Pass Aleph 3

[martin clark]

Would it be stupid to say it takes so long to come on song because it has so much cooling area?

No, not at all - there is an awful lot of thermal mass to be brought up to temp; and while that happens- all of it is designed to reject heat efficiently (by convection.)

I think the more interesting question is to wonder why being 'warm' makes such an obvious difference - when it really, really should not. At all.

(to the extent that - if temp is the determinant - covering the thing with a teatowel, to bring it up in 10mns from a cold start should be no different to wasting 2hrs and 0.4KwH on 'warmup')



PS - This is no slight on what it may or may not be in terms of enjoyment!

 

[MUTTY1]

Now you know what do with a spare duvet cover. I recently bought a decent D amp module and thought the sound tailed off the warmer it got! But it didn’t put a thump through the speakers so progress of sorts.

Yer, we all think of our car and it’s cold oil. Doesn’t translate necessarily…

 

[Tony L]

I think my conclusion from tonight is that once its up to temperature it is up to temperature, i.e. it needs that first hour or so (probably under) to get up to 46-48c, but then its happy and does what it does. I could hear no further improvements as time went on.

PS Agree with Martin that it would be good to understand why this is the case. IIRC Pass claimed the noise and distortion was measurably lower after warm-up, but again, why? FWIW it is an exceptionally quiet amp, I can hear no hum or hiss.

 

[S-Man]

I think the more interesting question is to wonder why being 'warm' makes such an obvious difference - when it really, really should not. At all.

Capacitor ESR versus temperature?
The big ones will probably take an hour or so to reach their "operating" temp. This may well also apply to Naim stuff (non-ventilated case), to a lesser extent.

My Aleph never came on song. The caps were outside the box so were nearly at room temp. Makes me wonder!?

 

[Tony L]

My Aleph never came on song. The caps were outside the box so were nearly at room temp. Makes me wonder!?

I’d always assumed it was the mosfets that needed to hit their zone, but looking at the physical design if you wanted to build a capacitor oven this is exactly how you’d do it! When I first got it out of the box I ran it for a while with the the thick alloy lid off as I wanted to be sure all was ok in there, with the lid off the caps were barely over room temperature. Given the lid (it’s about 1cm solid alloy), when in place, gets to a couple of degrees c below the heatsinks I assume the whole internal ‘cube’ is cooking at that temp. It puzzled me that NP hadn’t put vents top and bottom the way say the 303 has to allow airflow over the boards. Never occurred to me it was deliberate!

 

[S-Man]

if you wanted to build a capacitor oven this is exactly how you’d do it!

Brilliant!

 

[martin clark]

Capacitor ESR versus temperature?
The big ones will probably take an hour or so to reach their "operating" temp. This may well also apply to Naim stuff (non-ventilated case), to a lesser extent.

My Aleph never came on song. The caps were outside the box so were nearly at room temp. Makes me wonder!?

- yes, useful temp-related drop in ESR was my first thought -except- these PSU caps are huge/good parts in the first place;even if end-of-life - with several in parallel that term surely has to disappear? Sensitivity to c. single-fig mOhm drop (when the output Z is defined order of magnitude higher...) -doesn't work for my small imagination.

But @Tony L 'capacitor oven ' that's certainly true yes!

And big caps depend to an extent on radiative (IR) loss, they run warmer if you remove the plastic sleeve (because the IR emissivity reduces). I've this documented somewhere...

I wonder if there's some useful change in g'oss / equiv-to- gm/ whatever-it-is for mosfets - I haven't dicked with them enough to really understand such things at all. But curious; and happy to learn!

 

[earlofsodbury]

I wonder if the PTC Thermistors are playing a part in the whole warm-up equation - they're not perfectly linear devices and I seem to recall NP saying he purchased a huge quantity of CL-60s years ago and puts them in everything. Others have pointed out the rating curve for them may be far from ideal for some of his designs. It's all on DIYAudio, and somewhat over my head at times...

They're a cheap and easy soft-start / inrush-limiter but I'm not keen on them myself and will probably pull them from the J2 and put a switching circuit of some sort in instead.

 

[martin clark]

Good point.

The PTCs are at least on the mains primary side, so the PTC impedance even worst -case is reduced from the amp boards ' DC-side impedance POV' by 1/[pri:sec turns ratio^ 2]

back-of-a-postit-note makes that pretty darn negligible - to my view. No worse than the secondary winding impedance,for sure, likely less; and not likely to take an hour to change in a way users obviously discriminate.


(Ive obvs new things to learn in the Pass-approach realm & that's a good thing)

 

[S-Man]

- yes, useful temp-related drop in ESR was my first thought -except- these PSU caps are huge/good parts in the first place;even if end-of-life - with several in parallel that term surely has to disappear? Sensitivity to c. single-fig mOhm drop (when the output Z is defined order of magnitude higher...) -doesn't work for my small imagination.

But @Tony L 'capacitor oven ' that's certainly true yes!

And big caps depend to an extent on radiative (IR) loss, they run warmer if you remove the plastic sleeve (because the IR emissivity reduces). I've this documented somewhere...

I wonder if there's some useful change in g'oss / equiv-to- gm/ whatever-it-is for mosfets - I haven't dicked with them enough to really understand such things at all. But curious; and happy to learn!

It may well be a combination of the FETs (not just the o/p ones) and the caps.

I agree that a few mOhms shouldn't make much difference, BUT maybe it changes the ESR versus frequency curve??
Unlike captain arse-gravy, I do hear differences between different brands of caps (with similar specs, including ESR). These come across mainly as tonal differences, and I wonder if this is due to the ESR versus frequency. BTW, I don't like this from the engineering POV, but my ears tell me I prefer some to others and I'm looking for explanations.

On the topic of radiative losses I always thought that not much happens below about 80C (which means that black cases for hifi don't cool any better than silver ones) and since it's related to T cubed it ramps up very rapidly with higher temps. Very useful with regard to the sun!

FWIW I didn't use any inrush limiting with my Aleph. Currents are halved at UK mains versus US.
NP should have done the inrush limiting with JFETs

 

[earlofsodbury]

NP should have done the inrush limiting with JFETs

 

[Tony L]

Here’s the quote on warm-up from the surprisingly wordy Aleph 3 manual:

“It takes at least an hour of warm up time to get the best performance out of the amplifier. It will take that long to reach operating temperature and exhibit lowest distortion and noise. This is not a subjective judgement, but based on actual distortion and noise measurements. You may find somewhat greater residual noise coming through the loudspeaker when the amplifier is first turned on, but it should decrease as the amplifier warms up.”

No mention as to the ‘why’ though.

Worth linking to Steve Guttenberg’s interviews with Nelson Pass. He goes into a lot of the thinking behind the Aleph and First Watt ranges which do seem to be a pretty linear progression. A genuinely interesting bloke IMO, someone who still seems to follow paths out of pure curiosity and to see what he can learn.

 

[S-Man]

I always thought noise went up with temperature.
Anyone (Martin?) know why NP would measure the opposite in the Aleph?

Also, the gm of the input pair will reduce as temperature increases - leading to more distortion, I would have thought.

 

[palasr]

but I’m not getting that magical extra thing the Leaks do that is so hard to define in words.

In my experience, this is exactly what separates vacuum tubes (valves) from solid state. There's apparently some type of magic in thermionic emissions.

 

[martin clark]

the only way I can think of offhand that noise might drop is say , reduced Esr in the warm supply caps/ quieter supply ...hang on, isnt this the door we came in through..? ; )

 

[Tony L]

FWIW now I’m getting used to it and better understand the warm-up time I’m really liking this amp with the JR149s. It is a very nice match indeed, it just sounds rather boring when cold.

It occurred to me that it likely isn’t any more powerful into the JR149s (or LS3/5As which I’ve yet to try) than the Leak TL12 Plus on its 16 Ohm tap given the 149s impedance plot:

I assume it will halve output as impedance doubles, so what I have here is a 15 Watt class A amp! The TL12 Plus are 12 or 14W depending on how much of a salesman Harold was being at the time.

PS I will in time try it with the TV system’s Spendor S3/5R as they are a more modern 6-8 Ohm load IIRC.

 

[mandryka]

FWIW now I’m getting used to it and better understand the warm-up time I’m really liking this amp with the JR149s. It is a very nice match indeed, it just sounds rather boring when cold.

I'm not sure I could live with that, the idea of having to let it warm up. I haven't noticed this with the Krell or Radford.

 

[Tony L]

I'm not sure I could live with that, the idea of having to let it warm up. I haven't noticed this with the Krell or Radford.

Agreed. I’d not want that length of warm-up, or to be honest the inefficiency, in a main system in regular use. The upstairs near-field rig is very much an ‘extra’ system that I really enjoy, but maybe twice a week or so, and even then it has options as it is in reality much more than one system. I’ll enjoy having the Pass knocking around as an option, an alternative reference point etc. I’m just not a ‘one system’ hi-fi listener these days. I went through a stage of regretting sinking the money into it, but I’m now growing to really like it and the alternative perspective and additional options and potential it brings. I’ll learn some stuff too, which is always good. It is also nice to have an amp knocking around that can deal with low impedances too, e.g. if I ever wanted to try a pair of Maggies or whatever I could drive them very well (the 303 and little Leaks would need a fainting couch!).

 

[earlofsodbury]

Over the years I've owned a fair few items that seem to benefit from warm-up, but it's never occurred to me that waiting for it is an issue, chiefly because when the day is done I'm going to be plonked in front of the hifi anyway, and I usually kick-off with some radio listening where the last few % of SQ is unimportant.

 

[mandryka]

Over the years I've owned a fair few items that seem to benefit from warm-up, but it's never occurred to me that waiting for it is an issue, chiefly because when the day is done I'm going to be plonked in front of the hifi anyway, and I usually kick-off with some radio listening where the last few % of SQ is unimportant.

Maybe I could get into the habit, like I've got into the habit of turning the system off when not in use.