regulated AC current

Discussion in 'General Electronics Chat' started by Alexvatt, Aug 26, 2013.

  1. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    Hi,
    I’m looking for AC power supply with regulated current. I need to control current within 10+-5A range. The load is electrochemical cell with resistance of 1.25+-0.6 ohm. I hoped to find off-the shelf solution but was not lucky so far. Any advice is very welcome.
     
  2. LDC3

    Active Member

    Apr 27, 2013
    920
    160
    An electrochemical cell works better with DC current. After all, you don't get AC current from a battery.
    What voltage does the power supply need?
     
  3. #12

    Expert

    Nov 30, 2010
    16,278
    6,791
    If the resistance is that stable, all you have to do is provide a voltage that is stable to get a current that is stable. Just apply 12.5 volts and the current will be 10 amps, +/- 3.24 amps.
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    This was neither feedback nor a suggestion, so I have moved it to a more appropriate forum.
     
  5. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    Resistance fluctuates within the range that I mentioned but I need fixed current.
     
  6. wayneh

    Expert

    Sep 9, 2010
    12,108
    3,038
    Are you changing your specifications? Earlier you said 10A ± 5A. As noted, a fixed voltage supply will do that.
     
  7. #12

    Expert

    Nov 30, 2010
    16,278
    6,791
    You said you needed a fixed current +/- 5A.
    I told you how to get a fixed current +/- 3.24A
    What is the allowable variation now?
     
  8. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    Depends on application and I need AC current for mine. I suppose around 10V.
     
  9. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    I need to be able to fix current at 5A and run a test than increase the current to 6A than 7A.... During the test resistance fluctuates within specified range.
     
  10. #12

    Expert

    Nov 30, 2010
    16,278
    6,791
    Volts are not a measure of current. The bottom line is: you don't know what you need.

    If you will tell what you are trying to do, there might be some help for you here. If you just keep guessing, there will be no useful answers.
     
  11. wayneh

    Expert

    Sep 9, 2010
    12,108
    3,038
    I bet there's hydrogen involved. ;)
     
    shortbus, #12 and blueroomelectronics like this.
  12. WBahn

    Moderator

    Mar 31, 2012
    17,743
    4,789
    At any given time, say when you are running a test and you want the current to be 7A, what is the acceptable range of actual current?

    Are your current numbers RMS, amplitude, peak-to-peak?
     
  13. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    My current numbers are RMS and acceptable range is less than +-10% ideally but I can probably live with 15%.
     
  14. WBahn

    Moderator

    Mar 31, 2012
    17,743
    4,789
    One way to do it is to just put a precision and stable resistor in series with your cell so that the variation due to cell resistance changes are acceptable. For conversation sake let's call it a ballast resistor. Let's target a value that will nominally keep the current within ±5%. Since your range of cell resistances is 1.25Ω±0.60Ω, the allowed 5% tolerance consumes a bit over 0.06Ω, leaving us with another 0.54Ω to absorb. Thus we could put a 10Ω ballast resistor in series with the load. Choosing a resistor type that is stable is more important that choosing one that is accurate. But let's say that our ballast resistor varies by 1%, or 0.1Ω, over the course of a test. These are presumably independent changes, so the statistical change would be the Pythagorean sum of the two, or 0.61Ω Thus the total load resistance seen by the supply is 12.25Ω±0.61Ω, which is ±5.0%.

    Now, your supply has to put out 10x the voltage, which means a more costly supply and more wasted energy. If you decided to trade that off for less performance, you could target a nominal regulation of ±10% and choose a 5.1Ω ballast resistor which nominally gives you load stability of ±9.5% and let the soft performance margin of being able to tolerate up to ±15% allow you to accept the fairly rare times when you would actually go outside the ±10% preferred range.
     
    #12 and Alexvatt like this.
  15. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    It's a great idea and it should work in my setup! Thank you very much!
     
  16. WBahn

    Moderator

    Mar 31, 2012
    17,743
    4,789
    Glad to hear it. I would use the largest value ballast resistor that will work as that will reduce the variation even more and build in margin for other variations coming from the supply itself.
     
  17. BobTPH

    Active Member

    Jun 5, 2013
    782
    114
    Since it is AC you could use a capacitor or inductor as the ballast instead of a resistor and waste no power at all.

    Bob
     
  18. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    Bob, it’s a good point and I did not take into accout that the cell has some capacitive reactance that will likely change in the cource of the experiment and I would need to compensate for this change as well. I don’t quite understand how can I compenaste for the change in the cell resistance by adding a capacitor.

    Alex
     
  19. BobTPH

    Active Member

    Jun 5, 2013
    782
    114
    It works the same as the resistor, you just substitute the capactive reactace for the resistance. This is used in cheap LED drivers running off the mains.

    Bob
     
  20. Alexvatt

    Thread Starter Member

    Aug 24, 2013
    31
    0
    Does it go in paralled with the load? How do I choose the capacitance?
    Alex
     
Loading...