[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Mosfet H bridge questions

The idea is to eliminate a chip altogether, as I want to use a programmable input signal (+5VDC.) The frequency will at it's lowest be 15kHz, and go as high as several hundred, so a gate transformer will be an option, but I know noting about designing one, hence my autotransformer idea, which would in theory adjust itself more or less to varying input voltages, I don't know if it is feasible to take a 5V signal up to 400V + the turn on voltage of the FET, hence I didn't research that route (though I suppose I could use a standard gate transformer with the secondary of the high side grounded to (boosted by?) the Vinput to the bridge, but I don't know if it works that way... This would also allow for a more or less independent voltage input if MOSFETs work the way I think they do.) The key elements here is that it be full bridge, have flexible voltage input with minimal changing of parts, be driven by a 5V signal with a completely varying duty cycle and asymmetric input possible (so no driver chips,) and have a relatively low parts count. Any additional thoughts or suggestions? 

Scott Bogard.

On 5/30/2010 4:58 PM, Henry Hallam wrote:

I don't think the autotransformer idea is likely to work very well.
You can use an isolated gate drive transformer if the switching
frequency is going to be high enough, or something like the IR2214
(http://www.irf.com/product-info/datasheets/data/ir2114ss.pdf) which
is a chip designed for exactly this purpose.

Good luck

Henry

On Sun, May 30, 2010 at 9:17 AM, Scott Bogard<sdbogard@xxxxxxxxx>  wrote:

Hi All,
     I'm trying to wrap my head around just exactly how one would use an N
channel MOSFET/IGBT on the high side of an H bridge.  What I am trying to
accomplish is to build a more or less "universal" bridge that with
substituting components per necessary ratings could be used for everything
from a flyback driver, to a leg of a PWM for a VFD (apparently VFDs are
actually feeding a PWM signal that varies duty cycle with time to simulate a
sine wave, they don't actually put out a sine wave at all...) even
eventually a crude DRSSTC, using the same PCB.  I plan on using an arduino
or other similar programmable controller with multiple outputs to generate
signals for these various applications.  Here is what I came up with, I want
to know if it is at all viable, and keeping parts to a minimum is a key
point for me being a mechanical guy.
http://bogard.110mb.com/generic%20H%20bridge.bmp
If it doesn't load take out the %20's and replace them with spaces.  The low
side signals can be driven off a tap of the autotransformer, which I don't
really know how to approach designing or if it will even work in this
application (or even if it is really necessary...)  Obviously the number of
turns will be dependent on input voltage (more like 12 or 24V for flybacks,
400V for PWM applications.)  I will pick the wire gauge based on the current
capacity of the transistor, but as how to know I have an inductance that
will keep current low enough that my transistor will be able to survive, I
don't know how to approach this, or how to size my transistor for voltage
considerations (would it see 120V or only the saturation voltage?)  This
stuff is not my field so be gentle so very wise collection or electrical
engineers.

Scott Bogard.
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla





_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla