Date: Sat, 24 Sep 2005 20:31:52 -0700
From: Russell E Owen
Subject: Re: 2.5m obslog 20050923 (MJD 53637)
At 11:42 AM +0000 9/24/05, observer@sdsshost2.apo.nmsu.edu wrote:
>TCC errors: clock_loss_signal at all three axis, but only lasted for
>a few seconds and gone. The error message disappeared without being
>cleared. This occured at lest four times during the night.
This message comes from the MCP and so should appear for all axes at
the same time (unless something really weird is going on).
I suspect tracking will be fine during brief outages. It's a question
of how quickly the MCP's clock drifts, and I don't know the answer.
It's most likely to harm imaging. The only affect it'll have on
spectra is degraded pointing, making field acquisition harder.
But the message is still very disturbing. The MCP reads the GPS
clock's 1 Hz signal, and that should be totally reliable.
At risk of stating the obvious, the possibilities seem to be limited to:
- The GPS clock
- The link from the clock's 1 Hz signal to the MCP
- The MCP
My first guess would be the link or the MCP, but that's only a guess.
The clock would have to be really sick not to output a 1 Hz signal.
It's an old clock, but they tend to be reliable.
The clock is easy to check, but I'm not sure how to check the other
possibilities. To check the clock:
- Issue the SET TIME command from the TCC. If this completes silently
then the clock is happy.
- Look at the display on the clock (in the computer room). I'm not
sure what, if anything it shows if it's unhappy, but a manual is
around if you're curious.
-- Russell