F5 with 2SK2013/2SJ313
 

I got interested in Toshiba's 2SK2013/2SJ313 MOSFETs last year when estman, a forum friend, sent me generously a set of dozen j313s. They impressed me with their superior linearity of transfer characteristics but not until recently (when I got the N-channel complements - k2013) I could not try them in classical push-pull circuit.
Using them in F5 (3 pairs per channel at +/-20V, 400ma per MOSFET for total bias of 1.2A per channel) transformed it into amp that leaves no wishes open.

Schematics and pics are here:
Novi amp sa 2SK2013/2SJ313 - DIYAudio.rs

I'm sure the Mastah will also enjoy your creativity.

Can't help mentioning the 6 amp peak output current limit, 5 pairs would be costly but balance the SK56 more, which is also expensive.
(If Dorian Grey would still be around he'd likely start a lengthy itinerary about 10 pair/ch balanced TXF5. :clown: )

:Olympic: Bravo again, the dirty dozen rules the forum galaxy.

Maybe we should also try 100 pcs ZVP/ZVN3310 in parallel ....

;)

Patrick

Just like other Japanese trannies, they are actually quite cheap if you buy in sufficient quantity.
The price of this pair I get from Avnet is even quite a bit cheaper than IRFP240/9240.
So everyone should be able to easily afford a dozen unmatched pairs without dropping a sweat.

Thanks Jacco,
yes, 3 pairs per channel seems thin, but it's all I got right now ;)
Nevertheless, amp drives happily my speakers (Canton ct-1000 and Epos M5 which go down to 3.5 Ohms) and in burn-in test (24 hours at max power - 18Vpeak, driven with 10khz sine wave and loaded with RCL combo of 3.5 Ohms) temperature on MOSFETs never rose more than 26 degrees Celsius above ambient, so I think it's safe to build and use this amp.

Hi Patrick,
remember, we are still waiting for the result of "beast with 1000 JFETs" ;)

ZhouFang, Zen Mod contacted you already about these MOSFETs.... ;)

You cannot use JFETs in the F5 output stage without large changes.
But you can parallel Zetex devices without problems.

But if you really want a detailed yet punchy F5 with lots of reserves, you should try UHC Mosfets.


Patrick

After this one, I tend to even more agree with device's-linearity-is-everything philosophy, and I finally have not a slightest complaint about the amp.

I might yet build a SuSy circlotron with LU1014 out of curiosity, but fiddling with ZVP/ZVN and similar small devices is definitely not my thing ;)

:Popworm:









:cheers:

:worship: :cheers: :cool: