Recovery mode question

As I leave SGP running when I sleep, if I have recovery mode enabled I never see what it does… Can you tell me if it parks the mount on recovery or just keeps it tracking in the same position until told to start again?

Here’s the reason for asking this…If I enable recovery and want it to keep trying for say 1.5hrs… If it keeps in tracking during this time there’s a chance that the target will pass the meridian and I’ll have a pier collision… As I’m assuming it won’t meridian flip when the time comes?

If it successfully recovers from what ever triggered the recovery in the first place it should go into the ‘center target’ action if you have it enabled correctly.

But yes the scope will continue to ‘track’ unless it comes up against the slew limits.

If the recovery fails the scope should go to park and quite tracking… IF you have it configured to do so.

OK so as I thought… yes I have it set to park if recovery fails…meanwhile while the software is in recovery mode (waiting for x minutes and then trying to recover, then doing that for x minutes) … it is easy to imagine that the target during this recovery time could pass meridian and then the scope could hit the pier during this time.

Would it not be safer to park as soon as it goes into recovery mode and then try to recover each time from the park position? This way everything would be safe in the interim… does that make sense?

is there are reason why this is not done?

The primary reason we don’t do this is because the primary cause of failure is passing clouds. A full park and restart might be heavy handed here, but, like many things in SGPro this is subjective and what is best for one rig is not for another.

That said, SGPro will not perform a meridian flip if the meridian is passed while in recovery (for obvious reasons). That said, SGPro will quit recovery if it passes the meridian. While this may create rare situations in which the sequence was not given sufficient time to recover, it is better safe than sorry. A full park here would alleviate this and it’s something we will need to think about. Maybe, just stop tracking?

Actually Ken I think just stopping the tracking would be a good compromise… It’s a shame if the sequence is aborted because the meridian is passed - There may still be ample opportunity to get the sequence going again depending on at what point the clouds have come and where the target is at the time… but as it stands now there would be no chance to pick it back up again.

Yes stopping tracking sounds like a decent idea :slight_smile:

Doesn’t SGP determine that recovery has succeeded by the fact that guiding is successfully restarted? If that is the case, would stopping tracking prevent recovery from happening?

It seems like when recovery is active, the only thing that should be suppressed is image taking. If a meridian flip occurs during recovery, perform the flip. If “end time” occurs during recovery, perform the end time logic, which might park the scope. When a recovery is successful (currently defined as auto guiding resumes successfully), SGP should perform an auto focus, a plate solve and a centering, if so optioned by the user. However, it’s possible that a long period of recovery might prevent PHD2 from successfully recovering the guide star. In long focus, OAG, there is often only one or two stars available to PHD2.

Asking PHD2 to define a successful recovery may not be the way to go. At the end of each recovery period, SGP could take an image and then scan it to find stars. If stars are found, recovery can proceed with auto focus, plate solve, centering and then resume guiding. Recovery doesn’t end until PHD2 reports that it has successfully resumed guiding.

Charlie

I don’t think a flip can be performed with out visible stars as plate solving is involved? … That’s probably the issue that concerns me really… if you can’t flip then you can easily be left tracking into a danger zone for equipment.

@swag72

You can perform a meridian flip without stars. The scope simply slews to the coordinates specified. Without stars you can’t center, of course. You want the meridian flip to occur during recovery so that the scope doesn’t accidentally track into the pier during an extended recovery period.

Charlie

right, but doesn’t SGP use plate solving as part of the meridian flip? they had taken that out but then recently put it back in cause centering after the meridian flip was not working with enough accuracy.

@swag72 , you are using one of those really fancy avalon mounts, right? doesn’t it have any kind of meridian limits? if it does you could make the mount stop before a pier collision occurs. for my part i had several pier crashes with my mach1gto but APCC adds all kinds of safety features which prevent pier crashes.

rob

@pfile - I use a Mesu mount… I don’t want to set limits etc that will park everything and end a sequence, I just think that a safe way to work during recovery mode would be good.

I too thought that the flip only happened with plate solve…

OK. yes i think recovery mode should also honor the meridian flip settings as well, but there are other reasons why the mount could crash into the pier; in my case at least once windows totally crashed and the computer rebooted and there was nothing to stop the mount from tracking… another time SGP locked up, same result. once the usb/serial locked up… on one of these occasions the RA motor burned up, so i started taking it seriously :slight_smile:

so i always have meridian limits defined in APCC and additionally the mount firmware now has a deadman’s timer. if the software does not reset it periodically, the firmware just stops the mount in place. since then i’ve never had any pier crashes for any reason.

rob

Having a ‘blind’ meridian flip, without solving post flip would be useful. A recovery mode during a meridian flip period caused me some grief once too and the mount stopped at the slew limit, which then caused the guider to fall over when the clouds parted.

It could be made to work; in nearly all circumstances there would be an image to solve from the last good exposure to set the goal. The meridian flip occurs but without centering post flip and the recovery mode picks up again. If the clouds pass, it could perform the center routine and continue.

+1 for this ideea.
I also lost some almost perfect nights because of a very short cloud passage close to the meridian flip time. The recovery process just aborted the sequence.

Regards,
Horia

Is this something that is needed to go into the request forum or has it been noted here? @Ken and @Jared

I thought about this some more - what may be required is a ‘center after recovery option’, just as there is a autofocus after resume option. The meridian flip then can be independent of recovery mode and in effect, ignore it.

1 Like

Moving this to feature requests. There is something here, but it requires some more thought.