[time-nuts] Re: Envornmental control

[Tom Van Baak]

John,

I'm cc'ing the list since I'd be interested in other
stories about temperature control.

My upstairs lab/office has a free-standing, in-room,
4000? BTU air-conditioner and it seems to keep
the room to a couple of degrees F at all times all
year round. There is no need to heat - the clocks,
counters, and PC's take care of that ;-)

I tried something like 70 F at first but it could not
meet that in July and August so the last few years
it has been set at 78 F which works all year-round.

So no complaints there. The only problem is that
it blows exhaust air through a 4-inch hose to the
outside. This creates a partial vacuum in the room
which is filled by stealing air slowly from the rest
of the house. So there is a slight draft in the room
as well as more dust that the rest of the house.

If I did it over again I would never use an in-room A/C
again and instead use a split system or a regular wall
or window mounted A/C where the inside air loop is
separated from the outside air loop.

Note also the A/C is on the same UPS backup as
the instruments. You do not want a situation where
the A/C fails, the clocks keep running, and the room
heats up to thermal max.

My downstairs lab, where the good clocks are, uses
a 5000 BTU wall-mount A/C ($79 on sale) so I don't
have the draft or dust problem. The room is insulated
and sealed.

The A/C fan is on at all times. Once you have air
flow do not vary that (to sensitive instruments, varying
airflow can be as bad as varying temperature). The
A/C compressor, though, has it's normal on/off cycles

These thermostatic cycles began to show up in my
data so I switched to having the A/C on at all times.
This improves short-term temperature stability a lot
but it also degrades long-term seasonal temperature
stability and invites freeze-up during winter months.

The solution I'm using now is quite unusual. The A/C
fan and compressor are on full-time. What I vary is a
700 W electric space heater placed in front of the 
A/C intake. Now it goes against my Dutch heritage
to have an A/C on full blast and then moderate the
excess cooling with an electric heater but this works
extremely well. The room is now stable to better than
1 C; often 0.1 C. This will improve even more when I
implement PID PWM control of the heater current.

/tvb

----- Original Message ----- 
From: "John Ackermann N8UR" <jra at febo.com>
To: "Tom Van Baak" <tvb at leapsecond.com>
Sent: Wednesday, March 09, 2005 12:08
Subject: Envornmental control


> Hi Tom --
> 
> I think I recall you've now built an environmentally controlled clock 
> vault.  I wonder if you can tell me some of the basics of the design, 
> and in particular what you're using for the heating/cooling system.  I 
> may try to accomplish something like that at home but haven't had much 
> luck with the basic questions of how you generate and pipe the airflow.