ST L6599A resonant converter

[Chrishost]

I'm looking to design an LLC resonant converter with this IC
http://www.st.com/en/power-manage

Would it be possible to configure this for a full bridge?

Would I be able to use the high and low side drivers to drive 2A optocouplers (3180) in order to drive higher power mosfets or igbts?

[Chrishost]

I suppose I've got to go half bridge since I can't seem to find anyone that makes a full bridge IC. I have to understand bootstrapped high side drivers and the voltage divider capacitors before going forward.

My understanding is that a half bridge needs the voltage divider in order to invert the load with a single winding primary transformer. Many schematics do not include these capacitors. When looking at a half bridge resonant converter the chip manufacturers nor do any of the articles that I can find mention these capacitors and actually show a simple voltage source.

https://e2e.ti.com/cfs-file/__key/communityserver-blogs-components-weblogfiles/00-00-00-03-59/3036.figure1.jpg

Am I missing something here or is it just assumed that those caps are supposed to be there?

[ocset]

The capacitor divider in an half bridge LLC is to  reduce inrush at startup.
If you use a half bridge llc driver to drive a full bridge llc, then you need to drive all 4 fets through a similar pulse transformer…..this is because you need the delay to each fet to be the same, and  as low as possible…..this is because the dead time is important in llc converters…..specially at light load when the frequency goes high….you need to have enough dead time to allow enough magnetising current rise to enable zero voltage switching.

i think infineon possibly do a full bridge llc driver but i could be wrong.

Also you can use bootsrapped high side drivers too, yes.
You cannot use optos to drive llc fets...optos are just too slow.
Whatever you do the driving with, you need it to be the same for each fet.... or rather, the delay to each fet must be the same, AND LOW ENOUGH DELAY NOT TO AFFECT YOUR REQUIRED DEAD TIME. (sorry about capitals)

[ocset]

here is LLC deisgn for you...attached

[ocset]

here is another...i tried to get another to you but it was >2000MB.
i tried to delete the double attach  but it wouldnt let me

[Chrishost]

Thank you for the links. I've actually successfully made a gate drive transformer before so that works for me.

How would you size these divider caps? I'm still not clear on this. Do they serve the same purpose as they would in a conventional half bridge converter?

[ocset]

you mean in an half bridge LLC?...the two  series caps, taken  as if in parallel, their value is the same value that you would give if you were just using a single resonant capacitor....so they are each half the value of what the single capacitor would be......the two divider caps just stop the llc overcurrenting  at startup.

Pg 12 of the attached tells you about the two capacitors in llc

also here
https://www.infineon.com/dgdl/Infineon-ApplicationNote_MOSFET_CoolMOS_Primary_Side_MOSFET_Selection_for_LLC-AN-v01_00-EN.pdf?fileId=5546d46147a9c2e40147d3430e927e5d

[Chrishost]

So I went with an ICE1HS01G-1 resonant half bridge controller instead. I have its outputs going straight to the inputs of 12A low side drivers (mic4452yn). The mic driver ouput goes to a gate drive transformer with a coupling capacitor. The transformer is connected to a large IGBT (1200v/300A ) with an 8 ohm gate resistor. Im having issues getting good waveforms.

[Chrishost]

1. Setup
2. 25khz at IGBT
3. 60khz at IGBT
4. 60khz at mic4452yn output before transformer
5. 60khz at mic4452yn output with no load.

There is no power/load on igbts. When using a cap that is a lot smaller than IGBT gate capacitance the waveforms at the  cap are perfect but the other waveforms in earlier stages are similar to the pictures. What is the most likely culprit? Could this be related to the smps that is providing both IC's VCC?

[Chrishost]

LLC Resonant controller IC https://www.infineon.com/dgdl/Infineon-ICE1HS01G-DS-v02_00-en.pdf?fileId=db3a304327b8975001281ab6d8201a99

IGBT http://www.alldatasheet.com/datasheet-pdf/pdf/181/MITSUBISHI/CM300HA-24H.html

Gate Driver https://www.mouser.com/ds/2/268/mic4451-779120.pdf

Could this have anything to do with miller capacitance of the IGBT?

Thanks

[T3sl4co1l]

Dear God, what are you planning with those industrial modules?

They're specified for +/- 15V gate drive for a reason.  And 1 ohm driver resistance.

I've used IXDD614CI family drivers on similar transistors.  With a 10W (+/-15V) supply to back it up.  And desat protection.

Unipolar drive, 8 ohms?  You're tickling the beast.  It laughs at you.  Soon it will crush you... 

Tim

[Chrishost]

Just a learning experience at this point. I think it would be cool to get a 1000w+ resonant converter working. Those IGBTs are all I have haha. Also I have more tolerance for error with those. FYI my GDT is a step up so I do have +/- 15 (yeah I know that probable worsens the issue). Yeah I guess I'm going to have to shorten wire lengths and focus on parasitics. Rings like a bastard with lower resistance.