I recently purchased a small fan (40mm square, 12v dc) to mount in a temperature sensor housing in order to ventilate the sensors. I assume the fans are normally used in a computer.
There are three wires coming out of it, red, black, and yellow I assume the red and black are + and - 12v. What is the yellow for? is it some kind of pulsed output to enable one to control the speed? if so how is it usually done? some kind of diode pump arrangement perhaps?
Don
Donwill coughed up some electrons that declared:
It is the tacho line - open collector arrangement I believe.
So feed it with a small current from a pullup resistor and you should get pulses proportional to RPM.
There's a 4 wire fan too - the 4th wire controls fan speed.
There are some diagrams here that may be of interest:
Many thanks Tim the link is very useful. The sensor housing is going to be installed in a not yet built conservatory, I was going to drive the fan with the aid of a 1Watt 12V solar panel which is going to be installed in the roof of the conservatory. I wondered if the yellow lead was some kind of speed indication and if so, the output would be related in some way to solar radiation intensity and could therefore be used to control an automatic watering system for when we are on holiday; it would be more accurate than a timed system I think. I will have to integrate the pulses by some, as yet undetermined means to enable a timed valve opening for watering.
Time to get the old scope out I think. :-)
Cheers and thanks Don
Thanks again
Many thanks Tim the link is very useful. The sensor housing is going to be installed in a not yet built conservatory, I was going to drive the fan with the aid of a 1Watt 12V solar panel which is going to be installed in the roof of the conservatory. I wondered if the yellow lead was some kind of speed indication and if so, the output would be related in some way to solar radiation intensity and could therefore be used to control an automatic watering system for when we are on holiday; it would be more accurate than a timed system I think. I will have to integrate the pulses by some, as yet undetermined means to enable a timed valve opening for watering.
Time to get the old scope out I think. :-)
Cheers and thanks Don
If you are going to build something why not make it a soil moisture meter and get a direct reading of when to water?
A couple of bits of noncorrosive wire stuck to a plant label should do for the sensor.
You get two pulses per rev for the original tacho driven ones. Some also have a self-heating thermister in the airflow, and if the airflow isn't cooling the thermister (e.g. fan running but a blocked filter), they fake a lower speed. (I guess they start removing some pulses, because with better sense analysis circuits, you can tell that the fan is running at full speed but the airflow has been lost, but I don't have one of those to hand to check out exactly how it signals this).
The last time I saw a machine that detected lack of airflow it used a cup in the airflow that acted against a microswitch. Simple and fairly reliable.
What you care about is the cooling ability, which is what the thermistor measures directly, and it's a technique which is present in some kit I had which dates from 1970's (although not integral with the cooling fan back then).
I've never seen what you describe used. It fails to measure the cooling ability (e.g. if you're drawing in hot air, it will not notice that there's no cooling). Also, forced air paths are prone to dust build-up, and such a device could well end up getting stuck in position after a few years running and dust build up, and wouldn't come close to the reliability required in most commercial computer kit. Even back in the 1970's, the cost of the mechanical part to do this would way exceed the cost of what is a very simple and much more effective thermistor circuit.
Donwill has brought this to us :
Poke a couple of stainless steel probes in the soil, connect to an op-amp, then arrange the op-amp to turn a solenoid valve on when the conductivity between the probes falls - simples.
It was a big Siemens Unix/server machine. I remember it well as the disks were overheating and they sent the Germans over to have a look as they couldn't see what was wrong. It was fairly obvious once i had a chance to look inside but I did have trouble telling their engineers that they had put one of the fans in backwards and it was just circulating the air around the fans and not pushing it through the machine. They couldn't believe they had made the c*ck-up.
Airflow is the important bit as it detects fan faults before the system overheats. The environment board detected over heating and shuts the system down, something you want to avoid if its just a fan fault.
In message , Donwill wrote
Would the fan stop every night when it gets dark and hence you would zero pulses?
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.