Control Amplifiers?

[bluesman]

On 1/25/2021 at 9:02 PM, davide256 said:

This seems to be 1 example of a "control" amplifier, minimal circuitry between inputs and amplifier stage, volume is regulated in the amplifier circuit.

If I remember correctly, that CJ is “just” a power amp with variable input attenuation.  It’s a wonderful device, as are all the CJ pieces I’ve ever heard.  But if you like your Hafler 200 and don’t want to replace it, any of those passive attenuators will give you what you want.  The leap in SQ with a CJ would probably be quite apparent though, unless your DH200 has all the mods that were popular when it first hit the market and you really love it.

 

You might consider any variant of the Axiom or even a Nobsound (which is a great little box that punches well above its $60 price class).  The upgraded Axiom was about $500 as I recall, but I’ve liked the Nobsounds I’ve heard enough to live happily with one. 

[bluesman]

29 minutes ago, davide256 said:

 

@bluesman thanks for bringing up the Axiom. Their site does raise a question, looks like when you change amplifiers you may have to order a new passive pre because of a need to match input impedance and sensitivity?

You do not need to buy a new preamp under any circumstances.  FIrst, almost all modern preamps have output impedances >20 kOhms.  Once it gets above this, an increase to any practical amount will not affect gain significantly because the marginal load on the source is already so low at 20 kOhm that even an increase to 100+ kOhm won't change gain.  Of course, there are exceptions.  This is why their website clearly says that Axiom "... can easily readjust the gain ratio for you at a cost of $35" if it becomes necessary.  But a new amplifier is not likely to make it necessary - and if it is, that $35 buys a simple minor change.

 

The source output voltage is much more likely to change significantly than the amplifier input impedance, and an occasonal amp will come along that has a much higher or lower input sensitivity than most.  These factors are more likely to make a recalibration of gain desirable / necessary.  But changing your amplifier while keeping the same sources is not going to make a difference except that you might find the preamp gain knob set a bit higher or lower for the same SPL.

[bluesman]

1 hour ago, sandyk said:

 You obviously mean Input Impedances.

What I meant to say instead of "almost all modern preamps have output impedances >20 kOhms" was that "almost all modern amplifiers have input impedances >20 kOhms".  This is supported by my next sentence in that post: "Once it gets above this, an increase to any practical amount will not affect gain significantly because the marginal load on the source is already so low at 20 kOhm that even an increase to 100+ kOhm won't change gain".  Sorry about the error - I was on my way out the door when I saw the post that precipitated this response and I was careless in entering it because I was in  a hurry (which is admittedly no excuse).

 

Almost all modern preamps have output impedances at or below 1000 Ohms and many are below 100 Ohms. So the difference in load between 20 kOhms and 100 kOhms is negligible as far as the source is concerned.  If a preamp can generate sufficient voltage to drive the latter, it can do so into the former as well.

[bluesman]

2 hours ago, sandyk said:

Actually, there are many these days that have lower input impedances to achieve an improved S/N which is needed for high resolution digital sources .

The designs from Silicon Chip magazine are fairly representative , with their Ultra-LD Mk.3  from as far back as 2011 having an Input Impedance of a little under 12kohms , and many of the designs by U.K. Amplifier Designer and Author Douglas Self having a 10K input impedance. (Audio Power Amplifier Design Handbook 5th ed - D. Self (Focal, 2009) WW)

The functional difference between amp input impedances of 10 kOhms and 20 kOhms is zero for most preamp/amp combinations.  Instead of saying "...the difference in load between 20 kOhms and 100 kOhms is negligible as far as the source is concerned", I could just as correctly have said that the difference in load between 10 kOhms and 100 kOhms is negligible with a source impedance of under 1 kOhm.  Specialty pieces may have different requirements, but for the vast majority of amps and preamps you can buy today, I'm, pretty certain that this is true and correct.

 

Neither a project amplifier in a 10 year old magazine nor one text published a decade ago with content that's even older seems like a sound basis for advising the OP on mating any preamp/DAC and amp he's ever likely to buy.  The generally stated and accepted rule of thumb is that input impedance should be at least 10x the source impedance to avoid exceeding the source's ability to sustain the necessary voltage drop across the load.  Sure, there are exceptions - but there aren't many and they're not being sold at audio shops across the country or around the world.   See here and here for two of many simple discussions on this.

[bluesman]

13 hours ago, sandyk said:

Furthermore, most Vacuum Tube amplifiers , or Passive Preamps that the participants in this thread seem to prefer will have a much higher output impedance than typical Solid State Preamps OR DACs

I fear you've been as careless in this quote as I was yesterday. I assume you mean "...most Vacuum Tube preamplifiers". And many modern tube preamps have very low output impedance. Prima Luna's DiaLogue preamp has an OI of 256 Ohms.  Here's the Prima Luna info page on this.  All tube preamps with cathode-follower output stages have very low OI, e.g. 200 Ohms for the classic Audio Research LS8.  Many modern tube preamps also have low OI, e.g. 300 Ohms (600 Ohms on the balanced output) for the current LS28.  FWIW, ARC recommends 20K ohms minimum load and 2000pF maximum capacitance.  Other excellent low OI tube preamps include the Mystere CA11 and 21 and the now discontinued Manleys.

 

As for your attached graph, the accompanying text makes it clear that it is only the purely resistive components that affect thermal noise - "the impedance is not relevant for noise calculations".  The difference between impedance and resistance is critical here, since output resistance of a preamp changes with frequency and can vary widely (which is why it's really output impedance and not output resistance, despite the erroneous and misleading practice of offering only a single spec value).  Read this quote from the Stereophile review of the Lamm LL2 for some data:

 

"The LL2's output impedance was a low 245 ohms in the midrange and treble, this rising to 3.3k ohms at 20Hz, presumably due to the finite size of the output coupling capacitor. As a result, the preamp's frequency response suffered from a premature low-frequency rolloff into the low 600-ohm load (fig.1, lower pair of traces). Into the higher 100k-ohm load (fig.1, top traces), the response is flat from 10Hz to 50kHz and just 1dB down at 200kHz."

 

Some fine tube preamps will be affected by loading from amps with relatively low input impedance.  Here's Stereophile's observation on the Shindo Aurieges:

 

"The output impedance at 20kHz was very low, at 23 ohms. It rose to 139 ohms at 1kHz, which is still low, but at 100Hz it measured a very high 3700 ohms. It rose even further as the frequency dropped, to 11k ohms at 50Hz and 16.5k ohms at 20Hz...With line preamplifiers [SIC - I think they made the same hasty error we both did in priorn posts! This should be line amplifiers.] having an input impedance of less than 100k ohms, this increasing output impedance at low frequencies will roll off the bass."

 

This will roll off the bass a bit below 30 Hz but is "corrected" with simple EQ - here's the Stereophile's measurement and discussion.  Virtually all of the "offending" preamps have capacitor coupled output stages. So the message is simply to either apply a little EQ or avoid these with power amps that have unusually low input impedance, as you wish. The effect on overall gain is negligible.

 

FWIW, I have and have read all 718 pages of Self's book.  I'm partial to more useful works like The Tube Amp Book (Pitman) and Building Valve Amplifiers  by Jones - now in its second edition (2014).  But I try to understand as much as I can, even if a lot of it is knowledge I'll never use.

[bluesman]

43 minutes ago, davide256 said:

 

So trying to boil this down, some tube preamps are likely to cause problems if the connected amplifier is less than 100K ohms input impedance? I do see that the

PV10A I used to have was measured at 690 ohms output impedance by Stereophile. And that the amplifier sections I've owned before, Bryston B60 at 30K ohms,

 Hafler DH200 at 22K ohms and PrimaLuna Prologue Premium at 35K ohms are below that 100K value.

First and foremost, impedance is frequency-dependent resistance.  A single quoted figure with no reference frequency is probably useless in identifying potential negative interactions, since the preamp's output impedance probably rises at the very lowest frequencies and OI is almost always quoted as the resistance at 1 kHz.  If it reaches a very high value by 50 Hz or lower, it can result in attenuation of the lows in proportion to the rise in impedance by frequency.  This can be as much as 6 to 10 dB in extreme cases, but such cases are as rare as hens' teeth.

 

So in summary, a very few preamps have sufficiently nonlinear output impedance to cause minor frequency spectrum anomalies when driving the few amps that have extremely low input impedance.  As long as the OI/II ratio is greater than 1:10 across the spectrum, there's almost certainly not going to be an audible problem.  Just check the specs on whatever you buy to be sure it's not an outlier.

[bluesman]

1 hour ago, sandyk said:

Your idea of more useful  is clearly based on your preference for glowing ancient technology with it's attendant added maintenance costs, using replacement tubes most likely different to those originally supplied, and usually with quite a variance in specifications to the original tubes . 😊

The added maintenance cost is easily offset by the drastic reductions in our heating bills. 🤪

[bluesman]

1 hour ago, sandyk said:

 

 You fail to take into account the added cost of Air Conditioning in Summer, unless you live in a colder part of the country

 I guess that the hiss from the vacuum tubes is likely to mask the hiss from the A/C when running in Quiet mode though.😉

The added tube life from air conditioning exceeds the associated HVAC costs.  You obviously haven't priced good KT88s recently.  The hiss is just analog dithering.

[bluesman]

1 minute ago, sandyk said:

 I wouldn't want to also watch TV every night using a system like that at something around $280 for 4 matched tubes !  🙄

 Add to that the slow degradation that occurs without you realising it for quite a while .

If I didn’t know better, I’d be afraid that you’re being serious.  Besides, our HT rig uses 4 KT100s in each of its 11 channels.  We never installed heating in the house, so we used the money we saved to have theater seats made.  We then discovered that we don’t need any lights on the main floor.  And those solar panels in the walls generate all our electricity.  Most importantly, the tube glow turns out to be very romantic lighting.

[bluesman]

41 minutes ago, sandyk said:

 You aren't connected to the A.C. mains grid ?

No need - the tube glow powers the solar panels in our inside walls. The panels power the tubes.  I assembled large arrays of piezo mic capsules that generate power from the vibration of my speakers, and I’ve achieved perfect coupling.  So there are no losses and we use no fuel to heat and light the place, cook etc.  

 

The only downside is the need to play music or videos 24/7/365 to keep it all going.  The day I stop the music, I’ll drive my Chevy to the levee and proclaim it to be Don McLean day.

[bluesman]

20 minutes ago, sandyk said:

It's a shame that Michael J Fox isn't up to another Back to the Future movie. They could have found another power source for the Delorean. 🤣

I bought the last flux capacitors, too. They’re now my preamp’s output couplers - the SQ is amazing!