Oops! Powered-ff mount while still connected


Mike Dodd
 

Last night I had occasion to check my 1200's pointing accuracy. I followed the standard procedure:

1. Connect with APJog
2. Park to Park5
3. Level both axes, tighten clutches
4. Unpark from last parked
5. Park to Alt/Az (0, 90 -- due east) to confirm my normal park position

It all looked fine, so I went indoors to wait for dark. Then I did this remotely from inside the house:

1. Connect SkyX
2.Unpark from last parked
3. Slew to a star in SkyX
4. Center the star in MaxIm (crosshairs on) using the driver direction buttons
5. Sync in SkyX (driver has "use recal for sync" turned on
6. Start ACP, connect scope, run my pointing plan

Everything went fine, and pointing was very accurate. Then I began to shut down the observatory. I disconnected the scope in ACP, and did the other things necessary.

The final step was to kill power to the mount. That's when I discovered SkyX and APJog were still running. So I powered-up the mount again, disconnected SkyX, and reconnected APJog.

The mount was physically still parked in its correct 0/90 Alt/Az position, but the driver showed angles that were totally wrong. When I tried to park to Alt/Az (correct coordinates displayed in the confirmation dialog), the mount began to slew away.I aborted the slew, then used the drivers buttons to move it back.

Using the driver, I reset the mount's time from the PC, but that made no difference in the displayed coordinates.

Finally, I shut off power and went to bed.

QUESTION: What shall I do to recover from this situation?

I can try parking in Park5, but what if it ends up in the wrong position? The clutches haven't been loosened since I leveled it in Park5 yesterday.

Thanks for all advice.

--- Mike


Dale Ghent
 

Seems like quite an involved procedure you have there.

Why not just unpark the mount, manually slew it to some east or west part of the sky, and do a plate solve and sync (recal)? There's no need to hustle out to your mount and physically level the axes before doing anything, or to manually slew to a specific star only to solve it by hand. Keep your slippers on and stay out of the cold by letting the software do it for you :)

As for your power-off issue, the mount's CP would come back up in an uninitialized state after being powered on. Are you using the AP ASCOM driver in TSX, or are you using TXS' bult-in AP mount driver? If you're using the AP ASCOM driver:

1. Power on your CP per normal
2. Open APJog and connect to the mount. This will spawn the AP ASCOM driver which will initialize the CP
3. Using APJog, nudge the axes so that you're pretty close to a Park 5 position.
4. In the AP ASCOM driver, park the mount at Current Position
5. Then UNpark the mount from Park 5. This is the step that resets the CP's notion of how the axes are physically oriented
6. Slew the mount using APJog to some clear and cozy portion of the east or west sky, then use your preferred app to do a plate solve+sync.

For step 3, you can manually go out and loosen the clutches and rearrange the axes so that they're in a park 5 position, then do the park to current, followed by the unpark from 5... but any plate solver should do just fine if you've instead nudged things close enough to park 5, and the correction via the plate solve+sync operation will be exact.

On Nov 11, 2021, at 08:32, Mike Dodd <mike@...> wrote:

Last night I had occasion to check my 1200's pointing accuracy. I followed the standard procedure:

1. Connect with APJog
2. Park to Park5
3. Level both axes, tighten clutches
4. Unpark from last parked
5. Park to Alt/Az (0, 90 -- due east) to confirm my normal park position

It all looked fine, so I went indoors to wait for dark. Then I did this remotely from inside the house:

1. Connect SkyX
2.Unpark from last parked
3. Slew to a star in SkyX
4. Center the star in MaxIm (crosshairs on) using the driver direction buttons
5. Sync in SkyX (driver has "use recal for sync" turned on
6. Start ACP, connect scope, run my pointing plan

Everything went fine, and pointing was very accurate. Then I began to shut down the observatory. I disconnected the scope in ACP, and did the other things necessary.

The final step was to kill power to the mount. That's when I discovered SkyX and APJog were still running. So I powered-up the mount again, disconnected SkyX, and reconnected APJog.

The mount was physically still parked in its correct 0/90 Alt/Az position, but the driver showed angles that were totally wrong. When I tried to park to Alt/Az (correct coordinates displayed in the confirmation dialog), the mount began to slew away.I aborted the slew, then used the drivers buttons to move it back.

Using the driver, I reset the mount's time from the PC, but that made no difference in the displayed coordinates.

Finally, I shut off power and went to bed.

QUESTION: What shall I do to recover from this situation?

I can try parking in Park5, but what if it ends up in the wrong position? The clutches haven't been loosened since I leveled it in Park5 yesterday.

Thanks for all advice.

--- Mike






Mike Dodd
 

On 11/11/2021 10:04 AM, Dale Ghent wrote:
Seems like quite an involved procedure you have there.
As I wrote, "I had occasion to check" the pointing accuracy. This is not my normal startup procedure. I normally power-up everything, unpark, and start an ACP imaging run. No sweat. But last night was special: I needed to do the Park5 level, then a recal.

As for your power-off issue, the mount's CP would come back up in an uninitialized state after being powered on.
Why? I thought AP mounts have a "park anywhere" feature where powering-off the mount means it's parked, and it remembers where it was when powered-on again. Did the fact that it was connected via ASCOM mess up this feature? IOW, is the "park anywhere" feature valid only with the hand box connected?

Are you using the AP ASCOM driver in TSX, or are you using TXS' bult-in AP mount driver?

I'm using the AP ASCOM driver Ray wrote. My first connection was with the APJog utility. SkyX was the second connection.

If you're using the AP ASCOM driver:
[...]
For step 3, you can manually go out and loosen the clutches and rearrange the axes so that they're in a park 5 position, then do the park to current, followed by the unpark from 5... but any plate solver should do just fine if you've instead nudged things close enough to park 5, and the correction via the plate solve+sync operation will be exact.
Yep, that's what I did yesterday, Except (since I knew the mount was pretty close to being correct), I Parked to Park5, loosened the clutches to level the mount there, tightened the clutches, then did Unpark From Last Parked. Everything was fine, and my initial slew to a star put it only 8-10 arcmin from the center of the camera sensor. I tweaked to exact center, then did a recal in SkyX. All other slews resulted in the star being within 2-3 arcmin of center.

Per your procedure, I will again use a level to get the Park5 position exactly right, instead of nudging close enough. :-)

What I don't understand is why/how the mount got so confused about its position. Its site coordinates, clock, LST, and time zone are still correct. But the RA/Dec and Alt/Az are way off.

If the Park5 procedure is the only way to recover, I can easily do that tomorrow (rain later today).

--- Mike


Roland Christen
 


is the "park anywhere" feature valid only with the
hand box connected?
No, the park anywhere feature is correct regardless of what is used for initialization.
In your special case we don't know what was sent to the mount when you powered it back on. One would have to dig thru your ASCOM log file to figure out why the alt-az was off.

Rolando


-----Original Message-----
From: Mike Dodd <mike@...>
To: main@ap-gto.groups.io
Sent: Thu, Nov 11, 2021 9:32 am
Subject: Re: [ap-gto] Oops! Powered-ff mount while still connected

On 11/11/2021 10:04 AM, Dale Ghent wrote:
>
> Seems like quite an involved procedure you have there.

As I wrote, "I had occasion to check" the pointing accuracy. This is not
my normal startup procedure. I normally power-up everything, unpark, and
start an ACP imaging run. No sweat. But last night was special: I needed
to do the Park5 level, then a recal.

> As for your power-off issue, the mount's CP would come back up in an uninitialized state after being powered on.

Why? I thought AP mounts have a "park anywhere" feature where
powering-off the mount means it's parked, and it remembers where it was
when powered-on again. Did the fact that it was connected via ASCOM mess
up this feature? IOW, is the "park anywhere" feature valid only with the
hand box connected?

  Are you using the AP ASCOM driver in TSX, or are you using TXS'
bult-in AP mount driver?

I'm using the AP ASCOM driver Ray wrote. My first connection was with
the APJog utility. SkyX was the second connection.

> If you're using the AP ASCOM driver:
[...]
> For step 3, you can manually go out and loosen the clutches and rearrange the axes so that they're in a park 5 position, then do the park to current, followed by the unpark from 5... but any plate solver should do just fine if you've instead nudged things close enough to park 5, and the correction via the plate solve+sync operation will be exact.

Yep, that's what I did yesterday, Except (since I knew the mount was
pretty close to being correct), I Parked to Park5, loosened the clutches
to level the mount there, tightened the clutches, then did Unpark From
Last Parked. Everything was fine, and my initial slew to a star put it
only 8-10 arcmin from the center of the camera sensor. I tweaked to
exact center, then did a recal in SkyX. All other slews resulted in the
star being within 2-3 arcmin of center.

Per your procedure, I will again use a level to get the Park5 position
exactly right, instead of nudging close enough. :-)

What I don't understand is why/how the mount got so confused about its
position. Its site coordinates, clock, LST, and time zone are still
correct. But the RA/Dec and Alt/Az are way off.

If the Park5 procedure is the only way to recover, I can easily do that
tomorrow (rain later today).

--- Mike






--
Roland Christen
Astro-Physics