AR3, AR3a and AR2 owners - passive bi-amping

Purpose of this tread is to share with AR owners of vintage speakers a method of "passive bi-amping" that is cheap and non-invasive.

"Passive bi-amping" means you leave the xover circuits within the sealed box as intact. That is, you do NOT open the speaker boxes (non-invasive).

Many people will offer that you will get better results by removing the passive xover circuits. Removing the passive xovers will force you to tri-amp as there are really 3 speakers to drive.

Two weeks ago, I tried the scheme I will describe with stunning results:

1. better imaging
2. clearer/cleaner high frequencies (I can separate instruments better)
3. better definition in the bass
4. amps are running cooler

I'll be honest, I can't explain why this happens, but the scheme I will describe is cheap, easy to do and probably most importantly, ... easy to undo!

One more disclaimer: I got "stunning results" on MY AR'3a's with MY exisiting system. Your milage may vary!


Just a little background: the vintage AR's (3's, 3a's, 2's) were never
designed to be bi-amped like modern speakers, but due a few chance
coincidences they can be safely bi-amped.

Here is what I did:

1. I always listen to my 3a’s on an old Heathkit AR1500 and I still do! Only
now the Heathkit just powers the woofers, terminals 1 and 2 on the 3a’s.
Naturally, you MUST remove the strap between T and 2. Further, the treble
control is turned completely off. In short, the Heathkit is no longer producing
high frequencies. (AR1500 is rated 100 watts RMS at 4 ohms)

2. I take the audio signals from Tape Output jacks. This tap is before the
tone controls so it has the complete audio signal.

3. The Tape Output goes into an old Pioneer audio/video amplifier (rated at
30 watts - music power - that was in my basement collecting dust) at the VCR
Audio Input. Now this amp has the bass tone control turned completely off.
In short, the Pioneer is NOT producing low frequencies and it is
powering the tweeter and mid-range via terminals T and 1 on the 3a’s (again
the strap must be removed).

4. So that leaves me with two independent volume controls. One on each amp
and by adjusting these two controls I control the mix of high and low
frequencies. It took me no more than 3 minutes to figure out how these two
volume controls work together.

5. Now before you attempt this, you must make absolutely certain that both
amps have common grounds on the speaker outputs AND those grounds
MUST be connected to terminal 1 on the 3a’s. The “hot” or “red” lines go to 2 and T.

Notice that the shielded audio cables from Tape Output to the VCR Input
insures that ground on both amps is pegged to the approximately the same
level. (It's fairly common for solid state amps to have common grounds,
especially if they support standard headphone jacks.)

A good test to do before you connect is: With the power OFF on both amps and
with the audio cables connected between the amps, measure the resistance
between the grounded speaker output on one amp to the grounded speaker
output on the other. If you get a reading higher than zero, stop right here.
The amps are NOT compatible with this scheme.

I mentioned above that most solid state amps have common grounds, but NOT all.
The Dynaco 120 is an example of a ss amp without common grounds. I have been
informed that there are tube amps with common grounds, but I have never seen one.
Mine all had taps on the output transformers for speaker impedance.

Observations:

1. My old 30 watt amp has more than sufficient to power to drive the
mid-range and tweeter. In fact, it probably has too much power, so we must
be careful NOT to over drive. That is, turn the volume down whenever we
change program source.

2. I did NOT open the 3a’s! The crossovers are still there, but doing
less since the amps are producing frequencies more consistent with the
speakers they are driving.

3. There is no direct dc path between the "hot" outputs of the two amps.
Xover for the mid-range and tweeter provides only capactive coupling with
the amp.


In closing, I'd encourage AR's owners to try this scheme. I mean, if you gain nothing, returning to the way it was involves ... replacing the strap between T and 2. I specifically do NOT recommend that you run out and purchase a new amp. If you don't have an old, low power amp/receiver lying around, ... your son probably does.


Regards,
Jerry

Thanks for this advice! I've wanted to try this with my 3a as I have numerous amps to use - the tip on how to check the grounds is what I needed to know in order to do this safely.

I was planning to use an integrated amp which has a pre-out set of RCA jacks, and then use a Y-splitter to send one line back into the integrated amp for the mid-treble and the other end into the other amp for the bass. I'll have 120wpc on bass and 90 wpc on the top end - should be enough!

packrat, that little test to verify that the amps really have common grounds (speaker common one amp to speaker common on the other WITH the amps connected via the shielded audio cables) is simple, yet effective. If the amps really have common grounds the only resistance you'll measure is that of the audio cables shields and that should be very close to zero.

packrat, you'll be amazed at how little power is actually going to the mid-range and tweeter. 90 wpc is PLENTY actually way more than plenty. You might end up using very, very little of the volume control.

If you are driving your AR-3a's with 120 wpc, you may NOT get as dramatic results as I. Some people believe that the entire issue is power. My "gut" feeling is that the AR's will stress most amps.

Lastly, I don't understand your need for the "Y". Why not just use your integrated amp to drive the mid-range/tweeter both channels and use the dual audio pre-out to send both channels to the amp for the woofers? This is "horizontal passive bi-amping", but I like it because it allows us to have a smaller dedicated amp for the tweeter/mid-range.

Regards,
Jerry

Interesting idea; but I'm too scared to try it with my AR3a's. Anyhow, I have finally found the perfect setup for my AR's. They sound better then ever.

The x-overs in these early ARs were not designed for bi-wire systems. The terminal trpping was there to remove the tweeter from the circuit to add a JanZen Z-30 or 130 array in its place. I'd think doing as you suggest is a formula to fry the omewhat fragile AR tweeters from that vintage.

Brian,

You are the first to mention the tweeters. Most people worry about frying the amps, but my amps have never, ever run so cool.

Under this passive bi-amp scheme the xover which frequency directs between the mid-range and tweeter is still intact. I'd worry a lot more about the tweeter, if I removed the xover and directly drove the tweeter.

You are right though, in the single amp configuration it's just about impossible to over drive the tweeter, because the woofer drains sooo much power. Here we need to exercise a little care particularly upon progarm source changes.

Even a 30 watt (music power) amp can hit those speakers pretty hard.

Regards,
Jerry

onplane said:

.....Lastly, I don't understand your need for the "Y"...

Click to expand...

The amp is a Yamaha CA-1010 - I'm not sure but I always thought that I'd need to set the rear switch control to 'seperate' and have to use a Y-splitter to send the preamp signal back into the amp section and also into the 2nd amp.

I think this is passive bi-amping, and a better way to do it would be with a pair of terminals for the woofer and the mid-tweeters, with the positive and negative circuits seperated internally. Then I think the amp grounding is not critical. With the stock ar3a crossover the ground level of both amps must be the same.

I think you're more likely to fry a tweeter using a low powered amp running into clipping than a high powered amp that is never clipped. Of course cranking it without regard to SPL levels just for 'fun' would fry the mids and tweets pretty fast, and probably bottom the woofer.

Hi, packrat!

I don't know anything about your Yamaha. If you left the switch in it's normal position, can you get the full audio signal at the pre-out jacks? If so, and if the audio continues to your speakers, seems like you don't need anything other than a set of audio cables.

Clearly the preferred approach is to open the sealed boxes and electronically isolate the woofer. Then drill a hole and make a 4th terminal.

This, however, will take considerable time and effort. With my scheme you can "preview" the results. If you are satisfied, then you can make that investment in time and effort.

BTW, AR recommends that you do NOT remove the passive network, because it does more than just separate frequencies. It shape's the entire speakers audio response taking the non-flat drivers and cabinets into account.

A recent upgrade, which an individual suggested on another board:

Both amps have a ground lug for turntables. Run a “short”, heavy gage wire between these lugs. This will minimize ground loops and prevent instability in case the audio cables are disconnected.

I did this and notice NO change whatsoever. Nevertheless, I think it’s a good idea.

Lastly, I agree with your comment on clipping and the tweeters. I strongly suggest that you always turn down the volume controls prior to changing sources.

Regards,
Jerry

Rather than start a new thread, I figured I would resurrect this one.
Well, with Onplane's (Jerry's) help and guidance, I "Jerry-rigged" a pair of AR3's I picked up recently. Even though I am a big fan of the AR sound (currently have 2's. 2a's, 2ax's, 4's, 4x's, 5's, 3a's, I found the earlier 3's to be severely lacking in upper-mid and high end output. Whether this is by design or deterioration of capacitors and/or the pots, I have no idea, but whatever it was I didn't like what I was hearing. As per Jerry's instructions, I bypassed the tweeter pot and soldered the mid to the top of the pot. This still did not yield the sound I was looking for. So I hooked up a pair of Microstatic Tweeter Arrays and listened for awhile. They sounded better, but there was a certain lack of cohesiveness in the sound. So......I figured, since I have quite a few amps available, I would try the passive bi-amping that Jerry has been raving about, both here and on the Classic Speaker Pages.
I'm using a Carver 6250 Receiver to power the woofer, and this in turn feeds the audio signal via the tape outs to an Advent 300 receiver which drives the mid/tweeters. After double checking all connections, I sat back and gave a listen. In a word................WOW!!!!!!!!!!!!!!!!
To these old ears, the bass is tighter and better defined, the mids are detailed, crisp, and smooth, and the high end is extended and smooth as well. The sonic signature of the classic New England sound is still intact.....but it now has more "air" and is more transparent. I never would have believed that this arrangement could result in the sound I'm hearing right now. And the sound at low volumes, which for me, is a real weakness of vintage AR's, the 3's in particular, is now superb!
I am aware that there are purists who would consider this arrangement to be blasphemous, but, since it is an easily reversible mod, I think people who have the capabilites to do this, should by all means do so.
If you have AR speakers that can be "Jerry-rigged," and have a spare amp laying around, I heartily recommend passive bi-amping. Hearing is believing.
Thanks, Jerry!

joelongwood said:

After double checking all connections, I sat back and gave a listen. In a word................WOW!!!!!!!!!!!!!!!!

To these old ears, the bass is tighter and better defined, the mids are detailed, crisp, and smooth, and the high end is extended and smooth as well. The sonic signature of the classic New England sound is still intact.....but it now has more "air" and is more transparent. I never would have believed that this arrangement could result in the sound I'm hearing right now. And the sound at low volumes, which for me, is a real weakness of vintage AR's, the 3's in particular, is now superb!

Click to expand...

Bob, great observation! After bi-amping I find that music is so clear and detailed, that I now listen at lower levels and ... am totally satisfied!

For what ever reason, the benefits of bi-amping appear to be more pronounced at lower volume levels.

Frankly, I would have predicted just the opposite. That is, the off-loading of heavy current demand (caused by the woofers) in the amp powering the mids/tweeters would cause more pronounced differences at high volumes where the current draw is higher.

I really don’t know what’s going on. It might simply be that with bi-amping we have soooo much more control over the balance, that we can have it “our way” ... at any volume level.

Regards,
Jerry

onplane said:

I really don’t know what’s going on. It might simply be that with bi-amping we have soooo much more control over the balance, that we can have it “our way” ... at any volume level.

Regards,
Jerry

Click to expand...

I Have no idea what's going on either........all I know is that I have read countless times (and experienced it for myself) that vintage AR speakers need to be fed large amounts of power to "come alive." Right now I'm listening at a lower than moderate level, and, instead of dull, dead sound, that has me reaching for the volume control, the sound that is reaching my ears is vibrant and alive. I hate to sound like a broken record, but I never would have believed this type of sound was possible with these 40+ year old AR's.
Bob

Passive bi-amping is the practice of splitting the signal into high-pass and low-pass between the preamp and amplifiers, but using passive filters rather than active ones, and then feeding the amplifiers directly to the driver terminals with NO speaker level passive filters.

Tom Brennan said:

Passive bi-amping is the practice of splitting the signal into high-pass and low-pass between the preamp and amplifiers, but using passive filters rather than active ones, and then feeding the amplifiers directly to the driver terminals with NO speaker level passive filters.

Click to expand...

I really can't argue with your definition, Tom, but I did a quick google search of "bi-amping" and came up with a myriad of definitions, including this one by John Atkinson of Stereophile Magazine:

Passive bi-amping consists of using two full-range amplifiers in parallel to drive the drive-units. The speaker's internal crossover therefore still provides the appropriate frequency shaping for each unit.

Click to expand...

In any case, as I said before, call it what you will........to my ears, it made one helluva improvement.

Joe----You gotta listen to me before you listen to anyone else, especially somebody at Stereophile. Really. I doubt that Atkinson knows that passive bi-amping as I define it is a practice that dates back to at least the 1950s, it's amazing how technically and historically ignorant many hi-fi writers are.

Anyway what counts is that you're satisfied.

Tom Brennan said:

Passive bi-amping is the practice of splitting the signal into high-pass and low-pass between the preamp and amplifiers, but using passive filters rather than active ones, and then feeding the amplifiers directly to the driver terminals with NO speaker level passive filters.

Click to expand...

What are the benefits of moving the crossovers to a point in front of the amp stage? Of course, this would involve some serious mods to the AR's - and would you need to cut the mid and tweeter loose and tri-amp? How would you control the signal at the mid/tweeter point with the crossover coming prior to the amp stage?

Having run my AR's with several (increasingly larger) amps, I am getting how more headroom is really opening up these speakers - I'd like to try them as-is on a 150+ amp! I can see how splitting the load between two amps will make the speakers more "efficient", effectively increasing the available WPC on each amp to drive the individual speakers.

I'm trying to understand exactly what is involved here, Jerry has been pushing this "mod" a lot on the forum, and I am hesitant to try it - fried drivers are not easily replaced in these AR's, especially considering I've already invested quite a bit in their restoration. Joining the grounds I get, and I can see the wisdom of pulling back on the WPC aimed at the mid/tweet array. What I don't understand from Tom's post is how the signal is tailored for the mid/tweet with the crossover moved prior to the amp stage, with only two amps providing signal to the three drivers.

"What are the benefits of moving the crossovers to a point in front of the amp stage?"

Passives at this level have pretty much the same benefits as active crossovers----lower IM distortion, increased dynamics, improved transients, better amplifier/speaker coupling.

"How would you control the signal at the mid/tweeter point"

Use amps with level controls or insert a passive level control.

"fried drivers are not easily replaced in these AR's,"

Caution is understandable, many bi-ampers use protection caps between amps and tweeters.

Tom Brennan said:

"What are the benefits of moving the crossovers to a point in front of the amp stage?"

Passives at this level have pretty much the same benefits as active crossovers----lower IM distortion, increased dynamics, improved transients, better amplifier/speaker coupling.

Click to expand...

Well, ozmoid, it's NOT really as simple as moving the xovers before the power amps. The "process" Tom describes would use a "passive" vs "active" network to perform frequency separation. The passive xovers in the speakers get an answer that depends upon the inductance and resistance of the individual drivers. If you simply moved the caps, coils and resistors and did NOT compensate for the drivers, you would get a far different answer.

Tom Brennan said:

"How would you control the signal at the mid/tweeter point"

Use amps with level controls or insert a passive level control.

Click to expand...

Tom, I think you misunderstood the question here. What ozmoid is asking is how do you take the passive network, and with two amps DIRECTLY power three totally different drivers (woofer, mid, tweeter).

The answer is you can't! To do this you need to tri-amp.

Tom Brennan said:

"fried drivers are not easily replaced in these AR's,"

Caution is understandable, many bi-ampers use protection caps between amps and tweeters.

Click to expand...

Tom, the problem with this solution is we are going in a circle. For example, the entire passive network for the tweeter consists of ... a single cap!

ozmoid, the risk of frying those sensitive drivers is actually greater with a single amp. It's well documented that more tweeters are "blown" by using underpowered rather that over-powered amps.

What happens when we underpower is the amp starts "clipping" and this creates huge amounts of high frequency distortion. It's that massive amount of distortion (far beyond what is normal in music) that fries the drivers.

Reason I recommend a low power amp for the mids and tweeters is to limit damage that would occur if someone turned the volume full ON once the power dissipating resistors (pots) have been removed. These pots perform no useful function once we bi-amp and are further a souce of "unwanted" resistance.

Hope this helps ...

Regards,
Jerry

PS: My woofer amp is rated at 170 wpc @ 4 ohms
Mid/tweeter amp is rated at 30 wpc (on a good day)

"If you simply moved the caps, coils and resistors and did NOT compensate for the drivers, you would get a far different answer."

Yes, a passive filter at the line level will have different values than at the speaker level. There are websites that show you how to figure such circuits.


"Tom, I think you misunderstood the question here."

No doubt of that, people should make disgrams and post them.

"Tom, the problem with this solution is we are going in a circle. For example, the entire passive network for the tweeter consists of ... a single cap!"

Yes, I've often thought the same thing and have never used protection caps . But some people do and it's good for the bi-amper to know about such things.

onplane said:

The answer is you can't! To do this you need to tri-amp.

Click to expand...

THAT's what I was trying to figure out - how the crossover would handle the signal distribution in that configuration - it can't.

Considering giving this a go, trying to decide which pair to use. I may actually just open up a pair and bring an additional binding post out to allow full separation of the driver network. Nothing adding another "strap" won't fix if I move back to one amp.

Thanks, Jerry.

Todd

ozmoid said:

THAT's what I was trying to figure out - how the crossover would handle the signal distribution in that configuration - it can't.

Considering giving this a go, trying to decide which pair to use. I may actually just open up a pair and bring an additional binding post out to allow full separation of the driver network. Nothing adding another "strap" won't fix if I move back to one amp.

Thanks, Jerry.

Todd

Click to expand...

ozmoid, that would be the "correct" way to passively bi-amp.

For the 4th terminal you could use a brass screw, a couple of nuts, locking washer and a wing nut.

The beauty of "Jerry-rigging" is you can try bi-amping WITHOUT opening the boxes. Then if you don't care for the sound, you simply replace the straps.

Regards,
Jerry