IGBT driver / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

IGBT driver

previous 1 2 3 next

Move Thread

LAN_403

Linas

Sun Nov 02 2008, 08:53PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

and here is idea, , i don;t know or it's will work

Dylan wrote ...

Having said that , your 3 channel smps looks great, a work of art! And It would be a shame not to use it

damn right !!!

from line take just 6.1mA not loaded ! (IR supply resistor take 6mA, so for fet's go 0.1mA

)

rp181

Sun Nov 02 2008, 09:16PM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Dylan, you might want to flip those capacitors. When you use the capacitors, the elctricity electrolisizes the electrolyctic and produces gas. The hydrogen vents are on the top, and if its upside down, it will push out the liquid, rather than the gas.

Wolfram

Mon Nov 03 2008, 11:12AM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Sorry for the late reply, I had been meaning to reply to this thread some time ago, but I forgot about it.

Linas wrote ...

i believe, that much simple use independent drivers with semiconductors, it give more functions, and better speed,but stick have two ends

Simpler? I have counted the visible components on your SMPS, there seems to be somewhere between 50 and 60 components, I'll go with 50 to be fair. Your driver schematic consists of 19 components, four driver boards make up a total of 76 components, some can probably be omitted since the drivers for the lower IGBTs have a common PSU, but we are still talking about something like 120 components in total. Then there's the drive signal isolation, which will add something like 5-10 components.

Steve's GDT-based IGBT driver in this schematic

consists of 17 parts, 30 if you also count the GDT, the gate protection TVSes, the gate resistors and their antiparallel diodes.

I also think GDT drive is safer, as seperate high side drivers could accidentally turn on all IGBTs at the same time due to EMI. I don't know how likely this is to happen in practice, probably not very in a well-shielded design, but it's something I'd be a bit worried about. Also, with more parts, there are more parts to go wrong.

Anders M.

Steve Conner

Mon Nov 03 2008, 11:37AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

rp181 wrote ...

Dylan, you might want to flip those capacitors.

I don't buy this. Electrolytic capacitors get used in all orientations, all the time, in commercial products. I'd say you were getting them confused with SLA batteries, but even those can be used in any orientation.

Linas

Mon Nov 03 2008, 01:17PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

well, i don't know,, but having independent driver's you can control every think what you want. I will add delay for turn on, and but not for turn off (yep, it's possible with resistor and diode)
Maybe i will add Vsat protection ....
and about falls in driver, TI ic's give 2ns delay so it's hard to make fall's at so short times, also have noise filters

And all electronics surrounded by mu-metal !!!

Linas

Sun Nov 09 2008, 01:20PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

hm, can someone show, IGBT Vge voltage with GDT and with semi driver, because my wave form are bad

pic is 60KHz wave form....

LithiumLord

Sun Nov 09 2008, 03:29PM

Registered Member #1739 Joined: Fri Oct 03 2008, 10:05AM
Location: Moscow, Russia
Posts: 261

Really good in fact! The thing you see is some ringing on the leakage inductance, try adding a bit of charging resistance and it will work well!

Linas

Sun Nov 09 2008, 07:03PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

wave form is rubbish. And i guess i know why. typical in large IGBT gate charging is like 1-2uC as example CM300DY-24H have just 1500nC, CM60HA-24H have 3000nC,and my one IGBT have 4300nC, so for high frequency need have high dI/dt driver.( at 200KHz my smps wasn't strong enough, even my fet's was not sufficiently powerful(IRFZ24N-IRF9Z34N!!!)

Dylan

Mon Nov 10 2008, 10:31PM

Registered Member #1024 Joined: Sun Sept 23 2007, 10:56AM
Location: Northern NSW, Australia
Posts: 95

I have to say the waveform you have above is actually very good in my opinion!
--- EDIT---
Ahh , I see what you mean, I was only looking at the nice sharp rise and fall :)
------------
And LL is right , you should be able to get rid of that ringing with appropriate gate R.

Anyway, as per your PM, here is a sample waveform from my (derivative of Steves new driver) driver, driving my fullbridge of bricks at 180kHz (my highest F secondary).
Its a bit blurry, but it is at 10v/div.
This is about as fast as I would like to drive this bridge, so this is the worst case waveform I could give you.
3020601010 48fe337c50

The total (80%) rise time is around 1.2us, but note the soft rise above 10v, the critical turn on/off components are roughly 180ns(0>10v rise) and 160ns(high>0v fall).
Ill rummage around for the datasheets and update this with the full gate C this is driving....

----- update ------
Mmm, datasheet doesn't mention gate charge, but full gate capacitance on bridge is 48,000pF.

----- update ------
Forgot to mention , gate R is 5.1ohm with 1N5819 parallel

----- update ------
Mmm, looking at wrong datasheet , full gate capacitance on bridge is 76,000pF.

----- update ------
It is also worth noting I am only driving the gates at +/- 18v! they are specified up to +/- 30v .

Linas

Tue Nov 11 2008, 05:07AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

wow, impressive. at 150KHZ my fet's was heating very fast, extremely fast, also waveform was just 12V of 30V (18V fall on mosfets )
If you can, add 60KHZ waveform pictures please, i want to see difference between your driver and my driver at low frequency

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob