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

Re: [TCML] IGBT?



Gotchyea, the rise over fall time makes perfect sense now. Thank you for the link. 

Well I guess I will have to get some IGBTs and test them out, when the time comes. =) 
But having bricks would indeed be much nicer, just from looking at them I can see that they are rugged. 

Does anyone have any of the cm600's or close to that for sale? 

Thanks, 
John "Jay" Howson IV 

----- Original Message -----
From: "Finn Hammer" <f-h@xxxx> 
To: "Tesla Coil Mailing List" <tesla@xxxxxxxxxx> 
Sent: Wednesday, September 21, 2011 2:42:48 AM 
Subject: Re: [TCML] IGBT? 

About how much current you can push trough a semiconductor. 

Manufacturers test theit devices to a level where they are willing to 
back them up with at least their name and reputation. Not saying thay 
will ever replace them for you. 
In tesla hobby community, it has become current practice to test the 
devices to destruction, or in the case as with real professionals like 
Terry Fritz, to just under destruction. 
I recall seing scope shots of the voltage drop across these IGBT´s at a 
point where they started to drift into a situation where their resistance 
would rise, and heat is generated violently. As far as I understand, the 
devices are forced to turn off ever so slightly, because the gate voltage 
is lowered due to some paracitic capacitance acting on the rising 
current, and therefore cannot be counteracted upon. 
The device is going out of saturation, I believe is the term applying to 
that situation. 
So to the average joe, like myself, it is "suck it and see" and I have 
successfully driven CM600´s beyond the 8kA mark without blowing tham up. 
But no telling how long they will last. 

I do have one coil running now in museum duty on a daily basis for the 
6th year in a row, and if I was more scientifically inclined, I guess I 
would calculate the primary current in it, to support my words with hard 
data. 

So this didn´t help much, but then, with all due respect, you can´t 
expect detailed help before there is a committed design to select devices 
against. I would suggest bricks anytime, though, coz thay are rugged-fer 
sure. 

Now the trisils: 
When you get to a design that is actually going to be built, I will find 
a nice little program that Terry wrote back then, it calculates, and 
plots, the current trough trisils and IGBT, depending on the voltage, 
primary inductance and the size of the tank cap. 
The current trough the trisils can only rise to so much due to the LC 
time constant, and the turn on time of the IGBT, so these are the 
parameters you play with. 
The Trisils just short the voltage across the IGBT for long enough to 
turn them on, so that they themselves can short out, and take over. As 
soon as the voltage is kept low by the IGBT´s, the trisils will recover 
and be armed and ready for the next shot. 
The 10/1000 is a standard lightning related pulse, it should be 
gooogleable. 
Ok, just did: 
It describes a pulse that rises in 10uS and decays in 1000uS. That is a 
very wide pulse, and in radio frequency the device will be able to 
tolerate much higher peak curents. 
http://www.semtech.com/images/datasheet/si96-03_ag.pdf [http://www.semtech.com/images/datasheet/si96-03_ag.pdf] 

Cheers, Finn Hammer 
----- Original meddelelse ----- 

> Fra: jhowson4@xxxxxxxxxxx 
> Til: tesla list <tesla@xxxxxxxxxx> 
> Dato: Ons, 21. sep 2011 02:03 
> Emne: [TCML] IGBT? 
> 
> Hello all, 
> 
> I am attempting to figure out how to properly choose an IGBT for a 
> SISG board. 
> Not sure if I will be building one yet, but more of a learning and 
> trying to figure everything out kind of thing. so I have been using 
> my current coil as a base design, just for the purposes of figuring 
> this out on paper. 
> 
> The schematic calls for an IRGPS60B120KDP 
> But the current to pulse frequency characteristics are not listed in 
> the datasheet. 
> 
> How does one determine if the IGBT will be acceptable for my 
> particular coil set up? 
> Java TC tells me that my primary current is close to 400 amps peak 
> current (not that the coil that might be built would have that 
> primary current) 
> I read some where (wish I remembered where) that these were fine up 
> to an 800amp primary peak current...is that way off base, and if not 
> where did that number come from? 
> 
> 
> I am having a similar issue with the SIDACs which I have been 
> informed the proper replacement is called a TRISIL 
> specifically this one 
> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00001359.pdf 
> 
> Once again I am not sure why. Probably because I don't understand the 
> format. Like in the section labeled repetitive peak pulse current the 
> value is defined as 10/1000 us being 30A at first i though that it 
> was a simple fraction expressing that it could handle a 30 amp pulse 
> every 10ns, but apparently that is not right per a random forum post 
> I found... somewhere... So what exactly does 10/1000 mean? 
> 
> 
> 
> Thank you to anyone willing to help me learn, 
> John "Jay" Howson IV 
> 
> _______________________________________________ 
> Tesla mailing list 
> Tesla@xxxxxxxxxx 
> http://www.pupman.com/mailman/listinfo/tesla 

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