Subject: Re: [sloan-25mlog 4603] 2.5m obslog 20190830 (SJD 58726)
From: John Downey
Date: Sat, 31 Aug 2019 15:44:29 -0600
The telescope was slightly front heavy, but the problem is really a =
minor bug in the MCP (I've emailed Jeff about it, should be an easy fix =
remotely).  He may ask that it be tested tonight, depending on when he =
gets the email and makes the change.
The bug is in the sequence where the MCP stops an axis in response to a =
stop button or "S" command, or presumably and "axis stop" from the TCC.  =
It needs to set the brake, then open the servo loop and abort the =
amplifier state.  If it doesn't set the brake first, then minor =
variation in balance with altitude become very significant.  I've not =
seen a counterweight setting yet that will balance this telescope at 30 =
ALT as well as 90 ALT - any setting is a bit of a compromise, but with =
the motors active until a moment after the brakes set, you never see a =
problem.  With the current setting, the ALT axis is kind of in neutral =
for a moment between hitting the stop and having the ALT brakes come on.
The solution, for now, is to use the "B" command in mcpop before hitting =
a stop button.  When it's done this way, the brakes come on and then the =
loop opens, etc. and the telescope doesn't have a delay between the stop =
button and the brakes.
I've rebalanced for 60 degrees, which is about normal.  Ideally I would =
want it to be at a neutral balance at zenith and slightly back heavy at =
lower ALT like 20-30, but we've yet to achieve this with any counter =
weight setting.  Typically it ends up being a little back heavy at =
zenith and a little front heavy at low altitudes.  The idea with moving =
the weights differentially was to try to minimize this feature, but we =
do not yet know if its possible.
Incidentally, the numbers on the LCD display are in volts.  A setting of =
279 is 2.79 volts.  The string potentiometer on each weight has a =
voltage applied across it, and as the resistance changes with weight =
position, the measure of voltage drop across the pot changes.  I don't =
know what the MCP screen displays, I had always assumed a distance of =
some kind, but perhaps it is intended to be some guess as to the weight =
of what was installed.  Setting for a cartridge is a bit heavy in that =
case, but we all know this system is not very precise anyway.
> On Aug 31, 2019, at 6:54 AM, John  wrote:
>=20
>=20
>> Night:
>> =3D=3D=3D=3D=3D=3D
>>=20
>> Cleaned the drive surfaces in preparation for another fiducial data =
set.
>>=20
>> When we slewed from zenith to stow, we saw a lurch when a stop button =
was
>> pressed, altitude went down from 30:00 to 29:54, signifying that the =
telescope
>> is quite top heavy. We also noticed that the counterweights are not =
even. Two of
>> them are 30 lbs heavier than the rest, which we assume is on purpose. =
We will
>> ask John to look at the balance tomorrow.
>=20
>=20
> Counterweights being uneven is intentional.  I'm going in today anyway =
so I will see if I can duplicate the problem and adjust the balance if =
needed.
>=20
> ____________________________________________________________
> Logs of Nightly APO Activities
> =
https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html