Subject: Re: [sloan-25mlog 4973] Re: 2.5m obslog 20200210 (SJD 58890)
From: John Downey
Date: Tue, 11 Feb 2020 08:53:40 -0700
When I went out to try it, wind_AZ_ready did not clear, as expected.
With just the podium stop button in, I powered the windbaffle controller =
off, counted to 10 and powered it back on.  Worked fine after that.
We want this fixed as much as the night staff does, perhaps more. =20
The more we looked into how it was built in 2016, the more we became =
convinced that it probably only worked this long due to no less than 4 =
grounds some of which float independently - ie:  have significant =
resistance relative to one another.  In addition, there are at least two =
cases where a power signal is tied to a return (better word for the =
ground) that comes from an entirely different power supply.  In another =
case, the signal that switches servoing on and off for the ALT amplifier =
is tied to the PLC input return, but then gets 24VDC dumped into it when =
the ALT brake is off - this should have been done with a relay instead.
The concern at the moment is that if we start resolving these issues =
without a full plan for correcting all of them, we will break something. =
 Such a plan takes time for thought and measurements, which we are =
doing.  We did not have the plan ready for the previous engineering =
time, we might for March engineering.  As there is no backup system for =
this one (like there is for the lift pendant) we are proceeding with a =
great deal of caution.  With this if we break it we will be down until =
fixed.
> On Feb 11, 2020, at 7:47 AM, John Downey  wrote:
>=20
>=20
>>=20
>> Note:
>> =3D=3D=3D=3D=3D
>> When we tried to swap cart 2 out in the morning, wind_AZ_ready would =
not clear.
>> After spent 15 minutes or so in cycling e-stops and wind baffle power =
switch,=20
>> it snows too heavily for a cartridge change, we decided to leave the =
issue to
>> day staff since they may want to see the problem in the real time.=20
>>=20
>=20
> We see the problem in real time just about daily.  The hard part is =
solving it.
>=20
>=20
>=20
> ____________________________________________________________
> Logs of Nightly APO Activities
> =
https://trac.sdss.org/tracmailman/browser/private/sloan-25mlog/index.html