Eliminating Turn off Losses in IGBTs... UPDATED

Hi Genome,

This looks like a very interesting concept. Reducing losses by 75% is an impressive accomplishment. I'd like to run the LTspice model, but it needs your irgp50b60pmod.spi file. When I try to download it, I get the following error message:

Forbidden

You don't have permission to access /patent/irgp50b60pmod.spi on this server.

Apache/1.3.33 Server at

Port 80

Can you change that so your fans can download the file? Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler:

That's a bit of a bum.... I had a look and, being thick, not only does it not like me either but it seems to hate me.

Here's the text version....

I will try and work things out otherwise.

.SUBCKT irgp50b60pd1 1 2 3

• SPICE3 MODEL WITH THERMAL RC NETWORK

**************************************

• Model Generated by MODPEX *

*Copyright(c) Symmetry Design Systems*

• All Rights Reserved *

• UNPUBLISHED LICENSED SOFTWARE *

• Contains Proprietary Information *

• Which is The Property of *

• SYMMETRY OR ITS LICENSORS *

*Commercial Use or Resale Restricted *

• by Symmetry License Agreement *

************************************** *Model generated on Jun 7, 04

• MODEL FORMAT: SPICE3

*Symmetry IGBT Model (Version 1.0) *External Node Designations *Node 1 -> a *Node 2 -> g *Node 3 -> k M1 9 6 8 8 MSUB L=100u W=100u .MODEL MSUB NMOS LEVEL=1 +VTO=5.14841 KP=4.6111 LAMBDA=0 CGSO=3.92729e-05 RD 7 9 0.000923964 RS 4 8 0.00104286

*D1 3 1 50b60pd1

*.MODEL 50b60pd1 d *+IS=1.45652e-10 RS=0.0257735 N=2 EG=0.600503 *+XTI=0.5 BV=600 IBV=0.00025 CJO=2.89747e-10 *+VJ=0.4 M=0.430499 FC=0.5 TT=1e-09 *+KF=0 AF=1

Q1 4 7 1 QSUB OFF .MODEL QSUB PNP

+IS=3.21641e-16 BF=40.1409 NF=1.22694 VAF=443.169 +IKF=2586.69 ISE=9.8821e-12 NE=2.21576 BR=0.990766 +NR=1.44702 VAR=123.998 IKR=999.935 ISC=1.00646e-11 +NC=1.99793 RB=0.534983 IRB=487.231 RBM=1e-05 +RE=1e-08 RC=0.00102494 XTB=0 XTI=2.99904 +EG=1.13276 CJC=1.4649e-09 VJC=0.4 MJC=0.353063 +CJE=1.4649e-08 VJE=0.4 MJE=0.2 TF=9.99991e-10 RDS 7 4 1e8 RER 4 3 0.00993049 RG 6 2 2.86055 RL 10 11 1 D2 12 11 DCAP

• Default values used in DCAP:

• RS=0 EG=1.11 XTI=3.0 TT=0

• BV=infinite IBV=1mA .MODEL DCAP D IS=1e-32 N=50

+CJO=1.5982e-09 VJ=0.4 M=0.716994 FC=0 D3 0 11 DL

• Default values used in DL:

• EG=1.11 XTI=3.0 TT=0 CJO=0

• RS=0 BV=infinite IBV=1mA .MODEL DL D IS=1e-10 N=0.4 VFI2 12 0 0 FI2 6 7 VFI2 -1 EV 10 0 7 6 1 CAP 10 13 6.40436e-09 RCAP 10 14 1 D4 0 14 DL VFI1 13 14 0 FI1 6 7 VFI1 -1 .ENDS irgp50b60pd1

*SPICE 2-Layer Thermal Model Subcircuit .SUBCKT irgp50b60pd1_igbtt 2 0

R_RTHERM1 2 1 0.157214 C_CTHERM1 2 0 0.002201 R_RTHERM2 1 0 0.163184 C_CTHERM2 1 0 26.22805

.ENDS irgp50b60pd1_igbtt

• SPICE3 MODEL WITH THERMAL RC NETWORK

*SPICE 3-Layer Thermal Model Subcircuit .SUBCKT irgp50b60pd1_diodet 3 0

R_RTHERM1 3 2 0.3601412 C_CTHERM1 3 0 0.000311 R_RTHERM2 2 1 0.8647836 C_CTHERM2 2 0 0.0013691 R_RTHERM3 1 0 0.473519 C_CTHERM3 1 0 0.0681324

.ENDS irgp50b60pd1_diodet

That's great! Thanks. Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler:

RatsRatsRatsRats!

I made a self-inflicted injury. I downloaded your page using HTtrack, and always tell it up to rename the files to the DOS 8+3 format. So your irgp50b60pd becomes irgp50b6, and I had to rename all the include statements in the LTspice files.

But that sure messes everything up! Now it can't find the subckt, and probably a bunch of other things got screwed up also.

Moral of the story - not sure yet. This may take a while to sort out. Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler:

Whoops.... fixed it.

My html referred to the wrong directory so the server did a no-no.

I think I've fixed it...... can you give it a quick go?

Many thanks for the slap.

DNA

That's even better. Now I can tell HTtrack to leave the filenames alone, and forget about trying to fix my DOS 8+3 blunder. Thanks, Genome!

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler:

Oh... good.

Once again thanks for checking. I don't think anyone else has bothered with the simulation files but it would have been a little bit annoying if they couldn't get hold of the model required.

And thanks for the HTtrack idea.... Good stuff, I ripped a couple of my favourite specialist fetish websites and got all the piccies.

Monster!

Hmmmm, now how do I set this thing up for....

DNA

Thanks Genome. LTspice now works fine!

If you want to download the entire boot site, load HTtrack then load the HTtrack.opt file.

Tell it where you want to put the download directory, and enter the url.

Next, press the "Set Options" button.

In the "Scan Rules" tab, click the images box and ensure the text area shows "+*.gif +*.jpg +*.png +*.tif +*.bmp". If there are multimedia files, then click the "mov..." box also.

This tells HTtrack it can go anywhere on the web to retrieve those files.

But do not tell it to download "htm", "html", etc. files. If the web site references those files on a different web site, HTtrack will go there and get them. They may themselves reference an unlimited number of files in other directories, so you may end up trying to download hundreds of gigabytes.

In the "Limits" tab, set the maximum mirroring depth to 20. That should ensure you get pages that are way deep in directories.

In the "Build" tab, tell it the folder structure you want, and tell it "no external pages". This keeps it from wandering all over the web and downloading everything in sight.

In the "Mime Types" tab, ensure all entries are erased. Those are left over from previous releases, and HTtrack now figures out everything internally. But if you put anything in those fields, HTtrack gets confused and will probably ignore those file types. I think I got most of the critical things, but go through the rest of the tabs and try to set them to your preferences. Then save your settings for the next time.

Now you can press "OK" and proceed to download your web site.

Good luck, and let me know if everything worked OK.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler:

I forgot two of the most important things:

In the "Experts Only" tab, set "Travel Mode" to "Can Go Down".

Set "Global Travel Mode" to "Stay on the same address".

Now, it can only get html files in the current and child folders, but can go anywhere to get the images.

You need to do this to prevent HTtrack from trying to download the entire web. But you also have to give it permission to go upstream or to other sites if necessary to retrieve the image files.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler:

Don't get me wrong here but I think the right answer might have been....

'Fiddle about with it until it works.'

Which is the right answer for fiddleable software....

I might have to practice on

She is just too Deirdre Barlow!

I'd have a bash at

But they have too many piccies of blokes in rubber.

What's that all about then? Why does a woman in fine latex look nice whereas a bloke looks like a prat.

Oooops, bugger.

I don't know you.

DNA

I'm not from Nam, but I can't seem to be able to download it. I'm on verizon.

Cheers

Sometimes I can't reach sites for unknown reasons, but going through a proxy solves the problem. Try entering the url in

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

SPICE Analysis of Crystal Oscillators: Noise-Rejecting Wideband Sampler: