advertisement


Rega Amplifiers and “Simulated Class A”?

laughingboy said:
There's a YouTube video where Michael Fremer tours the Rega factory. I am linking to Roy Gandy talking about Terry Bateman's discovery of what John Curl discovered in the 1970s (at 22:35 in the video).

If I understand correctly (I know almost nothing about electronics), he is saying that Rega used to use transformers to improve sound quality, but they now understand why that worked and are starting to use pairs of FETs in some amplifiers. But he's talking about input stages, not output stages. Does this shed any light on the matter?
Click to expand...
Roy appears to be describing a recent update to the phono stages to an American who keeps interrupting him with flippant comments.
 
Last edited:
Arkless Electronics said:
Extrapolating back from figures Robert gave in a PM it seems the Elicit R makes roughly 1WPC in class A, a very high figure for a class A/B amp and why it gets so hot. Seems they've gone for that "First Watt" thang and this can only be a good thing.
Click to expand...

Hot because each transistor is dumping just over 4W at idle, i.e. nearly 9W per side.
 
Arkless Electronics said:
Extrapolating back from figures Robert gave in a PM it seems the Elicit R makes roughly 1WPC in class A, a very high figure for a class A/B amp and why it gets so hot. Seems they've gone for that "First Watt" thang and this can only be a good thing.
Click to expand...

Agreed and nice to a company doing things a bit different to the plethora of typical A/B amps out there.

If anyone is interested the Elicit R was pulling 360ma at idle after 1 hour and clocking up 60W on the meter.
By contrast, a more typical Class A/B design in the Cyrus 8 was way down at 85ma @ 15W. (note the 8 also has a Dac board).

The current crop of Rega amplifiers are among the very few where I can perceive a positive sonic difference over the SS amplifier herd.

These are the Elicit heatsinks, one per side.

Screenshot 2021-07-13 18.50.14 by Rob Holt, on Flickr
 
Shane2468 said:
Couple sellers on ebay flogging their IO.

S.
Click to expand...

There will be more.

Everyone here egging each other on to move up the Rega ladder.

That much for the iO being a giant killer (which is an entirely unfair description to carry for a beer budget amplifier ... it just does not happen that way and was the result of a little forum frenzy here). However, it does seem to do a lot right for a made in the UK product at that level.
 
mindlagoon said:
I was poking about inside my Elicit-R recently, as it was running damn hot and cut out a few times. The output stage has +/-54V rails and 0.22R emitter resistors, which in my opinion are too low for good bias stability.
Click to expand...

Depends on how it’s designed, my class A amplifier is a push pull complementary sziklai output stage and has 0.05R resistors and it’s fine.
 
Arkless Electronics said:
More than double that in fact. Because it's giving about 1W in class A.
Click to expand...

I'm puzzled here. I measured a bias current through the emitter resistors of about 80mA, and rails of +/-54V, so that suggests 4.3W dissipation in each transistor at idle. I don't see how a complementary output stage can run in Class A, when each device is only conducting for 180 degrees. Am I missing something?
 
If you have 360 mA quiescent, you can output up to 720 mA before one of the outputs turns off.
IxIxR = 4W peak for 8 Ohm ie 2W rms
 
I'm working back from Roberts measured 60W quiescent draw with 54V rails and making some allowance for inefficiency of PSU, losses from emitter resistors, drivers etc etc and getting that to around 280mA quiescent current. Depending on various unknowns we then get around 0.9 to 1.3W or so class A hence my rough figure of 1W
 
But Robert's figures don't add up:
360mA X 240V = 86.4W

That's quite a lot more than the 60W measured.
 
S-Man said:
But Robert's figures don't add up:
360mA X 240V = 86.4W

That's quite a lot more than the 60W measured.
Click to expand...

"Power factor".... VA versus Watts etc?

At the end of the day it's pretty pedantic as to whether the unit gives 0.6 or 2W in class A... either is a lot for a class A/B amp and the reason it gets hot
 
Arkless Electronics said:
Or it could also be that the Wattmeter measures true average Watts rather than RMS Watts.
Click to expand...

No idea, it's just a generic £20 Wattmeter.
However the relative (large) differences between the readings from the Cyrus and Rega are what matter and confirm Rega's description.
 
You must log in or register to reply here.


advertisement


Back
Top