From: Craig Loomis
Subject: Re: 2.5m obslog 20040815 (MJD 53233)
Date: Mon, 16 Aug 2004 16:04:26 -0600
    Looking for a quick way of crudely checking whether the lamps are
taking too long to start up, we took a run of 1s guider images after
turning the MgCd lamps on. I don't know what the real criteria are, so
I'll just point you to a plot. The counts are the mean of an 800 pixel
rectangle taken from fiber 11.
http://www.apo.nmsu.edu/users/cloomis/mgcd.png
  - craig
On Aug 16, 2004, at 10:43 AM, jegunn wrote:
>
> I think it might be more instructive to put the OLD ones BACK in.
> We did not have this trouble before, and it may be that the new lamps
> are somehow different. We DO know which the old ones were, don't we?
> ;-)
>
> thanks
>
> --jim
>
> On Mon, 16 Aug 2004, Larry N. Carey wrote:
>
>> Just a clarification about changing out the HgCd lamps. Is everyone
>> aware that all four HgCd, all four Neon, and all 4 Quartz lamps were
>> installed new on August 5th? In particular, the HgCd cost about
>> $2,000.00 to replace; so If they are to be changed out again, i
>> suggest that it be specifically to do a one nite test (with all other
>> conditions being constant) to determine whether or not they are
>> indeed the problem. Sound reasonable? (I'm guessing that everyone is
>> also aware that we have been blowing fuses on the quartz lamps, for
>> which i don't believe we have an explanation at this time.)
>>
>> Unless i hear otherwise, we will change out the 4 HgCd lamps this
>> afternoon. Then switch them back to the current ones tomorrow morning
>> unless they prove to be the problem.
>>
>> larry