VDmos
Hendrik Jan Zwerver
I have tested your VDmos program and here are my findings so far
(using the IRF7455 as example).
The program seems to have difficulty matching the gate plateau voltage
of the gate charge plot.
The program comes up with an IS parameter while there was no input for
the the body diode. I find this strange.
The transfer characteristics is not correct for the mosfet. Vgs=3.5V
the Id must be 150A not 80A
The auto ranging of the plots can't be turned off. Nothing happens if
I change the plot range manually.
If I hit the Plot button I get something different then the curve fit
button. I don't understand this.
The program calculates too low fet capacitance values. It could be my
fault.
There is one flaw in the datasheet. The test condition for the gate
charge was mentioned to be 15A it should be 150A.
I have used my tool (VDmostool) to model this fet and at least the
gate plateau voltage, transfer and body diode DC forward is a better
match then what I could do with your VDmos program.
.MODEL IRF7455 VDMOS(KP=152.2453 RS=0.0001 RD=0.0051 RG=3.0 VTO=2.09
+LAMBDA=0.001 CGDMAX=1422p CGDMIN=80p CGS=3380p TT=92n
+IS=2.98E-08 N=1.542 RB=0.0091 m=0.375 Vj=1.15 Cjo=2489.25pF)
What is missing in your VDmos program is ability of the body diode
DCforward, reverse recovery and nonlinear Cjo fitting (this has impact
on the switching rise and fall times)
For the rest I like the graphical input of the data, it's very power
full. Since I'm not a software engineer I don't know how to do that.
I'll leave my prog the way it is. I need to do some code cleanup and
then I'm finished.
Best regards,
Hendrik Jan
coli...@gmail.com
Thanks for your feedback. I have just changed the IRF7455 to IRF1234
in version 1.50. Those parameters are for testing only and not for the
real device.
I also agree that the gate charge curve fitting still needs work, and
I'll look into the fitting of body diodes and others.
The auto-ranging is turned off automatically when you edit the X/Y
numbers. You can also turn it on/off by clicking the box. I don't
understand why it doesn't work on your machine. Once it is turned off,
it remain off until you turn it on.
Colin
hendrik jan zwerver
I like the idea of using a real device to test the software, but maybe
it's just me.
The problem with the auto ranging is that if I change the x-y manually
nothing happens
The plot does not re-scale. If I hit enter I get the windows "ping"
sound indicating an error.
What I don't understand is why sudenly after more then two years you
come of with a VDmos program while I have asked in 2006 for
development support for my VDmos program.
Eventually I got Helmut and some other key persons from the group
testing my program and gave me feedback.
VDmostool as it is right now is the result of more then two years
development and testing from persons from the user group.
All the need features are now in the program.
Now it looks like things starting all over again from the beginning.
The difference between us is that I don't get paid for the creation
and maintaining the tool.
What I suggest we can do to save you a lot of time is to send you
documentation and code how to calculate certain model parameters.
Especially for the diode It's easy and accurate. How ever most of the
documentation is already on the user group.
After your program has became the same level of modeling accuracy and
functionality I then see no reason to continue further development of
my tool.
I have for my program only one additional feature on my wish list what
I want to put in before the final release.
If I give you all the support you need and all my code, are you
willing to mention my name in the about box?
Maybe we can get this a step further. If I have the same compiler as
what you use including the code for the program I can help you
developing.
Best regards,
Hendrik Jan