diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Pass Labs (http://www.diyaudio.com/forums/pass-labs/)
-   -   2SK82 + Lightbulbs = Fun (http://www.diyaudio.com/forums/pass-labs/192665-2sk82-lightbulbs-fun.html)

Michael Rothacher 15th July 2011 04:25 PM

2SK82 + Lightbulbs = Fun
 
1 Attachment(s)
This is my first go at a 2SK82 lightbulb amp. :)

Nelson, are you willing to sanity-check my graph before I embarrass myself too badly?

Thank you,

Mike

Nelson Pass 15th July 2011 04:47 PM

1 Attachment(s)
That's a lot better than I got:

:cool:

Nelson Pass 15th July 2011 05:18 PM

Probably I should add that I was concerned about the dissipation. As far
as I can tell, Sony didn't offer curves showing instantaneous dissipation
beyond about 80-90 watts, and they had a similar chip rated at 65 watts.

Out of concern for reliability, I didn't push it beyond 1/2 that.

If I read your data correctly, you are at about 100 watts dissipation on yours,
and I would certainly expect better numbers from that, but in addition I think
you've found the "sweet spot".

:cool:

Michael Rothacher 15th July 2011 06:49 PM

1 Attachment(s)
I don't have the output centered. I'm at 17.5 Vds for around 50W dissapation. The rest is across the bulbs, and I have a little negative supply for the bias.

A little further tweakery reduced the 1W THD still further. I think that's about as sweet as I can dial-up. I'm going to build another one and give them a listen.

Thank you, Neslon. I'm encouraged. This sweet spot stuff is pretty fun.

Mike

P.S. Does the low (2X) voltage gain figure seem correct? Also, I believe my samples have the drain connected to the body as the other way around had no gain and worse distortion.

Nelson Pass 15th July 2011 07:46 PM

1 Attachment(s)
That might explain something, although I used them with 2SJ28's in
complementary followers with case = Source and got the following curves:

:cool:

Nelson Pass 15th July 2011 07:48 PM

Also, my single-ended result had 3 dB of gain. I might note that I expect
300 light bulbs to come out around 22 ohms. With 2 in parallel, that would
be 11 ohms, so something's off.

:cool:

Nelson Pass 15th July 2011 08:01 PM

Breaking news: the 2SK82 was rated at 95 watts (25 deg C case temp).

Another similar Sony part does have the case = Drain. I'll have to check
my results when I get back from the coast.

:cool:

Michael Rothacher 15th July 2011 08:30 PM

Quote:

Originally Posted by Nelson Pass (Post 2640168)
Breaking news: the 2SK82 was rated at 95 watts (25 deg C case temp).

Quote:

Originally Posted by Nelson Pass (Post 2640155)
Also, my single-ended result had 3 dB of gain.

Good news there!

Quote:

Originally Posted by Nelson Pass (Post 2640155)
I might note that I expect 300 light bulbs to come out around 22 ohms. With 2 in parallel, that would be 11 ohms, so something's off.

I'll check all that and try another device to be sure. I have my gates set at -3V.

Thank you. Have fun at the coast!

Mike

Nelson Pass 15th July 2011 08:42 PM

1 Attachment(s)
I pieced together a 2SK82 curve and compared it to my notebook values
when I tested the parts. It's not a very good match. I'm going to assume
that case=Drain for the time being. Meanwhile, here's that curve which is
good:

:cool:

sheep 15th July 2011 08:59 PM

I read that the 2SK82 has 2 dies and they have to match or one will die!
Did you get them from the Singapore seller on Ebay?


All times are GMT. The time now is 02:24 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2