Power Supply kit - supply voltage on TL081

Discussion in 'Power Electronics' started by Standisher, Oct 15, 2016.

  1. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Just looking at the schematic for a Hiland Power supply kit and am a bit baffled by how the circuit delivers supply voltage to the opamps (TL081). The kit is supposed to be fed by a 24VAC transformer which, I think, after rectification, would supply about +32V DC (accounting for rectifier voltage drop) to the positive rail.

    The TL081 has a maximum supply voltage of 18V yet, in this circuit, they appear to be receiving their supply directly from the 32V rail. Sure I'm not understanding something here and would be grateful if someone could explain how the opamps come to get a supply voltage within the maximum 18V rating.
     
  2. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi,
    The TL081 will accept +18v and -18v volts across it power pins. ie: total of 36v.

    R2 and C2 conduct on the negative half cycles and provide a 'low' voltage on the TL's negative power pin.
    Assume that D9 is a 2.7 Zener, this means that there is only +32v and -2.7v across its power pins. ie:total ~35v.

    OK.?

    E
     
    Standisher likes this.
  3. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Thanks for your reply @ericgibbs. I'm not used to using dual rail opamps so I have a lot to learn here. D7 & 8 are listed as 5.1V zeners, D9 & 10 are listed as 1N4148. Does that change anything in your post? Given that that transformers don't always provide precisely the rated output (I seem to have one which, though rated at 24VAC secondary actually delivers about 29VAC on a 240V UK mains supply) I'm concerned that using your calculation, the DC output would therefore exceed the TL081 max.
     
  4. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi S,
    The UK mains supply is supposed to be 230Vac.
    As you say a 24Vrms would give a peak of 34V, take from this 2 diode forward voltage drops of say 2 *0.7v = 1.4v [it is full wave bridge]
    Giving a peak on the smoothing C1 cap of 34v - 1.4v = 32.5v
    With a 5.1v zener this would give 32.5v + 5.1v = 37.6v total across the TL's pins.

    This exceeds the specified rating of the TL's , however the voltage on the smoothing cap will not maintain 32.5v for the full positive half cycles, it will fall to to a lower voltage.

    I expect the person who designed the PSU has kept the TL's power pins just within the permissible limits.

    Build the kit and measure the smoothing cap voltage, let's know what you measure.

    E
    What is the value of R1 resistor.?
     
    Standisher likes this.
  5. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Thanks ... I appreciate that UK mains is "supposed to be 230Vac" (+/- 6% I believe) but in practice power meter always shows circa 242 Vac. I will build this kit, but think I will 'socket' the TL081's in case they fry. Suspect I may end up needing to get a lower rated transformer :(
     
  6. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi S,
    If you get a lower voltage rated transformer, the nearest of the shelf type would be 20v, which would not give the same highest regulated voltage from the power supply.
    Please post the specification of the PSU.

    Another way of dropping the C1 peak voltage would be to add power diodes in series with the bridge output.

    E
     
    • A02.gif
      A02.gif
      File size:
      21.9 KB
      Views:
      11
    Standisher likes this.
  7. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Last edited: Oct 15, 2016
  8. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi Erik,
    I get a 'server not found' error on that link.
    Just post the Vout and I out maximums for the kit, then I can suggest a diode type.

    E
     
    Standisher likes this.
  9. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
  10. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    Standisher likes this.
  11. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Thanks, The 1N5408 looks ideal (I will not want to exceed 2A on load). I had not heard of Bitsbox before. Their prices seem good and, as you say, the P&P is excellent. Will be using them in the future for sure.
     
  12. crutschow

    Expert

    Mar 14, 2008
    13,041
    3,243
    You might also consider reducing the value of the negative voltage by using a lower voltage zener such as around 2-3V.
    I suspect the circuit will work fine with a lower negative voltage as the voltage just appears to be a non-critical negative bias for some of the devices.
     
    Standisher likes this.
  13. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Thanks for the addition suggestion @crutschow. Unfortunately, I think I'll have to forget that transformer. As I don't have the kit in my hands yet I rectified the transformer output (which measured 30 VAC) with a 2KBP08 and a 1000uf cap (largest I had available) and the result was 40 VDC! I don't think therefore that a few diodes in series on the positive and a lower voltage zener on the negative would lower the voltage sufficiently to avoid opamp burn out. Obviously I'm happy to be corrected if you think I'm wrong.
     
  14. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi S,
    That 40Vdc was not a realistic test, in the actual final circuit build a number of components are drawing current from the smoothing cap C1 and it will measure lower than 40Vdc.
    On load, the added diodes will reduce the voltage on C1.
    As Carl says a lower voltage zener diode would also reduce the TL's power pin voltages.

    Ref your intention of using DIL IC sockets, I have always done so in all my products, for the cost of a DIL socket IMO it is worth while.
    I have seen a number of commercial PCB's ruined by someone trying to remove a dead IC.

    E
     
    Standisher likes this.
  15. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    So, you think that 3 (or maybe 4?) 1N5408, in series on the positive side and replacement of the 5.1V zener (D7) with a lower value would possibly reduce the voltage sufficiently? Couple of questions if I may:
    1) I assume the Zener (D8) can remain 5.1V value?
    2) Can you recommend a suitable lower voltage zener? I'm having difficulty finding one perhaps because I'm assuming it should be of a decent rating (1W or 1.3W).

    Yes I pretty well always socket ICs; pleased to hear the pros do too.

    Regards
    Erik (almost your namesake:) )
     
  16. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    • A01.gif
      A01.gif
      File size:
      15 KB
      Views:
      3
    Last edited: Oct 16, 2016
    Standisher likes this.
  17. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
  18. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    @Standisher
    hi S,
    I ran your psu in LTSpice, image A07 shows it set at 24Vout at 2Amp, with a 26Vac input which is equivalent to a mains supply of 250Vrms.
    Note the 3 added series diodes on the bridge out.

    In image A08 it has a 2.7V zener, the current limit circuit is effected and is unacceptable and would mean changing other components to make it work.

    At 250Vrms, the OPA supply rails are just within the OPA specified limits.

    Eric

    For ref I have added image A09, 250Vrms, no series diodes.
     
    • A07.gif
      A07.gif
      File size:
      61.7 KB
      Views:
      4
    • A08.gif
      A08.gif
      File size:
      57.6 KB
      Views:
      4
    • A09.gif
      A09.gif
      File size:
      61.6 KB
      Views:
      4
    Last edited: Oct 19, 2016
    Standisher likes this.
  19. Standisher

    Thread Starter Member

    Jan 16, 2015
    47
    3
    Thanks for going to the trouble of building in LTSpice @ericgibbs ! Really appreciated. I have come by a lower VAC transformer and was still considering whether to use that (the sacrifice of a few volts is not critical to my needs) or to push ahead with the transformer we have been discussing, incorporating the modifications. Your simulation has pushed me towards the use of the lower voltage toroidal, particularly as the zener diode reduction can't be applied and results otherwise are right on the edge. I have learned a lot through this thread and the extremely helpful contributions you have made.

    Regards
    Erik (AKA Standisher)
     
  20. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    hi E,
    What is the specification on the 'new' transformer.?
    Vin. Vout Amps .??

    E
     
Loading...