A new way to look at amplifier power??

12926 Views 128 Replies 23 Participants Last post by Oliver, Aug 11, 2009

Fellippe

Discussion Starter · Aug 28, 2007

The "watt isn't a watt" argument has been controversial, and I don't intend to start a war on this issue, but here's a conversation between an EE guru friend of mine (who is way stronger than myself in this area) with an interesting theory...

enjoy:
_______________________________________________

You know, CG, I was having the convo with Vu, and it seems like there is a "torque" factor involved with amps that is simply not quoted. As an EE it's hard to rectify POWER that is unquantifiable....very disturbing!

I equate HP to watts...we all get 200 hp, 400 hp, 530 hp (Alpine Diablo!)....but what about torque?

These Phoenix Gold MS amps I have now have a lot of torque....more effortless power. Every amp will give you power when you crank, but how many have it down low off idle?

I think as cheesy as the car analogy is, it might be going somewhere in the right direction...

All of the home setups I've heard so far have been, 2 , 5, 10 watts...maybe a 20 watter.

I still don't get it, lol.

His response:
______________________________
I'm not sure this analogy is as far off as you think. I'm just spitballing here, but it makes sense.

Recall that Power = Current x Voltage. So to make 100 watts, you can either create 1 amp at 100 volts, or 1 volt at 100 amps or anything in between.

However, the nature of cone/dome type speakers is that current through a voice coil creates a magnetic field, which interacts with the driver's permanent magnet to propel or retract the voice coil/cone assembly. But that force is based on current through the coil - not voltage across the coil.

Therefore, an amp which creates power by producing high current has better control over conventional loudspeakers than an amp which produces high voltage but low current. So in this case, as you hypothesized, current would be analogous to torque where voltage might be analogous to HP. And just as engine RPM relates the two under the hood, the impedance of the speaker relates the two inside the amplifier.

It would be interesting to take one of your Phoenix Gold amps and a similarly powered cheapy amp and do a test which is the equivalent of a torque curve for a motor - maybe plot the output current versus input voltage and the output voltage versus input voltage and see how the area under the curves differ for the two amps...

Interesting...

-- cg

See less See more

61 - 80 of 129 Posts

G

· #61 · Aug 29, 2007

chad said:
I just feel that rating an amplifier at sine wave output is a bit unnecessary for a device designed to amplify music, not a high duty cycle tone. This article kind of sums it up, it's not directed at you but for others to paruse. I'm prettyy sure you know what I'm talking about.

http://www.crownaudio.com/pdf/amps/i-tech-livesound_jan07.pdf

Rail voltages at 200V and can be sustained for a few seconds max. But do they need to sustain any longer? That begs the question.

If you asked me about this 3 years ago I would have fought tooth and nail about the new rating system. I liked the "old iron amps." But after seeing the new breed of amplifier in action I can say that it really makes no difference, the new amps, say one that can make huge peak power but only sustain a fraction of that will SMOKE a similar amplifier that can sustain more then the newer ones. So yes a 5KW amp MAY be able to sustain 1KW but will chew an amp alive that can sustain 2KW when it comes down to the real deal. Pushing a bunch of 18's every time that kick drum hits.

It goes right along with the earlier thread of how much juice is REALLY getting pulled down those power leads to the boot.

Chad

I would seriously entertain a suggestion along these lines :

"We know that any complex signal can be decompossed into nothing but a linear combination of sine waves, aka tones. But a single sinewave is not very representaive for two reasons :

1. Intermod distortion requires more than a single freqeuncy.
2. The crest factor (ratio of peak power to average power) of a single sinewave is not representative of music.

We recongize that amplifer design can "exploit" the crest factor of typical music, by storing sufficient amounts of internal energy. This may allow an amplifier to supply peak-bursts, for short periods of time, even though the long-term average power delivered is unaffected by internal energy storage.

Therefore we propose a test consisting of an industry-standard signal, somewhat more complex than a single sinewave, ...."

Know what i mean? First, demonstrate a mastery of simple engineerig concepts. Then, demonstrate why the new proposal is not just more marketing fluff, but real valubale stuff. Then, demonstrate that you're not just going to try to fool the consumer by inflating the results of the existing tests, just to sell your product. Finally, propose the new method.

And yes, i know, my criticisms are kinda aimed wildly ... at manufacturers as well as hobbyists!

See less See more

chad

· #62 · Aug 29, 2007 (Edited)

werewolf said:
I would seriously entertain a suggestion along these lines :

"We know that any complex signal can be decompossed into nothing but a linear combination of sine waves, aka tones. But a single sinewave is not very representaive for two reasons :

1. Intermod distortion requires more than a single freqeuncy.
2. The crest factor (ratio of peak power to average power) of a single sinewave is not representaive of music.

We recongize that amplifer design can "exploit" the crest factor of typical music, by storing sufficient amounts of internal energy. This may allow an amplifier to supply peak-bursts, for short periods of time, even though the average power delivered is unaffected by internal energy storage.

Therefore we propose a test consisting of an industry-standard signal, somewhat more complex than a single sinewave, ...."

Know what i mean? First, demonstrate a mastery of simple engineerig concepts. Then, demonstrate why the new proposal is not just more marketing fluff, but real valubale stuff. Then, demonstrate that you're not just going to try to fool the consumer by inflating the results of the existing tests, just to sell your product. Finally, propose the new method.

And yes, i know, my criticisms are kinda aimed wildly ... at manufacturers as well as hobbyists!

They have this industry standard signal in place, I'll find out what it is. Some folks chose to test with a more rigourous signal though, which, is fine with me.

I'm totally with you though

*edit* Lab Gruppen's testing is a 1K tone 33.3ms on and 66.6ms off. But again, that's not the standard and this is for their max voltage bench "stress test" I'm pretty sure it's the same thing 20-20K for ratings testing. I need to gather more info before I can feel comfortable.

G

· #63 · Aug 29, 2007

Fair enough

SQ_Bronco

· #64 · Aug 29, 2007

tard said:
i can definately understand the situation with the distortion explanation. if i can find the article, i'll post a link. maybe you could evaluate it and see if there's something i overlooked in what was being said in it.

http://www.caraudiomag.com/specialfeatures/0202cae_high_power_amp_test/

is this the article you are thinking about? There have been threads on a number of forums (including this one IIRC, definitely ECA) talking about the problems with the test methodology...

exmaxima1

· #65 · Aug 29, 2007

89grand said:
I'm curious as to what good fuses on the speaker outputs do.

One of the first things I learned out of school was that "transistors are the world's fastest fuses". It is doubtful that a fuse will blow fast enough to save an output stage. Plus, many transistors fail from exceeding their SOA (safe operating area) due to elevated temperatures. When the temps go up, the current ratings go WAY down. It would be near impossible for fuses to reliably protect against that.

Regarding the power advantage of a high-current amp versus a lesser design, keep in mind that the instantaneous DC resistance of a speaker can be much lower than its impedance---until the speaker starts moving and generates a back EMF (raising its impedance), the current can be very high. A high current amp design may very well perform better for transient response.

If you look at some woofers, such as Carver's massive subwoofers, the DC resistance is only a fraction of its AC impedance throughout its usable frequency range. Though the amplifier may be marketed as "2000 watts", that is predicated on a 4 ohm load while the actual working impedance may be more like 20-40 ohms over the subwoofer's range. The amplifier does not work so hard at those impedances.

Matthew

See less See more

chad

· #66 · Aug 29, 2007

Fellippe said:
Look Wolf, I'm not even claiming to be an EE know it all....just throwing out ideas that sound good on paper. On a daily basis I see the difference between using single phase 120/208, three phase 120/208, and three phase 277/480, and how the effects of going up in voltage are beneficial in reducing current.

I see where you are coming from but remember one thing, the amp switches voltage, the current delivered is derived from the loudspeaker impedance.... which does not change.

In your app you are raising the voltage, lowering the current but the "impedance" of the device also is raised to accomodate the higherr voltage present. If you took a 120V motor and ran it at 220V.... see what I mean.

In a lamp (because they are not bulbs) that is intended for 220V use has a thinner/ less resistive filament then one designed for 120V operation.

And yes, lamps intended for use overseas are not as durable as ours.

chad

· #67 · Aug 29, 2007

exmaxima1 said:
One of the first things I learned out of school was that "transistors are the world's fastest fuses". It is doubtful that a fuse will blow fast enough to save an output stage. Plus, many transistors fail from exceeding their SOA (safe operating area) due to elevated temperatures. When the temps go up, the current ratings go WAY down. It would be near impossible for fuses to reliably protect against that.

Regarding the power advantage of a high-current amp versus a lesser design, keep in mind that the instantaneous DC resistance of a speaker can be much lower than its impedance---until the speaker starts moving and generates a back EMF (raising its impedance), the current can be very high. A high current amp design may very well perform better for transient response.

If you look at some woofers, such as Carver's massive subwoofers, the DC resistance is only a fraction of its AC impedance throughout its usable frequency range. Though the amplifier may be marketed as "2000 watts", that is predicated on a 4 ohm load while the actual working impedance may be more like 20-40 ohms over the subwoofer's range. The amplifier does not work so hard at those impedances.

Matthew

Excellent post!

It's always a rough battle in the "to fuse or not to fuse" speakers, I am very much against it. One thing is for sure... you will either be blowing speakers or replacing fuses all the time and a fuse knows no impedance rise of the VC due to heating. Best bet is to scale the gear for the intended job so you don't have to fuse.

Also keep in mind (not you matt) that there was an amp pictured on here a bit back with what appeared to be speaker fuses. I'm pretty darn sure those are rail fuses, which, is not a bad idea when the "world's fastest fuse" fails to short to the rails

World's fastest fuse, man, I haven't heard that one in a long time

G

· #68 · Aug 29, 2007 (Edited)

exmaxima1 said:
One of the first things I learned out of school was that "transistors are the world's fastest fuses". It is doubtful that a fuse will blow fast enough to save an output stage. Plus, many transistors fail from exceeding their SOA (safe operating area) due to elevated temperatures. When the temps go up, the current ratings go WAY down. It would be near impossible for fuses to reliably protect against that.

Regarding the power advantage of a high-current amp versus a lesser design, keep in mind that the instantaneous DC resistance of a speaker can be much lower than its impedance---until the speaker starts moving and generates a back EMF (raising its impedance), the current can be very high. A high current amp design may very well perform better for transient response.

The impedance curve of the speaker reveals ALL. At DC, the coil of any speaker is not moving. Most drivers rated at 4 ohms nominal impedance have a DC coil resistance of 3 ohms. If your amp can drive 3 ohms ... meaning, it can supply enough current to sustain the driven voltage across a 3 Ohm resistor ... it will be indistinguishable from any other amp, even if the other amp has 1000x the current delivering capacity, able to drive a 0.003 ohm resistor to the same voltage.

But the real point is : the current is determined by the load, not the amp (until the amp current-limits), no matter if it's advertised as "high current" or not. This is a simple consequence of the amplifier emulating an ideal voltage course.

Finally, "DC" and "transient" are completely opposite concepts. "Transient" is, by definition, a short-term (therefore high-frequency) effect, while "DC" is a long-term (therefore low-frequency) effect. Consequently, the ability of an amplifier to drive the DC coil resistance will have zero impact on transient performance.

A better argument would have been electrostatic speakers, whose high frequency impedance can drop to scary-low levels (due to their capacitive nature). Here, an amplifier may exhibit transient difficulties if it can't deliver the required current.

But in any case ... the driven voltage is determined by the amplifier, the drawn current is determined by the load.

See less See more

chad

· #69 · Aug 29, 2007

werewolf said:
But the real point is : the current is determined by the load, not the amp (until the amp current-limits), no matter if it's advertised as "high current" or not. This is a simple consequence of the amplifier emulating an ideal voltage course.

One thing I like to explain to people is that a "high current" amplifier is simply an advertisement that states that the amplifier can handle low impedance loads without the power supply/final section falling to it's knees.

Yes an old PG MS275 is high current, but at 4 ohms puts out just as much power AND CURRENT as any other amplifier with the same rating... because the rail voltage is the same. BUT the MS275 has a substantial power supply AND the output section that allows it to be stable at insanely low impedance loads, thus allowing the marketing department to claim "high current" when in fact any understanding of ohm's law would make that a given

Chad

durwood

· #70 · Aug 29, 2007

chad said:
One thing I like to explain to people is that a "high current" amplifier is simply an advertisement that states that the amplifier can handle low impedance loads without the power supply/final section falling to it's knees.

Yes an old PG MS275 is high current, but at 4 ohms puts out just as much power AND CURRENT as any other amplifier with the same rating... because the rail voltage is the same. BUT the MS275 has a substantial power supply AND the output section that allows it to be stable at insanely low impedance loads, thus allowing the marketing department to claim "high current" when in fact any understanding of ohm's law would make that a given

Chad

Exactly, just like the old soundstream amps too. Note how the power ratings go down when you switch the amp to "high current" mode IF you keep the load the same.

Those of course are the same amps that advertise themselves as "Class A" even though they are not true class A. Marketing...

G

· #71 · Aug 29, 2007

Fellippe said:
I couldn't get a degree in EE without knowing Ohm's law, Kirchoff's Current & Voltage Laws, Thevenin/Norton with dependent and independent sources, small/large signal analysis of BJTs, MOSFETS, etc.

There's no confusion here, Werewolf.

Where the confusion lies is hearing differences in similar rated amps, and seeing some amps with lower ratings outperform those with higher ratings. You choose to believe a watt is a watt (which is fine), and I after having these experiences begin to QUESTION whether that's true.

Wrong. I never said "a watt is a watt" ... in fact, i've argued VIOLENTLY to the contrary. We can set up a 1kHz sine wave, delivering exactly 1 watt into a louspeaker, and compare it to a 1kHz square wave, also delivering exactly 1 watt to the loudspeaker. I promise ... I can hear the difference. And I know exactly why ... it's because all watts are NOT the same. Regarding the ability to hear differences in amps ... first, prove the difference exists in a double-blind test. Then, prove that the difference cannot be attributed to : gain, power, frequency response, noise or distortion. Until then, what you've heard is a frequency response, gain or power error ... nothing more. This has been demonstrated time and again ... it has nothing to do with an amp trying to force a voltage and current to a load. But it has everything to do with the FACT that amplifiers sound different!

My problem here is that I propose a theory from someone who thought it out a bit, and all that anyone has been able to come up with is to wait around to find a way to use Ohm's law to disprove it???

No need to wait. If the theory is disproven by Ohm's Law, as I have done, the theory is junk. I'm sorry to disappoint you.

Is real world resistance always fixed??? Is that first poster who suggested the resistance is the same automatically correct? (Are you automatically correct?)

No, resistance is not always "fixed". But as i've tried hard to demonstrate, if we don't understand how a "fixed" resistor works, how can we hope to understand something more complicated?

The concepts of current and voltage take a good amount of time to truly "get". I=V/R is simple math, but understanding what current really is and what voltage ("potential difference") is takes time. Did anyone care to read or even challenge the thought that "current" controls speakers better than "voltage" does.....the only argument was to spin it that they'd always be equal, invalidating the hypothesis.

Look Wolf, I'm not even claiming to be an EE know it all....just throwing out ideas that sound good on paper. On a daily basis I see the difference between using single phase 120/208, three phase 120/208, and three phase 277/480, and how the effects of going up in voltage are beneficial in reducing current.

You're barking up the wrong tree if you think I don't "get" simple EE, and I'll go out on a limb and say that your mastery of EE concepts reeks a lil more Ph.D. than P.E.

thanks for the compliment!

See less See more

G

· #72 · Aug 29, 2007

Fellippe ...

I'm not trying to be a jerkoff. All i'm asking, as gently as i can, is this :

Before we discuss a "new way to look at amplifier power", can we please understand the "old way"? It's been my experience that many who love to debate amplifier tests and measurements, do not even understand the fundamental foundations of basic electronics.

If that doesn't include you, my apologies.

MarkZ

· #73 · Aug 29, 2007

exmaxima1 said:
One of the first things I learned out of school was that "transistors are the world's fastest fuses". It is doubtful that a fuse will blow fast enough to save an output stage. Plus, many transistors fail from exceeding their SOA (safe operating area) due to elevated temperatures. When the temps go up, the current ratings go WAY down. It would be near impossible for fuses to reliably protect against that.

There are instances where the SOA of the transistor isn't ideal for the design, and so the current may be the bigger factor if you're pushing it on the low slope part of the SOA curve. The rail fuses aren't going to do much save the output devices (although I guess they could for a mono amp). Also, if the output devices fail short then the output fuses might be able to save the speakers.

Oliver

· #74 · Aug 29, 2007

So if a high current amplifier is the way to go, then you will need to get some speakers rated as low as possible ohmage wise.

Some 2 ohmers for the components, or a combination of 4 ohmers in parallel.

A sub with DVC 1 ohm coils in parallel for 1/2 an ohm, should do the trick.

And by all means, please, let us know what your findings are!!

MarkZ

· #75 · Aug 29, 2007

Hic said:
So if a high current amplifier is the way to go,

How did you arrive at that conclusion based on what everyone has said here?

Oliver

· #76 · Aug 29, 2007 (Edited)

Quote[The concepts of current and voltage take a good amount of time to truly "get". I=V/R is simple math, but understanding what current really is and what voltage ("potential difference") is takes time. Did anyone care to read or even challenge the thought that "current" controls speakers better than "voltage" does.....the only argument was to spin it that they'd always be equal, invalidating the hypothesis.]quote

Not everyone, there is one person who is suggesting this and a test to determine it, or help validate his thoughts would be what I have suggested.

If this is still confusing to you I would submit the above ^^^^ words, beginning of sentence[" Did anyone...."

MarkZ, does this help....
Quote[Therefore, an amp which creates power by producing high current has better control over conventional loudspeakers than an amp which produces high voltage but low current. So in this case, as you hypothesized, current would be analogous to torque where voltage might be analogous to HP. And just as engine RPM relates the two under the hood, the impedance of the speaker relates the two inside the amplifier.]quote

See less See more

Oliver

· #77 · Aug 29, 2007 (Edited)

excerpt/Quote[...and how the effects of going up in voltage are beneficial in reducing current.]quote

I would submit that Georg Ohm proved this!!

Unless I misunderstood his law, I thought there was an inverse relationship between the two

Or, in other words, if one of them goes up, the other one of them goes down.

Fellippe

Discussion Starter · #78 · Aug 29, 2007

Hic said:
and how the effects of going up in voltage are beneficial in reducing current.

I would submit that Georg Ohm proved this!!

Unless I misunderstood his law, I thought there was an inverse relationship between the two

Or, in other words, if one of them goes up, the other one of them goes down.

Well yeah..

I was just saying that the benefit of reduced current is smaller wiring/conduit.

A wire is sized for a certain amount of amps, regardless of input voltage...e.g. 120, 208, 277, 480, etc.

But that's another thread.

Fellippe

Discussion Starter · #79 · Aug 29, 2007

werewolf said:
Fellippe ...

I'm not trying to be a jerkoff. All i'm asking, as gently as i can, is this :

Before we discuss a "new way to look at amplifier power", can we please understand the "old way"? It's been my experience that many who love to debate amplifier tests and measurements, do not even understand the fundamental foundations of basic electronics.

If that doesn't include you, my apologies.

It's cool, no beef.

What are your thoughts on Darlington transistors for the output stage?

How does that rank up with other types of output stages?

durwood

· #80 · Aug 29, 2007 (Edited)

.

61 - 80 of 129 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.