If mine fails... I found this 1
Hi Michael!
Do you care to say if this is appropiate.
Nearly finished with my build - but have to travel to Scotland for a quick job(communication on Saipem7000). But I am waiting for a couple 1 ohm res - soon to come. Patient-patient.
But if my HeatSink for the SIT(found on E-bay) fail - I have spotted one that should fit in box and maybe in spec. On the Elfa-Distrelec- it is specd: Termisk resistans = 0.5 °C/W. Included is a picture of the HeatSink.
I have received a couple spare SK82 - so I`m safe if the first blow. And I didn`t know where to get the proper sink earlier.
So Michael - if you could help if my first SK82 goes up in smoke.
Best to all - and thanks for all support this far in my build.
Olav
Hi Michael!
Do you care to say if this is appropiate.
Nearly finished with my build - but have to travel to Scotland for a quick job(communication on Saipem7000). But I am waiting for a couple 1 ohm res - soon to come. Patient-patient.
But if my HeatSink for the SIT(found on E-bay) fail - I have spotted one that should fit in box and maybe in spec. On the Elfa-Distrelec- it is specd: Termisk resistans = 0.5 °C/W. Included is a picture of the HeatSink.
I have received a couple spare SK82 - so I`m safe if the first blow. And I didn`t know where to get the proper sink earlier.
So Michael - if you could help if my first SK82 goes up in smoke.
Best to all - and thanks for all support this far in my build.
Olav
Attachments
Heatsinks Revisited. I'm wondering if the Wakefield 433K would work.
433K - WAKEFIELD SOLUTIONS - Heat Sink | Newark/element14 Canada
It seems to have similar thermal efficiency to the aforementioned Aavid D180-20.
D180-20 - AAVID THERMALLOY - HEAT SINK, T SLOTS | Newark/element14 Canada
Regards,
Dan
Looks Like surplussales has a good deal on the 486K. That's gotta be big enough!
486K - WAKEFIELD SOLUTIONS - Heat Sink | Newark/element14 Canada
Good News
Hi Juanitox and thanks for positve remarks!
Yes - I think I will run them(mono-amps)with auto-bias always- AND soon order 2 - 100000 mF to mount across the source-resistor together w 10 mf polyprops.
So - the danger is bigger if you run without sourceresistor - lucky me who went for the autobias.
Today I tried the two 400watt Antek and they delivered accurately 35 Volt AC - which should give approximately 50 Volt DC.
Really looking forward to that moment - estimated time till my build is finished is 3-4 hours ++!
Best
Olav
Hi Juanitox and thanks for positve remarks!
Yes - I think I will run them(mono-amps)with auto-bias always- AND soon order 2 - 100000 mF to mount across the source-resistor together w 10 mf polyprops.
So - the danger is bigger if you run without sourceresistor - lucky me who went for the autobias.
Today I tried the two 400watt Antek and they delivered accurately 35 Volt AC - which should give approximately 50 Volt DC.
Really looking forward to that moment - estimated time till my build is finished is 3-4 hours ++!
Best
Olav
Here is a similar version:
http://www.diyaudio.com/forums/solid-state/6338-my-first-ever-class-amp-21.html
@Olav
.5C/W is good. The heatsink pictured looks reasonable by calibrated eyeball.
@dantwomey
The 486 is what I used in the original. It's plenty.
@All
The SITs are pretty robust current-wise, but running them too hot isn't a good idea long-term. In fact, the cooler the better, IMO. Remember, we're already pushing it on the Max Pd. I never hesitate to oversize my heat sinks a bit when possible. Also, do a really good job with the device mounting, clean the surfaces, and use generous amounts of goop. The heat will also affect the gate leakage and perhaps upset your bias. Thermal stuff is often ignored by us DIYers; the more amps you build, the more you'll pay attention to those matters.
.5C/W is good. The heatsink pictured looks reasonable by calibrated eyeball.
@dantwomey
The 486 is what I used in the original. It's plenty.
@All
The SITs are pretty robust current-wise, but running them too hot isn't a good idea long-term. In fact, the cooler the better, IMO. Remember, we're already pushing it on the Max Pd. I never hesitate to oversize my heat sinks a bit when possible. Also, do a really good job with the device mounting, clean the surfaces, and use generous amounts of goop. The heat will also affect the gate leakage and perhaps upset your bias. Thermal stuff is often ignored by us DIYers; the more amps you build, the more you'll pay attention to those matters.
Last edited:
I second this!...... even without a source resistor and forgotten battery this SIT survived half an hour of good music at 3-4 A. It was my luck that the sinks were big.....!
At any case Good Work Sony!
Bench Supply
Maybe a bit OT, maybe not given this kind of amp's power supply range. There are $200 double and $130 single linear high current regulated bench supplies about. I have a double 30V 5A unit for series linking up to 60V high or up to +/-30V symmetric. Also because I could find one of the fan-less variety in doubles. Maybe good for checking out such amps fast, listening, and then deciding to keep and build them their own costly supplies. And for a number of other projects of course. For Matching MOSFETS, SiC JFETS, SITS, also. But don't use it as a circuit's CCS, you remember it did not work correctly for member Protos amp. It takes to fit an NTC primary side not to trip 10A breakers (I put one there). That trafo is 500VA easily. Construction is so-so, but OK for the price. Its meters look set true. Setting the scope's trigger for single shot, I captured the power on event. Looks smooth, no over voltage or ringing. That one's a 5V/Div 100mS/Div ramp up.
Maybe a bit OT, maybe not given this kind of amp's power supply range. There are $200 double and $130 single linear high current regulated bench supplies about. I have a double 30V 5A unit for series linking up to 60V high or up to +/-30V symmetric. Also because I could find one of the fan-less variety in doubles. Maybe good for checking out such amps fast, listening, and then deciding to keep and build them their own costly supplies. And for a number of other projects of course. For Matching MOSFETS, SiC JFETS, SITS, also. But don't use it as a circuit's CCS, you remember it did not work correctly for member Protos amp. It takes to fit an NTC primary side not to trip 10A breakers (I put one there). That trafo is 500VA easily. Construction is so-so, but OK for the price. Its meters look set true. Setting the scope's trigger for single shot, I captured the power on event. Looks smooth, no over voltage or ringing. That one's a 5V/Div 100mS/Div ramp up.
Attachments
hi michael , today i decide to try negative bias with no source degeneration , first i make a big error in wiring and turn on my sit with no bias and a great 5amp on one side .. rewired turn on and Ouf! my 2sk82 is not dead .. i now got a great 22V and 2amp on each side.. for the sound i'v got a mixed feelling . first the gain of the amp is far bigger , low level details seems better now ,sounds cleaner too. i found it to be very dependant of the 10uf input cap , with a epcos MKTS it sounds very fast and a little dry sometimes, and with a Clarity cap Esa it sounds more refined and warm. do you think your hand made inductors will sounds different than the Hammond too ?
.. rewired turn on and Ouf! my 2sk82 is not dead .. i now got a great 22V and 2amp on each side.. for the sound i'v got a mixed feelling . first the gain of the amp is far bigger , low level details seems better now ,sounds cleaner too. i found it to be very dependant of the 10uf input cap , with a epcos MKTS it sounds very fast and a little dry sometimes, and with a Clarity cap Esa it sounds more refined and warm. do you think your hand made inductors will sounds different than the Hammond too ?
Cheap array full Chinese, but decent help if you know how to use well. Agilent and Tek in next life.
Handy. Do you link the middle +/- or just the series button mode takes it well with 1-2A+?
Edsyn iron in mid photo, sweet.