[Steve]

"Steve, I agree with your points...just saying that those are the goals.

Oh no, I know what you were saying. Just that the way you said it left the impression that there are speakers out there which met those goals."

Hi Steve,

     I saw your post and realized the meaning could easily be mis construed. When I first posted, I did put "good luck" right after my last sentence, but then I changed it to the next line thinking I would soften my comment. 

As I think you understood me, I meant to say, in a nicer way, that one would not find such a speaker, at least that I am aware of.

Take care and thanks for bringing that to my attention Steve.

     

[MJK]

I think it's fine to have a double standard.  I like the tubes in the living room, they're friendly to the material.  When you sit down and really listen hard though, you can tell that they put a sexy evening dress and some smoky make-up on whatever you're playing.

Adam,

Have you tried running the Adcom with a 2 or 3 ohms series resistor between the amp and the speakers? Might be an interesting experiment to see if the SS is closer to the tube sound with a little added output impedance.

Martin

[mfsoa]

So if tube amps suffer from high output impedance and if one has multiple impedance taps, what's the downside to just using the lowest impedance tap?

Like using the 2-ohm tap on an 8 ohm speaker?

Thanks

[Steve Eddy]

So if tube amps suffer from high output impedance and if one has multiple impedance taps, what's the downside to just using the lowest impedance tap?

Like using the 2-ohm tap on an 8 ohm speaker?

Well, generally speaking, the output tube would see an increased plate load which can result in increased distortion.

First, transformers reflect impedances by the square of the turns ratio. Second, a given tube ideally wants to  work into a given plate load for best performance.

In a typical series feed SET amp, the output transformer's primary serves as the plate load for the output tube. The top of the primary connects to B+ and the bottom of the primary connects to the tube's plate. The ideal plate load for the output tube is much higher than the loudspeaker's nominal impedance so output transformers are essentially step-down transformers.

Let's say the tube needs to see a plate load of 5,000 ohms. If the loudspeaker's nominal impedance is 8 ohms, then you need a turns ratio of sqrt(5,000/8) or 25:1. And that's all the various output taps are on an output transformer; different turns ratios which will give the same primary impedance for 16, 8, 4, and 2 ohm loudspeaker loads.

So for the same 5,000 ohm primary impedance, the turns ratio for the 2 ohm tap will be sqrt(5,000/2) or 50:1. Connect an 8 ohm load to the 2 ohm tap and that 8 ohms gets reflected back to the primary by the square of the turns ratio, which in this case will be a factor of 2,500. So with an 8 ohm load on the 2 ohm tap, instead of seeing a plate load of 5,000 ohms, the output tube sees a plate load of 2,500 x 8 or 20,000 ohms. Four times higher than it should be.

se

[slbender]

While what you said is true, OK, but maybe it not always about the maximum the power.  Running a 6BQ5 ( 5,200 ohm rated load ) at 5,000 ohms reflected back gets you 4.5 Watts with loads and loads of 3rd, 5th, 7th, and 9th harmonics, so why not just use transistors, mosfets, ICs, or Tripath TA-2020 digital Amp IC's ?

Lets try thinking outside the box.  I guess what is really important isn't the tap and the ratio, but the goal... Is it always about the maximum power transfer ratio ?  Yes, that gets the most power to the speaker, but so what ?

Who said a tube needs to see a given reflected back impedance... some ancient GE Tube Manual written in 1962 ?

Why not setting the design operating point to the minimum distortion as a valid operating point ?

Or how about a conservative point where the bias current is set for way lower dissipation than the maximum design rating - also another valid design point ?

Now why not design a SET with an operating point that limits the maximum output tube dissipation ( long tube life ); and also has slightly lower than nominal filament voltage ( Decades of filament life with nearly the same emissions ); and is designed for minimum distortion at the expense of maximum power output transfer - a very decreased plate load ( say a 5,000 ohm impedance reflected back as a primary plate load on a 600 ohm total tube plate internal resistance - thinking of six small triodes per channel in parallel for sake of hypothetical discussion; and ignoring the Miller Effect ).

You may get a rather clean 3 Watts output, instead of 15 distorted Watts output, but your tubes will be stable, run cooler, and using a 15 Watt OP transformer at a 3 Watt level, the bass will be totally improved by about two octaves all other things being equal; so will the high end, and many types of transformer saturation distortion, and the mostly junk distortions that run back into the power supply and across to the other channel will also be reduced, all without resorting to the use of global feedback.


-Steven L. Bender

So if tube amps suffer from high output impedance and if one has multiple impedance taps, what's the downside to just using the lowest impedance tap?

Like using the 2-ohm tap on an 8 ohm speaker?

Well, generally speaking, the output tube would see an increased plate load which can result in increased distortion.

First, transformers reflect impedances by the square of the turns ratio. Second, a given tube ideally wants to  work into a given plate load for best performance.

In a typical series feed SET amp, the output transformer's primary serves as the plate load for the output tube. The top of the primary connects to B+ and the bottom of the primary connects to the tube's plate. The ideal plate load for the output tube is much higher than the loudspeaker's nominal impedance so output transformers are essentially step-down transformers.

Let's say the tube needs to see a plate load of 5,000 ohms. If the loudspeaker's nominal impedance is 8 ohms, then you need a turns ratio of sqrt(5,000/8) or 25:1. And that's all the various output taps are on an output transformer; different turns ratios which will give the same primary impedance for 16, 8, 4, and 2 ohm loudspeaker loads.

So for the same 5,000 ohm primary impedance, the turns ratio for the 2 ohm tap will be sqrt(5,000/2) or 50:1. Connect an 8 ohm load to the 2 ohm tap and that 8 ohms gets reflected back to the primary by the square of the turns ratio, which in this case will be a factor of 2,500. So with an 8 ohm load on the 2 ohm tap, instead of seeing a plate load of 5,000 ohms, the output tube sees a plate load of 2,500 x 8 or 20,000 ohms. Four times higher than it should be.

se

[Steve Eddy]

Who said a tube needs to see a given reflected back impedance... some ancient GE Tube Manual written in 1962 ?

The characteristics of the tube in question determine that.

Why not setting the design operating point to the minimum distortion as a valid operating point ?

Unless I'm missing something, isn't this pretty much how most folks go about doing it?

se

[AdamM]

frequency response based on an anechoic measurements doesn't reveal the entire story about how a speaker truly sounds.

ScottF: I think anechoic tests are really only part of the story.  Rooms contribute so heavily to the final outcome, it's really hard to draw too much conclusion from those graphs.  Is it so weird how a 'bad' graph can sound good?   A violin playing 'A' will sound very different from a pure sine wave of the same frequency. The violins graph will be all crazy-alps. I think we'd all agree what sounds better.

People dig certain voicings and i bet that's rarely ruler flat. There's trends too - look at the 'thin' sound of the stuff coming out of Britain in the 80's.  Chuck a Cure album on some old Missions and you'll wonder where the bass is, but people liked it.

For my particular test case here, a flat response is pretty important, i'm mixing to it. I don't want it candy coated, i want the real deal. Sounding marvelous and  sounding 'right' are two different things. Your system adds colour, or tries not to.  Different horses for different courses.

Thanks for the tip on the swamping resistor. Cheers.

Have you tried running the Adcom with a 2 or 3 ohms series resistor between the amp and the speakers?

MJK: Thanks for the tip, i'll give it a try and post some results

So if tube amps suffer from high output impedance and if one has multiple impedance taps, what's the downside to just using the lowest impedance tap?

mfsoa: I'll try running it at 4 and 8 ohms and graph the response.  My OT doesn't have 2 ohm taps, unfortunately.

I won't use these amps in the studio, too crispy, but i'll finish graphing those comparisons because it's an interesting topic.

Cheers,
/A

[aerius]

Why not setting the design operating point to the minimum distortion as a valid operating point ?

Unless I'm missing something, isn't this pretty much how most folks go about doing it?

se

I wish.  We have lots of "designers" doing wonderful things like putting 10k plate loads on a 6SN7, biasing a 12AX7 at a couple milliamps in a hybrid preamp and expecting it to drive the gate capacitance of the MOSFET, using output transformers with insufficient primary inductance, running tubes way over spec (I'm looking at all the morons who think 200uF on a C-input 5AR4 power supply is ok) and so on and so forth.