Voltage regulator doubt

Discussion in 'The Projects Forum' started by imperatormk, Jun 15, 2011.

  1. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    Hello everyone,

    I need a circuit that will let me to regulate the output voltage from 0-~24V while having a load of ~1A (and surely less). After a few days research, I found voltage regulator circuit schematic and modified a little to get a variable power supply, but I am not sure if it is OK. My main concerns are heat dissipation and stability.

    Any suggestion/advice appreciated. (Schematic in attachment.)
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    That's not going to work due to power dissipation in the TIP31C.

    When the output is as high as it can go, you will have ~24v less the two ~0.7v emitter drops, so 24v-(2*0.7v) = 22.6v. Power dissipation in the TIP31C will be (36v-22.6v)*1A = 13.4 Watts at that point.

    When the output is as low as it can go, with a 1A load, you will have power dissipation in the TIP31C of 36 Watts. The absolute maximum rating of the TIP31C is 40 Watts, IF the temperature of the case is held at a stable 25°C. This implies a requirement for a near-perfect heat sink.

    The voltage output will vary somewhat over load, and over variances of the 36v input supply.
     
  3. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    So I see.

    ...Any other circuit design then? Circuit with a TL2575 for example?
     
    Last edited: Jun 15, 2011
  4. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    You could look for a bigger transistor, and maybe increase the Zener voltage a bit, but probably you would do better with an IC regulator.

    If you do want to persevere with your design, the first thing to note is that as shown the adjustment pot. will disable the 24V Zener as soon as the voltage is turned down from maximum.

    It might make more sense to connect the track of the pot (the ends) across the Zener, then wire the slider (the arrow) to the base of the first transistor only.
     
  5. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    Generally anything that could work without a heatsink would suit me. Could you give me some example IC or transistor?
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Adjuster,
    The purpose of the Zener was simply to prevent the output from rising above 24v.

    The TL2575 would probably do the trick just fine. Switching DC-DC converters are far more efficient than linear regulators, particularly when a wide range of output voltage is required.

    Here's a circuit I designed awhile back using Linear Technology's LT1170/LT1171/LT1172 family of regulators; it would work quite well for your application. Changing R2 from 910 Ohms to 730 Ohms would limit the output voltage to around 24v. See the attached.
     
    imperatormk likes this.
  7. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    Looks good. I`ll take a look if I quit from the TL2575.

    Great. Just one question - do I need a heatsink?
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    It certainly wouldn't hurt. I would at least make a copper pour area on the PCB design to act as a heat sink. I would not use the DIP package for this project.

    You will need to use an inductor of between 150uH and 330uH (inclusive) for your desired input and output voltage range.

    The TL2575 has a maximum current rating of 1A, which is the same as your output specification. You usually don't want to use a component at their rated maximum output current, so I will retract my prior suggestion that the TL2575 will "work just fine". Look at the LT1171 solution that I posted instead.
     
  9. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    Okay, what would be the max current it can output without causing problems?
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    One usually de-rates components to make fairly certain that they can handle the desired load.

    You could likely go up to 700mA.
     
    imperatormk likes this.
  11. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    That would be fine I guess...

    So, let`s build the thing :)
     
  12. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    OK, I built the circuit and after one blown chip, it works flawlessly. BTW, why the chip died when I attached an ammeter in series with the input voltage source?

    http://imageshack.us/photo/my-images/204/tl2575adj.png

    Here is the schematic and the ammeter location... :)
     
  13. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    If you placed the meter where shown in your schematic, you effectively added inductance between Cin and pin 1. That extra inductance destroyed the IC.

    Cin must be connected between pin 1 and GND with leads that are as short as possible.
     
    imperatormk likes this.
  14. imperatormk

    Thread Starter Member

    Mar 1, 2011
    96
    2
    That of course makes sense. Thank you :)
     
  15. castley

    Member

    Jul 17, 2011
    31
    0
    You cannot vary the voltage with the pot connected as shown.Connect the pot accros the zener diode and connect the pot wiper to the base of the transistor.Also the pot must be a much higher value or there will be no current thru the zener. I would use at least a 10 K ohm value. Check that the output transistor can dissipate
    the wattage, which will probably be at least 2 watts.
     
  16. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Gee, this thread was resolved over 3 weeks ago, and the schematic you're referring to from the initial post was superseded by several other schematics.

    Wanting to help out is greatly appreciated, but one needs to both read through the entire thread to become familiar with the OP's needs, and also to see what the last few entries were. If it's been several weeks since the last entry and the OP hasn't been heard from, it's not likely that the topic will be re-visited by them.
     
Loading...