Novice attempting to bi-amp. Help


I purchased two Carver TFM-42 amps with the intention
of bi-amping my Carver Amazing III speakers as they are
power hogs and have read the this configuration will
really open them up. However, I have no idea how to
wire this set up. ( speaker cables and interconnects).
I would appreciate any advice from those experienced
with a bi-amped set up. The speakers do have
dual binding posts for bi-wiring.

Thanks,

mar4004
mar4004

Showing 4 responses by sean

G13 hit it on the head. I too would recommend horizontal bi-amping for this specific situation. Sean
>
Vertical bi-amplification will work "better" if you have speakers that are a benign load on the higher frequencies and you have an abundance of power in both channels. Since the highs are relatively "coasting" in terms of power draw, the "excess" power that the supply is capable of is now available as reserve to help feed and control the much more demanding woofers. However, i must point out that this is NOT true if the amp that you are running is of true "dual mono" design since each channel has its own dedicated transformer and filter caps. As such, the supply is not being "shared" so one channel could not draw from the other. This method is more convenient than it is anything else, as it allows the amp to be placed closer to the speakers should one desire to do so.

If you have speakers that are reactive on top and multiple and / or low impedance woofers down low, each section of the speaker is pulling equally hard on the power supply. In this instance, horizontal bi-amping may provide greater benefits as you have a dedicated power supply that can cope with the demands of each frequency range on their own. This minimizes sagging of the supply for both frequency ranges simultaniously and presents a more stable stereo image. This is not to mention that this can result in less clipping being seen by the high freq drivers, which is a good thing in terms of reliability. I say this because the woofers would no longer have the capabilities to starve the power supply feeding the mids and tweeters, thus assuring them of always having adequate power to reproduce the signal.

If you gain a drastic amount of channel separation by vertically biamping, something is wrong with the design of your power amp. I say this because horizontal bi-amping is theoretically the same as running a single amp in normal stereo mode. The benefit is that you now have a dedicated power supply to each frequency range rather than to each channel. If you can hear a very noticeable difference in channel seperation using vertical over horizontal, your amps are simply poorly designed, lack channel seperation and are high in crosstalk between channels. If you are able to obtain excellent channel separation and imaging in stereo mode with one of these amps, you should be able to do the same with two of them in horizontal mode.

Obviously, the one drawback to vertical bi-amping is that you obviously need two identical amps to do this. One can horizontally bi-amp with two different amps, so long as their gain is relatively matched.

Obviously, the ultimate set-up would be to bi or tri-amp using monoblocks. This allows dedicated power supplies for not only each channel, but for each frequency range used. Channel separation and dynamic headroom are no longer part of the equation as you should have achieved the best possible in such a situation. That is, if the amps were selected so as to easily provide enough power under any conditions to each of the drivers it is connected to. It is possible to use a dedicated amp for each driver and still starve the driver IF the amp was too small to start off with to achieve the desired listening level. Sean
>
Bear, good idea about the dummy load and listening for "bleed through". Simple and quite effective. Do you think that removing the interconnect from the input of the "dead" channel and using something along the lines of a Cardas cap be of any further assistance while doing this test ?

While i have never done any "controlled" testing and taken measurements, i would think that -30 to -40 db's "should" be pretty easy to achieve. What are some figures of amps that you would consider to offer good to excellent channel separation ?

I agree with getting the passive crossovers out of there. It is amazing how much even a single "high grade" cap in a simple 6 db crossover can destroy the sound of a speaker.

I also agree with your comments regarding the Carver amps. Those speakers are capable of performance well beyond those specific amps, especially if the passive crossovers are bypassed.

Mar, in vertical bi-amping, let's say that we have amp A and amp B. You would hook up amp A to one speaker and amp B would go to the other. Amp A's right channel would drive the woofers and the left channel would drive the ribbons. Amp B's right channel would drive the woofers on the other speaker with the left channel driving the ribbons.

Obviously, i selected left / right channels just for demonstration purposes as one channel should be the same as the other and they don't have preferences as to what they drive. Sean
>
I would think that shunting the input should negate all but the highest levels of crosstalk. After all, the input is where the "leakage" would have the most chance to be picked up and amplified as that is where the circuitry would offer the most gain. Shorting the input jack to ground would in effect shunt the "leakage" to ground at the same time, making the test far less realistic in terms of real world operating conditions. That is why i thought a Cardas Cap might be worthwhile, as you would only hear the "internal leakage" within the amp itself with minimal influence from outside sources.

One other thing just came to mind. I would guess that someone with HIGHLY sensitive speakers might hear more "crosstalk" simply due to the fact that the speakers allow more to be heard with less amplitude signal sent to them. As such, someone with K-Horns, K.A.R's, etc... may think that their otherwise "excellent" amp is "leakier" than someone with an amp that is actually much poorer with a pair of 86 db speakers. I guess that is why it is "good" to have test equipment and dummy loads to document such things. You at least end up with consistent and repeatable conditions that can be used from component to component. This allows one to have a specific baseline to work from. Sean
>