A tibute to the ACA & the ACA mini?

Jump to Latest
#21
Of course you can get Class A in a small chassis. You just don't get much power - or you get a screaming hot chassis. The itty-bitty heat sink on the Schiit Rekkr may be good for a few watt. Maybe 1 Wpc in Class A. But then again, nobody's claiming the Rekkr is Class A. Not that I can tell anyway.

"Exotic, fully discrete, ..." Ooohhh!!! Ahhhh!!! This is Stoddard at his best.

Tom
 
#25
jjasniew said:
Many years ago, I tried to introduce water cooling to these physically tiny transient loads we used to test power supplies. By mounting a copper, fluid filled pipe very close to the FETs thermal tab, you could use it right up to SOA limits all day - as long as the water flowed.

But, they didnt like "water" on an electrical test bench (testing voltages in the single-digit range) so they s-canned my idea, put a tiny little HS on the device and added thermal shutdown protection. It was still usable in transient operation; a transient current device that was now only usable transiently...Better score that measurement quick!

The cooler / pumps were a common design for CPU cooling and they were way overkill. To be usable in a HiFi amplifier, fluid flow would have to be of course enabled silently. But if you could do it, the amp could be the size of a mint tin, or the current size of this product but with more WPC, harder bias - whatever this one cannot do because of the HS limitation.

I say if the computer gamers are as enthusiastic to drive a market for these water cooling systems, why not "Us"? If the coffee guys are enthusiastic enough to have a direct cold water line to their espresso machine... I say that because we do not know how much water it would take to cool during a 3 hour listening session. 1 or 10 gallons? It'd be simple to figure out empirically.
Click to expand...
An amp modelled on a `hit and miss' agricultural gas engine. What an awesome concept.
Or, maybe mount a Stirling engine driven fan on the heatsink... very steampunky. the ultimate mechanical/thermal governor
 
#26
andynor said:
Why on earth would you even want A/B on such sensitive drivers?

Class A/B is for weaklingsezzz.
Click to expand...

andynor - what don't you like with the performance?

Most class A designs seems to be tied to the "no feedback" etc. paradigm with quite poor clinical performance.

I want a clinical sound 🙂 i.e. no sound... I like the spec and measurements of the Rekkr.

I challenge all the talented designers here to come up with a competitive bid for a small amp... I would if I could... but don't have the skills...

//
 
#27
TNT said:
Most class A designs seems to be tied to the "no feedback" etc. paradigm with quite poor clinical performance.

I want a clinical sound 🙂 i.e. no sound... I like the spec and measurements of the Rekkr.

I challenge all the talented designers here to come up with a competitive bid for a small amp... I would if I could... but don't have the skills...
Click to expand...
Maybe you could reach those spec with an F5 running on lower voltage and lower bias with more feedback. There is feedback in a lot of the class A designs you can see around here. I don’t know if you have ever heard a properly assembled F5 but it is quite impressive and neutral, or maybe clinical as you said.
 
#28
tomchr said:
Of course you can get Class A in a small chassis. You just don't get much power - or you get a screaming hot chassis. The itty-bitty heat sink on the Schiit Rekkr may be good for a few watt. Maybe 1 Wpc in Class A. But then again, nobody's claiming the Rekkr is Class A. Not that I can tell anyway.

"Exotic, fully discrete, ..." Ooohhh!!! Ahhhh!!! This is Stoddard at his best.

Tom
Click to expand...

From what I've found so far... It's an AB design with a capacitor/diode voltage doubler. Low bias and heat dissipation, with a large-ish music peak ability and good specifications - hence the miniscule heatsink and very limited air-flow.

However, it does have a very short signal path and thick&wide PCB tracks where it counts... something that I appreciate a lot in an amp design.... above anything else - measurements are important, but not at the expense of what I like to see in an amp.

My ACA monoblocks sound nice... but they do run as hot as a smelting furnace.
 
#30
jjasniew said:
Many years ago, I tried to introduce water cooling to these physically tiny transient loads we used to test power supplies. By mounting a copper, fluid filled pipe very close to the FETs thermal tab, you could use it right up to SOA limits all day - as long as the water flowed.

But, they didnt like "water" on an electrical test bench (testing voltages in the single-digit range) ...

The cooler / pumps were a common design for CPU cooling and they were way overkill. ... If the coffee guys are enthusiastic enough to have a direct cold water line to their espresso machine... I say that because we do not know how much water it would take to cool during a 3 hour listening session. 1 or 10 gallons? It'd be simple to figure out empirically.
Click to expand...

(1) My current and past expresso machines support direct piping.. that's the easy part. The hard part is the drain. So, for the time being, I use water from a water purifier and a large pitcher.

(2) What kind of coolant do they use in PC liquid cooling pipes? Could a copper pipe be sufficient to carry the heat away from the heat sinks?

(3) How about immersing the heat sinks into a non conductive liquid?

I've thought about using PC cooling designs into a Class A chassis so that it could fit inside cabinets. The coolant could be a bucket of oil, or such, as they use for loads into RF transmitters.
 
#33
TNT said:
andynor - what don't you like with the performance?

Most class A designs seems to be tied to the "no feedback" etc. paradigm with quite poor clinical performance.

I want a clinical sound 🙂 i.e. no sound... I like the spec and measurements of the Rekkr.

I challenge all the talented designers here to come up with a competitive bid for a small amp... I would if I could... but don't have the skills...

//
Click to expand...
https://www.firstwatt.com/pdf/art_leaving_class_a_2019_redux.pdf
 
  • Thank You
Reactions: pfdavid
#34
TNT said:
andynor - what don't you like with the performance?

Most class A designs seems to be tied to the "no feedback" etc. paradigm with quite poor clinical performance.

I want a clinical sound 🙂 i.e. no sound... I like the spec and measurements of the Rekkr.

I challenge all the talented designers here to come up with a competitive bid for a small amp... I would if I could... but don't have the skills...

//
Click to expand...

Class D then... vanishingly low distortions. Powered with an SMPS ... then you'll have a full package.
 
  • Like
Reactions: DorinD
#37
TNT said:
I checked shiit europe.... shipped to my home I estimate it will be corresponding to 262 USD inc. (€ -> skr -> USD)

//
Click to expand...
The Gjallarhon were not listed in the European shop on Saturday so I ordered 2 directly from US.
With taxes and delivery they will end up at around 410 euros each to FR, so just slightly less than the EU shop but not enough to bother.
Except that I just checked and if they are now listed in the EU shop, they are still not in stock and/or available for purchase.

​

 
#39
Extreme_Boky said:
Class D then... vanishingly low distortions. Powered with an SMPS ... then you'll have a full package.
Click to expand...
Or Class AB powered by an SMPS.

I don't get the fascination with Class A. All you get out of that is a lot of heat. There was a point in time where Class A was a way to get lower distortion. But a well-designed Class AB can achieve the same low distortion without all the excess heat.

Then again, the ACA is not about low distortion. As far as I understand it, it's about getting some amount of low-order harmonic distortion that some find pleasing.

Tom
 
  • Like
Reactions: DorinD
You must log in or register to reply here.
Top Bottom