Passive Bi-Amping

[Alan Brown]

Do many others do this? I did a quick search which revealed a little interest in the subject, but are there many of you out there who have made this work?

I'll start the ball rolling - I use 70 watt monos in the main, with a passive pre. I have 40 watt monos from the same range that I badly needed to sell a year ago to free up some funds. The amps designer suggested I try a bi-amp set-up for fun before I sold them on, and - I've still got them. The difference was astonishing, natural, the music engaged me - even arrested my attention...I don't know how to describe sound with words, but it is sublime.

The difference between 'upgrading' the 40 watt amps to the 70 watt amps was not subtle at all, a good, value for money improvement. But the difference I heard between running the 70 watt amps on their own and adding the 40s into the mix was simply vast - a night & day contrast. So now I (have to!) run 40 watt monos on treble, 70 watt monos on bass.

It isn't just a question of power - I've heard my speakers on the end a 500 watt MF. Not a patch on this setup - not even nearly. Somehow this system, configured as it is, is greater than the sum of it's parts.

The surprising part is this: The speakers are a compact 2-way design, the xover is a simple 1st order. Admittedly the efficiency is low, around 83dB, but even so they sound so natural and unforced, even at low volume.

I have heard of folks who found this approach not to work for them, but I think a few more should try it.

Are there a few more of us out there who do Bi-Amp?

 

[Tenson]

It makes most sense for inefficient speakers. When an amp distorts with HD, IMD or clips (which 70watts will do with dynamic material in an 83dB speaker) the distortion created is high frequencies that go right to the tweeter. So if you keep the amp for each driver separate then it keeps any distortion only in the band of frequencies that driver covers. It can also allow you to choose a most suitable amp for the range covered like a big powerful amp for the bass and a less powerful class A amp that has less distortion when producing low power signals for the treble.

I haven't done passive bi-amping but I do like active systems with are a step even further by limiting the signal the amps receive to each band as well and giving them a direct 'grip' on each driver with no filters in the way.

 

[Alan Brown]

I would agree that 'grip' on the drivers is a good thing, I found just about everything improved with 'isolating' the drivers.

I saw a point re bi-amping that seemed to make sense recently, that 'reaction' from multiple drivers as they reversed their direction of travel was not sent back to the amp as load. Therefore in a bi-amp setup each amp is free to transmit its signal to it's own driver without having to deal with the 'reaction' of other drivers at different times. Therefore a cleaner, more detailed signal is sent to each driver, which in my experience results in a wonderfully communicative speaker.

I hope that makes sense, the inverted commas probably give away my lack of technical understanding. The principals there make sense in my mind.

I should imagine there are many points to be made about the differences between active and passive bi-amping wrt the pros/cons of introducing active components and the effects of passive xovers. But I wont be making those points....

I bet active setups exhibit a similar range of strengths, but passive bi-amping is compatible many a modern speaker.

I think more should give it a go. View it as active on a budget..I also think there can be additional strengths on the part of bi-amping as opposed to active if the speaker is chosen wisely, and has a simpler xover.

 

[Levi_501]

Question, is passive Bi-amping different from what I would call 'normal' Bi-amping ?

I Bi-amp, always have done. I have two Maias controling the Kefs

 

[Alan Brown]

'Passive' i.e. - not an active xover.

I think that there are more than 1 way to bi-amp too, e.g., 1 amp per side, or 1 amp on treble, 1 amp on bass. I presume the the latter is 'vertical bi-amping', though I may have got that wrong.

I don't really know which of these applies if an amp is replaced by monoblocks.

Bi-amping works for you as well then?

 

[Levi_501]

Bi-amping IMHO is fab !

I currently use one for mid and treb, the other for base, although I have thought many a time to have one for the left speker and the other for the right.

 

[Tenson]

I wouldn't. Separating the frequency bands is half the point. Having one per speaker is a bit silly really.

 

[Seeker_UK]

I wouldn't. Separating the frequency bands is half the point. Having one per speaker is a bit silly really.

Isn't that what you do with monoblocks though?

 

[Tenson]

Yes but I don't think monoblocks make much of an upgrade over a good stereo amp. If you have one stereo amp per speaker, with one channel for the bass and one for mid/treble then when the bass kicks, the voltage rails will be modulated and the other channel driving the mid and treble will suffer. Worse, if the bass channel sucks up all power from the PSU and clips the rails then there will be nothing left for the mid and treble (unless the PSU has far more current ability than the voltage swing can push in to the load). It just doesn't make sense. Have one amp for the left and right bass, and one for left and right mid/treble and you separate the mid and treble from any such problems. Although with the mid/treble amp dedicated to low power signals it would be wise to choose one with low distortion at low signal levels (i.e. low crossover distortion or a single ended amp).

 

[Alan Brown]

So the best approach is....bi-amp with monoblocks! Result!

Levi - Glad you rate it too! seriously overlooked as an upgrade imo. As regards your ideas re left & right amps - well, have your cake & eat it! Go left/right/up/down at the same time (Obviously need a really understanding wife to tolerate all those amps).

Tenson - I do disagree over the monoblock issue - I've heard the stereo version of my monoblocks, and whilst I can't vouch for what is inside, the monoblocks were a massive step forward in terms of dynamics & imaging.

Just noticed this bit:
"It makes most sense for inefficient speakers. When an amp distorts with HD, IMD or clips (which 70watts will do with dynamic material in an 83dB speaker) the distortion created is high frequencies that go right to the tweeter."
What power do those Kenzai's need to get going then? They are 83dB as well IIR, the same as my Royds. Sounds like they would really like to be bi-amped. With mono-blocks!

(edit: having said all this, If I had to I would certainly prefer to bi-amp by separating treble from bass rather than left speaker/ right speaker. It does seem the more effective of the two approaches.)

 

[Tenson]

My point was if the same amount of money was spent on components in a stereo amp as on monoblocks I think it will be just as good.

The Kensai work very well on the end of my 50watt RMS Cyrus 3. But that amp has a big PSU and will probably kick out as much as 300watts for short bursts which the Kensai love. As for bi-amping and such I don't want to say too much yet, but I have some plans

 

[dreftar]

The Kensai work very well on the end of my 50watt RMS Cyrus 3. But that amp has a big PSU and will probably kick out as much as 300watts for short bursts which the Kensai love.


ITS curious that the smaller the loud speaker gets the larger the amps need to be, or at least their power supplies.

 

[Tenson]

I wouldn't say needs to be. Benefit from is a better term. A lot of the time very slight clipping is inaudible, and differnt amps will clip in 'nicer' or 'worse' sounding ways. Its just that the perfectionist in me would like to avoid it all together.

 

[Levi_501]

I believe (although I open myself to be corrected if someone knows) that the amps I have (Rega Maia's) are actually built as two completely seperate amps within one box.

Hence there is two seperate mains transformers (the donut things) inside.

If you say power would drop for the treble if there was a sudden deep base sound; surely it would be more difficult for the amp to drive two base drivers when there is a sudden drop.

You may have uessed, I am not a sparky !

 

[Alan Brown]

Tenson,
I must be missing something here.

You don't bi-amp Kensai's because :"The Kensai work very well on the end of my 50watt RMS Cyrus 3. But that amp has a big PSU and will probably kick out as much as 300watts for short bursts"

How is it when it comes to my 70 watt monos : "When an amp distorts with HD, IMD or clips (which 70watts will do with dynamic material in an 83dB speaker) the distortion created is high frequencies that go right to the tweeter." , But when it comes to your 50 watt Cyrus it can produce 300 watts on demand?

A fifty watt amp is fifty watts, unless we start down the argos catalogue route of describing output in ridiculous RMS pseudo-figures like they do on their glorified karaoke machines. By those measures the large psu in my larger amps can produce something near 600 watts during peaks in the music. So if that is the case why would I hear any improvement through bi-amping? They are still seventy watt monos and it is daft, even misleading to say otherwise, even if it is just for a microsecond at a time.

I don't think clipping is the issue at all, even with an 83dB load - certainly I heard no clipping at all. How can an amp - yours or anyone's - produce signal larger than its own voltage rail without clipping anyway? Clipping therefore can't be an issue when it comes to bi-amping so why the improvement?

The thing is: Changing from 40 watt monos to 70 watt monos was great, as you would expect. But moving to Bi-amping with the whole set was just light years ahead - huge difference. That's not because I now have a thousand argos-watts on tap, it is because the channels are separate and the treble/bass is separated, etc.

There is more to passive Bi-amping than meets the eye, certainly a lot more than clipping. In fact, even the benefits already mentioned don't quite go all the way to explaining the improvements on hand. Bi-amping somehow (if done correctly) is greater than the sum of it's parts.

 

[Alan Brown]

Levi - in some ways you are bi-amping in both senses. Vertical bi-amping (separating treble / bass) with your two stereo power amps.

Their being dual mono in design means they are also already separating left & right channels. I don't know your amps I'm afraid, but if you want to improve them further without changing too much get a pre-amp with individual left & right volume controls. This makes a nice value for money improvement in my experience, although they are a pain to adjust each time!

 

[Tenson]

Hi Alan,

I wasn't saying I don't bi-amp the Kensai because they work fine with 50watts. I said they sound great, even with the 50watt Cyrus. It will likely clip, but it doesn't sound too bad, largely because the Cyrus has good transient capability for a 50watt amp. Bi-amping will no doubt bring benefits, although for the Kensai I have other plans in terms of an upgrade path.

With regards to 50watts being 50watts, this is not really true. For sure, the 'Peak Music Power' Argos type figures are useless in most cases as the only measure of power, but a 50watt amp with good transient capability from the PSU will do things an amp with a small PSU can't.

You are right that you can't have more power than the voltage rails will allow, but assuming reasonable voltage swing you still need current ability even if only for a short time. To take the Kensai and Cyrus 3 example. The Cyrus is rated at 50watts RMS - meaning it can do that sustained. But the PSU in there is something like +/-25V and 250VA (i.e. 250watts). Taking all things in to account you are looking at about a 60V swing. So into the minimum 6 Ohms of the Kensai there is potential for 600watts to flow! We know that the Cyrus can in fact deliver 250watts (or 125watts a channel) for a period of time until the output transistors start to heat up too much, which is long enough for a transient.

If the amp had a smaller PSU, less voltage and or less current it could still be rated at 50watts RMS but would not deliver those short term large peaks of power when called for.

The Naim Supernait for example is rated at 80watts RMS, but it boasts of having 400VA (watts) of peak capability. Its probably rated as being 80watts because it can't supply more than that for a long period of time without getting too hot.

I don't think monoblocks are much better than a stereo amp just because it separates the two channels power supplies. Most peak power is needed in the bass and that is mono in most recordings. So an amp with both channels running from one, double size PSU would be pretty much just as good as monoblocks running two half size PSUs I think.

Hope that explains myself better.

 

[Alan Brown]

Hi Tenson,

"If the amp had a smaller PSU, less voltage and or less current it could still be rated at 50watts RMS but would not deliver those short term large peaks of power when called for."
Sure this makes sense & I agree with it. There are some really good 'small' amps because of this, and really quite bad 'big' amps as well.

But an amp with bigger RMS power and a well designed power supply is better, therefore RMS power is still the only real benchmark for comparing amps and their potential during transient responses. Even so you still have one amp dealing with all those driver movements - at least 4 drivers - as they reverse their travel. This is part of the reason (aside from doubling up on power supplies & supporting components) why monoblocks have a further advantage over an integrated, you are halving what the amp has to deal with from the speaker(s) and doubling its capability to respond correctly to source material. Further, it deals with those driver 'reflections' from only one channel - A huge advantage for imaging and communicating effectively.

Exactly the same principles hold true for Bi-amping. Therefore some of the the arguments for Bi-amping are the same as those for monoblocks.

To say the kensai is great although it is likely clipping is selling it short by a long way. You definitely said that even with an amp a third larger than your Cyrus there would be clipping into an 83dB load. If you mean dynamic clipping then peak power will not help you anyway. Remember Bi-amping can be greater than the sum of its parts, the points I made above only go part way to explaining the actual results one can achieve.

What kind of clipping are you getting on the Cyrus - Voltage or current? I would think that with your amp into that load current clipping could be present quite a lot, and if that is the case then it doesn't matter one little bit how much peak power the Cyrus can chuck out for a transient response. You need (Clarkson/Steve Allen quote inserted here) More Power. And more amps, imo.

There you are, another late night response. Once my children are over the bug I'll have less time to compile this sort of post. It helps me get ready for sleep though - and, I suspect, anyone else who reads through it all! :SLEEP:

Seriously folks - Bi amping is great, Monoblocks are great. Combining the two is greater still.

 

[shrink]

IMHO using one amp per side in biamping is a far better idea than using one for treble and one for bass.

Tenson, you said that if you use one amp per side, then when the bass kicks, it leaves nothing for midrange and treble?

Except that typically in a biamped system, its the bass amp thats doing the midrange anyway, so the midrange is already going to be compromised, no matter what amp you attach it to.

In fact if you use one amp for both channels, the effect is even worse. if one amp, on one side is using one channel for bass, then its power supply only ever has to deal with 50% of the bass. If you use one amp for both L&R bass, then the power supply has to deal with ALL the bass. This increases the chance of the amp clipping, and screwing up the midrange in the process.

At least having one amp on one speaker, the load over the power supply is more even, the amp will strain less to provide the bass, and the likelyhood is the the power demands of the treble channel will rarely be enough to cause a problem.

all IMHO of course, but i believe this technique marries the best parts of monoblocks and biamping. The channel separation of monoblocks is acheived, but with the frequency seperatation of a biamped setup.

In your way, you have a whole amp just running a pair of tweeters, and a whole amp, maxing itself out running midrange and bass at the same time.

Unless the bottom amp was substantially larger and better endowed than the treble amp, your likely to compromise.

In any event, Ive always thought biamping was a waste of time. For the price of two whole sets of amps. You could likely just buy a single better amp, which will for the most part, always sound better than two cheaper amps.

Doesnt matter how much you biamp/triamp etc, the bottleneck will always be the ultimate performance of the cheaper amps your using. A better quality stereo unit (or better monoblocks depending on your setup) will pay dividends over wasting cash biamping.

That said, im happy that my speakers dont even allow it, so makes my life much easier

 

[Tenson]

Hi Tenson,

..an amp with bigger RMS power and a well designed power supply is better,

Agreed.

therefore RMS power is still the only real benchmark for comparing amps and their potential during transient responses.

I don't follow you here. Sustained RMS power tells you very little about how an amp will react to a short term transient requiring vast power.

Even so you still have one amp dealing with all those driver movements - at least 4 drivers - .... Further, it deals with those driver 'reflections' from only one channel - A huge advantage for imaging and communicating effectively.

The difference between monoblocks and a 'stereo' amp is usually just the sharing of the PSU. The actual amp modules are still one per channel. Otherwise it would be a mono amp. So each speaker still has one amp per channel, only powered by a single supply. Due to this, the back EMF (electro-motive forces) from the drivers are controlled just as well as they would be from monoblocks as the output transistors of each amp module, one per channel, present a very low impedance to each speaker.

You definitely said that even with an amp a third larger than your Cyrus there would be clipping into an 83dB load.

Quite likely yes, with dynamic material and high volume levels; something like uncompressed percussion in an orchestra or a well recorded rock or jazz group. You don't tend to get much clipping on very compressed music (well except that on the recording!), even with only modest RMS power because you don't need much average power for it to 'sound' very loud.