Vertical Bi-Amping

[mep]

Anyone here tried vertical bi-amping with two stereo amps that are the same make and model? Vertical bi-amping makes more sense to me than horizontal bi-amping. You get some of the benefits of having monoblock amps (one power transformer per channel) and now I'm powering the speakers with 2 channels of 220+ watts instead of one channel of 225+ watts.

It should be interesting to see if it's a step forward, sideways, or backwards. The only problem is speaker cables. My current pair for use with a stereo amp are bi-wire MIT Shotgun S-3 speaker cables. These will be worthless in a vertical bi-amp set-up. That means I have to fish through my box of cables for the retired and homeless and come up with two pairs which won't match of course. I will use one matching pair for the bass/mdiranges and another matching pair for the tweeters. Since many people are convinced that cables don't do anything besides make you feel good, this will be interesting as well. Mind you that I'm not just throwing in any old trash in here. I think one pair is
Canare which I will use on the bass and the other pair is some XLO something or others that have the bias circuit with the ta-ta bar purple LEDs which don't work anymore and I can't use the bias circuit anyway as the wires share a common wall wart and they will be too far apart. Probably doesn't matter either as some are convinced that bias circuits on speaker cables are just more nonsense.

By the way, for those that love specifications, here are the specs on my amps:

Power Output - 220 watts minimum RMS into 8 ohms from 20Hz to 20,000kHz with no more than .09 THD

Continious Power - 260 watts per channel into 8 ohms at 1 kHz and 360 watts per channel into 4 ohms at 1 kHz

Typical total harmonic distortion - .009% at rated power at 8 ohms at 1 kHz and .009% at rated power into 4 ohms at 1 kHz

Intermodulation Distortion - less than .09% at rated power into 8 ohms and less than .09% at rated power into 4 ohms

Frequency response - 12 Hz to 40 kHz +0-1dB

Signal to noise ratio - 110 dB IHF A weighted

Residual noise - 120 micro volts

Dampening factor - 1000 @ 1000 Hz

Rise time - less than 4 micro seconds

Slew rate - greater than 18 volts/micro second

Phase shift - 0 degrees at 20 Hz and -18 degrees at 20 kHz

 

[MylesBAstor]

Anyone here tried vertical bi-amping with two stereo amps that are the same make and model? Vertical bi-amping makes more sense to me than horizontal bi-amping. You get some of the benefits of having monoblock amps (one power transformer per channel) and now I'm powering the speakers with 2 channels of 220+ watts instead of one channel of 225+ watts.

It should be interesting to see if it's a step forward, sideways, or backwards. The only problem is speaker cables. My current pair for use with a stereo amp are bi-wire MIT Shotgun S-3 speaker cables. These will be worthless in a vertical bi-amp set-up. That means I have to fish through my box of cables for the retired and homeless and come up with two pairs which won't match of course. I will use one matching pair for the bass/mdiranges and another matching pair for the tweeters. Since many people are convinced that cables don't do anything besides make you feel good, this will be interesting as well. Mind you that I'm not just throwing in any old trash in here. I think one pair is
Canare which I will use on the bass and the other pair is some XLO something or others that have the bias circuit with the ta-ta bar purple LEDs which don't work anymore and I can't use the bias circuit anyway as the wires share a common wall wart and they will be too far apart. Probably doesn't matter either as some are convinced that bias circuits on speaker cables are just more nonsense.

By the way, for those that love specifications, here are the specs on my amps:

Power Output - 220 watts minimum RMS into 8 ohms from 20Hz to 20,000kHz with no more than .09 THD

Continious Power - 260 watts per channel into 8 ohms at 1 kHz and 360 watts per channel into 4 ohms at 1 kHz

Typical total harmonic distortion - .009% at rated power at 8 ohms at 1 kHz and .009% at rated power into 4 ohms at 1 kHz

Intermodulation Distortion - less than .09% at rated power into 8 ohms and less than .09% at rated power into 4 ohms

Frequency response - 12 Hz to 40 kHz +0-1dB

Signal to noise ratio - 110 dB IHF A weighted

Residual noise - 120 micro volts

Dampening factor - 1000 @ 1000 Hz

Rise time - less than 4 micro seconds

Slew rate - greater than 18 volts/micro second

Phase shift - 0 degrees at 20 Hz and -18 degrees at 20 kHz

Mark: I think you mean horizontal biamping where you use one channel of the left channel amplifier say for the midrange/tweeter and the other channel for the bass; same goes for the right channel amplifier. That way the power supply has less of a draw on it. Vertical would be if I used one amplifier for the mid/top and the other amp for the bass--say as the old Infinity RS1bs or IRSs did. Yes, I always preferred horizontal rather then vertical biamping on the Maggies. Part of that I attributed to the crossover slope used by Magnepan. Because of that, one could hear the "sound" of the bass amp thru the upper midrange and hence a solid-state amplifier on the bass could actually harden the sound.

 

[mep]

Myles-I think I had it right. Horizontal bi-amping is when you use one amp to drive the woofers of both speakers and the other amp to drive the tweeters of both channels. That would make it horizontal. Vertical bi-amping is when you use one stereo amp per speaker and one channel drives the woofer and the other channel drives the tweeter. I'm pretty sure that is correct. And yes, the power supply will not be as taxed and you have the added benefit of having one power transformer driving each channel with half the bass load which gives you the same separation you would get with a pair of monoblocks.

 

[ack]

Yes that's correct, but I think it's unusual to use stereo amps for vertical bi-amping; the preference is usually towards monoblocks in that case, because of the added control and higher power reserves available wherever needed. Wild audiophiles would then get four monoblocks to complete the picture Your approach would be more of a poor-man's quad-amping

 

[mep]

It may make it unusual, but no less a semi-elegant solution. You are putting twice the power to each speaker cabinet and using one power transformer per channel for a true dual mono set-up. You can call it a poor-man's whatever, but it's not quad amping. It's vertical bi-amping. Cary has a nice write-up on his website (or used to) on why it makes good sense.

 

[fas42]

A very wild guess, based on what typically happens in power supplies, is that vertical will give deeper, richer bass at the expense of treble quality, and horizontal will emphasis the upper midrange and treble, which will be cleaner as well.

Frank

 

[JackD201]

Strictly speaking I'm tri-amped. All this horizontal and vertical has just officially confused me so I'll just say what I do

My speakers are 94dB with a nominal impedance of 6ohms semi-active by virtue of a built in subwoofer (that's one pair of amps)

Mids and Highs which together are 8 ohm loads in configuration A get either a Lamm M1.2 Hybrids (110wpc into 8, 220 into 4) set in the high impedance position (8-16 ohms) or in configuration B, Lamm M1.1 tubes from the 8 ohm taps (90 wpc)

The midbass drivers which are acoustically suspended are paralleled and are thus 4 ohm loads. These get in get configuration A, Lamm M2.2 (220wpc into 8, 440 into 4 at the high impedance setting but 220 into 4 in the low impedance setting) and in configuration B the M1.2s slide back to bass duties but this time in the low impedance setting.

To my ears and all that have been here, both A and B outperform a single M2.2s sonically. Configuration A is apples and oranges power wise but configuration B, if you add things up are pretty close in power. In either case the MT amps don't have to deal with back EMF from woofers and the impedance of the MT modules are fairly flat. This makes the FR more linear in comparison but is only really obvious at higher levels.

 

[tdh888]

I used to biamp my old Snell Type A2 using identical amps (ARC D79B's ) But I was using an active Electronic crossover network (ARC EC 21,passive in design in the mids/high,Cutoffs designated by capacitor kit specific for your speaker make avail. from ARC, its electronic in the bass cutoff ) so each amp will only drive the specified frequency assigned to them. Does this biamping procedure has any advantage than the one were passive biamping is done ( using the amp fullrange w/ out the benefit of an active crossover and just using the crossovers of the speakers) The only disadvantage I see in active crossover is the crossover itself should be transparent or else it will produce additional coloration or signature to the sound.This is how I remembered Biamping (using either an electronic crossover/ or a passive external like what they use for the tympani's )was done before.

 

[DonH50]

There are pros and cons to either approach, but ultimately I prefer vertical as I can put each amp near the speaker and run very short speaker cables where the wire length matters most (IMO). However, this depends upon having speakers that can effectively utilize two similar amplifiers (my Magnepans are good for this) versus those that have pretty different power/drive requirements for the high and low sections. Otherwise, if you have the power for the bass, the treble is coasting... Not necessarily a Bad Thing, mind you!

 

[Mark (Basspig) Weiss]

A third approach is to use each stereo amp as a bridge mono. Bridging causes the effects of asymetrical power supply sag on asymetrical waveforms to cancel out, resulting in more linear amplifier operation and complete isolation between amplifiers/frequency bands.

 

[Vincent Kars]

This is what I understand as common used for horizontal and vertical bi-amping
http://thewelltemperedcomputer.com/HW/Connect/Speaker_connect.htm

 

[docvale]

Based on the classification of horizontal and vertical biamping from the latest post, I'd prefer the vertical one because of a better channel separation.

Instead, I'm not really sure you're driving really more power into your speakers, independently on your configuration. Unless you put an active crossover between the pre and the power amps, according to the speaker passive crossover design part of the power you're injecting is "gone".
I'll explain my consideration. To make it easier, let's imagine to have 4 monoblocks, 100W/each. You'd drive your speakers with a couple of monoblocks per channel. Is it like having 200W/channel? With an active crossover, each 100W monoblock would amplify the signal belonging to a single channel and a panel of frequencies. Without active crossover, each monoblock would amplify the signal for a single channel but ALL the frequencies: so, part of the amplification product wouldn't be seen by the frequency-specific speaker connectors.
Maybe you could have better control or dynamic range, but I doubt you'd double the power.

Having two sets of stereo amps, I'd check if their bridgeable. You'd actually quadruplicate the power, instead of duplicate, as each bridged amp would see half of the impedance of the speaker. Clearly you could do it only if the amps had an huge power transformers...

 

[ack]

This is what I understand as common used for horizontal and vertical bi-amping
http://thewelltemperedcomputer.com/HW/Connect/Speaker_connect.htm

Another reason your and mep's definition of vertical bi-amping (which, again, is correct, but just one of two possibilities - the other being monoblocks) is usually not desirable and monoblocks are preferred (on top of the reasons I gave earlier) is one of input impedance change - whereas each output channel of the preamp sees an input impedance R when driving one stereo amp's input channel, the same output channel (sporting the same output impedance as always) now drives two amplifier input channels (as shown in your diagrams) but with inevitably different total input impedance. This may or may not have an audible effect, depending.

 

[DonH50]

@docvale: An active crossover has been implied through this discussion. My viewpoint has been expressed several times in several fora, so I'll just re-iterate that I am not a fan of "passive" bi-amping, for the reasons you have stated (among others). Reduced crosstalk is a viable argument, though I have found over the years it is fairly easy to measure but almost impossible to hear. Generally, at line level it is set by the preamp instead of (or equally by) the power amp. The biggest benefit comes from decoupling large signals that are not common to the channels, introducing crosstalk through the output power supply. In most sources, such signals tend to be common anyway (bass, big booms, etc.)

@ack: True, but given the relative impedances involved, I cannot see any clear audible impact from doubling the preamp's load with two amps in parallel. There is generally at least a factor of ten, and more often 50 to 100, from preamp output to amp input impedances. A notable exception is a tube preamp driving a SS power amp.

All imo, fwiwfm, my 0.000001 cents, etc. - Don

 

[mep]

@ack: True, but given the relative impedances involved, I cannot see any clear audible impact from doubling the preamp's load with two amps in parallel. There is generally at least a factor of ten, and more often 50 to 100, from preamp output to amp input impedances. A notable exception is a tube preamp driving a SS power amp.

All imo, fwiwfm, my 0.000001 cents, etc. - Don

Don-And that is exactly what I was doing-driving the pair of SS amps with a tube preamp. The only difference I noted that I would attribute to the tube preamp driving the pair of stereo amps is that I had to turn up the volume slightly in order to have the same level I had when driving a single stereo amp.

However, my experiment is over for the here and now. When I bought the second Phase Linear 400 Series 2 from the original owner, it didn't work. One of the rail fuses was blown. I replaced the fuse and the amp worked perfectly last night. Today when I came home from work, the amp is distorting now in both channels. Both rail fuses are fine so now there is a new problem.

I can tell you all that when I was listening last night when both amps were working fine that I like what I heard from vertical bi-amping. I was looking forward to listening to this set up further until the gremlins bit me in the ass. The vertical bi-amp set up certainly made the soundstage wider and on some LPs I was getting a wrap around effect like I had speakers on the side walls in addition to the front speakers. All of this took me by surprise and I certainly wanted to hear more. For now, I can't.

 

[DonH50]

I'll spare you my long and painful experience and tirades about PL's reliability, particularly in the case of the original 700 I had for years in a bitter love/hate relationship. You may have lost something in the power circuit if both channels are distorting.

Though I have the perfect speakers for it, I last and longest bi-amped horizontally because of the different amps I was using. However, I have used vertical bi-amping with ML and Krell amps and it worked very well, just way more amp than I could afford at the time. As I said, were I to bi-amp now, vertical it would be. Ah, wish I was still in the biz where I could take home such great toys to play with! Although, durn your hide, now you have me thinking of rebuilding my old crossover network and trying with another XPA-2...

Hope it's not too painful getting your second amp running again!

 

[tdh888]

For Biamping is it best to use an active electronic crossover or just do it passive if the speaker is designed both ways in w/c you can actively or passively biamp it?

tdh888

 

[DonH50]

Most speakers require more than a first-order crossover, so active is generally the way to go as higher-order filters are more difficult to realize passively and you hve to add gain stages anyway. And, active circuits generally give you more flexibility with phase alignment through various filter topologies (yes, to you purists, I know you can implement the functions in passive circuits, but again you often end up having to add additional gain stages). Finally, active circuits make it easier to isolate the LPF and HPF sections.

If you are speaking of passive bi-amping in terms of sending the same signal to two amps, e.g. using two channels of an AVR, and then using the crossovers built into the speakers, for a lot of reasons I see no real advantage in that approach. Others do, and the debate goes on... That is a discussion best left to another thread.

All IMO - Don

 

[tdh888]

If you are speaking of passive bi-amping in terms of sending the same signal to two amps, e.g. using two channels of an AVR, and then using the crossovers built into the speakers, for a lot of reasons I see no real advantage in that approach. Others do, and the debate goes on... That is a discussion best left to another thread.

All IMO - Don[/QUOTE]


I agree w/ you Don. To me active Biamping should be done if you really want to get the benefit of using 2 stereo amp or 4 monoblocks in driving a pair of speakers. There should be an active electronic crossover network.

 

[LesAuber]

I agree that active is likely optimum. However, and I'm sure that one of might be able to shoot hopes in this, passive should achieve a significant portion of the benefits. While the amp does get a full range signal the speaker crossover should cause the amp to only be loaded over the crossover range. Amplifying a signal with no load shouldn't tax an amp much.

FWIW my 801's took on a more effortless quality to the sound when run this way using 4 channels of the Dreadnaught.