This will work if the input sensitivity of the amps in the same. Otherwise you will need something to balance them.
20 responses Add your response
We are both saying the same thing. If you do this you will have to make sure that the sensitivity (or gain , same thing) of the amp you are buying is the same OR HIGHER than that of your integrated amp. Why? Because the output of your integrated feeds directly into your power amp , if it is more sensitive (more gain) than your external amp there will be no way to balance their gain. The ideal solution would for them to be exactly the same but I wouldn't chance it, manufactures specifications are often inaccurate. I would buy an amp with considerably higher gain (more sensitive) because then I could even them out with a volume control in the input line. A simple one is easily constructed or get one at Radio Shack or more upscale source.
So that there is no imbalance between the tweeter and the mid/woofer, you must have the same voltage applied to both. To do that you must have the same output voltage from the two amps. The input voltage to the two amps will be the same, because you're running them from the same preamp. So as long as the voltage gain (output voltage / input voltage) of the two amps is the same you'll have the same output voltage. If the gains are different, you can place a volume control like an NHT PVC between the preamp and the amp with the higher gain so that you can attenuate the voltage.
The input sensitive is the amount of current required for the amp to reach full output. For example one amp might require one volt , another two volts. This is the governing factor in the choice of amps in this case I would think. Input sensitivity is always specified for a power amp , I am less familiar with voltage gain. I would assume that it is the mirror image of sensitivity, describing how much the initial preamp signal is amplified. But it is not necessarily the case that you should have equal output from both amps. The tweeter or tweeter/midrange will require much less power than the woofer so having the same output from both amps might overdrive the high end. In the past when amplifiers were offered with several amps on the same chassis the treble amps were always of considerably lower wattage than the bass amp .I think we are saying the same thing with different terminology. The critical factor is the bass amp having greater sensitivity
( higher voltage gain) so that it can be adjusted by a volume control as we have both suggested.
I totally disagree with most of the posts-it can't work. Forget getting the same amp gain. There will definitely be a delay caused by using 1 internal amp and 1 external. There is no way they are getting the identical signal at the identical time. I'm a firm believer that you shouldn't bi amp unless you use 2 identical amps.
It's not often that you see a post that is totally wrong but Elevick has achieved that distinction. The time difference will be totally insignificant and completely inaudible. If you look through equipment reviews and forums you will see that the majority of bi amp systems use amps that are not identical.
I have to agree, in principle, with Elvick.
A time delay in an amp, maybe call it 'latency' of 1 millisecond...... =.001 seconds will produce about 14 inches of apparent motion of the affected sound range and it can be forward or backward. Hook an inverting amp with a non-inverting and the problem multiplies.
To those who hear differences in fuses (It's HUGE!) to those who cry foul when any cabling is disturbed, I must add that 'time smear' is a known phenom.
Given the speed of the signal, I would not expect any time delay in this circumstance to be audible or significant.
On a related point, this reminds me of a review of a Bryston amp that I read some years ago. The reviewer was on record on a number of occasions as saying that a good amp circuit should be designed without negative feedback. The reason was that you can't take a signal and then feed some of it back to eliminate distortion as you can't correct a signal that has already occurred! In other words, the inherent time dealy involved made the exercise futile. In the Letters section of the next issue of the magazine, Bryston responded to some of the points in the review. I think it was one of Bryston's chief designers, Mr. Russell, who regretted the reviewer's continued insistance on spreading blatant misinformation, or words to that effect.
Time delay in speaker drivers is important because you have drivers pumping out sound that travels "only" at the speed of sound. It's also important in the reconstruction of analog from digital signals, jitter, for instance. However, these time effects are occurring at the interface between electrical and mechanical energy or between digital to analog conversion respectively. Signals traveling around the speed of light for a few extra inches inside an amp are another matter.
Anyways, I'll let the engineers argue this one. I will however, tweak the nose of the naysayers by saying that I biamp with different amps and I have no problems at all. All I require is similar gain. I hope that doesn't bother you too much.
But it is not necessarily the case that you should have equal output from both amps. The tweeter or tweeter/midrange will require much less power than the woofer so having the same output from both amps might overdrive the high end.
You have to have equal voltage at the inputs to the drivers or one or the other will scream at you. When you use a single amp you are providing the same voltage to both drivers. Yes, each driver will likely dissipate different amounts of power, because they have different impedances.
It seems to me that input voltage sensitivity is independent of voltage gain. I see no reason that would prevent two amps from having the same voltage gain but vastly different input sensitivity or vice versa.
Given the input voltage and the voltage gain one can calculate the output voltage. Gain equals output voltage divided by input voltage. The input voltage sensitivity tells me what input level produces the maximum output level so I don't see how that helps in biamping.
There will definitely be a delay caused by using 1 internal amp and 1 external. There is no way they are getting the identical signal at the identical time.
Do we have any idea how fast electrons travel in copper wire or a board trace? Even if we're down to say 80% of the speed of light (~186,000 miles per second * 0.8 = ~148,800 miles per second), do we really think a delay is an issue? We're talking about 1.3e-9 seconds per foot.
What am I missing?
Fact of the matter is that an electron is in a fuse about 8.5e-11 seconds and people claim to hear that. But who's counting?
All that has to happen to create time smear is for one amp to have a slight amount of latency....even 1/2 millisecond is about 7 inches of speaker misalignment....and it isn't the speed of LIGHT that matters but the speed of SOUND, which is what....about 1100ft/second?
I think this hypothetical is just as valid as some of the other....persuasions of hi-end.
At least amplifier latency can be tested with a signal generator and a duel trace o-scope.
Magfan, if you're talking about latency inside of the amp, how can the speed of sound come into play? There's no transducer in the amp converting the signal into sound.
The post was that the electrical signal coming from different amps (one internal and one external) would not reach the drivers at the same time due to the difference in distance the signal would travel. Again, this is not related to the speed of sound.
Amp 1 is 'quicker' or 'slower' to transfer a signal than #2.
Say....about 1/2 milisecond. 1/2 milisecond at the speed of SOUND translates into about 6 or 7 inches.
If the 'late' signal gets to the tweeter, than it will appear as out of phase/ time smeared info. A 3-way system would make matters worse.
Right, of course, about transducer in amp to convert electrical to sound, but IF there is some reason for the signal to take longer to go THRU the amp, there you are.
This, in all fairness, is theoretical. BUT, it is easily tested with a signal generator and dual trace o-scope. If you were only dealing with a length of wire, it'd have to be nearly 1800 miles long to produce such a delay. But, toss in some caps / inductors / RC / LC networks and it could happen.
It is not the distance a signal travels, but the time it takes.
If I were prone to Bi-Amp, I'd use identical amps.
Since this is testable, I have no worries that someone hasn't already thought of this, measured it and discarded the idea.
Another good reason to use identical amps is possible phase invert of one but NOT both. This'll throw most people and cause no end to image / focus problems until properly diagnosed.
Belden 1800F STP cable has a velocity of propagation of 76%, so my 80% guess was a little high. But we're still talking about a large percentage of the speed of light (NOT SOUND), so we're in the nanosecond per foot region.
Velocity of propagation
Granted that components within an amp will likely cause a greater delay than simple cable, but one amp would have to be radically different from the other for there to be more than a few 10s of nanoseconds difference between the two.
Test w/O-Scope is confirmation.
Your reasoning is overall sound, but some amps may contain circuitry which acts in a time delay manner and May be audible. Something over 1/4 millisecond should do it.
I was just suggesting time delay as a possible occurance. I doubt it is really an issue, but lesser things have been turned into issues.
One thing NoBody mentioned was to have amps which are a match in Timbre. Put Rotel and NAD together and You've got a catfight. Or how 'bout Krell and Mac?
The other issue I missed was that many receivers will have different circuitry going into internal amps vs the pre out. I highly doubt that the internal amp signal passes by the pre-outs before being utilized. This should further muddy the waters. A few milliseconds difference is all it takes to kill that "magic" we are all looking for.