloyd
     
  12. Doug

    Doug Staff Member

    Joined:
    Sep 25, 2014
    Messages:
    7,175
    Settings for the guider operations via Guide tab and SBIG w/AO are totally independent You need to set them up separately. That may be part of the confusion here?

    There's no reason you can't do normal autoguiding via the SBIG w/AO driver; just use the Guide tab instead of the AO control. It's possible that you might be able to switch the AO function on/off in ACP also. Bob could answer that one...

    Finally, you can switch all settings in MaxIm DL by using the Configurations feature. This can allow for rapid switching between camera and guider setups.
     
  13. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    Great Doug, thanks. I will make use of the configuration feature for rapid switching.

    I think the mistake I'm making is trying to switch drivers in the first place...your 2nd paragraph says it all.

    I made a video to demonstrate exactly what I'm doing and to show that my approach does not work. The issue: cannot switch between AO and Universal drivers for purposes of using the RGH as a guider without AO involvement.

    https://onedrive.live.com/redir?res...698&authkey=!ANgeVHtSV2isaCE&ithint=video,mov

    The steps:
    1. Connect to SBIG w/AO
    2. Take image with main imager
    3. Take image with RGH
    4. Switch to SBIG Universal
    5. Take image with main imager
    6. Take image with RGH -- From 1:45 - 2:55 you'll see the RGH does not complete an image
    7. Switch back to SBIG w/ AO
    8. Repeat steps 2 - 6 successfully
     
  14. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99

    I've witnessed this "slow track rate" several times and a reboot cures it every time. It's curious--2.5Hz must have some significance...a default / initial value or something.
     
  15. sixburg

    sixburg Cyanogen Customer

    Joined:
    Feb 10, 2016
    Messages:
    99
    I'd like to re-submit my question from May 4 on AO-X track rates. Over the last several sessions the rate is conspicuously 'stuck' at 2.5Hz. Since my last post it seems that the guide rate is either 2.5Hz, or in a range between 5 and 9Hz.

    What would cause the rate to stick at 2.5Hz? Guide exposures are 0.01 seconds, guide box is 16x16. The situation is just as described in the post on 5/4.

    The track rate does not vary with software in use: reacts the same with ACP or Maxim-only.

    Thanks for your help. By the way, firmware and drivers are current.

    -Lloyd
     
    Last edited: May 24, 2016

Share This Page