I have built an LED array and have it working fine. Only thing I have yet to resolve is some sort of control for the 2 12vdc fans I use for cooling. The LEDS are on an Arduino PWM controlled timed drivers. Because the fans are not they essentially run all of the time when I am not around to switch them on or off. I am considering a temperature controlled cct to control the fans. I have found the following project which appears to address my needs.

http://electronicdesign.com/article...l/dc-fan-controller-takes-bare-bones-approach

Any advice as to whether or not this would in fact do the job or is there a better way to do it? I dont want to spend a fortune, just want it to work.

If this works could I power both fans with one assembly?

Does this cct vary the voltage or the current to the fan?

If it varies the V and if my fans (muffin type) are rated at 12 VDC will it harm them to run with less?

 

If this works could I power both fans with one assembly?

Does this cct vary the voltage or the current to the fan?

If it varies the V and if my fans (muffin type) are rated at 12 VDC will it harm them to run with less?

It will vary both voltage and current. Your fans will be fine, but they may behave oddly at the lowest speeds where they will have low torque.

Another approach you might consider is a thermostat based on the LM35 IC and a comparator. A quad comparator would give you 5 power levels, each of which you could set.

 

Thanks for your input. With respect to the LM35 and a comparator dont know where to begin to look for a cct on that. Any guidance toward one would be appreciated. In the mean time I will surf away trying to find a solution along that line.

 

The datasheet is always a good place to start. You might also learn from a project I posted which used this general approach. I only had on/off control with a single comparator, but you'll see how the comparator controls a MOSFET which can in turn switch larger loads (like fans).

There are a LOT of fan control circuits out there in the computer modding world. The state-of-the-art is to use proportional pulse-width modulation (PWM) control. The width of the pulse is proportional to the temperature so that the fans run harder when hot, and slow down as the system cools. You can use a 555 timer IC to accomplish this.

The circuit you linked is an attempt to get proportional control with a much simpler circuit. It may work but I'm sure it has shortcomings. So does the on/off control of a thermostat, even if you add levels as I mentioned. Any of these simple approaches may well work fine for you, though.

Choosing a circuit comes down to what you need to do, what you feel comfortable building, what you want to learn and so on.

 

After looking at many ccts I have come down to what I think are two workable solutions although I am not totally sure.

First doesn't provide proportional control but gives me control over the temps at which the fans come on and shut off. My only problem in understanding and building the cct is the MOSFETS. The P/N's on the schematic appear to be surface mount and are tiny. It would be much easier for me to have something in a T0-92 or T0-220 form. Having said that, and overloading on trying to understand what are the important factors in determining MOSFET substitution I am having a hard time determining what substitute MOSFETS I could replace these particular MOSFETS with and still have the cct function as advertised.

http://www.discovercircuits.com/DJ-Circuits/12V_Cooling_Fan_Thermostat.htm

My second option is a proportional PWM controller based on a MIC502 fan controller IC. I tried finding a 555 cct like Wayne suggested and found many that controlled fan speed but finding one that took temperature into account was difficult. I think that this ones performs the same function with a few bonus features. I would like something that comes on around 40 deg C. I am assuming that the value of the thermistor and its associated resistors have something to do with determining that. Would the components the author uses here meet that need? The thermistor is a 10K @ 25deg. Is there anything else I need to consider?

It is the bottom schematic. There is a requirement for me to power 2 fans off of this cut. They are about .120 A each. If I install in parallel is there anything else I need to do to this cct to make it work?

http://www.bit-tech.net/modding/2001/12/03/pwm_fan_controller/1

If you can help it would be appreciated.

 

... I am having a hard time determining what substitute MOSFETS I could replace these particular MOSFETS with and still have the cct function as advertised.

It's not as bad as you think. When used a switch, as they are here and in many circuits, MOSFETs are a bit like a light switch and you don't have to worry too much about which to choose. You need to know:
• Whether you need a P-channel or N-channel
• The voltage level needed to control the gate. A normal mosfet needs about 10V and a logic level mosfet needs only about 5V to be turned fully on.
• That the device can handle the current load. Whatever real max current you expect, make sure the mosfet is rated for, say, 4X more. Safety factor.

As the voltages go higher and the switching frequency gets higher, the more obtuse MOSFET specifications come into play. These are all pretty much irrelevant in a simple circuit for switching a fan.

My second option is a proportional PWM controller based on a MIC502 fan controller IC.

That looks like a nice approach but IMHO it's overkill for your application. For a desktop computer you are sitting next to, yes. For something that runs most of the time while some LEDs are running, I think on and off control is enough. And actually I think you could use a simpler comparator circuit than the one you linked - you just need some hysteresis.

 

If your fans are brushless DC fans with the hall effect sensors in them..good luck.
They don't like to be fed anything but an analog voltage.
I tried in the past and all they would do is wobble/stall/shake.. Not spin.

There are many that can accept a PWM signal to control the speed. (don't confuse this with a PWM tach wire fan)

 

Thanks for the help! The first option becomes a little more complicated because of the mosfet compatibility which Wayne straightened me out on. Pretty much anything in a TO-220 form exceeds by far the max current of the ones that are used in the schematic so I should have no difficulty buying one locally. Also have to add a 5V supply to the project as well as the 12v. Have plenty of LM7805's around so that should hopefully look after that. I liked this one because of the LM35 for temp sensing. The math involved in selecting a thermistor/resistor network was giving me a headache for some of the comparator ccts I saw so this one eliminated that need.

The second cct still intrigues me. I realize that its overkill, but its not complicated and not that expensive. The MIC502BN is only a couple of bucks. The fans I am using though I believe are hall effect fans. On the companies website that manufactures the fans some come with optional PWM availability. The ones I have don't, so as Mcgyver mentions I may have an issue assuming of course i am interpreting their data correctly.