Fan control, PWM control of a DC DC converter?

Go To Last Post
10 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi guys,

I need to control the speed of a fan, do it efficiently and not lose the tacho signal from it.

Since the tacho signal is an open collector signal, I can't just PWM a transistor connected to the ground. Since there are some electronics inside the fan and I need feedback from the fan speed, I can't do PWN onto the +12V either. The fan used must be a 3 wire one, so no extra pin for controlling the motor.

Therefore my only option is to change its power supply voltage. I'll be using an LM2674 and mess around with its feedback using a bit of a hack as seen in the picture (similar values will be used, the simulation is for another regulator IC). The Q1 transistor will, when opened, pull the resistive divider down and increase the voltage to the output.

Now, in simulation, when I PWM the transistor, I can modify the output value of the voltage (from 4.8V to 12V). The feedback voltage doesn't look good however, see attachment. I'm not sure the regulator likes it. The output voltage is fine.

So my question is: Is this a good idea? Is there a better way, that lets me keep everything, efficiency included?

I'm using a pretty fast PWM (~40kHz).

Thanks,

David

Attachment(s): 

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

PWM-íng the plus lead works fine for me. The tacho output is just an open collector.

I usually add a schottky and a capacitor to create a SMPS, and do that to prevent RFI. Also for led-strips, ... same reason.

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tessa and Tina, You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yeah, but won't the PWM mess up the power for the tacho sensor?

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Try applying a variable voltage to the lower leg of R2 instead of connecting it to ground. The variable voltage can be made from a filtered PWM signal buffered by an opamp.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the tips! I've tried the thing jay suggested, works great!

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Cool 8) Does it work in a real circuit too?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi guys,

The setup works... a bit. The OP amp I used (LM258) seems to be having dificulties in real life pulling down to near zero - there's almost 0.6V there. I should have anticipated that. That's a problem.

Could you recommend me a dual op amp in an SO8 package that could pull a 1.21V through a 1k resistor to 0V as close as possible?

Or do you have any other solution?

The solution I use currently uses the output of a 10k-1k divider, where the 1k is connected to the output of the op amp.

Thanks,

David

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Oh, and I'd like to avoid chaning the divider ratio - currently, the max voltage output voltage of the regulator regardless of the voltage on the 1k end will not exceed 13V. Should I change the resistance to, say, 22k or so, the max voltage, should the op amp fail (or rather start functioning better) will go up to 24V or so, which would be bad for the fan and for the output cap.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Or just buy this for 28 dollars USD

http://controlresources.com/nimb...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Op-Amp? I don't see one in the circuit posted above.

Two thoughts:
If you use the search parameters on Nat Semi, or others, you can select for rail-to-rail op-amps, which are designed to have their linear range extend to very close to the power rails.

If you have a two screen development platform, then you can have Mouser/Digi-Key/etc. open on one screen to check cost and stock while looking at the various options available, (in theory), from the manufacturer.

Second: If the output of the op-amp is saturated to V+ or to Gnd, (PMW, Switch, etc), and not being run in its linear range, then one might consider using it to drive an NFet which has a very small Rdson, milli-ohms, for a low that is truely near zero.

JC