• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Temperature controlled fan

Hi,

I’m looking for a simple (minimalistic) way to switch a 12 volt DC low current fan ON when a temperature of for example 30 degrees Celcius is reached. When the temperature drops below 30 degrees Celsius the fan should be switched OFF.

For this particular purpose the exact temperature is not too relevant, som deviation is allowed as well as some hysteresis.

A control for setting the ON temperature would be nice, but is not absolutely necessary.

Any ideas?

Regards, Gerrit
 
I am glad I do not have to use a fan.

I have multiple holes in the amplifier bottom plate.
I have multiple holes in the amplifier top plate.
I use lots of space between tubes and other objects.
I space parts well, and use resistors that are large (dissipation ratings at least 3 to 5 times the power they will have to dissipate).
I run the tubes below their maximum ratings (plate and screen dissipation, plate and screen voltages, max filament voltage, etc.).
I use the amplifiers out in the open; no covers.
And whenever possible, I use 105 degrees C capacitors.
Reliability and simplicity.

What do I loose or trade off for those practices?
Max Power output and the need for fans.
 
Thermistor and a comparator will work. You could jump into the Arduino world and do it digitally with fewer parts.
I've designed and had this circuit in use since 2007, controlling the circulator pump on my hot water heating system.
Utterly reliable, it triggers on the pump when the thermistor senses the pipe being hot, and when the heater shuts off, keeps pumping the hot water through the radiators instead of sitting useless in the pipes.
It saves heating bills as well.

It can be easily modified for 12V fan use.

Flowmaster circulator schematic2.JPG
 
I am glad I do not have to use a fan.

I have multiple holes in the amplifier bottom plate.
I have multiple holes in the amplifier top plate.
I use lots of space between tubes and other objects.
I space parts well, and use resistors that are large (dissipation ratings at least 3 to 5 times the power they will have to dissipate).
I run the tubes below their maximum ratings (plate and screen dissipation, plate and screen voltages, max filament voltage, etc.).
I use the amplifiers out in the open; no covers.
And whenever possible, I use 105 degrees C capacitors.
Reliability and simplicity.

What do I loose or trade off for those practices?
Max Power output and the need for fans.
I'm also glad I don't have to use a fan.

BUT:
I have no holes in the amplifier bottom plate.
I have no holes in the amplifier top plate.

I design to run hot. Chassis temperature of 75°C is quite normal and fine. My MOSFET amp heatsink hits 95°C (class A).

If I needed to control a fan, I'd use PWM.
https://www.coolerguys.com/en-ca/products/coolerguys-12v-pwm-fan-controller
1666869031386.png
 
simple (minimalistic) way
https://www.amazon.com/Bimetal-Disc-Thermostat/s?k=Bimetal+Disc+Thermostat

https://www.aliexpress.com/w/wholesale-bimetal-thermal-switch.html

https://www.mouser.com/c/sensors/te...hermostats fixed temperature switch&m=Sensata

Two types:

Small "floating" ones which can be left "sensing air" or strapped/glued to your point of interest.
I embed one of these inside my transformrr windings. for example.

High-quality-5A-250V-bimetal-thermostat-thermal-protector-165-degree-normal-closed-temperature-control-switch-KSD.jpg_Q90.jpg_.webp


or traditional disk types, normally bolted to heatsinks and such.

51uGgS5HxuL._AC_SL1001_.jpg
 
I’m currently using this to control fans I’m experimenting with for cooling a BJ2 build which I unfortunately tried building in DIY equivalent of a 4U/300 case. I want to run more current, but its a struggle with this much heatsink.

https://www.ebay.com/itm/4036478472...IqXTk5SxwQsMpVH7N7m2syKQ==|tkp:Bk9SR-yb_ZaDYQ

It works properly and has a thermocouple that can be mounted where temperature reference is most important.

The relay clicking can be an annoyance if it is set at a temp threshold that causes it to cycle…at least on the bench. Inside the case its fine.

It requires a 12V power supply. I used a capsule transformer mounted on some perf board and piggy backed it to the controller board which mounts to the chassis with some brackets I made.

My biggest issue is that I can’t live with any if the noise from fans I’ve tried so far.
 
Hi all,

Thanks for all the input. However I still haven’t solved my “problem”.

The very simple 2 wire thermostat’s would be OK however, if they switch ON at for example 30 degrees, they switch OFF at something like 45 degrees. That’s not what I want. I want the contact to be OPEN below 30 degrees and CLOSED above 30 degrees. So once it’s switched ON at 30 degrees it should stay ON when the temperature rises. Something with similar specs I haven’t been able to find.

If this is impossible indeed, I may look into the more advanced regulator boards/controllers.

Regards, Gerrit
 
https://www.amazon.com/Bimetal-Disc-Thermostat/s?k=Bimetal+Disc+Thermostat

https://www.aliexpress.com/w/wholesale-bimetal-thermal-switch.html

https://www.mouser.com/c/sensors/temperature-sensors/thermostats/?q=thermostats fixed temperature switch&m=Sensata

Two types:

Small "floating" ones which can be left "sensing air" or strapped/glued to your point of interest.
I embed one of these inside my transformrr windings. for example.



or traditional disk types, normally bolted to heatsinks and such.


++1 At .29 cents each and so little to fail and a wide variety of temperatures these make a lot of sense. What is a good temperature to set for a fan kick on? I suppose it depends on a lot of things like chassis layout, what tubes, if you have heatsinks with solid state devices nearby, etc. but in the ballpark what temperature would people select in buying a bi-metal switch? 35 Degrees seems reasonable 95 degrees Fahrenheit?

https://www.aliexpress.us/item/3256803936726990.html?spm=a2g0o.productlist.main.7.162e324aAtSvys&algo_pvid=1ca959f6-eb42-4b4a-923b-9a830e3ca53e&algo_exp_id=1ca959f6-eb42-4b4a-923b-9a830e3ca53e-3&pdp_ext_f={"sku_id":"12000028111219917"}&pdp_npi=2@dis!USD!3.31!1.99!!!!!@212272e216668913902993422d0775!12000028111219917!sea&curPageLogUid=2lbQfIOpIJRG
 
Last edited:
The very simple 2 wire thermostat’s would be OK however, if they switch ON at for example 30 degrees, they switch OFF at something like 45 degrees.

Not at all.
No thermostat works that way.
Once triggered , it´s ON at a given temperature and stays ON until COOLED, not until even hotter, that´s nonsense.
We are talking a thermostat connected to a fan, exactly what you asked.

ALL N.O. (normally open) thermostats:

1) switch ON at any preset temperature to turn on fan.

2) fan does its work and cools the device

3) once cooled a few degrees BELOW trigger temperature (hysteresis) it turns OFF

Only detail I´d add, I suggest you add 2 click reducing elements:

4) a .1uF ceramic or a Zobel network (add 10 to 100 ohm in series with capacitor) across switch contacts.

5) a reverse biased 1n4002 diode across DC fan contacts.
 
Last edited:
++1 At .29 cents each and so little to fail and a wide variety of temperatures these make a lot of sense. What is a good temperature to set for a fan kick on? I suppose it depends on a lot of things like chassis layout, what tubes, if you have heatsinks with solid state devices nearby, etc. but in the ballpark what temperature would people select in buying a bi-metal switch? 35 Degrees seems reasonable 95 degrees Fahrenheit?
Your choice, but be realistic.

35C is almost room temperature in Summer 🙂
Maybe higher if it´s Naresh´s amp🙂

Tube amps run Hot by definition.
I would choose, say, 60C or so; high enough so it´s not ALWAYS ON, low enough to protect Electrolytics if things run out of control.
In general, if you can touch chassis and hold your hand on it, it´s fine for tube amps,