HI All,
As luck would have it I got this email this morning asking a question regarding amplifiers and output impedances vs power delivery. Thought you might find it informative as it relates to Dave's opening question.
From: Drew Daniels
Sent: Sunday, June 07, 2009 1:37 PM
To: jamestanner@bryston.com
Subject: Tech question
Greetings,
I am about to purchase a Bryston 4B SST2 for my mastering room. I have a question I hope is easy to answer. Please correct me where I am in error.
In electrical engineering school, I was taught about voltage source and current source audio frequency amplifiers. In the 60s I worked for Gene Czerwinski who used to routinely sketch amp designs using trifilar output stage driver transformers, on one side of a sheet of notepad paper and hand them to me to go build in the shop (they all worked first time). Gene had experience working for Bendix Corporation where he designed and built an all-germanium transistor 20 kilowatt amplifier for helicopter-deployed sonar use, as early as 1962.
Here is where I am hazy: should a voltage source amplifier be regarded as essentially (metaphorically?) having zero output impedance, and therefore be expected to deliver twice the power output when the load impedance is halved?
Why do (most) power amps give power ratings that seem in conflict with impedance equations? Is it more due to power supply stability or to reactance magnitude differences between loudspeakers rated at a characteristic 8 or 4 ohms? Thanks for taking the time to answer.
Cordially,
Drew Daniels
Grammy nominee - best sound,
Winner, French Jazz Academy Record of the year, 2007.
Hi Drew;
Thanks for your question, it is an excellent one.
Amplifiers today are almost always designed to be Voltage sources. That is, their output impedance is very low, (on the order of a few milliOhms), so their output Voltage for the same input signal varies very little from open-circuit load to nominal full load impedance. (A theoretically perfect Voltage source would not vary at all). In addition, if the power supply of the amplifier also had a zero ouptut impedance, the amp's power capability would vary exactly inversly with load impedance. That is, its power output capacity would double from 8 Ohms to 4 Ohms, and double again at 2 Ohms, etc.
In fact, of course, there is no such thing as 'perfect'. All amplifiers have finite output impedance, as do the power supplies in those amplifiers. Thus, no amplifier actually 'doubles' its output power capacity repeatedly with halving load impedance. If an amplifier has a large enough power supply it can come close enough for the manufacturer to quote the power at a low impedance like 2 or 4 Ohms, and then quote half those values with doubling the load impedance, (a bit of a 'fudge'). Bryston amplifiers are rated realistically but conservatively for specified load impedances.
In answer to your question about reactance, some amplifiers do in fact have problems with large reactances, and reduce their output Voltage and current drastically with large reactive phase angles. The manufacturer will not tell you that, of course, and the specs will never show it since power is measured on resistive loads. Bryston's output Voltage and current specifications relating to power output are valid for any reactive phase angle from +90 to -90 degrees, or from completely capacitive to completely inductive and all in between, or in any combination.
I hope the above is helpful, but please let me know if you have any other questions. Thanks for thinking of Bryston!
Sincerely,
Chris Russell
CEO, Bryston Ltd.