Why don't we see something like this more often? The driver stage is replaced by source followers. The schematic specifies LatFETs (which would stabilize the bias current), but it seems it was really implemented wit VFETs.
Help with replacement fets 2sk1529/2sj200
The VAS must drive the hefty input capacitance, but other than that, pretty neat. You get infinite current gain, so no nonlinear load on the VAS when beta of the output devices droops.
Help with replacement fets 2sk1529/2sj200
The VAS must drive the hefty input capacitance, but other than that, pretty neat. You get infinite current gain, so no nonlinear load on the VAS when beta of the output devices droops.
Last edited:
testing on a amp, Litchstark 780, IRF610 , IRF9610 driving 2SC5200, 2SA1943 works really really well. should be able to drive 4-8 pairs as well.
Last edited:
Member
Joined 2009
Paid Member
Well, for source follower the input capacitance is not too bad, it’s not the main issue, the main issue is when using common vertical FETs which because of their turn on voltage reduce overall efficiency as they can not swing so close to the rails. One solution is to bootstrap them.
I've used this topology in the Amp I finished last year. 2sk1530/2SJ201 driving 7 pairs of Sanken LAPT bjt's.
Circuit:
https://www.diyaudio.com/forums/sol...low-nfb-fet-front-bjt-ops-13.html#post5916541
Circuit:
https://www.diyaudio.com/forums/sol...low-nfb-fet-front-bjt-ops-13.html#post5916541
I built a 400 watt regulator once using an opamp and a IRF511 to drive 4x 2N3055's.
Had a lot of fun with it!!!
I could turn any spool of wire I had into a giant electro-magnet with it !! 😀
And it was variable !! 😉
FWIW 🙂
jer 🙂
Had a lot of fun with it!!!
I could turn any spool of wire I had into a giant electro-magnet with it !! 😀
And it was variable !! 😉
FWIW 🙂
jer 🙂
Well, for source follower the input capacitance is not too bad, it’s not the main issue, the main issue is when using common vertical FETs which because of their turn on voltage reduce overall efficiency as they can not swing so close to the rails. One solution is to bootstrap them.
You get the GS voltage drop no matter whether you use MOSFETs in the output or driver stage. Using them in the driver stage might be easier because you have lower current to deal with and you could use higher rails. You could also use LatFETs which have pretty low VGS.
However, the capacitance (actually both CGS and CGD) might still be an issue. Due to the low transconductance compared to a BJT, you get much more voltage swing at the gate and less bootstrapping of CGS. At the same time, the FETs in the example linked in post #1 are really output devices and have much, much higher capacitances than a classic BJT driver. Will using a small MOSFET help? Lower capacitance, but also lower transconductance?
Since the drivers and OPS are operating as voltage followers, won't the voltage swing be the same at VAS output no matter what devices are in the OPS?You get the GS voltage drop no matter whether you use MOSFETs in the output or driver stage. Using them in the driver stage might be easier because you have lower current to deal with and you could use higher rails. You could also use LatFETs which have pretty low VGS.
However, the capacitance (actually both CGS and CGD) might still be an issue. Due to the low transconductance compared to a BJT, you get much more voltage swing at the gate and less bootstrapping of CGS. At the same time, the FETs in the example linked in post #1 are really output devices and have much, much higher capacitances than a classic BJT driver. Will using a small MOSFET help? Lower capacitance, but also lower transconductance?
No, because MOSFETs have a gain much smaller than unity.Since the drivers and OPS are operating as voltage followers, won't the voltage swing be the same at VAS output no matter what devices are in the OPS?
You're right, I wasnt thinking it all the way through, with the voltage divider between 1/gm and load.No, because MOSFETs have a gain much smaller than unity.
- Home
- Amplifiers
- Solid State
- output stage MOSFETs driving BJT outputs