F5 with 2SK2013/2SJ313

[jacco vermeulen]

Originally Posted by GuidoR

don't think

400mA times 24V is 9.6W
Means the devices would heat up to >80C + temperature rise of the heatsink. (22.5C ambient)

Keeping the dies under 100C would require a 2C/W heatsink for each output device.
Check for yourself how large the heatsink would be.
(or a 0.33C/W heatsink for 6 devices)

[Tea-Bag]

Maybe a 12 or 15vac tranny would do well if you don't want a regulated supply. Just a SWAG.

[massimo]

Juma, as described in the Serbian forum, used a Fischer SK 56 (30 x 15 x 4 cm) heatsink and reported a temp rise of 21°C.
That heatsink will suffice to dissipate the 9.6W x 6 without becoming too hot (>50-55°C)

[jacco vermeulen]

Means a minimum die temperature of 94C at 25C ambient.
Raise the dissipation to 9.6W and die temperature will reach 110C !
(me head count, do the exact math yourself)

[Zen Mod]

[juma]

As Jacco noticed, power dissipation was my primary concern when I decided to use cap multiplers to lower the PS voltage to +/-20V DC (the transformer I used is 2x19V AC). Not that these MOSFETs are that fragile, but I don't like running things on the edge - reliability is my fetish, I drive Japanese cars only... Soon I noticed that amp sounds better with cap multiplers so I kept them.

[GuidoR]

Originally Posted by juma

amp sounds better with cap multipliers

Juma, you convinced me, thanks.

Jacco, as you quoted, I don't think, while you've done the math

[Samuel Jayaraj]

I've checked RS, Farnell and Mouser, but I can't find the output Mosfets. Any idea about their availability? Can they be substituted with K1058 and J162? Thanks.

[Zen Mod]

ask Zhou Fang

[juma]

Originally Posted by Samuel Jayaraj

...Can they be substituted with K1058 and J162?...

It will work, but it won't be the same amp...