Regarding heat sinks and cooling: It seems from Nelson's and other measurements that the heat sinks as is are dispersing ~12 W (16 V * 1.5A / 2 heatsinks) each at a Delta T of 50°C (or K) (75°C heatsink temp minus 25°C room temperature), so about 4K/W or, since they are about 200 cm^2 or 0.02m^2 (from my measurement of the fin lengths), about 12 W/m^2/K without any forced air flow.
Looking at
with a small fan and air at a few m/s you are not going to gain a factor 5, but more like a factor 2. But that still gets you a significant gain (factor 5?) in transistor lifetime, and also brings the heatsink temperature below the pain threshold, to maybe 50°C.
This also seems to be roughly in agreement with this graph
from the heatsink datasheet, at the Mouser link @birdbox provided, where 12 W corresponds to a Delta T of ~45°C (rising dashed curve and lower and left scales), and 200 ft/minute = 1 m/s air speed gets you 2.5°C/W (falling dashed curve and upper and right scales), so 30°C above ambient, or 55°C total, for 12 W.
TLDR for my last post (the edit interval on this forum is too short): a small fan or double stacking the heat sinks should get you into the 'safe to touch' regime, and significantly extend the lifetime of the output transistors.
If you put the amps in a case, and the metal side walls are larger than 8" by 4" (200 cm^2 single surface area), you automatically gain a factor two in heat dissipation by bolting the transistors to the walls.
If you put the amps in a case, and the metal side walls are larger than 8" by 4" (200 cm^2 single surface area), you automatically gain a factor two in heat dissipation by bolting the transistors to the walls.
Last edited:
The heat sinks on this amp is looking lot smaller than aca amp. But both amps have similar output mosfets and bias currents .. so do we need to use bigger heatsinks here , at least of the sizes used for aca amp ..
I'm not an electronics designer, but I think running parts at 80°C surface temperature is ok if you are not expecting half a century of lifetime. And actually it still might last half a century.
Regarding safety: people (including me) run tube amps with exposed tubes at significantly higher temperatures. Candles, stoves and fire places are also a lot hotter, but might give you more of a visual warning. Maybe put 'Caution Hot' stickers on the heat sinks?
Also, as Nelson pointed out, maybe you actually want to run at lower bias current and higher 2nd order harmonics?!
I really like the design; remember that part of the design idea (as I understand it) is simplicity and reduction to the essential.
Regarding safety: people (including me) run tube amps with exposed tubes at significantly higher temperatures. Candles, stoves and fire places are also a lot hotter, but might give you more of a visual warning. Maybe put 'Caution Hot' stickers on the heat sinks?
Also, as Nelson pointed out, maybe you actually want to run at lower bias current and higher 2nd order harmonics?!
I really like the design; remember that part of the design idea (as I understand it) is simplicity and reduction to the essential.
Last edited:
I’m with MJ. Build it as designed first. See how it performs with one’s speakers/systems. Then tempt fate with FAB experimentations. 
The heat sinks used on Zenductor II are the same size as the ACA Mini. In fact, it's the very same part as far as I can tell.
If you have little kids and want to keep the temps down a little, I don't think there's anything wrong with stacking on a little more heat sink instead of backing off on the bias for risk mitigation.
ACA Mini on left, Zenductor II on right.
Attachments
Last edited:
Do I have this right? 4 heatsinks on the 2-channel amplifier board, and 2 heatsinks on the 1-channel "monoblock" board . . . ??
Yep, that's a very fair point Mark. I missed that and removed that comment above. I don't see what the issue of stacking a little extra is though. Is that a bad thing to do? I got the impression it should not change anything sound wise. Is the smarter option to just back off on the bias?
Stacking the heatsinks is a completely fine idea. Will be curious to know how much difference it makes. (It may be notable)
Same Cinemag as elsewhere, or the Jensen equivalent, as I recall 4 times 600 ohm.
No surprise. That's why you're the MIGHTY ZM...Finally, after all these years of ZM preaching ........... Greedy Boyz going all Iron
😎
The CMOL-4x600T2? (https://cinemag.biz/splitter-xfmr/PDF/CMOL-4x600T2.pdf)
Edit: seems to be different from the CMOQ-4 that @6L6 mentions above. So probably not this one. I can find a Cinemag CMOQ-4LPC (not sure what the L means, but PC for PC board mount I assume) for 60 bucks.
'L' = low (50%) Nickel, 'H' = high (80%) Nickel?
Edit: seems to be different from the CMOQ-4 that @6L6 mentions above. So probably not this one. I can find a Cinemag CMOQ-4LPC (not sure what the L means, but PC for PC board mount I assume) for 60 bucks.
'L' = low (50%) Nickel, 'H' = high (80%) Nickel?
Last edited:
I played with 12 to 24v. If you have a 16 ohm load then 19v. 8 ohm 16v...
I recall posting the article in the BAF and PL forums, with the schematic. If you are unable to find them I'll post them here as well.