Bi-amping the AR-3a cheaply and non-invasively – WOW!!!

[tysontom]

>

>Using an active/electronic crossover changes the rules of the

>game quite a bit. Among the more obvious things, the crossover

>slopes are different (probably much steeper than the gradual

>6db/12db per octave design of the original 3a crossover). If

>you also change the original crossover points, significant

>differences are likely to be heard. Fortunately you have

>deemed the changes to be improvements.

>

Using an active crossover before the drive of an AR-3a definitely changes the rules of the game! AR recommended *against* the use of this type of arrangement, simply because the passive crossover network components play a role of not just establishing the crossover frequencies, but also in “shaping” the system response and balance at frequencies far removed from the nominal crossovers. This is true of just about any well-designed loudspeaker, if the best system characteristics are to be obtained from the speaker. It is a conscious part of the design of the speaker, so to nullify the passive crossover would change the balance of the speaker.

>

>That is probably part of the result, but you essentially

>redesigned the speaker system when you replaced the original

>passive crossover. I believe, as I do with Jerry's approach,

>that the speaker system's frequency response curve has been

>altered as well.

>This is definitely true when you re-design the crossover.

>

>In Jerry's arrangement, he has somehow achieved a significant

>difference utilizing the original passive crossover through

>the manipulation of the two amps' tone controls. The

>information he provided suggests that the controls operate at

>the extremes of the audio range, which would make them less

>likely to create a midrange problem, as I originally

>suggested. Some tone controls can affect a broad spectrum of

>frequencies however. Using them to modify crossover slopes is

>a crapshoot at best.

In the final analysis, what purpose is served bi-amping an AR-3a with a passive crossover? Bi-amping would only theoretically give greater power output available to the driver sections; but if you used a larger amplifier to start with, the issue of inadequate power would be solved without the potential of instability and distortion. Furthermore, by manipulating the tone controls for each amplifier -- if they are actively working -- the end-result would be probably dips and peaks in the response.

>

>The original crossover of the AR-3a has some woofer to

>midrange issues. It is possible that you have both addressed

>that problem somewhat in your bi-amping adventures....and,

>hey, if it sounds good...........:-)!

>

The woofer-to-midrange problem (a 2 dB rise in the upper bass) was fixed with the coil change to #9. This 2 dB rise in response made the AR-3 sound somewhat “thick” or heavy in the lower midrange.

--Tom Tyson

[onplane]

>>YES, the shields are connected to the common ground. This

>is

>>one of the things I measured BEFORE connecting everything

>>together. The whole ideal is to prevent "floating

>>grounds". We want both amps to have exactly the

>same

>>grounds and the shielded cables will force everything to

>a

>>common point.

>

>But are the amplifiers tied to "earth" ground?

>Neither receiver has a 3-wire cable, does it? Therefore, the

>grounds are basically "floating" on a common chassis

>ground, which would permit some offset to occur. Any offset

>in one amplifier would be felt in the circuit of the other

>amp, potentially causing oscillation, and most likely

>increased distortion.

>

>--Tom Tyson

Hi, Tom!

No 3 wire cables, but both amps I verified tie the grounds to thier chassis and the shields on the coax cables.

Once we connect everything together, just like when we connect a CD player or tape deck, we create a common ground.

I've been thinking about another experiment. I'll bet I'd see no difference if I ran just three wires to the speakers instead of 4. Well ... let me say, I'd probably keep the woofer common back to the low frequency amp, just because of the higher currents.

Regards,

Jerry

[onplane]

>In the final analysis, what purpose is served bi-amping an

>AR-3a with a passive crossover? Bi-amping would only

>theoretically give greater power output available to the

>driver sections; but if you used a larger amplifier to start

>with, the issue of inadequate power would be solved without

>the potential of instability and distortion. Furthermore, by

>manipulating the tone controls for each amplifier -- if they

>are actively working -- the end-result would be probably dips

>and peaks in the response.

>

>--Tom Tyson

Tom, it's a lot more complicated that just power. Here is a more detailed description of the factors involved:

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

http://sound.westhost.com/bi-amp2.htm

Further, in my case I achieved a terrific improvment in the sound of the AR3a's at zero "out of pocket" cost. My wife is just never going to understand the need for a modern, high power amp.

Regards,

Jerry

[ninohernes]

I have found that the biggest improvement in sound with AR speakers that have 3 terminals is removing the strap between terminal 2 and T. I have 12 gauge 4 conductor speaker cable. I tie them into pairs on the amp end. On the speaker end, I solder a spade onto the negitive pair and connect that to terminal 1, and I leave the positive wires separate, and solder spades onto those. One gets connected to terminal 2 and one gets connected to terminal T. So basically, I have sort of bi-wired them.

[onplane]

>I have found that the biggest improvement in sound with AR

>speakers that have 3 terminals is removing the strap between

>terminal 2 and T. I have 12 gauge 4 conductor speaker cable. I

>tie them into pairs on the amp end. On the speaker end, I

>solder a spade onto the negitive pair and connect that to

>terminal 1, and I leave the positive wires separate, and

>solder spades onto those. One gets connected to terminal 2 and

>one gets connected to terminal T. So basically, I have sort of

>bi-wired them.

Joe, this is clearly a form of bi-wiring. You get bi-wiring for free with my bi-amp scheme.

Joe, do you have an old low power amp/receiver lying around? I mean, you already have the speakers wired for bi-amping. Besides a low power amp, you'll need a set of coax audio cables. This should be a piece of cake for you. AND the results of your experiment might make for a great article in TONE.

Only thing you need to double check is that both amps use a common ground for speakers AND the coax shields.

Regards,

Jerry

[ar_pro]

Sorry, Jerry...you haven't really "bi-amped" your AR3a's. What you have managed to do is run an unusual concoction of dissimilar amplifiers, volume, and tone-control circuits into a 40 year-old passive crossover; compensating for a lack of measurement capability by adjusting by ear. That ain't bi-amping, and it's not good science, either.

Even disregarding the fact that the 3a's crossover has a desired effect on the system's designed sound quality, legitimate multi-amping would dispense with the passive crossover, and use three stereo power amplifiers and an independently-adjustable 3-way electronic crossover. An oscilloscope & signal generator wouldn't hurt, either.

[kkantor]

I really cannot fathom people's reactions in this thread.

It's "bad science." So? Some of the proposed explanations are tenuous at best. So?

These products exist to give joy to their owners. What could be more important?

[rrcrain]

The voice of logic and reason speaks at last.

Simply put, if it sounds better, or more pleasing to the owner of the speakers, who among us has the right to naysay what they have done?

I recently listened to a pair of old Ohm omnidirectional speakers being driven by a small Jolida tube amp. They sounded decent but definitely nothing to write home about. The shop owner connected a larger tube amp to the speakers and my mouth fell open. Absolutely everything these speakers were reproducing changed. They became brighter, mode defined with very decent bass.

my guess is that the first amp was working near its design limits whereas the second amp was well within its design limits. Perhaps, our gentleman with the AR3s experienced something similar when he biamped?

For those interested, read this article and post your opinions.

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

[tysontom]

>The voice of logic and reason speaks at last.

>

>Simply put, if it sounds better, or more pleasing to the owner

>of the speakers, who among us has the right to naysay what

>they have done?

>

Yes, the statement above is absolutely correct. I keep forgetting that this is just a hobby -- nothing more. And what KK said, "...these products exist to give joy to their owners," pretty much sums it up!

I don't agree with what Jerry is doing, but I am 100% guilty of over-reacting to it; in the end, if he feels that bi-ampimp improves the sound, I should step back and not criticize. Maybe it is the old adage: "one man's meat is another man's poison."

--Tom Tyson

[rrcrain]

Tom,

Given your history with AR, I found it rather normal that your a bit defensive of one of your "babys". Perhaps I would be the same way.

Perhaps one way to lessen the sting is to remember why the old AR speakers are still around; They are still darn good speakers designed and built by some of the best people in the industry.

Richard

[tysontom]

>Tom,

>

>Given your history with AR, I found it rather normal that your

>a bit defensive of one of your "babys". Perhaps I

>would be the same way.

>

Richard, I'm really not "defensive" about any of it, yet I know that I would never make that sort of change myself. I am probably more "protective." I did try a version of it several years ago, but I pulled a separate ground wire out from the woofer circuit and isolated it from Terminal No. 1. It was in essence two separate ground circuits. I used two AR Amps, but in the end, the difference was really neglible and not worth it, and I decided to return it to stock. What worried me was that some others might attempt it, and I was concerned about that.

--Tom Tyson

[onplane]

Tom,

As always thanks for your thoughts.

I will say one thing in defense of these great speakers. The very fact that we CAN "passively bi-amp" WITHOUT having to open the boxes (most of today’s 'modern' speakers come with this option), just goes to show are far ahead of their time the early AR's really were!

One other comment, and I don't mean this in the sense that I have "superior listening skills" to anyone, because I don’t! It is possible that I hear differences when passively bi-amping that others don’t hear, because of the type of music that I normally play.

Most of the time, I listen to jazz trios/quartets. With so few instruments and always a double bass and heavy percussion, it’s possible that the “significant” improvements that I hear, just stand out under this kind of music and are negligible in everything else.

Again, I’m NOT saying that I can hear better than anyone else. I’m just saying it's POSSIBLE the format I listen to, causes me to hear things that are invisible/negligible in other music formats.

Regards,

Jerry

[rrcrain]

I actually agree with you as I would never myself connect the ground or negative of two dissimilar amps together on a three post speaker. I was concerned doing what I did knowing the amps were rack mounted bringing their cases and negative terminals to a common ground. I was being overly paranoid, but a bit of paranoia served me well in my days as a bench tech.

I tried simple (no active crossover) biamping on my AR90s with my pair of amps and the change was "blah" to say the least. Given that, my only really good guess is the gentleman with the AR-3A'3 might not have had enough amp to push them. As I stated in a previous post, I've heard this sort of change with Ohm speakers when driven by a marginal sized amp.

But at the close of the day, he's pleased with the results; nothing blew up or caught on fire and he didn't spend a fortune. Neither you nor I believe he did it right and thats perfectly acceptable since he's happy...

[onplane]

Richard, you are correct, I am happy, very happy with the 3a's.

Further, I have a control to compare against, my set of TSW610's. My 3a's NOW out-perform the 610's and it's true ... I'm NOT sure why. Previously the 610’s were brighter/clearer than the 3a’s.

As for the common ground, please remember we connect the chassis of many dissimilar devices (CD players, tape decks, tuners, etc) together via the shielded audio cables all the time.

If there were "something insidious” going on, there would be some symptoms, such as:

1. humm

2. heat

3. reduction in stereo imaging

As for humm, with no input and both amps’ volume controls to the max ... NOTHING! No humm, no hiss, nothing!

Heat – both amps are running really cool

Separation – better than before or at least the clear IMPRESSION is that it’s better.

One more comment, it’s NOT easy to compare any bi-amping to straight/single amping, because of the time is takes to re-do the connections. So a control of some kind is really necessary. My point is while I say I experience “significant” changes, others may feel those changes are negligible.

It gets back to ... changes compared to ... what?

Regards,

Jerry

[ar_pro]

>I really cannot fathom people's reactions in this thread.

>

>It's "bad science." So? Some of the proposed

>explanations are tenuous at best. So?

>

>These products exist to give joy to their owners. What could

>be more important?

Come on, Ken...no one is trying to keep Jerry from his fun, but as almost certainly the best-trained, most-experienced contributor to this forum, surely you're not equating this so-called "bi-amp" project with anything of technical or practical value?

And wouldn't it be less condescending for someone with your real-world experience and background to constructively critique & explain what Jerry's actually done, as opposed to figuratively patting him on the head, whilst maintaining "it's all good"?

[rrcrain]

I've absolutely no doubt you've experienced major improvements, please don't doubt that for a moment. What I'm suspicious of is that others may not experience the same improvement using larger amps.

My concern with the grounds is specifically concerned with the phase relationship of the amps to each other. Ugly things "might" have happened if they were out of phase with each other being their output grounds were connected.

[Pete B]

>Richard, you are correct, I am happy, very happy with the

>3a's.

>

>Further, I have a control to compare against, my set of

>TSW610's. My 3a's NOW out-perform the 610's and it's true ...

>I'm NOT sure why. Previously the 610’s were brighter/clearer

>than the 3a’s.

>

>As for the common ground, please remember we connect the

>chassis of many dissimilar devices (CD players, tape decks,

>tuners, etc) together via the shielded audio cables all the

>time.

>

>If there were "something insidious” going on, there would

>be some symptoms, such as:

>

>1. humm

>2. heat

>3. reduction in stereo imaging

>

>

>As for humm, with no input and both amps’ volume controls to

>the max ... NOTHING! No humm, no hiss, nothing!

>

>Heat – both amps are running really cool

>

>Separation – better than before or at least the clear

>IMPRESSION is that it’s better.

>

>One more comment, it’s NOT easy to compare any bi-amping to

>straight/single amping, because of the time is takes to re-do

>the connections. So a control of some kind is really

>necessary. My point is while I say I experience

>“significant” changes, others may feel those changes are

>negligible.

>

>It gets back to ... changes compared to ... what?

>

>Regards,

>Jerry

I really don't see any point in arguing about this, some people take an all or nothing stance, do it perfectly or not at all. This is all or nothing, sometimes called black and white thinking and it is unhealthy.

So, yes, ideally you'd design or obtain a line level crossover that duplicates the loaded transfer function of the passive crossover that you would then bypass. But, this is difficult.

Let's ignore the common ground, actually a ground loop for a moment.

The first time I saw this done, bi-amping through the passive XO, was by a brilliant engineer that I worked with. His speakers were expensive 3-way B&W 802s and had bi-wire inputs. He said I like my 50W/ch amp but it is not powerful enough, so I bought another one just to get more power. I pointed out that since the mid-tweeter amp was not loaded by the woofer, there would be no current draw in the bass to pull down the PSU rails. It does help a bit beyond the simple power addition.

OK, so it works in terms of offering more power.

You're using the tone controls, and based on how most are designed with turnover points around 1 to 2 kHz, I'd say turning them all the way down is too much. Your going to alter the roughly 500 Hz mid to woofer crossover. You need to filter far away from the crossover point so as to not introduce amplitude or phase shift at the crossover point. I'd put in the low cut filter and turn down the bass just a bit, not all the way on the mid/tweeter amp. You could even leave them flat and simply gain more power.

It's also interesting that if the bass amp is driven hard and clips at say 400 Hz, the XO and woofer filter out the harmonics (800, 1200, 1600 Hz ...), and they never reach the mid/tweeter. So you get more out of the amplifiers.

About the common ground connection, you have to play this very carefully, it is risky. Make sure that all amplifiers are non-inverting and have a true ground connection, no bridged amps.

You have a ground loop, there are two paths now from the power output stage to the speaker ground. A simple way to understand this is to examine what happens if you pull off one power amp ground connection, obviously it could fall off or become corroded. The amp that has lost its ground will continue to drive the load, back out its input RCA jack ground, into the other amp, and out its speaker ground terminal. However, this is not typically a high current path, and there could also be inductance in this path that might lead to oscillation. Keep your grounds tight and clean. Even with clean grounds a fraction of the speaker load current will flow through the line level ground because you have a "ground loop".

The best solution, avoiding mods to bring out another speaker ground, is to use identical stereo amps for each channel, where the outputs share a common ground. You'd only need one speaker ground wire in this case. Right and Left would drive Woofer and Mid/Tweeter. Inputs driven mono.

Obviously, the ground issue is eliminated if you mod the speakers to bring out individual grounds.

Simple solution, buy a bigger amp. It's not unreasonable to use 200W/ch with the ARs.

You also mentioned shunting the woofer inductor and leaving the cap in. Amplifiers do not like pure capacitive loads, and you don't have a crossover anymore, don't do it.

It's good that your thinking and enjoying your system, that's what it's all about.

Pete B.

[onplane]

Hi, Pete!

>The first time I saw this done, bi-amping through the passive

>XO, was by a brilliant engineer that I worked with. His

>speakers were expensive 3-way B&W 802s and had bi-wire

>inputs. He said I like my 50W/ch amp but it is not powerful

>enough, so I bought another one just to get more power. I

>pointed out that since the mid-tweeter amp was not loaded by

>the woofer, there would be no current draw in the bass to pull

>down the PSU rails. It does help a bit beyond the simple

>power addition.

Pete, you've just got to read my new thread (this one is getting too hard to follow). In that new thread, I quote extensively from the B&W website.

>You're using the tone controls, and based on how most are

>designed with turnover points around 1 to 2 kHz, I'd say

>turning them all the way down is too much. Your going to

>alter the roughly 500 Hz mid to woofer crossover. You need to

>filter far away from the crossover point so as to not

>introduce amplitude or phase shift at the crossover point.

>I'd put in the low cut filter and turn down the bass just a

>bit, not all the way on the mid/tweeter amp. You could even

>leave them flat and simply gain more power.

Pete, you are right about not getting the tone controls too close to the xover frequencies. It turns out the tone controls on both amps work at the extremes of the audio spectrum. In short, they are a long ways from the xover frequencies.

I ran a test with the woofer amp shut down and turned the bass control all the UP while playing music with a decent balance of frequencies. Result: NOTHING - no change in pitch coming from the mid-range/tweeter I believe I can hear a very slight reduction in clarity with the bass control full on in the very high frequencies. That is a definite, “I think!”

>About the common ground connection, you have to play this very

>carefully, it is risky. Make sure that all amplifiers are

>non-inverting and have a true ground connection, no bridged

>amps.

>You have a ground loop, there are two paths now from the power

>output stage to the speaker ground. A simple way to

>understand this is to examine what happens if you pull off one

>power amp ground connection, obviously it could fall off or

>become corroded. The amp that has lost its ground will

>continue to drive the load, back out its input RCA jack

>ground, into the other amp, and out its speaker ground

>terminal. However, this is not typically a high current path,

>and there could also be inductance in this path that might

>lead to oscillation. Keep your grounds tight and clean. Even

>with clean grounds a fraction of the speaker load current will

>flow through the line level ground because you have a

>"ground loop".

Hmmmm, Pete, your comments here are making me think. I like this!

How about ... if I run some real heavy duty, very short wires from the common speaker terminals of one amp to the common speaker terminals of the other and then run just a single heavy gauge wire from that junction to the common terminal on the AR-3a's?

>The best solution, avoiding mods to bring out another speaker

>ground, is to use identical stereo amps for each channel,

>where the outputs share a common ground. You'd only need one

>speaker ground wire in this case. Right and Left would drive

>Woofer and Mid/Tweeter. Inputs driven mono.

>Obviously, the ground issue is eliminated if you mod the

>speakers to bring out individual grounds.

>

>Simple solution, buy a bigger amp. It's not unreasonable to

>use 200W/ch with the ARs.

>

>You also mentioned shunting the woofer inductor and leaving

>the cap in. Amplifiers do not like pure capacitive loads, and

>you don't have a crossover anymore, don't do it.

The more I think about this idea, the less I like it. Your comment, Pete, about capacitive loads is right on! In addition, it's clear to me that without the coil, I have no way to keep mid-frequencies out of the woofer. Lastly, it's also possible the coil is shaping the woofers response, so it's staying right where AR put it!

>It's good that your thinking and enjoying your system, that's

>what it's all about.

>

>Pete B.

Good comments and ideas, Pete! Please let me know what you think about "hard wiring" the commons right at the amps.

Regards,

Jerry

[kkantor]

I thought about my last message for a few days before I posted it, and I stand by it. I realize it is a complex subject, and I would love to debate it in more detail when I return from my latest travels, (toward the end of this month.)

To my read, a user was having a blissful sonic experience, which was achieved in an unorthodox and probably technically misconstrued manner. The pitfalls and potential dangers where (rightly and appropriately) pointed out. To these concerns, the user responded in a thoughful and reasonable way, assured his own safety, and continued with his joy.

At this point in the evolution, it was my read that some kind of an optimum compromise was reached. A user was having a blast, and the potential issues were laid out as a warning/guideline to others. After, it just became a game of "winning" by trying to destroy the other camp's philosophical position. Why? No greater truth was really going to be served, and this is not a peer-reviewed journal.

Finally, every time I see a person really getting off on their audio system, it warms my heart in a chilly world. I don't much care if the joy comes from scientific investigation, classic restoration or pure, amateur tinkering.

If Jerry had asked me for my professional opinion about his approach to biamping, and the plausibility of his theories about the reasons for his experiences, I would have been clear with my answer. But he wasn't asking, at least until pushed into a corner. He was too busy having fun, and was trying to share it. My instinct is to encourage first, educate second.

-k

>>I really cannot fathom people's reactions in this thread.

>

>>

>>It's "bad science." So? Some of the proposed

>>explanations are tenuous at best. So?

>>

>>These products exist to give joy to their owners. What

>could

>>be more important?

>

>

>Come on, Ken...no one is trying to keep Jerry from his fun,

>but as almost certainly the best-trained, most-experienced

>contributor to this forum, surely you're not equating this

>so-called "bi-amp" project with anything of

>technical or practical value?

>

>And wouldn't it be less condescending for someone with your

>real-world experience and background to constructively

>critique & explain what Jerry's actually done, as opposed

>to figuratively patting him on the head, whilst maintaining

>"it's all good"?

>

>

>

>

[Pete B]

Hi Jerry,

I really can't comment about your different ground connection because there's no way to predict the results not knowing the exact behavior of the looped ground system. You had some luck with a first configuration that worked for you, so I'd leave it at that.

Pete B.