• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 1844 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Really big capacitor with LM338

Jeff Starkell
Intellectual 505 points
Other Parts Discussed in Thread: LM338, LM3886

I'm adding voltage regulators (2 x LM338) to an existing power supply that powers a 12A rms current source. Regulating from +/-36Vdc down to +/-30Vdc. The original supply only produces +/- 3Adc however the current source is used intermittently for very short bursts (< 16ms), so I will rely on very large bulk capacitors to supply the required current for the short burst. I'm inclined to put the bulk capacitors on the output for the following reasons:

  • 35V vs 50V capacitors
  • Peak surge current does not have to flow through the regulator (data sheet states 12A for 0.5ms)

I will rely on the internal current limit to protect the device on power-up (charging the big caps).

Any reason not to take this approach?

  • 0
    TI__Mastermind 22610 points

    I presume that your 2xLM338 is a parallel arrangement.

    You did not mention what the continuous current draw might be, if any, between the pulses.

    As described (or as I have interpreted) I don’t see any obvious “show-stoppers”.

    Just be aware that the impedance that the load sees will be different depending on where the bulk capacitors are at in the circuit.

    I would strongly encourage you to check the device temperature during operation to make sure that whatever power dissipation (i.e. device heating)  there might be is being shared between the two LM338’s.

  • 0 in reply to Donald Jones
    Intellectual 505 points

    Donald,

    I appreciate the feedback.

    The LM338's are not in parallel. I'm using two independent 36Vdc switch mode supplies with floating outputs, each one is regulated via a LM338 down to 30Vdc and the +30V of the 2nd circuit is connected to the ground point of the first circuit as shown below. The protection diodes (for the 27,000uF caps) might not be populated to prevent backfeeding the regulator assuming the current limit on the LM338 kicks in after 0.5ms.

    Your comments are welcome.

    Current between pulses will be very low, typically the quiescent current of 3x LM3886 (~150ma)

    For cooling I plan to use a fan based Pentium cooler shared (3x LM3886 and 2x LM338T), so between pulses Pd~ 11W. I will generate 3 high current pulses (16 ms) every minute.

  • 0 in reply to Jeff Starkell
    TI__Mastermind 22610 points

    Appears to be reasonably straight-forward, with no obvious show-stoppers.