REGISTER NEW USERLOST PASSWORD? WELCOME, Logout

Protection or performance?

Paul's Posts — 31 December 2011

By

Protection or performance?

Sometimes we make great decisions other times not.  This is a tale of not, but includes a lesson.

In the early 1980′s a very gifted amplifier designer, Bob Odell, joined the company and with him a new design of a power amp – to be introduced as the 200C.  We spent a lot of time on the amp, tweaking, changing, doing what we could to maximize its sonic performance.

One bone of contention between myself and Bob was an output inductor – which had no measurable affect within the audio band but you could hear it none the less.  Bob insisted on including the ultasonic filter on the output of the amplifier because use of that filter made the amplifier unconditionally stable and safe.  I objected to the inductor for performance reasons – the filter made the amp sound too sterile.

Just before we were to release the final design we decided to settle the issue with a sonic shootout.   We performed the test on a system that had far more resolving power than ours: the Infinity IRS.  Our friend Arnie Nudell, one of the best listeners in the industry, had a pair setup in his home and to LA we went for the big shootout.  The question at hand: how much of a performance compromise did the inductor present?

We played the amplifier on the big IRS for some time – Arnie being polite but unimpressed.  I remember a particular album we were playing had a magnificent piano piece and frankly, it was good but not stellar.  The Audio Research reference amplifier was far more musical.

I then took a pair of clip leads, shorted out the output inductor and replayed that piano piece.  Arnie and everyone else in the room lit up like a veil had been lifted.  ”Now THAT sounds like a piano!”  Arnie was right, the piano had bloom and overtones not present with the inductor.

“McGowan, whatever you did, do that in production and you have a winner!”  Arnie proclaimed.

Bob Odell, on the other hand was as nervous as a cat about this decision.  He too wanted the sonic gifts of no inductor but knew the amp would be less reliable.  So I questioned him and found that we’d be fine as long as no one disconnected or connected anything to the amp’s inputs while it was on.  So that was it.  We went with the better sounding version and put a big warning on the back of the amp.  ”Do not make any connections to this amp when the power is on”.

It only took about 25 amplifiers going up in a puff of smoke to realize we had sacrificed protection for performance.  Seems most people either never saw the warning or ignored it.

We finally figured out a good compromise to make the amp stable and sound good but learned a great lesson as well.

The decision between protection and performance should not be made lightly.

email Protection or performance? Forward to a friend and help us engage more readers

Get new and fresh stories like this each morning by joining the folks reading Paul's Posts. Click here

Related Articles

Serving it up Rolling it off What’s all the fuss about? Young turks

Share

About Author

Paul McGowan

Paul McGowan is the CEO and co-founder of PS Audio Inc. a Boulder Colorado design and manufacturing company of high-end audio products and services. McGowan has been designing and building high-end products for nearly 40 years. Hobbies include skiing, music, hiking, artisan bread baking, kick boxing and cooking. He lives in Boulder Colorado with his wife Terri and his 4 sons.

(3) Readers Comments

  1. timequest

    December 31, 2011 at 9:46 am

    Do the “protection solutions” of today (the better designs) still impact performance (not necessarily on the scale that you and Arnie Nudell witnessed , but discernible nonetheless)?

    I would prefer, as I know many audiophiles would, to take my chances with the design approach that utilizes a straight-forward performance solution at the expense of “protection” (if that protection meant diminished performance). That said, would you consider sharing with those who are interested where your designs utilize “protection” compromises – thus allowing owners “tweaking options” (perhaps also providing instructions on how to bypass said “compromises” – along with, of course, a well-written disclaimer and voided warranty…!).

  2. December 31, 2011 at 10:40 am

    Ben, that’s a great question. Remembering that when this amp came out it was in the “dark ages” of high-end amp design and a long time ago. Yes, there are completely neutral ways to protect these days – the most popular is pretty black and white – exceed a certain level and the amp simply shuts down. Other schemes remove the input signal should the output exceed a certain point.

    The most common scheme measures the current across the emitter resistors on the output transistors and if a preset level is exceeded the input can be clamped (adds distortion) or the amp simply shuts off – or an output relay opens up and you get a momentary loss of sound.

    Most designers today don’t compromise performance for protection.

  3. Soundminded

    December 31, 2011 at 11:34 am

    I think what’s most disturbing is that people who develop high end equipment that claims to sound better but doesn’t measure better never seem to be able to determine why. Assuming at least some of it is true they don’t advance any theories backed up by new measurement’s to increase understanding, advance the art. One paper written by Cheevers about amplifiers was a joke, more full of holes as a serious technical paper than a wheel of Swiss cheese. This reduces development of high end equipment to nothing more than haphazard tinkering. There’s no way to isolate the superior element or characteristic and exploit it at the lowest possible cost and greatest effectiveness.

    Certain types of audio equipment strike me as very poorly measured, so called audio power amplifiers among them. The method of measurement is probably inherited from the 1920s or 1930s when one watt frequency response into a resistive load could probably tell us much between the differences in performance among various models and explain those differences but most of them these days tell us little or nothing we’d like to know. Loads are not purely resistive, they aren’t even passive. FR does change with output. This is the crux of what TIM was about. If FR had been measured at all output levels, TIM would have been a redundant measurement because it is based on the fact that as amplifier power output increases, amplifiers’ ability to reproduce higher frequencies decreases. This is very similar to the well known concept of gain-bandwidth product.

    If an amplifier is unstable with a specific input then it is probably inherently unstable to begin with and in this case may fail because with the right excitation the output stage went into spontaneous oscillation. This was hardly unusual for many early solid state amplifiers and certain brands of audiophile cables were known as amplifier killers because their reactances formed tank circuits causing the amplifiers to oscillate at ultrasonic frequencies.

    If the choice is between an amplifier that sounds mediocre or one that sounds great until it spontaneously blows itself up and possibly destroys an expensive pair of loudspeakers with it, I’ll take the mediocre one any day.

Leave a Reply