## PSN-L Email List Message

Subject: Re: Thermal Control for Sensors
From: ChrisAtUpw@.......
Date: Mon, 19 Nov 2007 09:33:24 EST

```
In a message dated 19/11/2007, skmort@............ writes:

Hi Chris,
Actually as I understand it, it is the switch  off temp that is fixed!   In
other words, once the temp has reach  its operational point, the control starts
skipping cycles to maintain a max  set temp?

Hi Stephen,

That is what I have assumed from your description.  Say you set it up in
Summer with an air temperature of 30 C.
The ground temperature minimum may be 20 C, giving  a comfortable
temperature difference of 10 C Deg.

You leave it at the same setting in the  winter when the ground
temperature has fallen to -5 C
The heater still maintains a top temperature of 30  C, giving a
differential within the case of 35 C deg instead of 10 C Deg. This  is likely large
enough to cause convection noise within the case.

If you can supply just enough heat to follow the  ground temperature
fluctiations + say 10 Deg, this would be ideal. It would also  use less power.

Maybe you could use two thermistors? If you send me  the circuit, I will
have a look.

Regards,

Chris Chapman

In a message dated 19/11/2007, skmort@............ writes:
<=
FONT=20
style=3D"BACKGROUND-COLOR: transparent" face=3DCourier color=3D#000000 siz=
e=3D3>Hi Chris,Actually as I understand it, it is the s=
witch=20
off temp that is fixed!   In other words, once the temp has reac=
h=20
its operational point, the control starts skipping cycles to maintain a ma=
x=20
set temp?

Hi Stephen,

That is what I have assumed from your descripti=
on.=20
Say you set it up in Summer with an air temperature of 30 C.
The ground temperature minimum may be 20 C, giv=
ing=20
a comfortable temperature difference of 10 C Deg.

You leave it at the same setting in the=20
winter when the ground temperature has fallen to -5 C
The heater still maintains a top temperature of=
30=20
C, giving a differential within the case of 35 C deg instead of 10 C Deg. Th=
is=20
is likely large enough to cause convection noise within the case.

If you can supply just enough heat to follow th=
e=20
ground temperature fluctiations + say 10 Deg, this would be ideal. It would=20=
also=20
use less power.

Maybe you could use two thermistors? If you sen=
d me=20
the circuit, I will have a look.

Regards,

Chris Chapman
```

[ Top ] [ Back ] [ Home Page ]