how does current work in an amplifier?

Specs are like statistics.

The only statistics you can trust are those you fabricate yourself.

Atma,
Just read your reply from last month to my specs comments.

You could be right. Bottom line? Don't believe everything you read, even if you think you know something about it....

figures lie and liars figure......

And yeah, my now ancient, long gone Carver Cube was 0.5% distortion and who cared? Sounded fine at the time.,

Dsper,

I've heard krell and Martin Logan and teh results were very impressive.

Haven't heard Krell and Maggie but my gut tells me that is a very good match as well!

Glad to hear things are on a happy path!

Mapman,

The last few weeks have proven to me that your words about matching components are true.

I had always wanted a Mac amp and got one and was happy with it until I got my Maggie 1.2's. At that point, the power into 4 ohms question came up.

I solved it by trading the nine month old Mac for Krell and paid a piece of change to make the trade.

I guess my only excuse is that I bought the Mac before I had heard Maggies. I will never forget that day - I was listening to Paradigm and Klipsch at a local dealer and he suggested that I might try the Maggies. The superior (to my ear) openess and resolution blew me away!

Anyway, it is now Krell and Maggie 1.7's.

I need to be happy with this for at least a while or the most beautiful woman in my life will shoot me dead!

Specs done well are useful for helping determine up front how well components might work together. I definitely recommend using specs for this purpose to get where you want to be faster rather than putting pieces together blindly and hoping for optimal results.

Less useful for determining how good a particular piece is. Since it always takes two or more components to actually produce music, this is really not very important anyhoo.

Magfan, isn't that the other way 'round? :)

Specs are advisory to the knowledgable and misleading to those who aren't.

Dsper - the owner at the store I deal with is constantly telling me to ignor specs and just listen. He's given me some extrememe examples of equipment that he's heard where the actual sound was the opposite of what was expected. It was something crazy like 9 wpc and a THD in double digits that he said was amazing.

He also noted that there isn't really any true regulations going on to verify specs. Many of the specs are taken at the one and only ideal frequency. Think Ford commercials.

Unsound, most of the studies I am aware of in this area are so new that they are not published yet. However, one of the people conducting them is a Nobel Prize neuro-chemical scientist. I understand that High Emotion Audio has based a lot of their work on his studies.

Ear is most sensitive to THD at higher frequencies. I'll let the guys that know fill in the blanks. Your ear is most insensitive at very low frequencies......which is why subs can get away with 5% distortion or whatever.....which, if it happened at 1000hz, would drive you from the room.

Don't worry about specs. Their is a relationship to feedback which some persons object to. Just for example, the Japanese amps of the 70s had amazingly low #s, but were high feedback designs. Again, the knowledgable ones will be along in a minute.

As for Emotiva? You could build the identical design of amp....#1 used all premium components, matched semiconductors, huge heatsinking on outputs, heavy gauge chassis, transformers it takes a forklift to move, and other 'quality' touches. MTBF is very high. Before being mass produced the piece was 'voiced' and sounded GOOD.

#2? Built with parts from Fry's or some catalogue. Just the values called out, with no matching, or thought of other characteristics of the part.....Caps have a resistive and an inductive component, for example. Everything else was built or sourced with price in mind. Meets specs? Sure. Sounds good, too....but not quite up to #1s level. More production variance and less lifetime. Sales price point?.....A heck of a lot lower than #1.

Just my opin.

Another question that comes to mind when selecting an amp is Total Harmonic Distortion.

I read on another Audiogon thread about what amp for Magnepan 1.7's. One suggestion was an Emotiva Audio XPA-2.

The EPA-2 is advertised at 300wpc @ 8 ohms and 500 wpc @4 ohms - at less than $1000!?!

However the THD is listed as less than 1% compared to Mac's, for example, which are listed at less than .004% THD.

So what's up with THD? When can you hear it? How do you hear it?

Martykl,

There is a Magnepan dealer in Columbus, Ohio who pairs the speakers with McIntosh as well.

The last time I was there, they had 3.6's with something like 400 watt monoblocks. Very real sounding without too much high end in the treble decay.

If I'm ever up in LaLa Land, I'll be sure to drop in. Woodland hills is a nice place.

I'm way South down the coast in SD. It was where wife and I could afford to buy when we got hitched....back in '86. I love it down here.

My local Magnepan dealer demos his speakers with Mac amplifiers. I haven't really listened carefully there for many years, but I don't recall hearing any obvious issues with the pairing. Presumably, the dealer (Shelley's Stereo in Woodland Hills, Ca) doesn't either, since he's paired the brands for quite a while in his showrooms.

Marty

I hear about few people who have Mac/Maggie. Not to say it wouldn't work or isn't a great combo.....just not 'common'.
Expensive, too, if you go to the MA6900 / MA7000 level.

I had a Rotel RB1070 of 130x2 @8 with NO factory 4ohm rating. That should have been a 'clue'. The amp simply didn't have the guts at higher levels. The Rotel was also 360 'bridged' to which I took to imply 180@4.

When I pulled the trigger on an upgrade I went with nearly 3x the power in a 'd' amp. Even though 'd' amps are time limited at max power, the dynamics are there, which is all that really matters, as far as power goes. IMO.

Hi Magfan,

I always wanted a Mac and figured it for good resale if it came to that.

I want the slam without the treble going over the top. Mac seems pretty mellow in that regard, especially if I decide to upgrade to MG 1.7's.

My 1.2's are less than one year old and I can get what I paid for them from my dealer if I trade up within the year.

Understand your comment about doubling power. My thought was the MA6900 or MA7000 at 200wpc or 250wpc.

The other side of this is maybe the panels have to go. I wonder about the new version of the Ohms?!

Atmasphere, all very interesting, and if you could kindly provide a link, I'd very much like to read it. With that said, I'm curious as to who determined, and what are the "perceptual rules", at what point(s) are variances to these "perceptual rules" deemed "violations" and does it really matter at what part of the brain we get our enjoyment? In the end I consistently seem prefer the sound that systems that have speakers that do better with ss amps provide over the alternatives. Obviously I'm not alone.

You bring up a good question and I'll be interested to read what others think.

I know that to go louder will start requiring wacky amounts of power. If you go for more power, don't even think of anything less than double+ what you now have.
Also, panels will only go so loud.
You MAY end up moving away from Mac. Sorry, but you are one of the few I've read who make that pairing.
If you can find a Bryston or Pass of similar power, you may find the 'missing slam'.

Guys, your knowledge is impressive, but gets confusing for the relative neophyte!

The main takeaways that I got from the last several threads are:

1) do not get hung up on current because Ohms Law demonstrates that current really does not tell the story - if there are watts, then there is current, and vice versa.

2) most music is played at the low end of the watts - rarely does one continuously use much more than 10 or 15 watts.

So.....if I am running a Mac MA6300 integrated at 100 wpc @ 8 ohms and 160 wpc @4 ohms....with the volume control no more than half way....and at this volume level, the meters jump to 100 wpc into 8 ohms just occasionally....but I would like it little bit louder...and I feel that I am missing something.....

Does this mean I am missing the top end of the dynamics due to the amp's powerguard feature kicking in to prevent clipping?

Does this mean that a Mac with more wpc power would give better dynamics (as well as more loudness)?

Or does it mean that the Magnepan 1.2's just lack that last bit of slam because they are not box speakers?

I realize that I am mixing dynamics and loudness here but I think that you can get my drift....

Your continued discussion and feedback, please.

And thanks!

Unsound, There has been testing in the laboratory that has quantified subjective listening objectively. What the testing has shown is that if the sound system violates human perceptual rules, the processing of the music moves from the limbic system to the cerebral cortex. Pretty interesting stuff.

One possible solution to giving an amp some 'figure of merit' based on reactive loads is to Standardize such a load and let amps fight it out.

Also, I've seen an amp tested the following ways.
Into a resistive load....Try it at 4, 8 and 16 ohms.
Into a +45degree reactive load at the various resistances
Into a -45degree reactive load at the various resistances.
Afte the above, you can draw a very nice plot....would be a 3-d plot and visually tell how an amp performs into real world loads at a glance.

However, for easy of testing, the 'standard' speaker load makes more sense.
It could be reproduced anywhere at any time by any competent tech and the results would compare with all others using the same method.

But, that's just my take.

Atmasphere, of course many of find some ss amps offer a balance of both objective and subjective performance that is preferable. I am confident that many speaker designers welcome the opportunity to avail their design talents towards making speakers for those ss amplification attributes for the sheer performance opportunities those combinations provide.

Unsound, certainly! I think we can add to that, heat, as well (although I know of transistor amps that run as hot as tubes). We are talking about a very traditional set of tradeoffs when these aspects are brought into the conversation: convenience vs performance, an issue you see in nearly every field of endeavor.

Mapman, Of course semiconductors have improved immensely since the beginning of the art! I'm not saying there is anything wrong with 'progress' in of itself. I'm just pointing out a simple fact.

Dob, I like the idea of EPDR; like other specs it does not appear to tell the whole story, but certainly would standardize a lot of things that can be hard to winnow from the other specs you see. IME, Magnaplanars seem to have an easy EPDR rating compared to a lot of 4 ohm speakers I have seen. This would be very apparent if such a spec were commonplace.

Atmasphere, with all due respect there were other incentives such as convenience and reliability amongst others.

Atmas,

I don't think the SS devices of 40 years or so ago can be compared apples to apples with their tube predecessors.

True, probably that SS was cheaper power technology than tubes, but most good vendors tended to maintain and add value with new features that could now be offered and still make a profit.

There is nothing evil about progress. It usually involves doing things more cost effectively in conjunction with providing greater overall value.

Its many years later now. Tubes have their advantages and disadvantages just like everything else. Their disadvantages overall led to their falling out of favor. Not to say that they are not capable of doing unique things in regards to sound quality overall still these days, but there is more than one way to skin a cat....

Athmasphere, I have a question for you.

First, background: In this discussions most of participants when refer to impedance mean only its modulus.

from Simon Thacher of Spectron paper
=====================================
... loudspeaker loads comprise complex impedances with both resistive and reactive (capacitive and inductive) components. Such impedance can be depicted as a vector with its magnitude (modulus) and angle (phase), and both vary with frequency. In general, the voltage and current waveforms, in complex impedance load, are out of phase with each other and therefore, to characterize accurately a speaker's load impedance, both modulus vs frequency and phase vs frequency plots must be known. Frequently, the phase angle is much more crucial to the speaker load than the modulus alone

To deal with this phenomenon, Keith Howard ("How Much Power Do I Need?" Hi Fi News July-Sept, Nov 2007 ) introduced the figure of merit he has labeled Equivalent Peak Dissipation Resistance (EPDR). This is, simply, the resistive load that would give rise to the same peak power device dissipation as the speaker itself. Using EPDR as a figure of merit, the speakers can be compared directly with each other.

Sorry for tbe long introduction: Question is simple: do you agree with Keith Howard figure of merit (EPDR) and if so why speaker manufacturers are not utilizing it, and amplifier manufacturers are not demanding it?

Thank you.

A nice thing to do if you have Magnaplanars and tube amps is to back the amps up to the crossover and use a speaker cable that is as short as possible- 6 inches is a good length, and make the connections as tight as possible. This can have a profound effect on the bass impact!

Unsound is right, tube power is expensive relative to transistors. This has been the case since forever; in the old days when transistors were coming in, manufacturers realized that they could build a transistor amp for about 1/10th the cost of tubes, but were able to charge about 90% of the retail. There was a tremendous financial incentive!

Speaker manufacturers, seeing that SS amps could double power into 4 ohms, began producing 4 ohm speakers that had larger voice coil gaps (easier to make). This cut their driver cost by a similar factor, and again it was possible to charge nearly as much- another financial incentive! IOW, it was cost and the ability to make more money that drove the rise of transistors and low impedance speakers in the 60s and 70s.

I've heard Maggies and tube amps work very well together.

The dealer did use a REL sub also though however.

I would strongly consider a tube amp were I to reacquire a pair of Maggies someday.

My caveat is tube amps and Maggies are both relatively finicky solutions in terms of what is needed to optimize results, so you have to be willing to deal with that.

Tony, except that tube amps tend to be more expensive and high effiency speakers tend to need help extending the bass, which doesn't come cheap either.

That is a perfect conclusion Magfan and Atma. Speakers and amps work as a pair. And I agree with Atma that speaker impedances dropped when SS became more common. High impedance, high efficiency speakers with a tube amp is probably the most cost effective quality sound/dollar. And here I am sitting at the opposite end of the spectrum with a massive amp, low impedance speakers. It sounds great, but I have to alert the power company whenever I plan to crank it up:)

AHHHHH! Atmo:
Good answer! Also, tube amps are considered a 'current source', right?
So the right speaker for a tube amp differs from the right speaker for a SS amp.

That's why the 'comparison' threads between tube and SS are not quite......meaningful.

I'd LOVE to hear a 100x2 tube amp with my panels. I'll bet it'd work just fine, thank you very much!

Yeah, Tony, That's why I doubled the 8vac I saw flicker by, TO 16v for my calculations.
I am well aware of power factor and the math involved. cosine of the angle and all. I've been a 'fan' of that approach to evaluating speakers as 'good' or 'bad' load for quite a while. Tube amps, for example simply do not like certain reactive loads......I don't 'member if it is capacitive or inductive......

People are stuck in the impedance / sensitivity paradigm for 'goodness' of load and usually fail to consider reactance.... / power factor.

And, from our FWIW department, a good cheap addition to your kit may be the purchase of a 'kill-a-watt' meter. Reads out KWH, voltage, PF, Watts / VA and current. Not bad for 25$

Anyway, I may have simplified the math, but in principle I think I'm right. People do not need these wacky current ratings which are 1. meaningless 2. not measured to a standard.
People also use far less power than they think they need. I doubt I've ever been more than about 15% of my amps RMS rating. Most people also really love their 'd' amps and the artificially hi power rating. The ASP1000 module, for example has only a 30 second rating while my ASP500 module has a 60 second rating. Even counting a 10x crest factor, I doubt I've even approach 200 watts total output....and that is LOUD.

The are some misconceptions about current. Here's the big one:

http://www.atma-sphere.com/papers/myth.html

Others have already touched on the serious aspect of this, the idea that for 200 watts you need 7.07 amps (if a 4 ohm load) which is not really all that much.

Another way of stating 'high current amplifier' is to say that it behaves as a nearly perfect voltage source. Confusing?? yeah. Here's how it works: if the amp makes 28 volts into 8 ohms, that is 98 watts. Now if it is a 'voltage source' it will make the same voltage into 4 ohms and that is double the power, since the current increases as the impedance decreases.

The *real* issue is, is this behavior correct for a panel speaker? For certain box speakers it certainly is, as there is a woofer in the box that has a resonant frequency and that is represented by an impedance peak. With a 'voltage source' amp, that would cause the amp to throttle back its overall power as it encounters the peak. This keeps the frequency response flat.

So is this the right response for a flat panel? The panel has an impedance curve that has nothing to do with a box, and it does have cancellation as you approach the low frequency cutoff. You might want more power, if anything, as you approach the cutoff, to maintain flat response.

Tricky.

This might be why there are so many advocates of tube amps on Magnaplanars; A tube amplifier will try to make something more like constant power into a load rather than constant voltage. This means the voltage and the current will vary with the load.

IME you don't need all that much current to make Magnaplanars sing, the amp just has to be comfortable with the load, which is not that hard. Otherwise they are fairly easy to drive.

Hi Magfan, I believe your DVM is giving you an average RMS voltage reading. It's response is not fast enough show you peak values. Also, if you were playing a sine wave, only then would it be an accurate RMS voltage reading. An oscilloscope is the only accurate way to look at peak voltage/power. Speakers are motors, not resistors. So they have impendance due to back EMF, not really pure resistance. The music signal is AC and so Phase Angle is part of the Impendance/Power Factor. On top of that, the crossovers play into the impedance and phase angle too. So you can see how things get really complicated from a math perspective.

Looking at the question from an 'ohms law' perspective, it is a red herring question.
Even my (alleged) current hungry Magnepans have only a 4 amp fuse on the mid / tweets. Say the low is good for 6 amps.....so the speaker can take a total of 10 amps.
I2R=watts.....so 10*10*5=500 watts at about 5 ohms. That's about it.
Well, the other way to go about it is to MEASURE your speaker. I stuck a DVM on my panels......About 8vac was about the highest value I saw flicker by. Say I double that to 16vac. Since Watts=E2/R That gives me just over 50 watts.
I'll give everyone the benefit of the doubt and double that twice for 200 watts per channel into my panels.
That, sir, is loud....no LOUDER
That is why my opinion, strongly held is that amp amps while not meaningless sure depend on the ability of it to drive those amps at a reasonable voltage to provide POWER.

Don't forget the SOR of a semiconductor. Safe Operating Range. You'd need a real bunch of 'em to sustain 50 amps at 50 volts..........=2500 watts.

The KISS answer to your second question is YES. An amplifier rated for a continuous 4 ohm load should handle 4 ohm nominal speakers better than an amplifier that is only rated for a continuous 6 ohm load. You're probably safe to make comparisons within a brand, but I don't think advertised specs really mean much when comparing different brands.

It just depends on the amp-speaker combination and how high a listening level you prefer. Once the power supply of the amplifier is maxed out, the output signal voltage is not going to be able to match the input signal waveform, ie. distortion. The bass requires the most power, so it is the bass that starts to sound flabby or weak. At least, that is my experience...

Actually, as 4 ohm speakers go, I think Magnepans are less current hungry than many. They do like lots of power though.

So.....all other things being equal (same room, same source, same cables, same other characteristics of the amplifier), am I to understand that an amp that delivers more current would do a better job of driving a 4 ohm speaker, like Magnepan MG12's, than would an amp with less current?

PS. I should mention that applying a dead short for smoking affect will require extensive repair of your amplifier. Also, a dead short will not hurt a transformer coupled tube amplifier, but would not be good for the tubes in an OTL version.

A high current amplifier has a very large transformer with a compliment of large capacitors. A SS amp has an array of semiconductors (transistors or mosfets) that are either cascaded or cascoded. These devices are connected to the transformer and capacitors and function as variable resistors to create a large, powerful signal that is identical to the small input signal coming in from the preamp. That large, powerful signal drives the speakers. The Capacitors and Transformer are energy storage devices. They provide the peak power needed as the music signal varies from low to high. Imagine having a big rheostat that ranges from -60 to +60 volts that is connected to a speaker. If you could turn the dial back and forth fast enough to mimic a musical signal- you could make music. A tube amp functions much the same way. The tubes get the preamp signal and vary their high level output voltage to mimic the musical signal. The output of most, but not all tube amps goes through transformers. That limits their current carrying capability and that is why tube amps generally have the same power rating for 16, 8 and 4 ohms. SS amps can output higher power levels into lower impedance speakers up to the point that the semiconductor devices overheat and go up in smoke. To demonstrate, apply a dead short across your SS amplifier. (Although many modern SS amps have overload protection devices which kind of takes the fun away). Very simplistic terms, but that is how any amplifier works.

The math is easily obtained, yet a little 'vague' in terms of practical application by the typical user of audio goods.
That's why I tried to use a real world example, certainly a more difficult to write, yet more graphic representation of the answer, hopefully, to the question of the posting a'gon member.

KISS method, lol.

Larry

I can't give you an in depth explanation. But if you look at the listed specifications for any given amp you should see the Amperes (Amps) that the Amplifier is capable of delivering. You should find a specification for both continuous and the peak number of amps the amp can produce.
The cut off, delineating a high current from an ordinary (non high current) amps has not been officially established AFAIK. But you can do your own informal survey of the big amps with reputations for handling demanding loads and following those specs as the amplifiers get progressively more powerful, to get a general idea of the range you are considering.
My own completely arbitrary idea of the lowest limit for a "high Current Amp" is about 20-30 amps continuous. but be aware their is alot more to it than this single figure.
I am prepared for the onslaught of scientific explanations to follow which will point out the critical importance of other specs to consider. Mine is a simplistic view.
BTW a really big amp will put out over 100amps non peak just so you know. Pay some attention to the slew rate as well which tells you how fast the amp can deliver that power when needed.

A formula that might help is the following:

Current in Amps SQUARED = Power in Watts DIVIDED by Resistance in Ohms

For comaprison:
100 watts into an 8 ohm speaker requires 3.54 Amps
100 watts into a 4 ohm speaker requires 5.00 Amps

Not sure if this helps at all, but...

EXAMPLE
Think of it this way--a garden hose of 1" diameter (diameter represents resistance) is presented with X amount of water (water represents current), and under these parameters, waterwill squirt 25 Feet.
Let's lower the resistance (increase the hose diameter to 4") and present the same amount of current (Water)....now the water simply trickles out the end of the hose.
So, and hopefully this is somewhat clear, the less the resistance of the speaker (impedance) the more current the amplifier has to product for all else to be equal.
In the absence of proper current the amplifier can and probably will, go into 'oscillation' trying to produce current with the transistors, for want of a simple example, and literally burn itself up.

Current is the ingredient which 'drives' the speaker, so to speak. For convenience in 'matching' speakers with amps, Dsper, manufacturers rate the speakers as to their impedance. Generally giving a 'nominal' figure. Speakers are generally REACTIVE not purely resistive, meaning that at different frequencies they present different resistance it's an approximation.
An amplifier is thought of as 'high current', if it keeps doubling into half resistance. Example: 100 WPC into 8ohms, 200 WPC into 4 Ohms, and 400 WPC into 4 Ohms.
The less resistance the speaker presents, the more current the amplifier must produce, in order to 'fill the pipeline.'
I hope this is clear.

If you have further questions you can reach me at lrsky@insightbb.com or I can supply a phone number, and we can talk in person. Actually, matching speakers and amplifiers is an important goal in audio in a host of areas, this being just one.

Good listening.

Larry