Using a high-power LED for a Voltage drop with a FAN or is it a stupid idea?

Status
Not open for further replies.

AlainD

Well-known member
Hi

I'm thinking of using diodes to get a voltage drop on an array of FAN's. The added heat isn't a problem.

Then it crossed my mind that high-power LED's (3watt) have a voltage drop of about 3V (depending on color) and a maximum current of 700mA.

Is it OK to use them for a voltage drop, if I keep the current below 700mA?
 
AlainD:
Your idea might work but, since you have not told us what type of fans, which LED, what is driving these circuits, and the voltage and currents involved it would be difficult for anyone to give much in the way of advice.
Please look at the forum rule at the top, it is very important to provide details.

THE DEVIL IS IN THE DETAILS!

Regards,
Ed
 
I've added a manually draw schematic.

I want 3 PC fans (Be Quiet 140mm 1000 rpm 12V 3.6watt 0.3A) to be driven at two speeds. A rather High speed, limited by a small resistor (1 Ohm here, maybe a bit more after testing) with one N-MOSFET (IRF3708) and another lower very quiet speed where's there's an additional LED diode (3W 3-4V and 700mA) to limit the speed. I will check is de flow is less than 700mA when there's an extra 3V drop.
 

Attachments

  • DSCN7934 (2) v1.jpg
    DSCN7934 (2) v1.jpg
    136.3 KB · Views: 29
AlainD:
Perhaps you should consider looking into the Teensy's PWM output. There are a lots of threads on this forum concerning PWM on Teensy.
With PWM you could choose which ever speed you would like.

Regards,
Ed
 
I looked at PWM. Still not sure about 3.3V vs 5V, according to some specs the fans may pull the PWM line to 5V.

I only need two speeds a very very quiet one and one that is audible but pushes more air.
 
I looked at PWM. Still not sure about 3.3V vs 5V, according to some specs the fans may pull the PWM line to 5V.

I only need two speeds a very very quiet one and one that is audible but pushes more air.

AlainD:
You already seem to want to use a MOSFet to control the fans speed. Why not just add an inductor and a capacitor to the output of the MOSFet and feed the input from the Teensy with a PWM Signal and there you have a PWM controller for your fans. You could also buy a specialized board or other IC that is made for this sort of application.
If you look at some of the many PWM threads on this forum I'm sure you could find more specific part to accomplish this goal.

Regards,
Ed
 
You can make a dummy load from a zener and BJT if you don't need the light output, something like:
zener_clamp.jpg
 
You can make a dummy load from a zener and BJT if you don't need the light output, something like:
View attachment 27677

For a dummy load a resistor or a couple of 2A diodes seem fine to me. I just thought it could be "nice" to have an extra light.

Probably going to drive the fan's slow enough so that I'm not hearing them. I think that 1 speed will be enough.
 
It will probably work. Even though many "better" ways are possible, if this is what you want then you might as well give it a try and experiment. If you get really lucky, maybe changes in the static air pressure will be visible in the LED light.

One minor caution would be the make sure you connect the diode the correct direction. While it's just a diode with higher forward voltage drop, do keep in mind all diodes also have a maximum reverse voltage and a certain amount of reverse "leakage" current before hitting that breakdown voltage. No diode perfectly blocks everything when connected in reverse. With normal (non-zener) diodes it's usually dozens or even hundred of volts, and leakage is usually nanoamps for silicon or microamps for silicon-metal (Schottky) diodes. With a high power LED, it's *probably* more than 12V breakdown and likely to be not more than than milliamps leakage, but no effort has been put into the LED's design to give you a useful breakdown voltage or reverse leakage spec. If you accidentally connect the LED backwards, these specs could turn out to be surprising.

If using a PC power supply, also keep in mind modern PCs can deliver several hundred of watts at 12V. Best to first experiment with a lab bench power supply with the current limit set to no more than you know you need, or at least something like a "wall wart" power adapter which is limited to a lot less power.
 
It will probably work. Even though many "better" ways are possible, if this is what you want then you might as well give it a try and experiment. If you get really lucky, maybe changes in the static air pressure will be visible in the LED light.

One minor caution would be the make sure you connect the diode the correct direction. While it's just a diode with higher forward voltage drop, do keep in mind all diodes also have a maximum reverse voltage and a certain amount of reverse "leakage" current before hitting that breakdown voltage. No diode perfectly blocks everything when connected in reverse. With normal (non-zener) diodes it's usually dozens or even hundred of volts, and leakage is usually nanoamps for silicon or microamps for silicon-metal (Schottky) diodes. With a high power LED, it's *probably* more than 12V breakdown and likely to be not more than than milliamps leakage, but no effort has been put into the LED's design to give you a useful breakdown voltage or reverse leakage spec. If you accidentally connect the LED backwards, these specs could turn out to be surprising.

If using a PC power supply, also keep in mind modern PCs can deliver several hundred of watts at 12V. Best to first experiment with a lab bench power supply with the current limit set to no more than you know you need, or at least something like a "wall wart" power adapter which is limited to a lot less power.
Thanks Paul

I'll use a small "wall wart" supply. Although the fan's are PC fans, it will be a stand alone setup based on temperature.
 
Maybe I'm missing something, but sounds like you are simply try to reduce fan RPM by means of lowering the voltage? Can you not use and adjustable buck converter, or an LM317?
 
Maybe I'm missing something, but sounds like you are simply try to reduce fan RPM by means of lowering the voltage? Can you not use and adjustable buck converter, or an LM317?

The plan was to have two voltages. One that's rather quiet and moves quite some air and another that's a few volt's lower and is quiet, but keeps on moving air. The fan would controlled by a teensy depending on a temperature probe (nothing, very slow or "faster"). Power would delivered by a "wall wart" supply of maybe 9V.
 
A quick follow up.

I've did some tests with a fan (be quiet! Pure Wings 2 140 mm case fan) and was very pleased:

7.7V 53mA --> Rather quiet, but maybe not for long times (and 6 fan's)
6.22V 45mA --> a lot better
5.2V 40mA

I even tested spinning up at a bit below 4V and this worked every time. This is nice because I can do software development while powering a fan from the 5V vin.
 
Status
Not open for further replies.
Back
Top