ALK Crossovers

[Honeybadger]

Has anyone taken into consideration that amplifiers typically make 2-3 time more power into 8 ohms than 24 ohms.

 

Just food for thought.

 

HB

 

The Interrelationship of Speakers and Amplifiers

We often hear how one amplifier is a good match for one speaker, a poor match for another. Why is this? What makes the association succeed or fail? Let’s look into the electrical relationship between the speaker and the amplifier to learn more about the interactions that occur. 

The Loudspeaker

What makes one speaker an easy load and another hard to drive? There are a number of answers, but the two most important are impedance and sensitivity. First, impedance[1]. Impedance is resistance that varies with frequency. The fact that the speakers’ impedance varies with the frequency it is reproducing, is one reason why the amplifier has such a tough job.

For example, let’s examine a speaker with a nominal rating of 8 Ohms. The term nominal, means average, thus is not only possible, but quite likely that significant impedance variations will occur throughout the speaker's frequency range. Indeed, a variation as wide as 3 to 40 Ohms is possible. The amount of variation, in addition to how low or high the range, determines how difficult the speaker is drive, and thus defines the amplifiers role. The amplifier must be able to deal with these impedance variations, producing the amount of power necessary to drive the speaker at any frequency. If the amplifier is not capable of dealing with the impedance swings, audible distortion occurs.

The Amplifier

Ideally an amplifier should be a constant voltage source. That is, for a given input signal, the amplifier should produce a constant voltage across the speaker terminals whatever the load. For example, if the amplifier is producing 20 Volts at the output terminals, Ohms law (R=V/I) tells us that there are 50 watts being fed into an 8 Ohm speaker (watts equal voltage squared divided by impedance). If we connect a 4 Ohm speaker, halving the original load, the same 20 Volts would now produce 100 watts, and further, 200 watts into 2 Ohms. From this example we clearly see that each time the load resistance is halved, the amplifier should ideally double its output. This high current capability is especially important if the loudspeaker impedance dips into a very low range.

To make this happen, the amplifiers’ the power supply must also double its current delivery to the output transistors for this equation to hold up. The continued doubling must stop at some point, the progression cannot go on forever and, if carried too far, could end in disaster. Either the power supply will run out of current and fail to maintain the amplifiers output wattage, or worse, go beyond the capability of the output devices, creating excessive heat and eventually destruction of the transistors.

Even a speaker with a nominal 8 Ohm rating can fall below 4 Ohms at certain frequencies.  If the current reserves of the amplifier are not sufficient to sustain its output wattage into low impedances, the unit will "run out of gas" sonically, at the time when the extra power is needed most. This helps to explain why a very high quality 50 watt per channel amplifier may sound less strained than another unit rated at 200 watts per channel. We also begin to understand the vast price differential among competing brands.

As we have seen, the power supply is one of the key factors in determining the current delivery capability of the amplifier. If the power supply runs out of steam, the amplifier can produce no more power, limiting the ability of the unit to deal with musical peaks and/or low impedance loads. Unfortunately, power supply components (large storage/filter capacitors, transformers etc.) are expensive, the most expensive parts within the amplifier. Consequently, we don't see inexpensive ultra-high current amplifiers. The economics of the situation simply won't allow it.

What about tubed amplifiers? Tubes, unlike transistors, are not capable of producing large amounts of current. Thus, most tube amplifiers would not be the ideal choice to power very low impedance speakers. Further, most tube amplifiers tend to be of lower power than a similarly priced solid state unit so the sensitivity of the speaker becomes another important consideration when mating a speaker with a tube amp. 

But that doesn't mean tubes aren't a great alternative. Indeed, we are big fans of tube gear. Many listeners feel that better vacuum tubes designs simply convey a more natural performance than their solid state counterparts. No question that tubes have a sense of warmth and harmonic richness that can make lesser solid state gear sound a bit threadbare by comparison. Those that tout tube designs also feel the genre is particularly adept at reproducing a very believable three dimensional soundstage.

Ok, so now we have a basic understanding of the issues relating to the electrical match between amplifier and speaker. From here we can look at a more subjective area of the matching process, that of tonality.

Let’s say that you had chosen a loudspeaker that has a tendency toward brightness. You found that, once you got it home, your very live room exacerbated this bright character. The choice you make in mating an amplifier to your speakers will have a significant impact on whether you further aggravate or help to downplay the tonal imbalance you perceive. To best determine a match, it's important to work with a dealer that can give you a detailed description of the tonal character of the various components that fit your budget and preferably one that will let you audition the amp with your system . Ultimately, the only way to be absolutely certain of the ability of an amplifier to mate with a given speaker, is to try it.

Footnote: This explanation of how amplifiers and speakers interact is an over simplification of a very complex topic.  This simple treatise is by no means intended as a complete technical explanation of the very detailed interaction that occurs between speaker and amp. I hope however, that it has served to give you a basic understanding of the important electrical relationship between the amplifier and loudspeakers. 

---

[1] A simple experiment will help to better understand this statement. Using a volt ohmmeter, measure the terminals on your speaker. You will find a discrepancy between the manufacturers specification and the reading on the meter. A speaker with a nominal rating of say 8 Ohms, may measure only 6 Ohms. Why is this? Taking a reading at the speaker terminals gives you the DC resistance of the speaker, not the impedance. Measure the speaker with an impedance bridge, which makes its measurements using an AC test signal, and the speaker will produce a range of readings that vary with the frequency of the test signal. 

OHMS LAW
Volts (E) = Amps (I) x Ohms (R)
Amps (I) = Volts (E) / Ohms (R)
Ohms (R) = Volts (E) / Amps (I)

R=Ohms, E=Volts, I=Amperes

[mikebse2a3]

8 hours ago, Honeybadger said:

Has anyone taken into consideration that amplifiers typically make 2-3 time more power into 8 ohms than 24 ohms.

 

Just food for thought.

 

HB

 

I would hope most of us know that as impedance goes down more power is drawn from the amplifier.

 

I don't want to assume so I have to ask what point or points are you trying to make..? 

 

miketn

[mikebse2a3]

On ‎11‎/‎12‎/‎2016 at 2:18 PM, Honeybadger said:

I am sure one of the technical types could figure out to the milliwatt the additional draw through the swamper. (It's not much)

 

HB

 

Wrong it is significant and should be considered by low watt amplifier users..... the "swamper" draws about 3 times more power relative to the autotransformer/driver as I proved by my calculations in my post on page 6 of this thread.

 

miketn

[Honeybadger]

1 hour ago, mikebse2a3 said:

 

Wrong it is significant and should be considered by low watt amplifier users..... the "swamper" draws about 3 times more power relative to the autotransformer/driver as I proved by my calculations in my post on page 6 of this thread.

 

miketn

The math speaks for itself.

[Honeybadger]

1 hour ago, mikebse2a3 said:

I would hope most of us know that as impedance goes down more power is drawn from the amplifier.

 

I don't want to assume so I have to ask what point or points are you trying to make..? 

 

Just an observation that seems relevant.

 

Does you amplifier make more power at 8 ohms or 30 ohms?

 

HB

[wvu80]

12 hours ago, Honeybadger said:

For example, if the amplifier is producing 20 Volts at the output terminals, Ohms law (R=V/I) tells us that there are 50 watts being fed into an 8 Ohm speaker (watts equal voltage squared divided by impedance). I

That formula does not indicate Volts SQUARED.

[Honeybadger]

Take it up with the author of the article.

 

http://www.gcaudio.com/resources/howtos/spkramp.html

[Marvel]

Dave - This will show all kinds of formulae...and is handy to have close at hand. I have carried a similar one with my calculator for over 25 years. You can find this and an online calculator here:

 

http://www.sengpielaudio.com/calculator-ohm.htm

 

 

 

 

 

 

 

[babadono]

^ good one Marvel. Having worked in electronics for 35+ years it is engraved on my gray matter. All these formulae are derived from the 2 basics:

Ohm's Law  E=IR   and

Watt's Law P=IE

 all the rest can be derived by algebraic substitution.

[Deang]

Al has a rebuttal on his Facebook page. 

 

I've seen math presented every different way possible on this, including sideways - and all I get is a headache. 

[John Warren]

On ‎10‎/‎17‎/‎2016 at 0:36 PM, dBspl said:

 

It's just a crude estimate based on a midrange impedance of around 30 ohms and a shunt resistor of around 8 ohms. 

 

The loss would be significantly higher on a product such as a Heresy where the midrange impedance approaches 100 ohms.

 

Kerry

 

Amp is set at 2V output, 1kHz signal first (top) montage is current (mA) sourced by the amp into a no-shunt autoformer loaded by Klipschorn mid as taps are changed from -3 to -12dB.

 

Second montage is same 2V amplifier settings at 1kHz but with 10 Ohm shunt across input side.    With the shunt across the input side the bulk of current sourced by the amplifier goes thru the shunt regardless of what the taps are doing.  That said, that's the point of the "swamping" resistor.  It makes the impedance somewhat less sensitive to changing the taps because the impedance load presented to the filter is largely determined by the current draw thru the shunt resistor. 

 

That current data is then used to tweak a SPICE model of the circuit to, eventually, determine power.   In the power calc plot, I placed a 60uF capacitor in the circuit between the amp and the shunted autoformer to provide AA type filter.  It assumes the mid is on the -3dB setting.

 

Link shows the analysis

 

http://www.northreadingeng.com/Forums/viewforum.php?f=4

 

[Deang]

Such a hot topic until you get real data. 

 

Thanks for taking the time to do this John. 

[mikebse2a3]

2 hours ago, Deang said:

Such a hot topic until you get real data. 

 

Thanks for taking the time to do this John. 

 

Thanks for posting your test John.

 

Dean why do you think John's data changes anything about this " hot topic" ..?

 

Low wattage amplifier users need to be aware of the real drawbacks of using the swamping resistor which my calculations and John's data indicates also.

 

miketn

[Honeybadger]

 

Paul Klipsch   "WHAT THIS COUNTRY NEEDS IS A GOOD 5 WATT AMPLIFIER".

 

If you run a crossover with a swamper, you will need one capable of driving an 8 ohm load.

 

 

 

[Marvel]

Refer to my post above...

 

My 3.5 watt amps worked fine on my LaScalas.

[mikebse2a3]

2 hours ago, Honeybadger said:

Paul Klipsch   "WHAT THIS COUNTRY NEEDS IS A GOOD 5 WATT AMPLIFIER".

 

If you run a crossover with a swamper, you will need one capable of driving an 8 ohm load.

 

 

DFH Vol 16 No 8

 

The rising impedance offers two improvements in this case which was it draws less power from the amplifier and has less distortion.

[mikebse2a3]

1 hour ago, Marvel said:

Refer to my post above...

 

My 3.5 watt amps worked fine on my LaScalas.

 

I'm happy for you

 

For others:

The room's size and acoustical live vs deadness will be a variable for some people with their power needs

Musical taste will be another variable for power needs.

Taste in reproduction listening levels will be another variable for power needs.

 

I just want you and others using low wattage amps to understand that AL's claim of no downside isn't true and that you are giving up some max spl headroom and possibly increasing distortion in your system.

 

miketn

[Deang]

With 2V, we're essentially looking at a 1/2 watt as opposed to a little less than a 1/4 watt. I'm not going to dispute that if you only have 3.5 watts to begin with, you're probably not going to reach live levels without the amplifier going into distress. However, those like Marvel, who listen in intimate settings at modest levels, won't have any trouble. This is rather obvious, since people here have been doing just that for well over a decade. 

[Deang]

19 minutes ago, mikebse2a3 said:

I just want you and others using low wattage amps to understand that AL's claim of no downside isn't true and that you are giving up some max spl headroom and possibly increasing distortion in your system.

 

2a3 users can pretend they're using 45s. :-)

[John Warren]

14 hours ago, mikebse2a3 said:

 

Thanks for posting your test John.

 

You are more than welcome sir!