damping factor

Status
Not open for further replies.
By your own criteria, nearly every single amplifier built is considered world class. $500 or $600 sounds very reasonable for selling a ton of them though.

Do you have proof that Pass understood the advantages and disadvantages of global negative feedback considering his later designs used zero global negative feedback? Remember that adcom amp was approx. 30 years ago.

------------------

Interesting that I am the one who actually brought out the maximum possible damping factor of 1.33 or so (Using voice coil dc resistance as 3/4 of the driver impedance for ease in calculating. 6 dc ohm for 8 ohm, and 3 dc ohm for 4 ohm). This is the amp to driver direct connection. Less damping factor when the dc resistance of the choke winding is added.
One can use both mechanical and variable electrical damping when designing their speaker.

Maybe you should be grilling Ken Kantor since he actually proclaimed 10-20 damping factor (the typical way of calculating df) is appropriate with little improvement beyond. :yes:

Cheers

Pos

As far as I'm concerned, solid state amplifier design was at its zenith by the mid 1980s.
Esoteric amplifier designs are great for keeping people on the treadmill of buying new gear every few years or even more frequently, and/or for audio snobs to justify their purchases.
As for myself, I'll stay with tried and proven amplifier topology.
BTW, I wouldn't grill Ken about anything. He is well known, and well respected in his field. His opinions on amp and speaker design carry a lot more weight with me than those of any audio snob who trying to convince me that the latest and greatest oddball design is the cat's meow and everything else is just tinny sounding junk. Sorry about the run on sentence.
 
Last edited:
Do you have proof that Pass understood the advantages and disadvantages of global negative feedback considering his later designs used zero global negative feedback? Remember that adcom amp was approx. 30 years ago.
I can say with certainty he did - and understanding the challenges created by using large amounts of NFB became the basis for the Threshold Stasis amplifiers. Understand Adcom products were targeted for production at a particular price point and had to be inexpensively manufactured. That was not and has not been the case with Threshold, Pass Labs and First Watt products where Pass has been in complete control of not only the design but the implementation of realizing performance at the highest achievable levels.

Here are two of six pages from the Threshold brochure circa 1981 which speak to the "principle of attempting to influence an event after it has occurred" Despite using no overall loop feedback, the Stasis amps had a DF of no less than 60 from DC to 20kHz.

Threshold Brochure

Threshold brochure part 2
 
By your own criteria, nearly every single amplifier built is considered world class. $500 or $600 sounds very reasonable for selling a ton of them though.

Do you have proof that Pass understood the advantages and disadvantages of global negative feedback considering his later designs used zero global negative feedback? Remember that adcom amp was approx. 30 years ago.

Cheers

Pos

Understand Adcom products were targeted for production at a particular price point and had to be inexpensively manufactured. That was not and has not been the case with Threshold, Pass Labs and First Watt products where Pass has been in complete control of not only the design but the implementation of realizing performance at the highest achievable levels.

I know we're getting a little off track here. But, what you guys are saying is that the only way an amp can be considered world class is if it is expensive, made in small numbers, and of esoteric design?
 
I'm pretty sure that everyone who has taken 2nd semester electronics knows about inductive reactance. I'm not sure what mystery you have exposed there.
And you just came out and said that a $500 (~$1500 today) amp can't be world class. So I am also to take it that all of these amps and receivers that we all hold dear and think sound great actually don't sound good at all?
 
Had to bring up chokes, didn't you? It was just the usual friendly disagreements over easy stuff, and now we have to think about magnetics. :D

If you look at the back end of most power amps, you'll find the very common series choke and generic 0.1 uF cap to ground. Usually a 10 ohm resistor in there too. Some time ago I was designing a very fussy amp for a very fussy customer. All was well until I got to the choke. It turns out that winding it on any kind of form containing metal (like steel leads or the end caps of a resistor) raises the distortion, sometimes by quite a bit compared to what a good amp can do. It's common to see them wound on a big 10 ohm resistor. Dynaco even wound them around the output capacitors! I wanted to save a bit of space by using a core, but no core is acceptable. The thing has to be reasonably heavy wire wound on a plastic bobbin or no bobbin at all. I've never calculated it, but that network must have at least some effect on the damping factor at higher frequencies.

BTW, IMO a $500 amp can be absolutely world class, as long as you don't spend $450 on a fancy machined chassis.
 
Last edited:
I'm not upset at all. You asked me point blank if I thought the Adcom GFA-555 was a world class amp. I answered yes. You asked my how I justified that answer. I told you how I came to that conclusion. You told me I was wrong.
I never said that every $500 piece of gear is world class. Hell, there are amps out there that sell for thousands of dollars that I consider to be unreliable junk.
However, I stand by my assessment that the GFA-555 is world class. As far as I'm concerned, obsessing over every little aspect of amp design, as someone who doesn't design amps for a living, just takes all of the joy out of listnening to music.
 
Last edited:
But, what you guys are saying is that the only way an amp can be considered world class is if it is expensive, made in small numbers, and of esoteric design?
No. The best engineering, however, does require developmental expense in addition to the use of higher quality parts, matching, testing, etc.

When you go to Mr. Pass and say, I want a 150 watt/channel amp to be delivered at the $1500 level - that will result in a very different design than his own Xs150 product! It all depends upon the objective and budget. :)
 
You misconstrued one of my answers to one of your questions. The question being the price to performance ratio of the 555. Somehow you took my answer to be that all $500 gear was world class.
 
i am still trying to figure out the correct tools to measure amplifier output resistance .

You need a sine wave signal at the frequency of interest. That might be a proper signal generator or test tones generated in something like Audacity and burned to CD or directly generated and sent out a sound card.

You need a way to measure the AC amplitude of what comes out of the amp. That should be an AC meter- it does not have to be true RMS, but it does have to read the frequency of interest. If the damping factor is high, a scope probably won't resolve what you're looking for.

Set the amp up with some arbitrary output level, say 1 volt rms, with no load. Since there's no load, the voltage will be the same at the output as on the other side of an imaginary series resistor in the amp that represents the output impedance.

Now, without changing anything, load the amp with 8 ohms or the load of your choice. Measure the new voltage right at the output terminals of the amp. Let's say you started with 1 Vrms and now it's 0.971 Vrms.

The first thing you know is the current- 0.971/8=0.121 amps.

You also know that on the far side of the imaginary resistor, the output impedance, the voltage is still 1 volt, so you can subtract and know that the voltage drop across that imaginary resistor is 0.029 volts.

You've got voltage drop and you've got current, so ohms law gives you the value of the imaginary resistor, the output impedance. 0.029/0.121=0.24 ohms. Damping factor is 8/0.24=33.3, pretty low, but I just made up the numbers.

Now, you need to decide what frequency to make the measurements at, what amplitude, if it matters and what load to use, because damping factor isn't one all-encompassing number.

No doubt if I've made a mistake here, and I do make a boatload of 'em, it will be swiftly pointed out.:yes:
 
Are the output transistors with no load actually conducting current? IOW are they in a suitable operational state for this voltage-division method to work? I know squat about amps but I would put some load on the outputs that could be ignored in the voltage division calculation. Just guessing but something 10x or so greater resistance than the test load. You could even factor it in if you didn't want to ignore it. Thoughts?
 
Are the output transistors with no load actually conducting current? IOW are they in a suitable operational state for this voltage-division method to work? I know squat about amps but I would put some load on the outputs that could be ignored in the voltage division calculation. Just guessing but something 10x or so greater resistance than the test load. You could even factor it in if you didn't want to ignore it. Thoughts?

As long as the amplifier is stable with out a load, all is good.

An amplifier can produce a voltage with out current flowing and that is the point of Conrad's test, measuring the voltage sag from no load to the load of your choice.

This is done when testing the regulation of power supply. For example, and slightly off topic a LM 317 has a certain ratio of no load voltage to full load voltage as specified by the manufacturer and this can be taken as its static impedance.

Then there is its dynamic impedance and so it goes with an audio amplifier.

Its fun stuff is it not Conrad?
 
An amplifier can produce a voltage with out current flowing and that is the point of Conrad's test

I agree but I'm wondering about how far we can carry assumptions about this voltage appearing through a chunk of non-conducting silicon and perhaps other paths. BTW love the thread. All my favorite carnivores are here. :D
 
Ha, if a tree falls in the woods when nobody is there, does it make a sound? In the same way, does an amplifier have an output signal if nothing is attached? Rest assured, it does. There will always be some small loading from the feedback network and output network, so there's no need to add anything. Feedback insures that the output signal remains constant, regardless of load, except for the output impedance that we're so desperately trying to measure.:D

Yes, fun stuff indeed, though I admit to having actually measured damping factor only a few times ever. It's just never been an issue for me. Actually I've learned more by driving the amplifier with a second amplifier through the load with different waveforms. You can measure damping factor that way too, but it's more involved for the average DIYer, or at least takes up more bench space.
 
A- An inductor is most assuredly linear. A frequency dependency is completely different than non-linearity. Any first-year engineering student should know that.

It is described by a linear differential equation. Its output doubles when its input doubles. It is impossible to get any frequency out that you don't put in. Etc.

B- That's the last time I waste 35 minutes writing a comprehensive article, just so people can argue ridiculous, irrelevant points. :(

C- Designing a transparent preamp is trivial. I do it all the time. And with cheap op-amps, too.

-k
 
Ken, nice load. How do you feel about the changes they made? Also, if I want to build one, is the DCR of the coils important? Unless they were quite heavy I'd expect it to be a significant fraction of the series resistors.

Thanks! The DCR of the coils is listed on the Stereophile schematic. How much each one matters can be determined (and, possibly, compensated) by noting any other series resistance and by doing a circuit sim.

I wish John Atkinson had talked to me before making the change, or at least, before writing about it. My choice for the top end was deliberate. I feel that a purely reactive top end is much more demanding on amp stability than a partially resistive top end. I was concerned with stability, he with current draw, which I don't think is an issue up there. But, both circuits are valid. I see speakers that approximate both behaviors.

-k
 
A) Read post 90 again. The schematic, test, and results clearly demonstrate you are obviously incorrect in your knowledge, and math.

C) Well, I do not believe you have designed and built a perfect preamp because you have not shown me the capability of understanding a choke and designing a proper power supply.
So why are you not advertising your preamplifier as perfect? Why have you not offered a $49.95, $99.95, $995.95 perfect preamplifier to the public?

Cheers

Pos

A- Careful... someone may spread post 90 on the interwebs, where it will become an engineer's meme, and you a laughingstock.

B- I don't believe >you< have designed the perfect preamplifier, because you don't understand electronics.

C- My preamps don't need power supplies. They work on quantum zero point energy. Just because you can't understand it, doesn't make it wrong, right?

D- I don't advertise them because many others can also design equally transparent preamps. As I have said, it is trivial.

-k
 
positron;8211910C) said:
Well, I do not believe you have designed and built a perfect preamp ...
Everything is relative.,right? My most listened to system is not my best. I enjoy listening to music on my iPhone using Shure 535se IEMS. On the other hand, I would like to think I have an objective perspective on what audio systems can do. I've heard Harry Pearson's systems over the past thirty years and remain *amazed* at how good audio quality can be.

When your vision of *perfection* is a Dynaco amp. you've clearly set the bar at your knee caps. :)

"I have found many amps that do an effectively perfect job amplifying a signal above their noise floor and below their clipping limit. In my listening rooms I have used a Dynaco 410, a Crown Microtech, an Aragon 4004...

I don't understand why Ken and others eagerly want to dumb down audio quality.
 
Status
Not open for further replies.
Back
Top Bottom