Suppression of (electrical) noise from CH Thermostat
 

OK, the title doesn't tell the whole story, because I am using a CH stat for
a non-standard purpose.

I'd better explain!

I've been having problems keeping the inside of my computer, and the
processor, cool - even after adding extra DC fans. So I have solved the
problem by fitting one of Mr Maplin's 120mm mains-driven fans in the side of
the (tower) casing - which blows quite a lot of ambient air past the
electronics, exhausting through the rear.

Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I have a
spare Danfoss TWP room stat after fitting a programmable stat on the central
heating. So I've mounted this TWP inside the computer case, and wired it to
switch the fan on when the temperature is above the set value.

This works quite well *except* for the fact that there's quite a lot of
crackling on the computer speakers for several seconds each time the stat
switches on or off.

It seems to me that I ought to be able to suppress this noise with a
suitable capacitor. Can some kind soul please advise on the spec/value of
the capacitor required, and whether to wire it in parallel with the motor,
or across the switch in the stat. FWIW the fan motor is described as
"impedance protected shaded-pole motor" rated at 22w 140mA (which implies a
power factor of about 0.65)

TIA.
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

On Sun, 12 Sep 2004 15:10:06 +0100, Set Square wrote:


Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I have a
spare Danfoss TWP room stat after fitting a programmable stat on the central
heating. So I've mounted this TWP inside the computer case, and wired it to
switch the fan on when the temperature is above the set value.


why not use a pot (potential devider) to slow the fan down ?

Martin Warby

In an earlier contribution to this discussion,
Martin Warby wrote:

On Sun, 12 Sep 2004 15:10:06 +0100, Set Square wrote:


Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I
have a spare Danfoss TWP room stat after fitting a programmable stat
on the central heating. So I've mounted this TWP inside the computer
case, and wired it to switch the fan on when the temperature is
above the set value.


why not use a pot (potential devider) to slow the fan down ?

Martin Warby

Do you mean insert a resistor in series with the fan - so that it *always*
runs at at least low speed - and use the stat to by-pass the resistor when
max cooling is required?

Would that work with this type of motor?

Wouldn't contact noise from the stat still get into the system?
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

On Sun, 12 Sep 2004 15:47:08 +0100, Martin Warby wrote:

why not use a pot (potential devider) to slow the fan down ?

On a mains fan? Not to mention the fact that the dumped power ends up
as heat and the power dissipation required is a bit above your average
variable "pot", which generally aren't mains rated either...

I guess you could use a suitably rated wirewound resistor, slowing the
fan down will help the noise, indeed just taking 10% of the speed will
make quite a difference.

Suppresion of the stat, 0.1uF in series with 100R across the contacts.
Either as suitably rated components or a contact suppressor like
Maplin RG22Y. Note the Specification part of that items description
appears blank in my browser check the device is mains rated...

--
Cheers
Dave. pam is missing e-mail

"Set Square" wrote in message
...
In an earlier contribution to this discussion,
Martin Warby wrote:

On Sun, 12 Sep 2004 15:10:06 +0100, Set Square wrote:


Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I
have a spare Danfoss TWP room stat after fitting a programmable stat
on the central heating. So I've mounted this TWP inside the computer
case, and wired it to switch the fan on when the temperature is
above the set value.


why not use a pot (potential devider) to slow the fan down ?

Martin Warby

Do you mean insert a resistor in series with the fan - so that it *always*
runs at at least low speed - and use the stat to by-pass the resistor when
max cooling is required?

Would that work with this type of motor?

Wouldn't contact noise from the stat still get into the system?
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.


Use the correct fan, you can get them with internal / external thermisters
and this regulates the fan speed.

On Sun, 12 Sep 2004 16:34:52 +0100, Set Square wrote:

In an earlier contribution to this discussion,
Martin Warby wrote:

On Sun, 12 Sep 2004 15:10:06 +0100, Set Square wrote:


Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I
have a spare Danfoss TWP room stat after fitting a programmable stat
on the central heating. So I've mounted this TWP inside the computer
case, and wired it to switch the fan on when the temperature is
above the set value.


why not use a pot (potential devider) to slow the fan down ?

Martin Warby

Do you mean insert a resistor in series with the fan - so that it *always*
runs at at least low speed - and use the stat to by-pass the resistor when
max cooling is required?

Would that work with this type of motor?

Wouldn't contact noise from the stat still get into the system?

I was thinking of a pot (a variable resistor looks like
http://www.maplin.co.uk/media/largeimages/2203i0.jpg).Set it up as a
potential devider as below

connect outer 2 tags to supply
connect inner tag to +ve lead of fan
connect -ve led of fan to -ve lead on outer 2 tags

Simply turn the pot till the desired speed is obtained

Martin Warby

On Sun, 12 Sep 2004 17:07:02 +0100, Dave Liquorice wrote:

On Sun, 12 Sep 2004 15:47:08 +0100, Martin Warby wrote:

why not use a pot (potential devider) to slow the fan down ?

On a mains fan? Not to mention the fact that the dumped power ends up
as heat and the power dissipation required is a bit above your average
variable "pot", which generally aren't mains rated either...

I guess you could use a suitably rated wirewound resistor, slowing the
fan down will help the noise, indeed just taking 10% of the speed will
make quite a difference.

Suppresion of the stat, 0.1uF in series with 100R across the contacts.
Either as suitably rated components or a contact suppressor like
Maplin RG22Y. Note the Specification part of that items description
appears blank in my browser check the device is mains rated...

DOH should have read post more throughly thought it was a DC fan but was
using a mains rated thermostat.I agree using a pot would be a BAD idea in
the circumstances.

I think if normal DC fans aren't powerful enough to cool the system then
the general airflow in the case needs to be considered

Martin Warby

Set Square wrote:
OK, the title doesn't tell the whole story, because I am using a CH stat for
a non-standard purpose.
snip
I've been having problems keeping the inside of my computer, and the
processor, cool - even after adding extra DC fans. So I have solved the
problem by fitting one of Mr Maplin's 120mm mains-driven fans in the side of
the (tower) casing - which blows quite a lot of ambient air past the
electronics, exhausting through the rear.

http://www.ebuyer.com/
Lots of stuff for doing this sort of thing, very cheaply.
Lots of options, from thermostatically controlled fans to little control
panels that can be set to run fans at certain temperatures from multiple
temperature probes, and have lots of blinking lights on.

In an earlier contribution to this discussion,
Ian Stirling wrote:

Set Square wrote:
OK, the title doesn't tell the whole story, because I am using a CH
stat for a non-standard purpose.
snip
I've been having problems keeping the inside of my computer, and the
processor, cool - even after adding extra DC fans. So I have solved
the problem by fitting one of Mr Maplin's 120mm mains-driven fans in
the side of the (tower) casing - which blows quite a lot of ambient
air past the electronics, exhausting through the rear.

http://www.ebuyer.com/
Lots of stuff for doing this sort of thing, very cheaply.
Lots of options, from thermostatically controlled fans to little
control panels that can be set to run fans at certain temperatures
from multiple temperature probes, and have lots of blinking lights on.


Thanks. Could you please point to a specific area on the Ebuyer site -
there's a lot of stuff to wade through!
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

In an earlier contribution to this discussion,
Martin Warby wrote:


I think if normal DC fans aren't powerful enough to cool the system
then the general airflow in the case needs to be considered

Martin Warby


I've had one or two "funnies" with the computer in the past which look as if
they might be DC power related - so I'm trying to avoid loading the internal
power supply any more, even though it has been uprated.
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

In message , Martin Warby
writes
On Sun, 12 Sep 2004 16:34:52 +0100, Set Square wrote:

In an earlier contribution to this discussion,
Martin Warby wrote:

On Sun, 12 Sep 2004 15:10:06 +0100, Set Square wrote:


Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I
have a spare Danfoss TWP room stat after fitting a programmable stat
on the central heating. So I've mounted this TWP inside the computer
case, and wired it to switch the fan on when the temperature is
above the set value.


why not use a pot (potential devider) to slow the fan down ?

Martin Warby

Do you mean insert a resistor in series with the fan - so that it *always*
runs at at least low speed - and use the stat to by-pass the resistor when
max cooling is required?

Would that work with this type of motor?

Wouldn't contact noise from the stat still get into the system?

I was thinking of a pot (a variable resistor looks like
http://www.maplin.co.uk/media/largeimages/2203i0.jpg).Set it up as a
potential devider as below

connect outer 2 tags to supply
connect inner tag to +ve lead of fan
connect -ve led of fan to -ve lead on outer 2 tags

Simply turn the pot till the desired speed is obtained

Until you see smoke and smell burning

Have you looked at the power rating of a normal pot?

1Watt IIRC


--
geoff

In an earlier contribution to this discussion,
Dave Liquorice wrote:


Suppresion of the stat, 0.1uF in series with 100R across the contacts.
Either as suitably rated components or a contact suppressor like
Maplin RG22Y. Note the Specification part of that items description
appears blank in my browser check the device is mains rated...

Many thanks Dave - looks like just the thing. I think the RG22Y is mains
rated. I captured the picture from the Maplin website and blew it up - and
although it's not wonderfully distinct, I'm pretty sure it says 250VAC on
it.
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

In message , Set Square
writes
In an earlier contribution to this discussion,
Martin Warby wrote:


I think if normal DC fans aren't powerful enough to cool the system
then the general airflow in the case needs to be considered

Martin Warby


I've had one or two "funnies" with the computer in the past which look as if
they might be DC power related - so I'm trying to avoid loading the internal
power supply any more, even though it has been uprated.

You could always go for water cooling, quite a few of them about
nowadays

It sounds like you have some fundamental problems to sort out

I have a desktop 2 gig machine with 4 HDs, a gig of memory, quality
graphics card etc in a case with poor ventilation, no problems with it
cutting out with overheating

--
geoff

On Sun, 12 Sep 2004 18:57:56 +0000, raden wrote:

In message , Martin Warby
writes
On Sun, 12 Sep 2004 16:34:52 +0100, Set Square wrote:

In an earlier contribution to this discussion,
Martin Warby wrote:

On Sun, 12 Sep 2004 15:10:06 +0100, Set Square wrote:


Works a treat - *except* that it's a bit (audible) noisy, and doesn't
actually *need* to run all the time. Now it just so happens that I
have a spare Danfoss TWP room stat after fitting a programmable stat
on the central heating. So I've mounted this TWP inside the computer
case, and wired it to switch the fan on when the temperature is
above the set value.


why not use a pot (potential devider) to slow the fan down ?

Martin Warby

Do you mean insert a resistor in series with the fan - so that it *always*
runs at at least low speed - and use the stat to by-pass the resistor when
max cooling is required?

Would that work with this type of motor?

Wouldn't contact noise from the stat still get into the system?

I was thinking of a pot (a variable resistor looks like
http://www.maplin.co.uk/media/largeimages/2203i0.jpg).Set it up as a
potential devider as below

connect outer 2 tags to supply
connect inner tag to +ve lead of fan
connect -ve led of fan to -ve lead on outer 2 tags

Simply turn the pot till the desired speed is obtained

Until you see smoke and smell burning

Have you looked at the power rating of a normal pot?

1Watt IIRC

see my earlier post I had not realised it was a mains fan (I presumed it
was running from the 12V DC line).

Martin Warby

Set Square wrote:
In an earlier contribution to this discussion,
Ian Stirling wrote:

Set Square wrote:
OK, the title doesn't tell the whole story, because I am using a CH
stat for a non-standard purpose.
snip
I've been having problems keeping the inside of my computer, and the
processor, cool - even after adding extra DC fans. So I have solved
the problem by fitting one of Mr Maplin's 120mm mains-driven fans in
the side of the (tower) casing - which blows quite a lot of ambient
air past the electronics, exhausting through the rear.

http://www.ebuyer.com/
Lots of stuff for doing this sort of thing, very cheaply.
Lots of options, from thermostatically controlled fans to little
control panels that can be set to run fans at certain temperatures
from multiple temperature probes, and have lots of blinking lights on.


Thanks. Could you please point to a specific area on the Ebuyer site -
there's a lot of stuff to wade through!

Sorry.
You're looking for the "case mods" section.
I think then the misc subsection.
For example, I've got a 4 channel fan speed controller for 3.99.
And 5 (sleeve bearing) 120mm fans for 72p each. (that are remarkable in
that they will reliably start down at 2.4V, so can be run from the 3.3V
line, in almost absolute silence.)

In an earlier contribution to this discussion,
Ian Stirling wrote:



Thanks. Could you please point to a specific area on the Ebuyer site
- there's a lot of stuff to wade through!

Sorry.
You're looking for the "case mods" section.
I think then the misc subsection.
For example, I've got a 4 channel fan speed controller for 3.99.
And 5 (sleeve bearing) 120mm fans for 72p each. (that are remarkable
in that they will reliably start down at 2.4V, so can be run from the
3.3V line, in almost absolute silence.)


Thanks - but does any this stuff work with a *mains* voltage fan - which is
what I've got?
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

In message , Martin Warby
writes
On Sun, 12 Sep 2004 18:57:56 +0000, raden wrote:

Do you mean insert a resistor in series with the fan - so that it *always*
runs at at least low speed - and use the stat to by-pass the resistor when
max cooling is required?

Would that work with this type of motor?

Wouldn't contact noise from the stat still get into the system?

I was thinking of a pot (a variable resistor looks like
http://www.maplin.co.uk/media/largeimages/2203i0.jpg).Set it up as a
potential devider as below

connect outer 2 tags to supply
connect inner tag to +ve lead of fan
connect -ve led of fan to -ve lead on outer 2 tags

Simply turn the pot till the desired speed is obtained

Until you see smoke and smell burning

Have you looked at the power rating of a normal pot?

1Watt IIRC

see my earlier post I had not realised it was a mains fan (I presumed it
was running from the 12V DC line).

Even then, you're pushing it

--
geoff

Set Square wrote:
In an earlier contribution to this discussion,
Ian Stirling wrote:



Thanks. Could you please point to a specific area on the Ebuyer site
- there's a lot of stuff to wade through!

Sorry.
You're looking for the "case mods" section.
I think then the misc subsection.
For example, I've got a 4 channel fan speed controller for 3.99.
And 5 (sleeve bearing) 120mm fans for 72p each. (that are remarkable
in that they will reliably start down at 2.4V, so can be run from the
3.3V line, in almost absolute silence.)

Thanks - but does any this stuff work with a *mains* voltage fan - which is
what I've got?

No.
DC fans are pretty cheap, and may be more suitable, as they can be
slowed down easily, whereas AC ones would require more complex electronics.

Thanks - but does any this stuff work with a *mains* voltage fan - which
is
what I've got?

I think the idea is that at 72p, you could possibly purchase a new DC one?

Christian.

In an earlier contribution to this discussion,
Christian McArdle wrote:

Thanks - but does any this stuff work with a *mains* voltage fan -
which is what I've got?

I think the idea is that at 72p, you could possibly purchase a new DC
one?


True - but I've just invested rather more than that in a nice mains fan,
with proper bearings - so that it doesn't make a lot of *mechanical* noise
like cheap fans do.

--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.

Set Square wrote:
In an earlier contribution to this discussion,
Christian McArdle wrote:

Thanks - but does any this stuff work with a *mains* voltage fan -
which is what I've got?

I think the idea is that at 72p, you could possibly purchase a new DC
one?


True - but I've just invested rather more than that in a nice mains fan,
with proper bearings - so that it doesn't make a lot of *mechanical* noise
like cheap fans do.

I'm surprised at the 12V fan (72p).
At 12V it's as quiet as any fan I've seen with the same airflow.
At 5V, it's barely audible, and still moves a decent amount of air.
At 3.3V, it's inaudible (20cm from fan) and creates a reasonable draft.

There is no "clicking" or electronic noise from the rotor at low speeds
like there is on many fans.