Tri-amping speakers - Which amps


I am seriously looking at running a tri-amped system with an active crossover and 6 channels of amplification (the speakers will not use passive crossovers). The speaker designer suggested 100-200 watts for the woofers.
With that in mind an obvious choice would be a 6 channel amp from someone like ATI or others. Another alternative would be to find 3 used stereo amps. Something like the B&K ST140 would seem like a readily available possibility, but that model is fairly old and repairs may be expensive. I guess I am 'testing the waters' with this setup and do not want to spend a fortune on amplification. I can always sell the active crossover and have the passive crossovers hooked up in the speakers, but a 6 channel amp would be hard to find other than new and I don't think the demand would be great to try and sell used. Any sugggestions/
cnut
Hi guys,

I just came onto this thread and find it excellent. I am also considering a tri-amp set-up with an open baffle dipole speaker system using a DSP to provide x-over, eq. and in-room correction. As i understand it from the manufacturer, i have great flexibility in the amplifier and cable selection and set up for the system. As the DSP can adjust gains for differences in amplifier input impedances, one can use different amps and by re-adjustment, change amps in the future if one wishes. I have listened to a prototype of this speakers in a non-treated room and the sound was impressive using mid-fi amplification and cabling. The one drawback, in theory, that i see is that, all inputs into the DSP is digitized so that this may not work for those that need their all analog domain intact eg. vinyl. Also, one is limited to the quality of the digital conversion of the input signal in the DSP and this may also add and additional A to D and D to A conversion depending on whether the digital front end source is inputted directly or analog converted prior to the DSP. But, at the proverbial end-of-the-day, what really counts is the sound coming out of the speakers in your room and to your ears and the rest is just that... theory.

Rafael
Cnut, the "flat" I was referring to is equal output from the speaker system over the entire frequency range with a given constant input.

Raw drivers will output more sound at some frequencies than at others. In many high quality speaker systems, the manufacturer has designed the passive networks to compensate for these non-linearities.

Now, there may be little of this in your speakers. Your internal networks may be nothing more than simple frequency separation and in that case, the active xover will give you far more control over the resulting sound and no degradation.

Cnut, you are correct that room dynamics can significantly impact frequency performance. Those controls for tweeters are there to compensate for rooms either full of sound absorbing materials or with many hard reflective surfaces. Active operation, however, will do a better job that those controls. You simply get much more control.

Since the manufacturer of your speakers suggests going active, that-s probably the way to go. My comments were really about folks who were contemplating going active on speaker systems where the manufacturer does NOT recommend active operation.

Lastly, I-d suggest the simple way to solve your problem is to purchase the two way monitors and then get a high quality powered sub. The high quality powered sub, by the way is active! Your active xover will supply the line level signal for the sub and then it will supply the line level for the mid-range and tweeters.

Cnut, if you go this way you only need a two way active xover and two stereo amps. Life just becomes so much easier/simplier.

Here is an example of a two way active xover:

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=245-862

There are more expensive units to be sure, but many folks have reported successful results with these.

Hope this helps -

Jerry
Jerry mentions the 'flat response' being the speaker designer's intention (in some cases). Doesn't flat depend largely on environment. As I understand it, ideally you test a speaker outside to determine how it truly measures. Tri-amping with an outboard active crossover is the recommendation of the designer of the speakers I am considering. He has several designs of 2 way monitors (with passive crossovers) that he can match with his own designed passive sub(s). The builder has indicated to me that for the 'best' sound from his design you need to leave out the passive crossover from the monitor and go the active tri-amp route. You take measurements of the speakers output in your listening room using a Radio Shack meter and aided by pc software use the controls on the crossover to 'tailor' the sound to correct any problems with the room.
While this would take some work and time, it does seem to me to be the best way to get the most out of your equipment. This route would add some expense such as an active crossover, more cables, and more channels of amplification, but if done correctly would appear to have a lot of benefit. Yes, I know I said "if done correctly". It is common now for the more expensive subs, and some of the cheaper ones, to have a built in amp & equalizer to correct for any bass nodes in the room. Bass problems are supposed to be the most common and problematic but why not address problems at higher frequencies as well? Some speaker mfg put controls on the back of their speakers for basic adjustment particularly for the tweeter. Is this merely for personal taste or are they helping to taylor the sound to the particular room?
Thus is the entire problem with tri-amping. How much $'s and complexity to achieve sound that often is far, far removed from the "flat" response that the manufacturer's seek.

B&W has the following on their website:

"The term crossover is a little misleading. A multi-way speaker contains a crossover network that not only divides the incoming signal into different frequency ranges, appropriate to the working range of each drive unit, but also equalises each driver's response to be flat (raw driver responses are usually anything but flat)."

They further have a very interesting article (it's short):

http://www.bwspeakers.com/index.cfm/fuseaction/local.faq/ObjectID/F5CA2EE1-3D20-11D4-A67F00D0B7473B37

There is a fair amount of confusion about passive xovers. Most people think their only function is frequency separation. Not true. In many high quality units the manufacturers have "shaped" the frequencies sent to the drivers to correct for non-linear drivers and cabinets.

Further, this huge "power loss" is somewhat a myth as well. Xovers don't really dissipate power (other than the trival internal resistant of the components). What they do is "block" frequencies unwanted for a driver.

In any event, Cnut, one way to insure you maintain the voice of your speaker systems is to passive bi-amp. That is, leave the passive components alone and use two amps. Each amp ends up with a much easier task because the impedance for each amp is just much better behaved/simplier.

My "gut feel" is people, who claim a huge improvement in sound out of tweeters, are simply sending more power to those tweeters. Yes, the sound is brighter/clearer, but the same could be done with EQ's and even tone controls.

Regards,
Jerry
Active speakers are VERY limiting in certain ways i.e. you are stuck with the one specific sound that the designer voiced into the amplifier / speaker interphase. They also require power to the speakers where an AC feed may neither be convenient and / or practical depending on placement.

Having said that, active speakers can have multiple technical advantages to them. Obviously, the bottom line comes down to sound. One would have to like their sound quite a bit as they come, as one can't "fine tune" the system by substituting amplfiers, cabling, etc...

There are quite a few design considerations that have to be taken into account with an active speaker, and quite honestly, i've never seen one that really caught my eye. Then again, i've never really looked at actives for several different reasons. I don't like having all of my eggs in one basket, as all the baskets i've seen aren't quite sturdy enough and / or lack versatility and / or the sonic traits that i desire : ) Sean
>
As Sean mentioned this will take a lot of 'homework' to properly think it through. That is why I asked for help on this board. I have already talked with the speaker mfg at length on a couple of occasions to get his input on amplification etc. He suggested getting a multi channel amp of my choice with a minimum of 60 watts with 100 being better. I think he was trying to lower the chance of me screwing up by removing at least one of the variables. He also said his speakers did not need mega watt amps that cost what you would pay for a new car to sound great. Howvever, he did say the better the source and amplification the better the whole system would sound. I believe the Linkwitz Orion 'package' uses a 12 channel 60 watt ATI amp. This is actually what started my interest in the active system but my wife said it looked too 'odd' for HER living room.
I had not thought of contacting the amp manufacturer to discus setting the bias higher for the mid/tweeter amps. That is a great idea.
I had thought of active speakers such as the ATC. I am hoping my project will cost less and allow more control flexibility over the sound than ATC.
Post removed 
For all the bother, you're probably better off looking at a pair of ATC active 50's or 100's. They're triamped internally and you'll avoid all the hassle of cabling, shelf space, etc. I'll bet that the results will be very similar for less money too.
One can typically raise the bias of ANY amp, so long as there is adequate heatsinking and air-space available. As such, you can use a multi-channel amp and have the manufacturer adjust the bias setting to operate in Class A up to X amount of power, switching over to Class B above that point.

In my experience, depending on speaker efficiency and impedance, somewhere between 6 - 10 watts of Class A power @ 8 ohms seems to work quite well with Class B operation kicking in above that point. Due to the lower impedance of your MTM section and low efficiency of your woofer section, i would recommend going higher on the bias scale rather than lower. That is, 6 watts of Class A bias @ 8 ohms is only going to net you 3 watts of Class A @ 4 ohms. While this would still be better than most, moving up to 10 watts @ 8 ohms would give you 5 watts of Class A @ 4 ohms. By the time you switch over to Class B, your listening level is so high that any form of crossover distortion becomes negligible.

As far as the woofer section goes, 100 watts wouldn't make an 84 dB woofer flinch, even if actively crossed and bandwidth limited. Besides that, passive radiator systems generate more reflected EMF than most other designs, meaning that you need even more power to keep the driver under control. I don't know how big your room is, how loud you like to listen, etc... but i would be looking at some very high powered, high current amps for that section if i were you. That is, unless you listen to smooth jazz or chamber music at background levels.

With the above info taken into account, you might want to think about a medium to high powered ( 100 - 200 wpc @ 8 ohms ) four channel amp for the highs / mids and a very robust ( 350+ wpc @ 8 ohms ) two channel amp for the woofers. Larger amps typically have more output devices and heatsinking, which allows them to share the higher load of a richer Class A bias with out as mech thermal stress on any given part. As mentioned above, sticking with the same manufacturer would have several advantages.

Before investing in any of this, do a LOT of homework. What you are attempting to do is neither inexpensive nor simple in most cases. Obviously, it can be done using less power and in a very simple & convenient manner ( one six channel amp ), but why go to all of this trouble unless you're really trying to obtain max performance? Sean
>
As Jerry mentioned I have considered bi-amping due to the advantages such as cost, space, & simplicity to name a few. But that would seem to be a compromise. Think I would always wonder if bi-amping was an improvement, wouldn't tri-amping be better. Have not totally discarded the idea however.
Sean, you mentioned most of the things I have had to consider if I go this route. I have thought about Class A or tubes for the tweeters and mids and bipolar amps for the woofer. Obviously you cannot have this with a multi channel amp. With 3 stereo amps you have to consider compatibility issues with the system. I am not totally familiar with an electronic crossover, but as I understand it cannot totally correct for all mismatches between power sources. Any suggestions on actual amps to consider? (I am serious about doing this but don't want to just throw money at it. I like a noticable improvement, and not to quickly reach the diminishing return category.)
In your last section you mentioned 'electrical resources'. Were you talking about available power from my wall socket? If so I think the increased efficiency of tri-amping will allow amps of about 50% the usual power requirements.
The monitors with mids and tweeters are an MTM design with 4 ohm impedance and 90 db sensitivity. The wooofer cabinets have a 10" active driver & 12" passive, 8 ohm and 84 db. The sensitivity of the woofer cabinet is low but the designer has told me that 100 quality watts should do fine.
Any more discussion would be greatly appreciated.
Jerry / Onplane: I would beg to differ as i've yet to hear even a 6 dB crossover ( a simple capacitor in series with the tweeter ) that sounds as good as a "naked" tweeter ( no capacitor ) actively crossed. This not only has to do with removing the capacitor from the high current area of the circuit, but also from the electrical changes that take place prior to the amplification stage.

My Brother made the same assumption that you did here, as he was running his system as you described for quite some time. When i finally got him to try using an antique "Pro Sound" active 3 way that i had laying around, he was flabbergasted at the difference. He couldn't believe he had wasted all those listening sessions with reduced fidelity when even a 30 year old active crossover ( $40 on Ebay ) could have provided so much better results. Bare in mind that this was using cheap plastic Radio Shack 1/4" to RCA adaptors on every cable in and out of the crossover too. Sean
>
Depending on the efficiency of the speakers, the crossover points and slopes used, the type of music being played, etc... this will determine how much power you'll need for each frequency range being amplified. Having said that, i've always found that "more is better", so long as quality isn't compromised in order to obtain the greater quantity.

One thing that should be noted is that system and amplifier efficiency will go up, so you'll be drawing less power from any given amp at any given volume. Most of this is due to limiting the bandwidth that each amp sees, but some of it will be due to getting rid of the "power sucking" passive crossovers too. Those "power sucking" crossovers also "cover up" low level resolution, so expect to hear greater transparency too.

Due to the lower power demands and greater resolution, crossover distortion from the amps may become more apparent to you. In this area, it helps to have amps that run in Class A longer than most commercially produced amps do. This increases signal purity and moves the point that the amp crosses over into Class B operation above the point of average power consumption. The only downside to this is that there is greater heat to deal with.

As was mentioned, it may make things easier for you by using amplifiers from one manufacturer. In doing so, you can have them increase the bias levels to where you want them and gain match all of the channels at one time. Depending on if the amps are new or used, and what the policies of the company are, they may / may not charge for this service to be done. As such, you might want to talk to the service department of the manufacturer's that you are considering using.

You should be aware that there is quite a bit of "fine tuning" that you'll have to do with a system of this nature. If you are someone that wants to maintain a high level of accuracy, you will at least have to take some minor electrical measurements using a multi-meter. If you're not worried about that kind of stuff, you can simply adjust things by ear until they sound "good" to you. I'm primarily talking about gain settings on the electronic crossover that you use, but you can make this as simple or complicated as you would like to carry it.

One more thing. Running multiple amps will require greater electrical resources. Are your AC resources up to the task? If not, factor that into the cost of such an installation too. Sean
>
It is late at night and maybe I am not reading your post well enough, but I can't see the name (and model) of the speakers. Would be a help answering your question.

Richard
The more common solution is to bi-amp. Here you remove the passive xover for the woofer and use your active xover.

Then for the mids/tweeters you leave the xovers intact. This way you need only two amps and balancing becomes that much easier.

Power requirements are radically different for these amps. You are correct woofers suck up tons of watts, so you need a big amp there. For the mids/tweeters it's a totally different story. I'm using a low power chip amp with terrific success. Some people prefer tube amps with their "silky sound" powering the mid/tweeters.

Unless you are in a theater and need the efficency, you lose little with the xovers in the mid/tweeter.

Hope this helps ...

Regards,
Jerry
Post removed