• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 712 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.

Original question:

UCC28950: Ucc28950 current transformer pacement. transformer saturation.

UCC28950: Ucc28950 current transformer pacement. transformer saturation.

Bob LaFrance
Prodigy 170 points
Part Number: UCC28950
Other Parts Discussed in Thread: UCC28070,

Greeting everyone.

I have been following this interesting thread.

I am also interested in using current sensor.  My application is 24vdc input with 600W output in a tight package.  I have a little trouble getting my head around a dc current transformer.  It's a slightly different kind of magic.  I can understand the volt-second balance and have used the Ion-Physics current transformers in the past.

I am looking t a 0.001 ohm current sense resistor with an amplifier on the output.   The current transformer method looks like it might be worth looking at.  I can keep the losses in the resistor to a couple of watt so I am interested to know what the current transformer approach will do for me.

You mentioned that the core must be reset.  When commutating from a zero state to a power state the current  always begins negative.  Won't this automatically reset the core every half cycle?  Would there be any benefit to using two cores - one for each source?  Looks like the current is zero for a good part of the cycle.  Or maybe one core with the source leads broken out and passed through the core in an anti-parallel fashion to drive core both ways?

Maybe there is enough strange magic here to prompt a short application note.

Thanks again.

Bob LaFrance

 

  • 0
    TI__Mastermind 25570 points
    Hi Bob

    I'll try to answer your question tomorrow,

    Regards
    Colin
  • 0 in reply to Colin Gillmor
    TI__Mastermind 25570 points
    Hi Bob

    A current transformer is a good way to sense high currents in a relatively lossless way. However there are a few things to be careful of.

    First - as in any transformer Ip*Np = Is*Ns. The secondary current flows in a burden resistor and generates an appropriately scaled voltage - for example something in the range 0V to 2V full scale. The user has to allow the core to reset during the off time of the power stage so that Vsec*Ton + Vsec*Toff = 0 - in other words you have to achieve a volt*second balance on the CT in order to prevent it saturating.

    There are a number of ways you can do this. A diode and resistor for example. You do need to allow for high voltages on the CT secondary during reset with small Toff.

    Of course, if the primary current is symmetrical about 0 - if you have put the CT in the primary circuit of a full bridge power transformer for example, then reset is much easier because as you say the positive current on one half cycle is cancelled out by the negative current on the next.

    There is some useful information in the UCC28070 datasheet at www.ti.com/.../ucc28070.pdf starting on Page 32.
    You may find the article on current sensing at www.ti.com/.../slup114.pdf of interest - although it is fairly old.

    Have a look at the reset circuits we use on our EVMs - the UCC28950 and UCC28070 for example both work well.

    Current sensing resistors will always dissipate power - that's their nature. If you use a 1mOhm resistor then you will have to use appropriate Kelvin connections to make sure that you are not including track and solder joint resistance in the measurement. You will also have to make sure that the sequencing is correct so that the amplifier is on stream when the controller starts - but you won't have to worry about resetting it. !

    Regards
    Colin