DC Voltage doubler using fan tach signal

Discussion in 'The Projects Forum' started by kwadrofonik, Mar 29, 2016.

  1. kwadrofonik

    Thread Starter New Member

    Mar 29, 2016

    I have a 24v 1.6A DC Brushless fan but only a 12v power supply. I was wondering if I could use the tach pulse from the fan to run a pumped DC voltage doubler similar to the clock generated from a 555 timer: http://www.electroschematics.com/648/555-voltage-doubler/. The added benifit is that the voltage will ramp up from 12v to 24v as the fan spins up.

    The first problem is that because I only have a 5v DC square wave from the tach, I have to pull down the PNP base with a resistor. The second issue is that all of the voltage doubler examples I can find use less than a 12v supply so my circuit has to be protected with more resistance. I can't seem to get it working and I was wondering if someone could recommend the proper wiring and resistor values from the tach signal to the shared base terminal and from the NPN base to ground.

    All I have left in my tickle trunk that can handle the current is a TIP31B (NPN-tran,5v base) and a BUZ71 (N-MOS, 7v gate) which seem to work alright together. Let me know if I need to get some more suitable transistors.

    Thanks in advance
  2. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    Welcome to AAC!
    Assuming the tach provides 3 pulses per rev and fan max speed is 6000rpm the max tach frequency is 300Hz. To sustain a current draw of 1.6A and a voltage > 20V at that speed will require voltage-doubler capacitors of at least 10000uF each, with low ESR to handle the large ripple current. If it were me I'd forget the tach and use a much higher frequency (say 50kHz) generator, so that smaller capacitors would suffice.
    kwadrofonik likes this.
  3. crutschow


    Mar 14, 2008
    1.6A is a lot of current to be generated by a doubler.
    It might be better to use a step-up switching converter.
    kwadrofonik likes this.
  4. kwadrofonik

    Thread Starter New Member

    Mar 29, 2016
    Thank you for your advice! My caps were 3300uF and the fans spin at 2000 rpm so the ripple was too much. I will throw the 555 doubler together with high enough rated components.

    The fan is actually two fans totalling 1.44A nominally and I am realistically aiming for 16-18v output. I appreciate your insight and if my attempts fail then I'll order a step-up power supply module online.
  5. AnalogKid

    Distinguished Member

    Aug 1, 2013
    Typical small fan tach outputs are two pulses per revolution, or 67 Hz in your case. 100x would be better. Interesting idea to use the fan tach as the clock, but the low frequency and high current make it difficult to do with real world components.

  6. kwadrofonik

    Thread Starter New Member

    Mar 29, 2016
    Difficult for guys like me maybe. Up for the challenge ak? A patent for an inline 3-wire fan booster using discrete components?