I had a problem with some IGBT models....

______________________________________________________________ Reference :R7621707218970340 Title :Big Dildo Requester : snipped-for-privacy@yahoo.co.uk Question : I could have been nice about this. I might have spent some time defining the problem as I see it and providing more detailed information. However in order to inform you about a problem I had to register. The information in my registration details is like your spice models for IGBTs. CRAP!!!! I will be informing my company that under no circumstances should they use your IGBTs in the product because they are not suitably rated for RBSOA. That might be a lie but I'm sure they won't argue too much. Have a nice day. ______________________________________________________________ Dear User, Please find here below latest information concerning the request R7621707218970340 Dear Sir,

You seem to be a little bit upset? This is a new system that we are just getting up and running and you may have had an unsatisfactory service for which we apologies. However, I am sure that your company needs a competitve source of products now and in the future and to dismiss one of the major supplies of components for a technical issue with our support hotline does seem a little short sighted if I may be so bold!

If you regain your composure in the near or far future, please do not hesitate to contact us again.

With our very best regards, _____________________________________________________________

Reply to
Genome
Loading thread data ...

Silly thing is that after that response I feel the need to explain further but I've forgotten my log in details and wouldn't be able to send in pictures to show the problem anyway.

Well... there you go. I'll bollox it......

If you are into spice....

Go to SGS Thompson ST or whatever........ download some of their IGBT models.

Go to IRF and grab hold of some of theirs.

Stick them up circuits....... compare them.....

IRF seems to mess up the turn on characteritics but the turn off is close.

ST seriously fucks the bollocks on turn off.

I'm no expert, but comparing things in the models I might take a guess.........

ST got shafted by the model provider.

I'd ask Jim Thompson to take a look at things and give them some advice but, apart from the name, they are a bunch of Frenchies.

DNA

Reply to
Genome

Many a slip between cup and lip. Seldom pays to go ballistic.

RL

Reply to
legg

I've had inquiries about making models, but was informed that I was too expensive ;-)

I guess crap models are OK with the manufacturer.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

Not even. I've seen worse...

--
Thanks,
Fred.
Reply to
Fred Bartoli

snipped-for-privacy@newsfe1-gui.ntli.net...

ut,

See I keep telling you you have to be an EXPERT to get anything useful out of spice, maybe now you'll believe me but then again....

Reply to
cbarn24050

Yes....... of course.

I presume that, by the same reasoning, when you have built one of your 'real' circuits and it is misbehaving you give up. I suppose you will claim that you actually investigate the problem and find a cause and a solution.

In this case the 'problem' was a result of some bad models, hence my disgust with ST. Of course that will be a reason for you to bounce up and down with glee saying 'told you so'.........

Sigh, it must be the shrapnel.

DNA

Reply to
Genome
00cwc.googlegroups.com...

of course.

Why would I do that?

I suppose you will claim

Correct

ust

th

No not really. You've been working on this for weeks, you could have built it long ago and it would most likely be working. My point is you're still going to have to build it and it's very unlikely to work like the simulation so you will still have to get it working the "old fasioned way" and so your simulation has achieved very little.

Reply to
cbarn24050

snipped-for-privacy@aol.com a écrit :

messagenews: snipped-for-privacy@h48g2000cwc.googlegroups.com...

course.

How do you know that (since you're such a spice expert)?

so you will still have to get it working the "old

--
Thanks,
Fred.
Reply to
Fred Bartoli

ice

. of course.

ion.

isgust

with

How do you know that (since you're such a spice expert)?

Because he can't model the transformer acurately, for a start it probably doesn't exist as yet. Then there are all the other unmodeled parasitics.

Reply to
cbarn24050

Hello Jim,

Sounds familiar over here as well. Some don't want to pay the consultant and try it on their own, thinking this would save money. Then a couple of years down the road major problems start to unfold and the phone rings. By that time the required changes will typically be painful. Relayouts, major architectural changes, new agency approvals, significant inventory write-off and so on.

What really surprises me is that most companies have no qualms about paying their attorneys several hundred per hour. Sometimes even for rather mundane work such as checking a standard consulting agreement. But competent technical help often seems not to be valued much. Until the comeuppance, that is...

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

Yep. The only smile is when they come back a year or two later and I stick it to 'em ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

Pointy-haired bosses at least somewhat understand what it is that attorneys do for them, whereas they really don't even begin to understand all this stuff about "models" and "SPICE" and "simulation?"

Reply to
Joel Kolstad

Hello Jim,

I don't. I'll help them but sometimes I am booked up and can't. The really sad cases are when it's too late. Like with a European company that had to shut down. It was clearly avoidable but they could not do it on their own.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

Hello Joel,

The successful ones surround themselves with engineers who tell them when it's time to call in an expert. However, some of them won't listen.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

"Joerg" schreef in bericht news:ydf6h.25005$ snipped-for-privacy@newssvr21.news.prodigy.com...

Ha! As if you don't have a hearing problem, and rethinking DOS as the ultimate OS for your PC's. Hahaha....

--
Thanks, Frank.
(remove \'q\' and \'.invalid\' when replying by email)
Reply to
Frank Bemelman

do

This is particularly frustrating when the expert that needs to be called in is oneself. ;-)

Cheers! Rich

Reply to
Rich Grise

Hello Frank,

Oh, I use Windows. Right now, actually. But without any doubt in my mind DOS is more reliable and sometimes you just need that. Also, there is stuff that simply doesn't exist without DOS. Or have you ever heard of a program to calculate wave digital filters that doesn't run under DOS? And here I don't mean some wild and skimpily documented collection of routines and snippets but the real stuff, where you key in the passband values and it spits out the coefficients in proper CSD representation.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

"Joerg" schreef in bericht news:eZL6h.7481$ snipped-for-privacy@newssvr11.news.prodigy.com...

Yes, you once told us about your favorite filter program. But even so, it may be time to throw it out of the window now, together with the warped 5.25 floppies. I'm not in the digital filter business, but I would expect that your compadres in this area use Mathlab or something.

formatting link

--
Thanks, Frank.
(remove \'q\' and \'.invalid\' when replying by email)
Reply to
Frank Bemelman

Hello Frank,

My compadres in academia use Matlab. The ones in industry mostly don't. Too many version changes, incompatibilities, etc. That's only what I hear, I do not use Matlab myself.

Throw out the filter program? And then? Of course I could tell my clients that they can't have that new gizmo to be "really" portable anymore because now I need a fat DSP that requires four big D-cells to run for a few hours. Or a car battery :-)

BTW Texas just came out with a new filter routine that is a little more comfortable than the one I had, no need to crank the CSD conversions by hand anymore. However, this one is sans graphical interface. Strictly DOS command line. If a company that size does that I guess there has to be a reason...

Thanks, but I already got about 2ft of DSP books on my shelf here.

It's not that I find Windows less useful. This afternoon I had to do a final presentation after completing a consulting task. In that one I had to show the achieved noise reduction and the only way to do that was to run three independent video loops in parallel via a projector. Now way that I could ever do that under DOS. But there are situations where DOS is simply better.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.