Shop OBEX P1 Docs P2 Docs Learn Events
Using a DAC, Mosfet and an OP-AMP — Parallax Forums

Using a DAC, Mosfet and an OP-AMP

Zap-oZap-o Posts: 452
edited 2010-12-26 00:33 in General Discussion
I was wondering if someone could offer up some advice.

I am running the propeller that outputs to a DAC. The DAC then places a voltage between 0v - 5v into a no-inverting op-amp with a gain of 3. The op-amp then fedds into a Enhancement N-Chan mosfet.

Untitled.jpg


My concern is that I will be wasting the lower bits of the DAC because the mosfet wont turn on till ~ 2.2Volts. I am trying to figure out how to bias the mosfet at 2Volts so that I can use the full spectrum of the DAC. Any advice will be appreciated.
Untitled.jpg
475 x 149 - 6K

Comments

  • Erik FriesenErik Friesen Posts: 1,071
    edited 2010-12-24 15:09
    Does your dac have a vref- ? Or, you could offset the op amp 2.2v and lower the gain. What is your circuit, including chip part nos?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-12-24 15:10
    What are you hoping to accomplish with the MOSFET? Proportional current or proportional voltage? Without feedback from the MOSFET to the op-amp, you willl get neither. With properly-configured feedback, other biasing will be unnecessary.

    -Phil
  • Zap-oZap-o Posts: 452
    edited 2010-12-24 15:25
    I am using a DAC8531 16bit running 0 - 5 volts.

    The mosfet is configured as a current controlled device.

    I figured the op-amp should be a gain of 3 because the mosfet ranges from 2v to 15v.

    Feedback at the moment is a ADC that measures the temperature of the heater.

    Is there a way I can offset the voltage using an op-amp. I have tried various methods and cant seem to pull it off. My last attempt was 2 op-amp circuits one was a non inverting gain of 3(mentioned already) that feed into a adder op-amp circuit. Its not proving to be the best approach.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-12-24 15:30
    If you're controlling a heater, there's really no point in using linear proportional control, due to the extremely long time constants involved. Since you already have temperature feedback. Why not just pulse the MOSFET on and off and skip the DAC and op-amp?

    -Phil
  • Zap-oZap-o Posts: 452
    edited 2010-12-24 15:47
    I need an analog output because this is a precise heater that changes temperature rapidly, about 20 degrees a min. I am drawing 10 A and I tested PWM. It was / is a noise maker interfering with my ADC readings.
  • kwinnkwinn Posts: 8,697
    edited 2010-12-25 21:11
    You should still be able to use PWM to control the heater. At 20 degrees per minute the temperature changes only 0.333 degrees per second. Having the micro control the heater and take the temperature measurement allows you to take a reading only when the heater is on or off, never when it is switching. This should get rid of the noise.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-12-26 00:33
    kwinn makes an excellent point about synchronizing the ADC sampling to the heater PWMing. That will eliminate any effects from the switching noise. Moreover, the MOSFET will always be fully saturated or off, never in its linear region, radiating nearly as much heat as the heater itself.

    -Phil
Sign In or Register to comment.