Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Help with a little controller project

  1. Nov 11, 2014 #1
    Hi everyone, I joined this forum to hopefully get some guidance on a project I have been working on.

    I'm trying to either buy (or make if it doesn't exist) a controller of some type that controls a larger output wattage that changes based on the input signal from a lower wattage source:


    I know absolutely nothing about electrical engineering, except for my research trying to solve this issue. So please be patient.

    An example of the controller in use would be:

    Setting the controller that if it detects 1 watt on the left, it draws 30 watts from the power supply for the output. If it detects 10 watts, it outputs 300 watts.

    It would be great if I could fine tune the relationship between the two so if upon testing, I find I need 1 watt detected to equal 50 watts output while if it detects 5 watts it outputs 300 and anything pass 5 also equals 300 output.

    I realize there will probably be some programming in here, but I hope there is a simple enough solution that even a simple liberal arts guy can do it =].
  2. jcsd
  3. Nov 11, 2014 #2


    Staff: Mentor

  4. Nov 11, 2014 #3
    Your description isn't clear enough. Could you tell us what exactly you are trying to control?
    Do you want to amplify an audio signal? Control a lamp? Or a motor perhaps?
  5. Nov 12, 2014 #4
    Thanks for your two replies. Its actually for a 12v 200 watt thermometric peltier that will change its power consumption based on a separate input of 0-10 watts via a separate wire. I'm trying to control the wattage going to the peltier via splicing one of my computer fan wires coming from the motherboard pwm as the signal. I don't want to suck up the power from the motherboard because it can't move nearly as much current as I needed. That's where the second leads from a power supply come in.

    3 connections involved:
    1) input wires for the controller to detected wattage 0-10 and translate that into a larger wattage of 0-200 watts
    2) input wires that will actually carry the 0-200 watts to the thermoelectric
    3) output wires that actually go to the thermalelectric (0-200 watts)
    I'm basically looking for a computer fan controller that can handle much higher wattages.

    Thanks for your help
  6. Nov 13, 2014 #5
    Since a peltier element can be driven with a pwm signal, a single Mosfet may be enough.
    This one for example

    But you should ask someone with more experience in electronics if it's safe to do this and how to best connect the mosfet.
  7. Nov 13, 2014 #6
    the 12V/200W Peltier makes sense- PWM type signal will work fine. Now -can you describe the input in similar terms a max V or I? Ex: 10W could be 0.1A and 100V or 100A and 0.1V - etc...typically an input is not "watts" -- it could be measured watts being delivered or consumed by another device.
  8. Nov 13, 2014 #7
    Windadct is correct, if you talked in Volts and Amps, be more standard for assistance. If one had the spec for the Peltier be easier to help. What you want to do sounds plausible. You say a feedback signal from some device/sensor is 10 watt , what is this? Need more than 10 watt information to know how to interface. Guessing this some type of Temp measurement.

    A block diagram might be a way to add clarity.
  9. Nov 13, 2014 #8
    Here it says that the control pin is pulled to ground with a frequency of 25 kHz.
    It requires a pull up voltage of 3.3 or 5V.
    Now I just wonder of e.g. a duty cycle of 70% means the signal is pulled to ground 70% of the time or it's not pulled to ground 70% of the time. Anyway, according to the datasheet of that mosfet I mentioned 5V should be enough to get the 17A flowing that the peltier needs.
  10. Nov 14, 2014 #9


    Staff: Mentor

    Does it have to be continuously variable? If not, one or more simple relays could switch lots of power to the fans while being controlled by the computer. You could even modulate the duty cycle of the fan for finer control.

    Your question used the word "amplifier" which suggests all sorts of things to EEs and may have led to unnecessarily complex answers. For example, 25 kHz modulation for a temperature control application.

    In a very crude sense, a relay can be considered a power amplifier but we don't usually say it that way.
  11. Nov 14, 2014 #10
    So - you have a "signal" that normally runs a 10W Fan? Is it a 12, 24V DC or 120V AC Fan - then then signal is typically a 0 to X volts ( with max power output 10W). I am willing to bet the fan dontrol if it is DC is already a PWM signal... may just need to use somthing like the MOSFET above - but we still need more details.
  12. Nov 14, 2014 #11
    You may have misunderstood me there. The 25kHz signal is already present in his computer on the fan connector. All he needs to do is attach a mosfet to it.
  13. Nov 14, 2014 #12
    Original Poster seems to have gone away. I hope he didn't Overheat next to computer. LOL.

    Anorlunda was great on Relay, Yep it's a Classic Amplifier so to speak. Great for Isolation as well.

    This all reminds of sticking 10 engineers in a room to solve a vague problem, We didn't get a Budget limit and Schedule on this one. :w

    The Simple answer might be to just clean out the dust bunnies in computer. 300 W cooler was pretty high. Still don't know what he's over heating....

    ASIDE: What are typical Efficiencies for a Peltier? or what Temp Gradient it can achieve under some set of conditions. Can't be too good. I've never used them in any of my designs. Water flow yes, Peltier not yet.
  14. Nov 15, 2014 #13


    User Avatar
    Science Advisor
    Gold Member

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook