Date: Tue, 14 Feb 2017 12:30:38 -0500 (EST)
From: jeg@astro.princeton.edu
Subject: Re: [sloan-25mlog 3365] 2.5m obslog 20170213 (SJD 57798)
Well, fiddling with the power supply cables is almost as dangerous
as fiddling with the controller->camera cables, since we deliver a
very large number of voltages to the controller, which are then
modified and sent to the camera. CCDs are relatively robust to
powerup/powerdown sequencing, but one can imaging situations in
which enough substrate current could be pulled to damage or
destroy a device. Also, with the sort of band-aid changes from
SDSS-I to BOSS, not all of the power connections are in the
DB-25 power cable, and this includes the dangerous ones. So
it should be an absolute policy that no cable to or from the
saddebag is ever reseated without powering down...as you say,
this is common sense for all cables, but it needs to be a rule
for these.
I think it would be a good thing to do to look into more positive
retention mechanisms for these cables, though I *think* that
they are all at least supposed to be retained by jackscrew/post
combinations--but I am not at all surprised that many are missing/
broken. The 80-pin SCSI connectors with their 2-56 jackscrews
are AWFUL, and the 50-pin cameralink-like ribbon things really
have to be mated very well to even work--and these cables carry
not only data but power. At one point we began a campaign to
put polyurethane tube `handles' on all the jackscrews so that
(a) it is impossible to use a screwdriver seat them and hence at least
attempt to avoid breaking them, and
(b) make it easy to loosen/tighten them by hand
but I do not think we got to all the cables.
--jim
On Tue, 14 Feb 2017, John Downey wrote:
> This was on the other side of the saddlebag, between the controller + power supplies and the saddlebag rather than between the saddlebag and camera.  But considering the equipment, I did want to make it clear that its always the best practice to power it off before reseating and that the operations recorded in the log did reflect that.
>
> Had we been doing anything with the cables going to the cameras, I probably would have been talking to Connie first.
>
> Honestly, I wonder that we do not have more problems like this with many of the heavy cables held in by friction alone.
>
> John
>
>> On Feb 14, 2017, at 6:44 AM, jeg@astro.princeton.edu wrote:
>>
>> Yes, please do not reseat ANY CCD cable without powering the system
>> down first; the detector could easily be destroyed. The jackscrews
>> on most of these connectors are woefully inadequate and fragile,
>> though the connectors themselves are more-or-less OK. But in any
>> case the screws and posts need to be carefully maintained. Maybe this
>> should be a regular summer maintenance activity, probably by Mike,
>> Connie, and myself.
>>
>> --jim
>>
>> On Tue, 14 Feb 2017, John Downey wrote:
>>
>>>>
>>>> R1 was reading zero signal during the cals for afternoon checkout. JohnD
>>>> re-seated some cables on the saddlebag, but ended up power-cycling both cameras
>>>> on SP1 to fix it.
>>>>
>>>
>>> A minor point for clarity, I just pushed in on the cables at both ends rather than reseating them.  In general I don't like to reseat cables on equipment while it is powered on.
>>> Most of the R1 cable connectors are missing the nuts or their screws are damaged - last fall we fixed this on B1 when we had trouble, but most of R1's screws were not working to hold the plug in it's socket last night.  This is something to add to summer shutdown I think.
>>> ____________________________________________________________
>>> Logs of Nightly APO Activities
>>> https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html
>>>
>
>