Subject: Re: [sloan-25mlog 4466] 2.5m obslog 20190410 (SJD 58584)
From: John Downey
Date: Thu, 11 Apr 2019 12:20:31 -0600
We did several iterations of moving from 5 deg to stow using both the =
button in STUI and the command line in the log window, and saw this =
problem occur once.
After reinitializing the axes it moved normally to stow.
One thing I did see is that the AZ brake indicator on the interlocks =
screen was commonly showing "neither" even during slews that worked =
normally.  I think this is a good clue to what is going on - the brake =
is physically either on or off, as it's run by ~90 psi air.  In this =
case that brake is on because the telescope is at a low ALT, the AZ =
brake comes on at 15 degrees.  When the TCC tries to activate the AZ =
axis to slew from 5 to 30, it appears to have no idea that the MCP and =
PLC aren't going to let it move the AZ.  Seems like a weird approach to =
me, but I'm used to thinking that by now.
What you are seeing with the "neither" is the monitoring GUI (whether =
the STUI version or the MCP version) waiting for brake state information =
from the MCP itself.  The MCP is an old and very slow computer, and the =
GUI's are running on much faster machines, and I'm guessing "neither" is =
what you see as the GUI waits.
As far as the ALT brake coming on, I can only speculate that TCC doesn't =
always wait gracefully for a response from the MCP.  I can't provide a =
definitive answer as to why it sometimes upsets the other axes and =
sometimes it doesn't, but I do expect that this kind of problem is one =
of the things that will presumably be fixed by upgrading the MCP =
hardware.  I've seen all axes go out of closed loop when using the MCP =
and you hit "brake on" for just one axis.  We've never addressed this =
because its just an annoying problem - it takes minimal time to reinit =
and it all comes back.  If we had lots of software support and an actual =
hierarchy for prioritizing problem solving, that is just the kind of =
thing that would get fixed eventually.
I've asked Deanna to put my home numbers in both Cloudcroft and Corona =
back on the phone list.  Try the cell twice then the land lines.  =
Sometimes I go to Corona and Jean stays in CC for the weekend, so you =
might get an irate wife rather than me, but that's a risk you'll have to =
run.
> On Apr 11, 2019, at 7:57 AM, John Downey  wrote:
>=20
>=20
>> Night:
>> =3D=3D=3D=3D=3D=3D
>>=20
>> High wind and dust.
>>=20
>> 03:57Z - The weather got much better. We mounted cart 1 and opened =
the
>> enclosure, but the telescope would not move up from 5 degrees to =
stow. The
>> interlock window showed the altitude brake in a 'neither' state and =
the 15-deg
>> stop retracted. We tried resetting the amplifiers, fiddling with the =
15-deg
>> stop, checked the air pressure, rebooted the MCP. Unsure what to do =
next, we
>> called JohnD and left a message on his phone. We also called James to =
ask if he
>> may have done something with the interlocks somehow - he didn't.=20
>>=20
>> 04:10Z - We put the telescope back into the enclosure when the wind =
gusts went
>> over the limit again.=20
>>=20
>> We called Kaike. He suggested to try to move the telescope with the =
MCP. We did,
>> and it worked. We cycled the telescope back and forth between 5-deg =
and stow.
>> Then mounted cart 6. The odd behavior presented again a few times, =
but not every
>> time. It is intermittent.=20
>>=20
>> What seems to happen is the altitude brake does not always disengage =
with
>> commands with the TCC, but disengages fine with the MCP. Whenever =
this problem
>> happens, the interlocks window shows the altitude brake as 'neither'.=20=
>=20
> For some reason my phone never rang last night.  I was there, it was =
on, it announced a voicemail but no missed calls so I didn't know what =
to make of that and ignored it.
>=20
> We'll try to replicate the problem, I have an idea what it is since =
I've seen something like this before.  It may be new behavior, but I =
think it is just the kind of clash you get when there are 3 ways of =
doing the same thing and a dozen interlocks that work a little =
differently with each method of movement.
>=20
> ____________________________________________________________
> Logs of Nightly APO Activities
> =
https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html