Thermistor

The thermistor should be as close to the area you want to measure as possible. Not sure why it would be inside a case, unless it was the inside of the case's temperature that needs monitoring ....

I take it this is a classified project, since you give us almost no detail. For quick response a thermistor needs to go where you're measuring. For a glacially slow response it can go elsewhere when the temp there follows the temp to be measured.

NT

In message , rick writes

We were not attempting close temperature control. The thermistor was potted inside a short length of metal tube and then clamped to the fins of refrigeration plant.

ive got temp monitoring on this PC and in warmer weather the internal temps sorta track the outside with a couple of hours lag and the fans going anyway ;-)

seriously, in air, inside a sealed case? forget it. In air inside a case with airflow from whats to be measured going through it, fine.

Are you sure this is going to work?(depends where and how you are using the fan) fans in a closed area don't usually cool,they add heat,(from the motor) unless they are moving cold air from somewhere else or moving hot air which is replaced by colder air. fans make you feel cooler by evaporating sweat, they don't actually change air temp much.

Not sure why you think that .... this is a home build project to control the speed of a computer fan ... thus allow cooling of a comms cupboard that gets too hot in summer.

The circuit is based on the cct here:

They use a bare board ... for safety & longevity building this into al aluminium case. That is why question of where to mount thermistor comes from.

Yes it will work ... it is not in a sealed box .. but in top of teh comms room cupboard. The fan will expel air out of the cupboard, fresh air coming in at bottom of cupboard. Ambient temp is fine ... just closed cupboard gets too warm in summer

well, you avoided giving us the relvant info, until now

It would work gluing it to the case, but woudl give a time lag in operation, resulting in poor temp control and wasted energy. I'd put it near top of cabinet in the airstream.

NT

Some good old classic components in that circuit :)

Andrew

I can't see anything on that page

NT

Basically a 741 op amp comparing a fixed divide chain to one with a thermistor in it - 'supose it could have been even more classic and used a

Andrew

It's the image from this article

thanks you two.

That circuit contains no feedback, making it bangbang operation rather than proportional.

Using transistors as temp sensors is cheaper.

NT

Sorry guys referenced wrong cct ...

should be this one .... a fair improvement on basic 741 this gives full PWM control of speed.

or

The question is the same.

wrong link .... reposted

wrong cct .... corrected

If you have issues with drop box ... its the Thermistor version here:

I can't see anything wrong with your idea. You need to get the thermistor somewhere near the top of the cupboard - it doesn't have to be directly in the airstream as what you want is to sense the cupboard temperature, not necessarily the airflow. If there is a lot of heat buildup I'd be tempted to duplicate R4, T1 & D1 to add a second fan. As larger fans are quieter and more efficient I think I'd use 2 150mm fans. They should cycle down to a point where there is very little noise.

Just a point - remember that if the thermistor doesn't sense changes in air temperature it can't work. That means you have to make sure that rising air passes it even without the fan(s) running otherwise it won't run them. It's surprising how easy it is to find "dead zones" where this can happen. :)

I posted wrong link. This is the correct schematic:

Full PWM control

Same question though on Thermistor position