Need help biamping SL3s (tube/ss)

[xenonaut]

Hi,

I've got a pair of SL3s and recently picked up a tube amp that I'd like to try to use to drive the panels. The tube amp is a Conrad-Johnson Premier 11A, which I'd like to try to mate with a Musical Fidelity A3.2CR driving the woofers. I'm doing this passively, using the internal crossovers (nothing fancy).

The problem, of course, is that the amps have different sensitivities. I assume the answer is that I need to put some inline attenuators before whichever amp has the greater sensitivity, but how can I accurately determine how much attenuation I need? Ideally, I'd like to have a simple, fixed attenuator at the correct resistance to balance the amps (and spare myself the expense of a high-end stepped attenuator).

Can anyone who has been through this offer any advice?

Thanks in advance,

Hal

 

[JonFo]

Simple thought is to obtain any kind of attenuator and put that in circuit between the pre and the amp you suspect has most gain, and then adjust until it balances (using pink noise and SPL meter) with the other driver.
See if your local audio store will lend you a passive preamp or attenuator for a few days.

Then measure the insertion loss of the attenuator with an ohmmeter and that's the value of the resistance you'd want in your fixed setup.

 

[xenonaut]

Hi Jonathan,

Thanks for the response.

I've got a Radio Shack SPL meter and a test CD with pink noise, so I guess that ought to be enough to get balance. But I can't think of an obvious/easy way to provide variable attenuation. I have a second (active) preamp that I could insert between the primary preamp and amp to use for attenuation, I suppose. How would I then measure the resistance? Center wire (+) on input side to center wire on output side? Will this work, with an active preamp? Alternatively, I could just look for an inexpensive potentiometer to rig inline; if I do this, do I want to be adding resistance to center or return?

I was hoping I could do something like this:

Connect amp A, set volume level, play pink noise, measure and note voltage(amperage?) across speaker terminals. Connect amp B, leaving volume setting untouched, play pink noise, make same measurement across speaker terminals. Then, based on comparison of measurements, calculated needed resistance to get equal signal strength. Does something like this sound feasible? If so, is it indeed DC voltage that I should be measuring, and if so, approximately what range should I expect to see?

Forgive me if these are naive questions, this is new territory for me. I had been biamping the SL3s horizontally, with a matched pair of the MF A3.2CRs, but I was hoping for a bit of that "tube magic" for the panels. But obviously, using dissimilar amps complicates things a bit...

 

[raanan]

What I did-when I was biamping my bedroom system SL3s was:
I powered the panels with a pair of 12 watt Triode Class A tube monoblocks,and in order to achieve balancing the respective volumes I ran the SS amp through an electronic crossover cutting at precisely 250 cps. which fed the SS amp that drove the woofers.The electronic crossover enabled me to control the respective volume of the woofers.
It is even possible that I may have taken the highs out from the crossover too-feeding the source directly into the crossover without using any preamp.
The tube monoblocks had stiff power supplies and I was surprised that they drove the panels really well-no lack of highs and dynamics.The same tube amps could not get more than a squeak out of my CLSs before they started distorting.

 

[khenegar]

biamp

if i were you i would look for another conrad amp same as you have and use it, conrad has a great bottom end. Have you listened without biamping you might be suprised. if you are using a sub no need to biamp. use your conrad and let the sub do the work where it counts. good luck and happy listening

 

[xenonaut]

Hi Raanan,
Thanks for the reply. Yes, it occured to me that I could add a crossover, but
(a) I'm trying to keep a minimalist/purist signal path and
(b) that still doesn't solve my problem of how to correctly compensate for the differing amp sensitivities.

 

[xenonaut]

Hi Khenegar,
Thanks for the feedback. I thought about this initially, as well.

In fact, I *did* try the Premier 11 by itself, without biamping, and felt that it was struggling a bit. I'm not sure if the problem was the complexity of the load or the CJ's lack of (adequate) current or damping, but in the end, the bass was loose and sloppy (compared to when driven with a total of 4 channels @ 130 watts solid state, from the pair of MF 3.2s).

There are also some practical issues, in that I am currently in Europe and it is extremely difficult to find a reasonable deal on a Premier 11 (I looked for 2 years for the one I have now). Also, my wife wants to be able to use the system sometimes. If I have a solid state amp on the woofers, I can just leave jumpers on the speakers and disconnect the C-J most of the time, so that she can switch on the system for casual listening without dealing with tubes, but I can remove the jumpers and wire in the C-J for "serious" listening.

So, in the end, the option of biamping with a pair of Premier 11s is probably not feasible for me. I agree, however, that this might be a good solution, were it not for my practical constraints. Another Premier 11, powering only the woofers, would see an easier load and probably give me adequate bass, as well as remove the need for level-matching the amps, of course.

 

[raanan]

Hi Raanan,
Thanks for the reply. Yes, it occured to me that I could add a crossover, but
(a) I'm trying to keep a minimalist/purist signal path and
(b) that still doesn't solve my problem of how to correctly compensate for the differing amp sensitivities.

Some crossovers have pots for highs and lows.

 

[xenonaut]

Some crossovers have pots for highs and lows.

Right, of course -- I was talking about actually determining what the proper levels/settings would be (more in the direction of Jonathan's suggestion of pink noise and an SPL meter).

Sorry if I was unclear.

But I believe profoundly in a minimalist signal path, so the less I have to insert, the better (from my perspective). That said, I recognize that there is a beneift to asking each amp to handle only a portion of the frequency spectrum, and this benefit could potentially offset some of the loss in signal purity.

But I'd prefer so have a single resistor inline to achieve the level-matching. Yes, maybe the amps work a little harder by reproducing parts of the spectrum which in the end are filtered out by the crossover network, but I believe there are still significant gains due to the simplified load of panel OR woofer (as opposed to panel AND woofer), as well as the obvious gain in net power.

 

[Spike]

A quick peek at the S'phile's review of CJ Premier 11 yields a measured voltage gain of 25.7db. For practical purpose, we'll say 26db of gain for the CJ. The spec for MF A32CR states 30db of voltage gain. So, the MF A32CR needs to be attenuated by about 4db (-4.3db if you want to be exact based on the measurements of ONE CJ amplifier).
I searched around but found no attenuator with 4db value so you'll need to have one custom-made. There are a few 3db attenuators around but you'll need to figure out whether the 3db of attenuation will work in your system. An easy way to experiment is to switch the bass control of the SL3 to the -3db position. This will simulate a 3db cut of the woofer to see whether your amplifiers come into balance.
Now, if you really want to attenuate at the input of the A32CR, you cat get these fixed value inline attenuators from Harrison Labs. From my experience, I cannot perceive any differences when there's -1db difference between the 2 attenuators so a -3db attenuator will probably work for me. You should try out for yourself.

Good luck
Spike

 

[Bernard]

Connect amp A, set volume level, play pink noise, measure and note voltage(amperage?) across speaker terminals. Connect amp B, leaving volume setting untouched, play pink noise, make same measurement across speaker terminals. Then, based on comparison of measurements, calculated needed resistance to get equal signal strength. Does something like this sound feasible? If so, is it indeed DC voltage that I should be measuring, and if so, approximately what range should I expect to see?

You could do it that way, but you really need an oscilloscope to do it. Basically you would calculate the gain of each amp by comparing input and output, and then calculate the drop required to match the gains. You would need a pair of resistors per channel. It is AC voltage that would be measured.

Alternately, get a couple of 100K potentiometers and put them before the louder amp; adjustment would be by ear.

 

[xenonaut]

Hi Spike,
Excellent! I should have thought of this myself (and doubtless would have, were I only a little more clever). But this is something I can start with -- I'm looking for 4.3 dB attenuation. I should be able to figure out how to make something for that -- maybe a single vishay resistor in the return line(?) of each channel? All I need is the correct value, but I should be able to find a formula for that somewhere.

Thanks a bunch!



Bernard,

Thanks, too, for your response. Unfortunately, I don't have access to an oscilloscope (and don't have frequent enough need to justify the expense).

Longterm, I don't want to have potentiometers in the path, but short term, I could use them to fine-tune the levels (Jonathan's suggestion of an SPL meter with pink noise is something I could manage) and then measure the resistance I end up with; I could then swap in a pair of fixed resistors.

Do I want my resistors in the send (+) or the return (-)?

Thanks again (everybody)!

 

[Spike]

Attenuation info

I should be able to figure out how to make something for that -- maybe a single vishay resistor in the return line(?) of each channel? All I need is the correct value, but I should be able to find a formula for that somewhere.

Here's the link to Attenuator design info (from Tweak Audio) you can use as reference. To arrive at the actual resistor value for a 4.3db attenuation is left as an exercise for the reader :music: Who knows, somebody knowledgeable may chime in and help with this exercise

Spike

 

[raanan]

Hi,

I've got a pair of SL3s and recently picked up a tube amp that I'd like to try to use to drive the panels. The tube amp is a Conrad-Johnson Premier 11A, which I'd like to try to mate with a Musical Fidelity A3.2CR driving the woofers. I'm doing this passively, using the internal crossovers (nothing fancy).

The problem, of course, is that the amps have different sensitivities. I assume the answer is that I need to put some inline attenuators before whichever amp has the greater sensitivity, but how can I accurately determine how much attenuation I need? Ideally, I'd like to have a simple, fixed attenuator at the correct resistance to balance the amps (and spare myself the expense of a high-end stepped attenuator).

Can anyone who has been through this offer any advice?

Thanks in advance,

Hal

Nothing drove my SL3s better than one Behringer EP1500.

 

[Tj Bassi]

Hal-

Try this link....

http://67.19.167.226/~tdacquis/forum/showthread.php?t=1837&referrerid=680

Back in '05 I tried the same thing you are doing, but with a pair of reQuests. At the end of the experiment, I was of the opinion that the woofer section of that speaker was not substantial enough to benefit from bi-amplification (as opposed to something like, maybe, a Prodigy).

Or to rephrase, the cost and effort of bi-amping the reQuests werent justified by the improved sonics.

Off the top of my head, I dont recall all the details of that experiment, but I should have notes/photos at home and will search for those later. Maybe something there might be of help....

Tj

 

[xenonaut]

SL3 Amplification

Nothing drove my SL3s better than one Behringer EP1500.

Hi Raanan,

I don't doubt that -- the EP1500 is a very powerful amp, some 10x more so than my humble Conrad-Johnson. I'm sure it drove the SL3s quite easily.

I think we simply have a different approach to system building. I've got a C-J tube preamp (PV12A) and am trying to integrate a C-J tube amp. At the moment, I'm really interested in putting tubes on the panels. But the 70W tube amp I'm using just doesn't have enough muscle to get a tight grip on the bass (though as Khenegar suggested, a pair of them might).

However, the Musical Fidelity A3.2CR (I have a pair of these) are ok with the bass, so what solid state amp to use is not really the problem. The question really goes to (a) how do I best determine the appropriate amount of attenuation to achieve a proper balance between the dissimilar amps, and (b) what is the least-intrusive intervention I can add to achieve that attenuation. According to Spike's calculations, I need about 4.3 dB of cut on the M-Fs ( I should be able to double-check this with Jonathan's suggestion of pink noise and an SPL meter).

So I guess the only question that remains is how to get 4.3 dB of attenuation with minimal impact on my signal. Spike's link to attenuator info is probably enough to get me started. Looks like I need to do a (logarithmic) calculation to determine the appropriate resistor values, and assemble an attenuator (which appears to require an "inline" resistor on the return, and a "shunt" across send/return).

 

[xenonaut]

Hal-

Try this link....

http://67.19.167.226/~tdacquis/forum/showthread.php?t=1837&referrerid=680

Back in '05 I tried the same thing you are doing, but with a pair of reQuests. At the end of the experiment, I was of the opinion that the woofer section of that speaker was not substantial enough to benefit from bi-amplification (as opposed to something like, maybe, a Prodigy).

Or to rephrase, the cost and effort of bi-amping the reQuests werent justified by the improved sonics.

Off the top of my head, I dont recall all the details of that experiment, but I should have notes/photos at home and will search for those later. Maybe something there might be of help....

Tj

Hi TJ,

Thanks for the info.

From the thread, it appears that you tried biamping with compensating for the differing amp sensitivities, and found that you had poor balance between highs and lows (not surprisingly).

If I read correctly, it appears that you came to the conclusion that your 200W ASL Hurricane monos were up to driving the speakers full range, and that there was therefore no benefit to biamping. I can see understand that conclusion, in your case, but my C-J Premier 11 (70W stereo) is not the "monster amp" that the Hurricanes are, and I find myself short on bass control. As has been mentioned in this thread, a *pair* of Premier 11s (giving me a total of 4x70W) would likely be adequate, but it's more practical for me to get the balance right on a tube/solid state biamp setup than it is to hunt day (and pay for) another Premier 11.

If you did anything else in the course of your experimentation (other than what is described in the thread), I'd certainly be interested in hearing about that.

 

[Bernard]

and assemble an attenuator (which appears to require an "inline" resistor on the return, and a "shunt" across send/return).

I don't understand your explanation, so try mine: An attenuator consists of two resistors connected in series from the output of your preamp to ground, and at the junction point of the two resistors a connection is taken to the input of your power amp. In order to minimize the number of cables and contacts, I suggest that instead of building an external box, you solder the attenuator network within the preamp or power amp, unless they are under warranty.

 

[xenonaut]

I don't understand your explanation, so try mine: An attenuator consists of two resistors connected in series from the output of your preamp to ground, and at the junction point of the two resistors a connection is taken to the input of your power amp. In order to minimize the number of cables and contacts, I suggest that instead of building an external box, you solder the attenuator network within the preamp or power amp, unless they are under warranty.

Hi Bernard,

Thanks for all your input.

My understand is something like the left image here. I guess I see what you are saying, and yes, this is a more accurate/exact description than mine. I was only expressing the realization that I needed two resistors, one which you could consider "inline" (though it is on the hot/send/center, whereas I thought it was on the ground/return/outside), and another, the shunt, which goes between the hot and the ground.

Though none of my equipment is still under warranty, I think I'd prefer to build a small external box (or tube, maybe) than to do the attenuation internally. I've read that it's better to put the attenuator on the amp side, so the preamp sees the load of the cable instead of the resistors, but I have 2 of the amps, am not sure if I'll keep one or both, and don't want to do anything that might affect resale value.

You have a beautiful system, BTW - I'll comment over there, so I don't hijack this thread....

 

[raanan]

For the bi-amping experiment-I tried using amps that I had here:2X LEAK TL12.1 for the panels,DPA 50S Class A for the woofers.The results were very good and musical.
Using the cheap Behringer gave even better results.Lively and transparent.
What I mean is that no matter how good the amps that you have,it may be a better idea to sell some of them after you have tried the cheap Behringer,which is a great match with the SL3s.