sub for 2sj50/2sk135
- Thread starter sasha
- Start date
Lateral FETs are expensive, USD$6.81 in 20+ is about the best you can do.
http://www.profusionplc.com/cgi-bin/gex/pcatdtl?ipartno=ECF10P20
http://www.profusionplc.com/cgi-bin/gex/pcatdtl?ipartno=ECF10N20
http://www.profusionplc.com/cgi-bin/gex/pcatdtl?ipartno=ECF10P20
http://www.profusionplc.com/cgi-bin/gex/pcatdtl?ipartno=ECF10N20
I am planing to make Acoustat TNT 200 clone. That amp is specifically
designed for driving electrostatics.
Anyway, after driver board is done I will try to experiment
with cheap Mosfets and include source resistors in design.
If it explodes no big deal, it is only couple of dollars.
You said bias problems. What about it? It won't be stable or what?
Regards.
designed for driving electrostatics.
Anyway, after driver board is done I will try to experiment
with cheap Mosfets and include source resistors in design.
If it explodes no big deal, it is only couple of dollars.
You said bias problems. What about it? It won't be stable or what?
Regards.
Years ago I used a TNT-200 for 5 years with a pair of Acoustat IIa speakers. It is a powerhouse. Did you literally clone this amp? Did you find a source or replacements for the VN01, VP01, etc., devices?
If you want 12 MOSFETs for 65 VDC rails I have them:
http://www.diyaudio.com/forums/showthread.php?postid=660932#post660932
Good luck. . .
If you want 12 MOSFETs for 65 VDC rails I have them:
http://www.diyaudio.com/forums/showthread.php?postid=660932#post660932
Good luck. . .
2SJ50, being a lateral MOSFET, has zero temperature coefficient at ~80mA, whereas IRFP240 has zero tempco at over 6A. Thus IRFP240 will probably require some form of thermal compensation for stable bias.sasha said:
Furthermore, Vgs is higher for vertical MOSFETs: about 0.4V for 2SJ50 vs. 4V for IRFP240.
For a TNT 200 clone, the SuperTex MOSFETs are tough to find nowdays.
Hafler eventually absorbed Accoustat.
Jim Strickland, who designed the TNT200, refined the concept Trans Nova concept and applied it to Hafler models 9505, 9303, and perhaps some others.
The service manuals (which contain schematics!) may still be available at the Hafler web site - I saw schematics there years ago.
The newer designs are more easily duplicatable with currently available parts.
But they still used lateral MOSFETs at the output. I think you'd be asking for trouble (bias control, larger capacitance, no true complementary pairs available, etc.) trying to use vertical MOSFETs there. But who am I to say ... You might discover something, so have at it.
I've heard the 9505 driving some Soundlab Pristines - very nice
I'm actually surprised more people aren't cloning these designs ....
mlloyd1
Hafler eventually absorbed Accoustat.
Jim Strickland, who designed the TNT200, refined the concept Trans Nova concept and applied it to Hafler models 9505, 9303, and perhaps some others.
The service manuals (which contain schematics!) may still be available at the Hafler web site - I saw schematics there years ago.
The newer designs are more easily duplicatable with currently available parts.
But they still used lateral MOSFETs at the output. I think you'd be asking for trouble (bias control, larger capacitance, no true complementary pairs available, etc.) trying to use vertical MOSFETs there. But who am I to say ... You might discover something, so have at it.
I've heard the 9505 driving some Soundlab Pristines - very nice
I'm actually surprised more people aren't cloning these designs ....
mlloyd1
sasha said:
Thanks for posting the link to the Hafler 9500 schematic. Any comments about how it sounds?
Years ago, after I sold my Acoustat IIa speakers, I kept the TNT-200 to use with a pair of speakers with dynamic drivers. The TNT-200 seemed to produce more than the input signal provided. I think it had some distortion products that made the Acoustat speakers sound more "spacious" but only confused the sound with dynamic drivers. At least to me at that time with the TNT-200 I had.
I'll bet the 9500 is clean as a hound's tooth. It does show the design influence of Bob Strickland. I had a chance to talk with him at a Chicago CES years ago. He was quite laid back, even modest, but made interesting comments about amplifier design and use of MOSFETs.
I would never clone a TNT-200 but the 9500 looks like it has great promise.
Years ago, after I sold my Acoustat IIa speakers, I kept the TNT-200 to use with a pair of speakers with dynamic drivers. The TNT-200 seemed to produce more than the input signal provided. I think it had some distortion products that made the Acoustat speakers sound more "spacious" but only confused the sound with dynamic drivers. At least to me at that time with the TNT-200 I had.
I'll bet the 9500 is clean as a hound's tooth. It does show the design influence of Bob Strickland. I had a chance to talk with him at a Chicago CES years ago. He was quite laid back, even modest, but made interesting comments about amplifier design and use of MOSFETs.
I would never clone a TNT-200 but the 9500 looks like it has great promise.
You'll be hard pressed to get Audio Power MOS-FETs for a few dollars each. In the UK they normally retail for around £7.00 EACH (ie $24.00 for a pair).
However, any complimentary 160V 7A 100W pair will do. Try looking for RFP10P16 / RFP10N16 or BUZ900 / BUZ905 which are still current devices.
If all else fails the 2SK135 / 2SJ50 pairs are still available in the U.K. at £7.00 each.
However, any complimentary 160V 7A 100W pair will do. Try looking for RFP10P16 / RFP10N16 or BUZ900 / BUZ905 which are still current devices.
If all else fails the 2SK135 / 2SJ50 pairs are still available in the U.K. at £7.00 each.
