More powerful amp for top or bottom?

The usual answer would be to put the more powerful amp on the bottom, since that part of the spectrum is where most music has its greatest energy levels.

However if you are referring to passive bi-amping, meaning that there is no electronic crossover ahead of the amplifiers, and you properly gain-match the amplifiers in some manner, be aware that you will not be able to utilize the majority of the power capability of the 570W amp.

That is because in a passive bi-amp arrangement the outputs of both amps have to swing a voltage range corresponding to the full-range signal. Therefore the maximum output power of the higher powered amp that can be utilized will be determined by the clipping point of the lower powered amp.

Regards,
-- Al

I'd be careful passively bi-amping with that much of a difference in output power. Bi-amping is a fussy business, usually best to stick with matching amps to ensure impedences etc. are closely matched. Presumably the 2 amps are from different manufactures which could cause problems.

You might find one or the other works better alone with good quality jumpers.

Meiwan...Can you please explain the types of problems that could arise from that great a difference of output power. Also the types of problems from using different amp manufacturers. I'll probably end up with a very similar situation so any advise is appreciated.

Bass definitely requires more power. However, it is almost impossible for both of these amps to increase in volume in a neartly exact fashion. Even if you find a good match at 10 watts, was volume increases one amp may increase at a different rate.

Strangely enough, when I was bi-amping Maggie 20.1's, sound was best with more power on treble and mids. Never would have expected that...

Tyler makes active speakers. Could you go that route. Give him a call

Even though both amps must cover the full voltage swing, since the impedance into the low frequencies will be extremely high due to the internal crossover, it will be putting out very little current and therefore very little power at low frequencies. Ultimately, the real nswer is try it both ways, and see which souunds best.

Can you please explain the types of problems that could arise from that great a difference of output power. Also the types of problems from using different amp manufacturers.

Assuming passive bi-amping is being referred to:

1)A lot of the power capability of the higher powered amp will be wasted, as I indicated above, since the maximum volume level that can be used will be limited by the clipping point of the lower powered amp. Honest1's point in the post above is correct, and the reduced current demand on the lower powered amp will often (depending on the design of the particular amp) raise its clipping point somewhat, but a large disparity in power capability will still mean that much of the capability of the higher powered amp is wasted.

2)If the amp that has higher gain does not have a volume control, a resistive attenuator (or equivalent means) will have to be provided to match the gains of the amps. That will also often be true even if the amps have similar power ratings, if they are not identical models.

3)If the amps have significantly different sonic characteristics, overall coherence may suffer.

4)With different model amps, the flexibility of being able to vertically biamp is lost. If identical model amps are used, both vertical and horizontal configurations can be tried. The vertical configuration might sound better, perhaps due to reductions in inter-channel crosstalk or other inter-channel effects within the amps, since the two amp channels process identical signals in that configuration.

5)If the output impedances of the two amps differ significantly, as may happen if one is solid state and the other is a tube amp, tonal imbalances might result, depending on the speaker.

The following two issues pertain to any passive biamp configuration, even if the amps are identical:

1)In any passive biamp configuration, if the preamp does not provide separately buffered outputs for each amp you have to be careful that the combined input impedances of the two amps remain much higher than the output impedance of the preamp. Preamps providing two sets of output jacks often have those jacks wired directly together, rather than separately buffering them.

2)If the preamp does not provide separately buffered outputs for each amp, the sum of the capacitances of the interconnect cables to BOTH amps will affect the signals to BOTH amps. If total cable capacitance is too high in relation to preamp output impedance, the upper treble may be slightly rolled off.

However, it is almost impossible for both of these amps to increase in volume in a nearly exact fashion. Even if you find a good match at 10 watts, as volume increases one amp may increase at a different rate.

I don't think that is true. What should be matched is voltage gain, not power output (which will vary with the impedance vs. frequency variations of the speaker). If voltage gain is matched at one output level, and does not remain closely matched at other levels, it would mean that at least one of the amplifiers is significantly non-linear. That is, a plot of output voltage vs. input voltage would not be a straight line, or very close to it. Non-linearity produces distortion, which would be unreasonably high in that situation.

Regards,
-- Al

I didn't realize the answer was so complex. For more clarity: both Amps are Wyred4Sound; the pre is a BAT VK30SE

Both Amps are Wyred4Sound; the pre is a BAT VK30SE

I took a look at the specs on those components.

Both W4S amps have the same gain, so that issue (no. 2 in my previous post) is ok.

Their input impedances are 60K, so two of them in parallel would have a combined input impedance of 30K, which is comfortably high in relation to the 300 ohm nominal output impedance of the BAT (even if the output impedance were to rise to 3000 ohms or so at deep bass frequencies, which is not uncommon among tube preamps). So you could split either the rca or xlr outputs using y-adapter cables (xlr y-adapters are available), to drive the two amps.

Both amps have extremely low output impedance, so item no. 5 in my previous post is a non-issue.

Because of the similarity between the amps, no. 3 in my previous post would seem unlikely to be an issue.

The cable capacitance issue can be avoided by using low capacitance cables, and/or avoiding long runs.

So the only potential issue, as I see it, is how much of the 570W amp's power you will be able to utilize. I would suggest contacting W4S and asking them if they can tell you what the internal voltage rails are for the two amps. If the numbers are not too different, I think you'll be fine.

Regards,
-- Al

Good work Al.

http://sound.westhost.com/bi-amp.htm

Depends on crossover frequency how much power you need per amplifier above / below the crossover point.....

Al,
Educate me, please. I thought that many amps increase in output as a logarithmic function, not linear. As input power from the pre is "amplified" that each step in volume is not linear but a function of the amps total output power? ie turning a pre-amp up 1/2 way will not necessarily utilize 1/2 of the amplifier's output capability.
Even if we look at this as a linear function, 1/2 volume on the pre amp would be 1/2 of the amps output capability. Therefor, 1/2 volume for a 100 watt amp will be 50 watts and 1/2 volume on a 240 watt amp is 120 watts. That's about a 4-5 db imbalance?
Thanks (no I'm not an engineer and yes, that's part of my problem...)

Hi Elevick,

Those are good questions.

The volume control on a preamp or integrated amp is indeed not linear, and will approximate to some degree being logarithmic, which is necessary because our hearing mechanisms are to some approximation logarithmic.

Power amplifier gain, as you realize, is often expressed in terms of db, which is a logarithmic scale. However, what the amp is really doing is generating an output voltage that is directly proportional to the input voltage, meaning that the input voltage is simply multiplied by some numerical constant, and providing the capability of supplying the amount of current that the speaker's low impedance will draw at that output voltage.

In this case the W4S amps have a gain of 27db, which corresponds to a voltage gain of about 22.4 times, based on the 20(log(V1/V2)) relation between db and the ratio of two voltages. So what the amp is doing, within the limits of the maximum output voltage, current, and power that it can handle, is generating an output voltage that is 22.4 times greater than the input voltage, at any instant of time.

Viewed in that manner, I think you will see that if some other amplifier had a voltage gain of say 30 times, and we provide the input to it through a resistive divider which in conjunction with the amp's input impedance attenuates the signal by a factor of 22.4/30 = 0.75, then the two amps will be gain matched for any signal amplitude and preamp volume control setting that is within their capability.

Best regards,
-- Al

Al,
couldn't you change the rate at which the perception of increase or decrease changes simply by changing the potentiometer type? From Linear to Log or the other way?
Isn't there a third type? Can't 'member what it may be called?

Hi Magfan,

Sure. The potentiometer, or equivalent stepped attenuator, can have pretty much any characteristic the designer wants. Although a linear characteristic (or linear taper, as it is often referred to), would be unsuitable for audio applications because it could not provide the combination of range and resolution that our hearing characteristics require.

BTW I'll add a further thought to my previous post, concerning the reference to a volume control as usually being approximately logarithmic. That is somewhat misleading, because what is logarithmic is the relation between the amount of attenuation the volume control provides and the amount the knob is turned. But for any given position of the knob, there is nothing logarithmic about it. For a given setting, all a volume control does is to multiply the voltage going into it by some fixed fraction, meaning that all signal amplitudes are affected equally.

So in that sense, for any given setting a logarithmic (or any other) volume control affects the signal in a linear manner, just as a power amp does. Except that the control has a gain of less than 1 while the amp has a gain greater than 1 (as well as the ability to supply lots of current, provide a low output impedance, etc.).

Best regards,
-- Al

Al,
That's a great answer!
Hope Santa gets you that 180gm LP you wanted...
Emery