Bob Cordell's Power amplifier book

The Wiki page make an interesting claim:
"...solid state amplifiers have been developed to be flawless..."

Scratching my head trying to understand what definition of "flawless" is being used?

If there are standards for measuring the following items, maybe someone could claim, "...plastic toothbrushes have been developed to be flawless...?" "...electric toasters have been designed to be flawless?" Who decides how good something has to be in order to be declared flawless? It is whatever is popular to believe over at ASR?
 
  • Like
Reactions: 1 users
Administrator
Joined 2004
Paid Member
Well Mark,
I would suggest that even though nothing is flawless, if the flaws are at levels lower than you can detect them they can be viewed as flawless. Wouldn't you say? Otherwise we are venturing into the realm of the mind (which disappears if you have no way of knowing what you are listening to).

To argue otherwise is merely a discussion for the sake of an argument.
 
Audio Science Revie. It's the engineering equivalent to diyaudio.

https://www.audiosciencereview.com/forum/index.php

Jan
I would respectfully on the "engineering level" there. I is more and more about total beginners and technically misleading. IMO diyaudio has higher "engineering level" than ASR. At least here are many members contributing with technical projects. At ASR, there is almost nothing like this and the "engineering level" of the posts is usually low or basic.
 
  • Like
Reactions: 5 users
Yes. Perhaps there is some engineering at ASR but I couldn't find any. It's all marketing, IMHO. You might call it "Do Nothing Yourself Audio". Even calling it "Science" is a bit generous. But I have no patients for academics that talk and write a great deal and spend zero time "on the bench". I confess that I've become too fat and lazy to do much myself lately, but at least I play with spice etc. But I will never spend hundreds of dollars on audio equipment that someone else built.
 
  • Like
Reactions: 1 user
...if the flaws are at levels lower than you can detect them they can be viewed as flawless.
Isn't there also a question of what characteristics could count as flaws? My question is where is the agreed upon list of characteristics that can't be measured in order to call, say, for example, a plastic toothbrush flawless? Or for any other product, for that matter?

Understood there may be people who have their own opinions as to what should count for amplifiers, but I haven't seen anything like at ITU standard for it (doesn't mean there isn't one though).
 
  • Like
Reactions: 1 user
AX tech editor
Joined 2002
Paid Member
ASR does not pretend to be about engineering or diy. It is about objective testing of equipment, and in that is very different from diyaudio where a lot of the discussion is purely subjective and personal.
ASR measurements can be repeated by anyone in any place and will largely lead to the same results.
As such, it is valuable for those looking for info.

If subjective/personal 'testing' is repeated by someone else in some other place, it is not unusual that results are totally different. As such, those results have no real meaning.

Jan
 
  • Like
  • Thank You
Reactions: 4 users
Administrator
Joined 2004
Paid Member
Hi Mark,
Isn't there also a question of what characteristics could count as flaws?
Sure. If you cannot perceive something, it for all intents and purposes does not exist. It doesn't matter what you want to cook up as a flaw. Then of course, the old familiar suspects will come up with "could be" and "we don't know everything" to try and spark a debate. Well folks, we do know what the limits of human sensory detection are. That are your limiting factors. Well researched, peer reviewed - go argue with those people if anyone wants to drag that up.

Then we have the environmental noise floor, and the noise floor imposed by our own bodies. Now that's if and only if our brains aren't playing tricks on us. Our brains hate missing information so it will fill the void.

It's always good to make a circuit perform better - within reason and looking at expense. At some level we are further ahead making a circuit more reliable and tougher to kill. Engineering is a balance. Chasing one factor at the expense of others never ends well for the consumer. So once we reach a point beyond what a human can perceive with audio reproduction equipment, we should either slow down, stop on that product and concentrate on other things that bring value.
 
  • Like
Reactions: 1 user
...we do know what the limits of human sensory detection are.
Anatech,
With all due respect, we have measured some things about human sensory detection. For hearing we have a number of 'threshold of audibility' numbers, and we also have some 'absolute thresholds' for single tones. Some people think such thresholds are absolute limits for everyone, and or that they apply to all auditory signals.

IOW, some people don't understand statistics in perceptual science, and or some people don't understand about the differences between LTI versus non-LTI systems.

In any case I don't want to argue about the above issues here. If there are questions, then people are welcome to welcome to PM.
 
  • Like
Reactions: 1 user
Administrator
Joined 2004
Paid Member
That's fine Mark,
All I am really saying is that beyond limits of our perception, there is no argument. If you're in the grey area, then sure. Some people can hear differences. When impairments are in areas that everyone can hear, then those are definite problems and those products need improvement.

What I said (clearly I thought) is that there are in fact hard limits we simply cannot "hear" or sense. Yes, we have grey areas above that where some can sense, others can not. We have no disagreement there.

What isn't useful are arguments about things that could be when we know those are beyond our limits to hear or sense. Those go on forever and you end up with a ton of mis-information. That holds no benefit for anyone.

Just be real and reasonable.
 
  • Like
Reactions: 1 user
This may interest Mr Cordell and some others.

Recently I needed to calculate some load-lines for a bipolar transistor used in the output stage of a class AB amplifier. After some work with pencil and paper I managed to find a couple of simple formulas for plotting load lines. I have attached a pdf describing the formulas and a python script that calculates the load lines.
 

Attachments

  • loadline.pdf
    100.2 KB · Views: 95
  • loadline.zip
    1 KB · Views: 50
  • Like
Reactions: 1 user
And, a typical example of load lines and SOA. Will a TIP35C be sufficient for the amplifier? Maybe for domestic use, but forget it for industrial uses.

soa.jpg
 
  • Like
Reactions: 1 user