• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 369 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Trimming the LMZ22003

TI Lady
Genius 4025 points
Other Parts Discussed in Thread: LMZ22003, TPS62050

I have a new application which I do plan to use two LMZ22003’s.  Is it possible to trim up/down the output by driving the FB pin externally using a 0-3V through a resistor?  I think it depends on the internal reference voltage.  Attached document shows the way I achieved this before.  The internal reference has a 1k resistor in series with it and one can use it as a summing junction to modify the reference voltage hence trim up/down the output.  I am hoping I can do the same with the LMZ device.  Please advise.

 3438.300W 8V Regulation by SC pin.pdf

  • 0
    Prodigy 5 points

    Amy, the LMZ22003 uses an external divider compared to a 0.8V internal reference.
    You should be able to vary RFBB by injecting a current offset into it by your external 0-3V signal.

  • 0 in reply to Ed Walker
    Genius 4025 points

    HI Ed,

    I will be using two of them at a common clock (600-800k, not sure exactly yet—but this is the range).  I know it is advertised the device will pass EN5502.  But, can you suggest an external filter ALREADY designed so I can use in case I need one?  

    Thanks!

  • 0 in reply to TI Lady
    Prodigy 5 points

    Hello Amy,

     

    EN55022 has both radiated and conducted.
    The AN-2125 shows conducted and radiated scans of the LMZ22 LMZ23 EVM.

    PCB layout and mechanical enclosure really determine the radiated profile in the LMZ modules.

    About the only thing you can do with this type of module is perhaps filter out some high frequency noise on the input bus or output.

                    For better radiated a mechanical metal shield should be considered.

     

    A simple PI filter made with capacitors and FERRITE BEADS is common.

     

    In the EVM schematic it shows a 150uF followed by a 0.47uF input cap, the PI filter would go before the 150uF.

    The output shows a .047uF followed by 2x220uF, the PI filter would go after that.

    MAKE SURE the FEEDBACK signal is taken BEFORE the PI filter, not after it.

     

    A Fair-Rite 2773021447 and some 0.001uF (1nF) ceramic caps on each side would make a good PI filter.

    Of course any filter might need to be TUNED to work the best for a given application.

     

    Here is an example:

    This figure shows a typical input filter. This input filter is a pi-filter that provides up

    to 40dB of This pi-filter could also be used as an output filter as well. In this

    example, C3 is a 10uF bulk input capacitor for the TPS62050. Looking at the

    specifications for the 10uF capacitor shows it has a SRF (series resonant frequency) of 1.925MHz. This means

    that C3 can not be used as part of the input filter since it has a poor high frequency

    response. To solve this problem, another capacitor, C2, is placed in parallel to

    complete the input filter. The bead and C2 form a L-C filter to attenuate noise from

    the input power source. The bead and C1 form another L-C filter that attenuates

    noise generated by the SMPS from propagating toward the input power source.

     

    Some designers, in an attempt to use common parts, try to use a power inductor as

    the inductive element in the input or output filter. A power inductor is a poor choice

    for a high frequency filter even if it has a high SRF. A power inductor is a high Q

    component meant to store energy with low loss. A ferrite bead is a low Q part meant

    to dissipate energy as heat. The two are not interchangeable.