• State
  • Locked Locked
  • Replies 10 replies
  • Subscribers 65 subscribers
  • Views 2445 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.

LM5100 FET Driver

Luis Pereira
Prodigy 10 points
Other Parts Discussed in Thread: LM5101, LM5041

Hi, this is my first post in this forum, so I will try not to be too much off-topic.

I have some doubts about the LM5100 FET driver (from National) that I would like to have some advices.

http://www.ti.com/product/lm5100a

I build a small prototype board just with the chip LM5100 and 2 capacitors, in order to do some tests with MOSFETs and IGBTs.

The current configuration is:

1 - An Arduino is programed to generate and sinusoidal wave (configured by pushbuttons) from 2-100Hz, with commutation frequency also selectable from 1kHz to 4kHz.

2 - The 2 outputs from the Arduino (inverted and non-inverted) of 5V are directly connected to the LM5100 inputs (LI and HI). The LM5100 board has 1 electrolitic 10uF, 63V capacitor between Vdd and Vss, and also a bootstrap electrolitic capacitor of 1uF, 63V between HB and HS.

3 - On the output of the LM5100 I already succesfully tested to directly connected some LEDs, and a pair of IRFZ44N. My plan is to use the configuration to test the limitis of some MOSFET and IGBTs, so later I can build something bigger.

Now my questions:

a - Can I use LM5100 for something more serious like a full bridge PWM controller of a 1000W DC Motor, or a AC inverter for 3Ph Induction Motor? Or it is restricted to DC/DC chargers and rectifiers?!

b - Concerning point 2, I have no idea if I should use electrolic or ceramic capacitors. I know that they should be low ESR... And what about recomended values? the LM5100 datashet has no reference to that.

c - I it correct to assum that an MOSFET Driver can be use without restrictions to drive some IGBTs?!

d - This test board I build just work with the supply voltage between 7 and 11V! Why it can´t reach the 18V indicated in the datasheet!?!? It is an "hand built" PCB, probably with some parasitic currents between tracks... but I imagine that could be relevant only for high frequencies...

 

Thanks for your support!

LPereira

  • 0
    Prodigy 10 points

    Try a larger cap for C-boot I suspect that you are losing the high-side gate charge at that low frequency.

  • 0 in reply to Larry Dawson
    Mastermind 7395 points

    You may want to take a look at the LM5041 evaluation board, which includes a LM5101

    http://www.ti.com/lit/an/snva074/snva074.pdf 

  • 0 in reply to Larry Dawson
    Prodigy 110 points

    I'm new in this kind electronics, so expect from me some stupid questions! :P

    Can you let me know what really is the influence of a bigger capacitor?

    I was already using 1uF. Some examples like Albert_83 indicated use 0.1uF.

    As far as I understand, bootstrap capacitor will charge during the OFF state of High side IGBT, and use the stored energy to turn it ON.

    That mean if it is too large, and the frequency to high, it won't have enought time to charge and "build up"  the necessary voltage?

    From other side, If it is too small, it won't store enough energy to turn ON the IGBT?!   Is that correct?!?  

    But some industrial frequency inverters can change the commutation frequency between ranges like 5-20KHz. I'm wondering how they do that...

  • 0 in reply to Avi Chami1
    Prodigy 110 points

     , thanks for your example.


    Did you know why they are using a small resistor (R15/R16) parallel with an inverted diode (D4) in the output of the driver!?!

    And also a 10K resistor between Gate and Source of the MosFET!??


  • 0 in reply to Larry Dawson
    Prodigy 110 points

    Larry Dawson said:

    Try a larger cap for C-boot I suspect that you are losing the high-side gate charge at that low frequency.

     , I'm not sure if you are answering directly to my question 4, but both High and Low side outputs of the driver stop working above the 11V, so I think that's not directly related with the bootstrap capacitor?!?

  • 0 in reply to Luis_Pereira_PT
    Mastermind 7395 points

    Hi Luis,

    You are welcome.

    The resistor and diode make for different behavior during turn on and turn off of the FET

    On turn on, when the driver sources voltage, the gate current is limited by the resistor.

    On turn off, when the driver sinks current, you want to take out the charge on the MOSFET gate as fast as possible, the diode makes a fast (no resistance) discharge path for the gate capacitance.

    I don't know the reason for the 10K. I usually would put a resistance like that to provide a path for self discharge of the gate capacitance when not driving the fet. But I do that when my driver has source capacity but no sink capacity. Since the LM5100 has source and sink capacity, I don't see the need for the 10K resistor.

  • 0 in reply to Avi Chami1
    Prodigy 110 points

    Thanks Albert.

    And about the bootstrap capacitor, it should be rated acording commutation/working frequency of the IGBT!? 
    There's any theorical calculation or I have to experiment diferent values for my application and measure the PWM signal vs gate-source voltage?! 

    I don't have any osciloscope... :(  

    I'm interested in DSO Quad due to it's low price, but I'm not sure if it won't be enough for this kind of measurements... 

    http://www.seeedstudio.com/wiki/DSO_Quad

  • 0 in reply to Luis_Pereira_PT
    Mastermind 7395 points

    First of all, the device has to work up to 14V, not 18V. Always look at the recommended operating conditions, NOT at the absolute maximum ratings.

    I don't know why it stopped working at 11V, though, and I don't think that it is related to the value of the bootstrap capacitor.

    If you have to debug, you need a scope. I don't know all your needs, and I have no experience with the small DSO you showed me. But for this application, it seems to be enough, and it will surely be better than debugging blind. If you can get it for testing, it will be the best. Also, try to check its price against 2nd hand, first brand scopes like Tektronix. Second hand scopes are sometimes cheap and are a good choice for beginners.

    Good luck!

  • 0 in reply to Avi Chami1
    Prodigy 110 points

    Thanks Albert!

    I think I will try it!

    It's cheap, portable, and have some good reviews on the web! 

  • 0
    Prodigy 10 points

    Hi,

    the resistors are present to reduce 'ringing' on mosfet gate, produced by high switching speed of synchronous driver.
    Did you get circuit running? pls pm if you want to compare notes, im working on something simelar.

    thanks