GaN FETs for high frequency SSTCs
Dr. Slack, Tue Nov 08 2016, 11:11PMI seem to have taken my eye off the ball for the last couple of years, but have suddenly woken up to the fact that GaN seems to be all over the market in sensible powers like 600v 20A cont 80A peak, with sub 10nS rise and fall times.
I feel a paper design coming on! I've always had a hankering to make a solid state magnifier, rather than a DRSSTC. The higher frequency may permit an air-cored transformer with manageable primary current without the need for resonance, and so rather simpler dynamic control.
I've not seen any threads of people working on GaN TCs, are there any?
Re: GaN FETs for high frequency SSTCs
Mads Barnkob, Wed Nov 09 2016, 07:01AM
There are some older threads where some issues with using them are discussed. I can not find them right now, but I recall it was about the different driving voltages needed for turn-on/turn-off.
The closest might be loveoceans SiC Tesla coil:
Mads Barnkob, Wed Nov 09 2016, 07:01AM
There are some older threads where some issues with using them are discussed. I can not find them right now, but I recall it was about the different driving voltages needed for turn-on/turn-off.
The closest might be loveoceans SiC Tesla coil:
Re: GaN FETs for high frequency SSTCs
Dr. Slack, Wed Nov 09 2016, 09:55AM
Yes, I'd seen SiC discussed, but that technology has been a long time coming and isn't delivering like GaN. I think a major practical difference is that GaN is compatible with relatively standard silicon processing, whereas SiC is not, so there's been an explosion available types. The gate voltage is typically a more conventional +/- 20v as well. SiC still has its high temperature advantage, but its speed lunch has been eaten by GaN.
Dr. Slack, Wed Nov 09 2016, 09:55AM
Yes, I'd seen SiC discussed, but that technology has been a long time coming and isn't delivering like GaN. I think a major practical difference is that GaN is compatible with relatively standard silicon processing, whereas SiC is not, so there's been an explosion available types. The gate voltage is typically a more conventional +/- 20v as well. SiC still has its high temperature advantage, but its speed lunch has been eaten by GaN.
Re: GaN FETs for high frequency SSTCs
Uspring, Wed Nov 09 2016, 10:44AM
Primary to secondary power transfer is proportional to primary and secondary current and also frequency. So a high frequency helps. Running outside the secondary resonance is somewhat of a problem since it implies less secondary current.
Uspring, Wed Nov 09 2016, 10:44AM
Primary to secondary power transfer is proportional to primary and secondary current and also frequency. So a high frequency helps. Running outside the secondary resonance is somewhat of a problem since it implies less secondary current.
Re: GaN FETs for high frequency SSTCs
hen918, Wed Nov 09 2016, 11:46AM
I like SiCs, especially for their low (relative to GaN) price, and their HUGE benefits over IGBTs and conventional MOSFETs, one only has to be careful not to put too much negative voltage on them, and they work just the same as normal 'FETs
There are also several manufactures now selling SiC, but my last look around a couple of weeks ago concluded that Wolfspeed (CREE) still leads.
GaNs are brilliant for really high frequencies, but for normal switching speeds I think SiC has the edge.
hen918, Wed Nov 09 2016, 11:46AM
Dr. Slack wrote ...
Yes, I'd seen SiC discussed, but that technology has been a long time coming and isn't delivering like GaN. I think a major practical difference is that GaN is compatible with relatively standard silicon processing, whereas SiC is not, so there's been an explosion available types. The gate voltage is typically a more conventional +/- 20v as well. SiC still has its high temperature advantage, but its speed lunch has been eaten by GaN.
Yes, I'd seen SiC discussed, but that technology has been a long time coming and isn't delivering like GaN. I think a major practical difference is that GaN is compatible with relatively standard silicon processing, whereas SiC is not, so there's been an explosion available types. The gate voltage is typically a more conventional +/- 20v as well. SiC still has its high temperature advantage, but its speed lunch has been eaten by GaN.
I like SiCs, especially for their low (relative to GaN) price, and their HUGE benefits over IGBTs and conventional MOSFETs, one only has to be careful not to put too much negative voltage on them, and they work just the same as normal 'FETs
There are also several manufactures now selling SiC, but my last look around a couple of weeks ago concluded that Wolfspeed (CREE) still leads.
GaNs are brilliant for really high frequencies, but for normal switching speeds I think SiC has the edge.
Re: GaN FETs for high frequency SSTCs
Dr. Slack, Wed Nov 09 2016, 01:50PM
@Uspring - running on the secondary resonance, but with a non-resonant primary. Moving the coupled section and resonant section apart to make a magnifier arrangement with a discrete transformer section.
Dr. Slack, Wed Nov 09 2016, 01:50PM
@Uspring - running on the secondary resonance, but with a non-resonant primary. Moving the coupled section and resonant section apart to make a magnifier arrangement with a discrete transformer section.
Print this page