What does Current mean in a power amp???

Ohm's law has nothing to do with whether or not an amplifier can drive a speaker. While an amp may be able to sustain high power levels into low impedances, that tells you nothing about how well it deals with various amounts and types of reactance across a wide frequency spectrum. It is possible for a speaker to have sharp phase angles at several different frequencies all at the same time.

If you can picture a combination of a "slalom course" AND a "torture test" for amplifiers, that is what some speakers present. As such, you can have a speaker that is highly capacitive at treble frequencies, highly inductive at low frequencies and generates a high amount of reflected EMF ( electromotive force or "voltage" ) all at the same time. All of this can be independent of the amount of "pure" resistance that the amp sees at any given time or frequency. This is why speakers are a VERY complex load and why we don't have specific "tests" that show whether or not an amp can drive every load known to man OR maintain consistent sonic characteristics doing so.

Something else that is not commonly considered is that amplifiers produce LESS power as impedance is raised i.e. kind of the "reverse" of looking for high current into low impedances. After all, impedance raises at the point of resonance on a woofer. As such, power transfer is reduced right at the point that you need it most. Since vented systems typically have MUCH higher impedance peaks at resonance, you have even less power available to control a driver that is already lacking "damping". Even though most manufacturers do not offer power output specs into 16 and 32 ohm loads anymore, one might be able to make a more informed opinion about the overall build quality of an amp if we did know such things.

If you think that this sounds "crazy", take a look at the impedance curve of a vented speaker with a large woofer(s). It is not uncommon to see impedance peaks at resonance along the magnitude of 40 - 100 ohms. What kind of power transfer do you think an average SS amplifier is going to produce into a 50 ohm load ??? Let me tell you, not much.

Most "good" sealed designs keep the impedance of the woofers below 20 ohms, which results in much more accurate and controlled bass. This is due to the increased ability of the amp to transfer power and literally "muscle" the cone when it does not want to see any type of signal at all. After all, resonance is nothing more than the speakers' point of self oscillation. If excited at that frequency, it is literally contributing sound on its' own. It is up to the amplifier to "force feed" it at that point and damp / control the ringing that is taking place. Obviously, a lower impedance at resonance allows the amp to generate more power. This in turn can effectively work to control the speaker and produce greater accuracy.

Having said all of that, rms power ratings are WAY to easy to fudge given the way that the FTC has things set up. I think that measured power output at CLIPPING at various impedances is FAR more revealing of how "sturdy" an amplifier is. After all, this is the point of maximum long term stress for an amp. As such, it is the ultimate test in terms of how much TOTAL power the output devices can pass and how much current the power supply can sustain. If an amp can come close to "doubling down" at the point of clipping from 32 ohms down to 2 ohms, it can probably drive just about any load that you can throw at it. To do so would mean that the amp was as close to a pure "voltage source" as we are currently capable of making. Even if the amp is sturdy enough to do something like this, there is no guarantee that you will like the tonal balance or level of refinement & detail that the amplifier produces. Passing the aforementioned test simply shows that it is capable of "brute force" into just about any given load and does not necessarily mean that it will "sound good". It will however, drive the load that you connect it to. Sean
>

My point was that Ohm's law ( used in a generic manner ) is applicable at very specific points. Obviously, one can measure and calculate what is taking place electrically at any given point in the operating curve and break it down mathematically. However, the measurements at any given point may have nothing to do with what is taking place elsewhere within the parameters of operation. This is what makes a loudspeaker a "complex" load. Even if one were to try to break the speaker / amplifier interface down using Thevenin's theory, which is far more complex, those parameters would still vary somewhat with frequency and amplitude.

With the above outlook taken into consideration, the amplifier / speaker interface is one of a dynamic nature that changes with amplitude, frequency and impedance. One can't model a "simplistic" speaker / amplifier interface based on just a few measurements that would otherwise work fine in a circuit with set parameters and limited variables. As such, trying to break the entire amp / speaker interface down to something as simplistic as Ohm's Law would be next to impossible unless the speaker maintained a constant impedance across the entire frequency spectrum at any given amplitude and the amplifier acted as a true "voltage source". Does anyone know of such a speaker ?

The parallels that one can draw using a resistor as a load and that of an actual speaker are far and few between. For each "benign" speaker load that you show me, i can show ten that are not quite so simplistic. As such, how an individual amplifier and the corresponding circuitry will respond to each load that it sees with varied frequency and amplitude becomes completely unpredictable. Hell, some amps even change frequency response aka "tonal balance" as the drive level increases on a dummy load ( non-reactive "perfect" speaker ) on the bench. Who could predict how such an amp would respond with the variables encountered with different speakers and their associated reactances during real world operation??? If you think i'm making this up, John Atkinson noted this in a recent review of a Rotel amplifer in Stereophile and Moncrieff had covered this 20+ years ago in IAR.

While some of the variables can be minimized by having a true "voltage source" as an amplifier and a speaker that was linear in impedance, this still does not take into account how this combo would actually "sound". As i've tried to stress, there are just TOO many variables to try and sum things up "simply". While many folks don't have the technical background to know why such things take place, they have enough experience to know that you will never know exactly how various components will mesh until you try them out within the confines of your system. We can call it system synergy, complimentary colourations, etc... but it still boils down to the fact that there is NO set formula other than the old "trial & error" method that most of us have had to do to get where we are today. If things could be summed up easily using a simple formula, i think that there would be no need for forums such as this. Sean
>

PS... I'm not trying to be argumentative or step on toes, so please don't take it that way. I'm simply trying to say that you don't know what to expect until you try it. I've been "confounded" way too many times before to know that things that should work sometimes don't and vice-versa.

Marakanetz: Since music is typically comprised of multiple instruments being played at the same time, the signal produced can be extremely wide in terms of frequency response at any given moment. All of these notes / instruments can be played simultaneously.

As you know, most speakers are "multi-way" units i.e. have woofers, mids, tweeters, etc... and some may have multiple drive units within each range. Besides the amplifier trying to load into all of the various drive units simultaneously at any given point in time, it also has to deal with the signal dividing crossover network. As such, it is possible for an amp to see very different impedances and levels of reactance at different points within the audible bandpass at the same time. With that in mind, we are now using a complex signal ( multiple notes over a wide frequency range ) driving a complex load ( multiple drivers working within limited frequency ranges ). As such, each driver / frequency range as provided by the crossover network could present a different picture with various challenges to an amp at any given time. You simply can't "sum" all of those variables into one simple equation.

On top of this, reactance of a speaker / individual drive unit can vary with different drive levels applied. Obviously, not all drivers saturate at the same point nor do they have the same amount of excursion capability. While better designs seek to eliminate such variables and / or reduce the potential for what would appear to be a "seam" between drivers in a multi-way system at any given drive level, such is not always the case.

If the "lack of seamlessness" situation does arise, you can bet that part of what you hear is not only attributable to the speaker design itself, but also to that of the specific amplifier being used. The amplifier is responding to oddities in terms of loading within the "problem" region. As such, some amps may respond in various manners to loading conditions and the result is why we have audible differences.

This is the reason that you may think that a speaker has a certain sound to it yet hear a very different presentation when you change amplifiers. The electrical constants of the speaker have not changed. The ability of the circuitry to deal with those electrical characteristics is what changed when you swapped amplifiers.

Given that both amps were being driven within their range of linear operation and were not clipping, how can you mathematically calculate the discrepancies heard and explain why one amp sounds "better" or "different" than the other with the same load on them ??? You can't and there is no specific formula to do so.

If i were to "pick a formula" for a "good" amp, the amp would have a helluva power supply. I am not talking about just a big reserve of staggered value capacitors placed at strategic points within the circuit, but a very powerful transformer with high rail voltages with extended duration current capacity. The circuitry would be very fast in terms of rise time. In order to to have a fast rise time, you have to have wide bandwidth as you can not have one without the other. This insures good linearity within the audible bandpass since the unit is fast enough to keep up with signals above and beyond that range. The amp would also have a very fast slew rate. This would mean that it could respond to changes in amplitude very rapidly. Between having a quick rise time to duplicate sharp directional and / or polarity changes in the waveform, the high slew rate would allow us to do that regardless of intensity or amplitude of those changes. On top of all of that, the amp would have a very low output impedance. This would minimize the ability of the speaker to actually "modulate" the output of the amp. Some call this the "damping factor" of an amp.

The end result would be an amp that was quite stable and retained consistent loading and sonic characteristics into whatever you threw at it. Obviously, "passive" parts quality and circuit layout does count, so one would have to take such things into consideration. If you can get those basic things right, chances are, you'll have an amp that will work well into just about any load you can give it AND sound pretty good doing it. Sean
>

Fbi: Hhang onto your hat. From what i hear through the grapevine, such a "feature" might not be that far down the road. The folks at A-gon have got some really "groovy" tricks up their sleeves from what i've been told.

Marakanetz: I have very limited experience with Plinius products. I have heard a Plinius preamp driving an SA-100 ( don't know if it was an original, Mk II, etc.. ) on one occasion. I was not familiar with the recordings but the front end was a three piece Metronome transport & matching tubed DAC with external power supply. The speakers were custom built sub / sat's with a ribbon tweeter and dual mid-woofers arranged in an MTM. Cables were Kimber if i remember correctly. While i don't know exactly which pieces contributed to what i remember hearing the most, all i know is that the sound was phenomenally hard and sterile and gave me listening fatigue in a matter of minutes. I've been told that the Plinius gear is very smooth and warm sounding, so it makes me wonder just how "bad" the rest of the gear would have had to have been to make things sound the way that it did. Sean
>