reducing voltage after rectification question

Discussion in 'The Projects Forum' started by drycounty71, Jan 16, 2012.

  1. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    Hello Forum,

    I am not an electronics expert but I know some and I need some help with a problem I am having. I have a transformer tap of about 35 vac and after rectification I am getting about 46 vdc or so. I am trying to bring this down to somewhere between 30 and 33 vdc for a 7824 40vdc max. in voltage regulator. I have tried diodes-not much help, a couple of zeners-not much help and a dropping resistor not much help-only because the more the resistance climbs, the more the voltage climbs as well. BTW, I am trying to find the best way to do this on the test bench without frying my equipment or at least the 7824 ic. *Side note and engineer from the transformer company suggested that 3 or 4 1N4003's or a couple of 5w 5v zener diodes in series before the 7824 to bring it down to safe range but again, I am trying this on the test bench ahead of time BUT I don't know the total circuit resistance of the +24vdc circuit to recreate so I can test the engineers suggestion. I am lost and a little frustrated. Any help would be greatly appreciated.

    Thanks in advance!
     
  2. K7GUH

    Member

    Jan 28, 2011
    191
    23
    You will need to know that characteristics of the load operated from this supply and you will need to know how the supply behaves under that load. You can simulate this with a resistor large enough to handle the heat. Once you have that info, you can determine what values of resistor, zener, and other components. If you don't already have a multimeter, even the $3.00 variety from Harbor Freight will see you through this task.
     
  3. BJT_user

    Member

    Oct 9, 2011
    35
    8
    Okay, right off the top of my head, I can think of a simple three-component pre-regulator that would fit your needs. The basic circuit would be a common collector (emitter follower) transistor configuration. You would need a decent high power NPN transistor like a TIP120 and a low poer zener diode of around 34v (NTE5036A?) for a stable bias voltage, as well as a 1k (1/4 watt) limiting resistor for the zener.

    Examine the circuit at http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/emitfol.html for a diagram. Look at the "Buffer for Voltage Regulator" circuit. The second resistor connected to the collector of the NPN transistor can be omitted since your circuit would theoretically never reach 0 impedance. In this configuration, whatever voltage is applied at the base would appear at the emitter, minus the emitter-base junction potenials. So 34 volts - 2 volts junction potenial would provide a relatively stable 32 volts at the emitter of the NPN transistor.

    Your 7824 voltage regulator circuit would connect between ground and the NPN transistor emitter pin. When I get time to draw up a proper schematic, I will post it here.

    Hope this helps.
     
    drycounty71 likes this.
  4. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    I do have two MM and I have tried my best to review schematic and add up the resistance values. I have come up with 2.3m ohms for total circuit. The kicker is that I only need to bring the vdc from the transformer to around 33 vdc. That is what is frustrating. The combinations that I try zeners, resistors the voltage will go up as I add a higher value. As of my last test, I added a 1.1m ohm resistor as load and tried to drop but the voltage came in at 47-48 vdc. I don't know what I am missing.
     
  5. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    Here is the PS section of my schematic as it stands without modification. schematic.jpg
     
  6. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi drycounty71,

    Here is another possible solution if you have two 7824s..

    Use two 7824's in series. The first one has a 10V zener diode in series with the common terminal to ground. This raises its output to 34V, which becomes the input to the next 7824.

    This has the advantage of distributing the power dissapation over two devices instead of one, and the first 7824 will now tolerate upto 50V on its input.

    Regards,
    Ifixit
     
  7. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    BJT,

    I plugged the suggested components in (had to substitute a J-fet for a 7824 in the simulator - not one available in this version) and here is the outcome. circuit-test.JPG
     
  8. BJT_user

    Member

    Oct 9, 2011
    35
    8
    Everything seems to be as it should be, drycounty71. I'm sorry I didn't get back with a schematic last night. Work ran late and I ran out of time. Here is what I have put together. (Hope it works.)
     
  9. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    BJT,

    Thanks for the extra mile on helping me with this. I have printed off the info and will source the parts locally. I did plug it in to the circuit simulator and it looks great. I will get back to you once I have it in place and tested.

    Thanks again!

    DC
     
  10. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    How much current do you need? Bear in mind that the minimum current gain (β, or HFE) for some power transistors may be quite limited, say 25 or so. Some devices will be better, and of course with a substantial voltage to be dropped it won't be necessary to adopt the reduced saturated gain value, often taken to be 10 times. Nevertheless, the base feed resistor will have to be low enough to supply the base current at the maximum load current, plus an allowance for the Zener diode.

    With a 1kΩ resistor dropping 10V, you have only 10mA total input. Suppose the Zener needs 5mA, leaving 5mA over for the transistor. As the emitter current Ie = (β +1)*Ib, the maximum current available with a low gain power transistor might be 130mA. If more current, is needed, either a higher gain transistor is needed, or a lower base resistor is needed.
     
  11. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    You could also try a lm317HV variant as regulator. This can handle up to 57 volt on the input. You are on the correct track with your current plan. But using a LM317HV may make your design more simple. In both cases remember to use a proper cooling finn the chosen regulator.
     
  12. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    That's a good point. The regulator will need a heat-sink, and so will the transistor, if used.

    The only possible exception to this would be if the current required were a very small fraction of the device ratings. Check the manufacturer's data-sheet for cooling requirements at the intended power levels.

    For a regulator the power level is found from the voltage drop across it multiplied by the current through it plus a little internal dissipation. For the transistor this is pretty much equal to the voltage drop across it multiplied by the current through it, plus Ib*Vbe.
     
  13. BJT_user

    Member

    Oct 9, 2011
    35
    8

    Good points, Adjuster. The transistor I suggest for the regulator buffer is a TIP120 which is a very common high power darlington. It has a typical minimum current gain of 1000 so even with 5ma base current, it would be able to regulate up to 5 amps DC at its emitter, which is more than adequate for 7824 regulator. Also, being in a TO220 case, inexpensive heatsink kits are common and easy to install.

    The zener diode will conduct at most 10ma and with a voltage drop of ~35v, will dissipate 350mw. So it needs to be rated up to 500mw, which is a fairly common rating for axial glass zener. I suggest a NTE5036A or equivalent.

    The 1k resistor dissipates 10ma at ~10v which is 100mw so at least a 1/8 watt resistor is needed, but since these are so small and fragile, I typically use at least 1/4 watt for their greater physical resilience.
     
    drycounty71 likes this.
  14. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    My current ic components are mounted to a rather large flat heatsink and I have space to conveniently mount the TIP120 and rewire as diagrammed below. Thanks all for your input.
     
  15. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    BJT, this is the more inclusive portion of the HV side of the PS.
     
  16. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    BJT, or anyone, any ideas on the HV side? I see a power mosfet only capable of handling 500 vdc. My HVT on transformer is pushing near that-too close for comfort. I was able to put a 58k resistor after the 82uf/220K junction and bring the voltage down to ~450vdc. Not sure this is the "cleanest" or "efficient" way to do this- thoughts please. Schematic is just above this post.

    DC
     
  17. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    The saga continues- I have a IRF830 power mosfet with the "mosfet B+ reducer" testing in the breadboard but the ic seems to keep frying. I have a 220ohm resistor going to gate in series with a 43v diode to drain and a 2k resistor going to source. The DVM registers 440 vdc at first and then goes back to 485vdc shortly there after. I read somewhere about someone putting diodes on this to keep that from happening. I am thinking I need something besides a IRF830. I am trying to reduce by about 40 volts. Similar to my setup is this picture. Credit for photo to Henry Hurrass KB7OCY @

    http://www.audiotesla.org/6146B Tweet.htm

    Based on article from source:
    www.geofex.com/Article_Folders/mosfet_folly/mosfetfolly.htm

    Both articles at bottom of pages.
     
  18. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    MOSFET IRF830 requires heat sinking for this to work.

    Have you mounted it on some form of heat sink?
     
  19. drycounty71

    Thread Starter Member

    Jan 16, 2012
    30
    0
    No as I have it temporarily on a breadboard and it never even got warm. The total time from readings was about 10-15 seconds and I felt the ic and it was room temperature.
     
Loading...