Quite certain you can, especially if using liquid nitrogen 🙂
If simply keeping the output devices' temperature close to ambient , cannot see how this may affect SQ. Class A places demands on idle current, not on temperature.
Then, otoh, we all know that devices are supposed to sound better after warm up. But what does this really mean? Is it temperature?
If simply keeping the output devices' temperature close to ambient , cannot see how this may affect SQ. Class A places demands on idle current, not on temperature.
Then, otoh, we all know that devices are supposed to sound better after warm up. But what does this really mean? Is it temperature?
Last edited:
If liquid helium is used, wife will have sweet little voice...
No! You must only use gaseous helium on wife, never liquid. Safety first 🙂
I vaguely recall Grey's watercooled Aleph, and that he had issues with it 'biasing up' as he says. Maybe the cooler you can get it, the more current you have to send through it....? I've experimented with water cooling, seemed too much of a hassle, as it was a temporary setup, and the wife wasn't too pleased about water lines all over the basement. But it seemed to work well, and sound great, albeit on one channel. Now if you're talking about using fans, that can obviously have an effect on the sound, due to fan noise, but there are ways around that.
I already installed the fans, without the fan, the heat sink temp is 50c, now, the heatsink is just warm to the touch, just like a class D amp 🙂 I used four 4 inch CPU fans, virtually no noise as each one is relatively weak, but using four is enough to cool the amp to Class D levels.
No, she'll have the same voice she always has when you spend $5,000 without telling her.
Forced air is the way to go.
No need to buy expensive noctua fans. Regular fans run on lower voltage yield to quiet running. Trick is to find low voltage to be reliable in start and stop.
I purchased a dozen of fans for less than $2 each on pe sale. They are designed for 24 volts dc, but will happily run on 12 and provide plenty of quiet air.
I did the same thing with some 12V inexpensive radio shack fans force cooling my F2 monoblocks.
I built a small variable voltage regulator I ended up dialing in between 6 and 7V.
It was very quiet and cooled the amps very well for about 6 months. Then the fans started chattering and making noise.
Bad luck?
I ended up investing in the Noctua fans as I didn’t want to have to keep taking the amp apart to replace fans (the DIY case I designed makes it difficult).
The Noctua are so quiet I have to put my ear against the amp to hear them and make sure they are working.
Just my experience.
I built a small variable voltage regulator I ended up dialing in between 6 and 7V.
It was very quiet and cooled the amps very well for about 6 months. Then the fans started chattering and making noise.
Bad luck?
I ended up investing in the Noctua fans as I didn’t want to have to keep taking the amp apart to replace fans (the DIY case I designed makes it difficult).
The Noctua are so quiet I have to put my ear against the amp to hear them and make sure they are working.
Just my experience.
My only concern about forced cooling is that you have to depend on the fans functioning properly in order to cool the amp properly. If a fan fails, it may be a good idea to have a temperature switch or something.
On one hand, it would be a great way to save on chassis cost as you could use smaller heatsinks, mount them inside the chassis so they don't have to be black anodized etc. All of the tricks that you see with pro amps. Fans, like the larger Noctua fans, are so quiet that I don't think you would be able to hear a difference between passive and active cooling in your seating position.
On one hand, it would be a great way to save on chassis cost as you could use smaller heatsinks, mount them inside the chassis so they don't have to be black anodized etc. All of the tricks that you see with pro amps. Fans, like the larger Noctua fans, are so quiet that I don't think you would be able to hear a difference between passive and active cooling in your seating position.
Are many of you using a big fan 'onto' the fins of the heatsinks as opposed to small ones blowing along the fins.
Obviously along the fins is how the airflow should go but only gives opportunity for small fans. Is a bigger fan blowing onto the fins more efficient?
Obviously along the fins is how the airflow should go but only gives opportunity for small fans. Is a bigger fan blowing onto the fins more efficient?
plenty of forced cooled gadgets are having exactly that arrangement - frontally situated fans
practically parallel to heatsink base plate
inverters, PA amps, cell phone tower amps, transmitters of all sorts
in the end, you'll end up with solution most suited for you, taking care both about resulting noise and aesthetics
speaking of that , I did found that little help is, in my case(s), best compromise
and I was even surprised how effective is
https://www.diyaudio.com/community/threads/babysitter-for-papas-koan.186554/
practically parallel to heatsink base plate
inverters, PA amps, cell phone tower amps, transmitters of all sorts
in the end, you'll end up with solution most suited for you, taking care both about resulting noise and aesthetics
speaking of that , I did found that little help is, in my case(s), best compromise
and I was even surprised how effective is
https://www.diyaudio.com/community/threads/babysitter-for-papas-koan.186554/
Have not tried blowing into the heatsinks, but mine is blown into the amplifier, the heatsinks are externally located and the heatsinks are now just warm to the touch.
I was initially planning to point it into the heatsinks, but figured, the transformer and smaller capacitors inside the amp would also benefit from a fan.
Each fan is very weak, so there is virtually no noise, even when 4 is on.
Attachments
Last edited:
Is that an old Threshold amp? I had a thought of having active cooling while having a standard pass looking amp. Although not as simple. Put holes between the fins (not on your threshold though) and presurrize the chassis with the fans. some of the air will go out other places, but some will go through the heatsinks. Otherwise you could just mount small heatsinks on the large heatsinks inside the chassis, little heatinks on the mosfets etc.
I recently completed a repackaging of a SIT3X https://www.diyaudio.com/community/threads/the-sit-3x-amplifier.353999/ in a diyAudio Store 300mm 4U chassis.
With convection cooling of the heatsinks then amplifier runs too hot, having a temperature rise of about 37C at the desired rail voltage (-70V) and bias current 1.4A which dissipated 98W into each heatsink. The inside is like an oven. The heatsink temperature rise is .38C/W
I did some fan cooling experiments today:
The first experiment used a pair of Noctua NF-S12A FLX fans running at full speed. The top panel of the chassis was removed and a the fans were mounted to a thin fiberboard panel as shown with upward airflow as shown in the first image below. The result was a heatsink temperature rise is .29C/W.
The second experiment used the same fans with the airflow down into the chassis and a simple "duct work" to direct the air outward and up through the heatsink fins as shown in the drawing and the remaining images. The result was
a heatsink temperature rise is .20C/W.
With convection cooling of the heatsinks then amplifier runs too hot, having a temperature rise of about 37C at the desired rail voltage (-70V) and bias current 1.4A which dissipated 98W into each heatsink. The inside is like an oven. The heatsink temperature rise is .38C/W
I did some fan cooling experiments today:
The first experiment used a pair of Noctua NF-S12A FLX fans running at full speed. The top panel of the chassis was removed and a the fans were mounted to a thin fiberboard panel as shown with upward airflow as shown in the first image below. The result was a heatsink temperature rise is .29C/W.
The second experiment used the same fans with the airflow down into the chassis and a simple "duct work" to direct the air outward and up through the heatsink fins as shown in the drawing and the remaining images. The result was
a heatsink temperature rise is .20C/W.
