Telescope.Connected = true Fails to connect telescope

[Gerald Ruch]

I'm having a puzzling problem. Here's the scenario:

1. I start ACP and Scheduler and click the "Dispatcher Running" checkbox in scheduler.
2. At dusk, the StartupObs.js script runs, starts everything, and connects to the scope.
3. The observatory runs all night happily executing plans.
4. At dawn, ShutdownObs.js runs, parks the scope, disconnects, shuts down support programs etc... everything looks great.
5. At dusk the next day, StartupObs.js runs but the scope doesn't connect. No exception is raised... it just doesn't connect and the Telescope.Connected attribute remains false.

I've added the following code snippet to StartupObs.js:

try {
if(!Telescope.Connected) {
Util.WaitForMilliseconds(5000);
Console.PrintLine("Connect ACP to the telescope, will auto-home if needed");
Telescope.Connected = true; // (Requires driver "home on connect")


// Diagnostic added by GTR - 23 July 2012
Util.WaitForMilliseconds(15000); // Pause 15 seconds to to allow telescope to connect
if(!Telescope.Connected)
Console.PrintLine("Telescope did not connect.")
else Console.PrintLine("OK");


} else {
Console.PrintLine("Telescope already connected");
}
} catch(ex) {
Console.PrintLine("**Failed to connect to the Telescope:");
Console.PrintLine(" " + ex.description);
Console.Logging = false;
return;
}

The above "Telescope did not connect." message appears in the console log. At this point, I can run StarupObs.js as many times as I'd like and the telescope will never connect and the same message will show in the log. If I connect to the telescope manually through the Telescope->Connect menu item in ACP, the telescope connects and StartupObs.js will run correctly ever after until dusk the following day.

[Bob Denny]

Gerald --

What type of mount? This sounds like an issue with the mount and/or its driver. When the startup script executes Telescope.Connected = True, it is making a call to the driver. If the driver is not raising an error, yet it is not connecting (later Connected = False) this is a bug in the driver for sure. Unloading and reloading the driver seems to "fix" it, per your description.

[Gerald Ruch]

Hi Bob,

Thanks for the reply. The mount is an Astro-Physics GTO-3600 with their latest driver installed. As a side note, the dome doesn't connect either when this happens. But, I'm assuming that ACP doesn't connect to the dome until the telescope connects, so that's not unusual.

Doesn't selecting the 'Telescope->Connect' menu item in ACP also call the driver to connect the telescope? Something is clearly different between executing Telescope.Connected = true in a script and selecting Telescope->Connect from the ACP menu...

We've got thunderstorms here so I couldn't do any further testing last night... I'll put pipe between ACP and the telescope for the next run to see if I see anything unusual.

~Gerry

[Edit]
I just logged into the observatory computer to fiddle around. Telescope.Connected = true was again refusing to actually connect the telescope. I took the following steps:

1. In ACP, I set the telescope to Pipe and pointed Pipe at the AstroPhysics GTO V2 Mount driver.
2. Run StarupObs Result: Telescope.Connected = true fails to connect to Pipe but no error is raised.
3. Point pipe to the Simulator instead of the AstroPhysics mount.
4. Run StartupObs Result: Telescope.Connected = true fails to connect to Pipe but no error is raised.
5. Choose Telescope->Connect from the ACP Menu. Pipe connects, the Simulator comes up.
6. Choose Telescope->Disconnect from the ACP Menu. Pipe and the Simulator disconnect and go away.
7. Run StartupObs Result: Telescope.Connected = true works fine, connecting both Pipe and the Simulator.

Not sure what to do next. I'll give the computer a reboot, that always helps right?
I'll also mess around with tricking Scheduler into firing the startup and shutdown routines to see if I can find a way to reliably reproduce the error.

~Gerry

[Jim McMillan]

Hi Gerry,

One other thing to try is to turn logging on in the driver. The driver author, Ray Gralak, should be able to analyze the various commands coming from ACP and the driver's reaction to them.

Jim

[Bob Denny]

How strange. No need to "trick" scheduler into doing startup and shutdown, there are test items

Snap2.png

[Gerald Ruch]

Quick update.

I'm unable to reproduce the error by using the Test Startup and Test Shutdown menu items from the Scheduler menu.

By a quirk of fate (lack of attention to detail?) the logging option in the AP driver has been on since ... forever apparently so I have LOTS of driver logs including short logs that I believe cover the events in question, but I can't verify that for sure. I'll have to wait until it happens again and be sure that I get the log from the exact time that I'm testing.

I did reboot the observatory control computer and the weather looks reasonable to run tonight, so I'll wait and see if the error reoccurs and post here if it comes back.

~Gerry