Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[John Schalk]

I had no idea that the amp charts showing plots of the outputs for a Lab, Crown, and PSoft were produced by Bink.  I've studied them several times as I feel that provide some very useful information with regard to how one would want to utilize class D amps for different duty cycles.  One of the things I have trouble with these days is deciphering what the actual output power is for new amps.  I realize that manufacturers have to use the "burst power" rating to compete with everyone else, but some have stopped listing the FTC ratings altogether making it hard to figure out just what you're buying.  My request would be for you to produce a chart like Bink's 8/4/2 ohm plot from time 0 to 100 seconds for any newer amps you test.

Note: I'm almost certainly done buying power amps so any data you produce will be for reference only, at least for me.  But don't let that stop you!

[John Roberts {JR}]

I had no idea that the amp charts showing plots of the outputs for a Lab, Crown, and PSoft were produced by Bink.  I've studied them several times as I feel that provide some very useful information with regard to how one would want to utilize class D amps for different duty cycles.  One of the things I have trouble with these days is deciphering what the actual output power is for new amps.  I realize that manufacturers have to use the "burst power" rating to compete with everyone else,

and because burst or short term power better resembles real world music signals, and similarly loudspeaker power handling characteristics.

but some have stopped listing the FTC ratings altogether making it hard to figure out just what you're buying.

perhaps because the FTC power rating is an archaic leftover from older amplifier technology. Class A/B amplifiers suffered worst case power dissipation at 1/3rd power, FTC instituted a 1/3rd power pre-conditioning stage to reign in inflated power claims (from the 1970s). Modern amplifier technology (like class D) exhibits completely different power dissipation behavior and different limiting constraints.

My request would be for you to produce a chart like Bink's 8/4/2 ohm plot from time 0 to 100 seconds for any newer amps you test.

This is a mature topic, and I wrote about it at length during the original power amp test rodeo performed by Bink (Michael Knowles). Short answer, there is no single power/duration metric that can provide all things to all people.

Note: I'm almost certainly done buying power amps so any data you produce will be for reference only, at least for me.  But don't let that stop you!

I am mildly irritated by the implication that amp manufacturers cheat or lie about amplifier performance specifications, while over the years some probably have. It was so bad back in the 70s that the FTC imposed regulations, since then not so much. A better use of this community (IMO) is to ask members about their experiences with new amp models to identify the few snakes in the woodpile (if any). Modern technology has made amplifier design easier, I wish we had these class D chip sets back when I was in the trenches.

JR

[Helge A Bentsen]

FWIW burst ratings seem to correlate pretty well with how a modern amp performs when listening to a system. Replaced an old iron amp recently with a new class D for a subwoofer application.
The old iron was specced at 225W/channel, the new one is 80W/channel/200W burst.
The new amp sounds better and goes as loud as the old amp, it's definitely putting out more than 80W when it's pushed with music.
I matched gains on new VS old relative to the mains.

[John Roberts {JR}]

FWIW burst ratings seem to correlate pretty well with how a modern amp performs when listening to a system. Replaced an old iron amp recently with a new class D for a subwoofer application.
The old iron was specced at 225W/channel, the new one is 80W/channel/200W burst.
The new amp sounds better and goes as loud as the old amp, it's definitely putting out more than 80W when it's pushed with music.
I matched gains on new VS old relative to the mains.

+1...

This is TMI but back in the late 80s at Peavey I designed a small audio amplifier with a trick power supply (PMA70+), without getting into too many details the amplifier could put out 1x continuous voltage (35W) 24x7, 2x voltage or 4x power (>100W) transiently for tens of mSec, and 50-60W for tens of seconds. As I expected it sounded more like a 100W amp than 35W, while being sized and costing like 35W.

 I had grand plans to design an entire series delivering progressively more transient power but sadly the technology didn't scale up cost effectively. The expensive high power components were still needed, and the savings in transformer/heat sink components didn't create enough of a value proposition.
======

Of course modern class D technology is a different ballgame with dramatically improved thermal performance. Of course class D has been around a long time (Peavey was making class D amps before I started there in the mid 80s). So while class D is mature the semiconductor industry has stepped up in the decades since then to deliver high voltage at high current switching devices that were only a wet dream back then. I suspect there are also canned class D controller chip sets available now, but I have been out of those trenches for years.   

Of course junior engineers can still screw up canned designs but it appears that modern technology is quite good.

JR

[Tim McCulloch]

So, Nathan-

What are you measuring to find?  What parameters affecting the electrical or thermal characteristics are you trying prove or disprove as having a causal affect on the output signal?

This is like Smaart - you can measure lots of things but determining where and how one is taking the measure is determined by the "why" of the measurement -  what is being done with the results of the measurement.

We can measure a lot of factors in a power amplifier in use or under test, but what does any or all of them have to do with the electricity delivered to the loudspeaker?

[ProSoundWeb Community]