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

Darlington pairs in parallel

  1. Feb 17, 2012 #1
    Hi all

    Im currently working on a high power led driver circuit, and at this point I need to test my design. Since leds are expensive (and time consuming to solder together) Im using this relatively simple circuit (http://www.edn.com/article/510832-Accurately_simulate_an_LED.php) as a substitute for real leds. It works great and all, for low power applications, but I need to simulate 700ma @ 336 volts. The only way to get darlington pairs to drop that much voltage across collector to emitter is to connect multiple pairs in parallel. Im currently working with TIP665 devices, see attached screenshot of the schematic.

    I have five in parallel right now.....initially testing at half power (165 Vce) and trying to run up to about 140 mA thru each one. My problem is that i slowly turn up the 250 ohm pot, and I see the cumulative current in all five branches increase to about 300mA, then at least one of the darlington pairs blows (shorts out), immediately followed by the 2 Watt 1 ohm resistor (which I'm only using to take easy current readings)....my Vbe is around 1 volt (datasheet says 8 max) so thats not the problem.....the transistors are supposed to be rated at 150 watts and 400 volts Vce, so they "should" be sturdy of enought to handle the task. Note, I first tested each of the five individually at the same Vce 165 volts and they ran 140 mA for five mins no probs.....Just that when they all hook up together...hmmmm

    This scenario has occurred twice now, so im thinking there is one of two problems 1) one or more of the transistors is hogging all the base current, and subsequently blowing...... or....2) the transistors simply cannot dissipate the power they claim, and are failing under stress

    I can't believe option 2 is correct, since each pair passes the individual test.

    P.s. Each of these pairs have heatsinks, mounted in a plastic tray covered in thin plastic liner, and filled with ice.....yes ice.....there is alot of wattage going out here hehe

    Sooooo....ANy helpful insight Darlington parallel configurations? anyone?

    Attached Files:

  2. jcsd
  3. Feb 17, 2012 #2
    Problem #1 is your emitter resistor is only 0.1Ω. You want 5 transistors to provide 700mA so each is supposed to provide 140mA. The voltage drop across the 0.1Ω resistor would theoretically only 14mV. That is not enough to compensate the Vbe difference of the transistors. One or two transistors is taking on the full load while the rest is idling. For 140mA full scale, you want to have voltage drop of say 1.4V across the resistor. That means the resistor at the emitter should be 10Ω. With 10Ω the power dissipation of each resistor is 0.2W. So use a 1W resistor and it will not get hot. With 1.4V across the resistor, the difference of Vbe between the transistor will not be as important and all transistors will carry roughly the same current.

    You need to adjust the voltage control circuit accordingly to accommodate the increase voltage of the emitter resistors.
    Last edited: Feb 18, 2012
  4. Feb 17, 2012 #3
    thank you!! will try that and report back, makes sense duh :)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook