Active Bi-Amping/Tri-Amping FAQ

[Born2RockU]

On 5/26/2011 at 3:58 PM, Rudy81 said:

I'm convinced!

Rudy.  Can you share WHAT Rack maker and model # it is ...or, send me a internet link.

I like this approach and am wondering what model/maker this rack is from . 

[Rudy81]

Unfortunately, I have no clue what manufacturer this is.  Like many other things, I found it on Craigslist for $75.  I was a little beat up and needed a fresh coat of paint.  I also made a base for it with heavy duty casters underneath.  That turned out to be a great idea.  I have needed to move it several times over the years.  Judging by the price of new racks, that was a great way to go.

 

 

[Kev the ironworker]

  There are two Luxman tube crossovers on ebay right now  (very rare and perfect for Bi-amping) …. guys been selling some good stuff (Audio Research, Power conditioners, Antique Sound Lab, … and pots ! good pots though … All Clad... guy's selling quality and has a 100 % rating...

[etc6849]

On 3/31/2018 at 12:00 PM, Chris A said:

get a handle on Class D amplifiers to simplify my amplifier rack (hopefully reducing the need for class A amplifiers that heat the room)

 

I'm glad to finally see measurements of the hypex ncore modules.  These measurements look really really good, and the amps are very small.  ATI also makes an 8 channel amp that uses OEM ncore modules...  now I see why.

https://www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-hypex-nc400-diy-amp.5907/

 

 

[NBPK402]

45 minutes ago, etc6849 said:

 

I'm glad to finally see measurements of the hypex ncore modules.  These measurements look really really good, and the amps are very small.  ATI also makes an 8 channel amp that uses OEM ncore modules...  now I see why.

https://www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-hypex-nc400-diy-amp.5907/

 

 

I have been reading the test/reviews for a while there. Interesting that Accuphase amps tested horribly, and some inexpensive products beat the more expensive and recognized Companies.

[Khornukopia]

.

Edited April 24, 2019 by Khornukopia
double post

[Khornukopia]

On ‎4‎/‎22‎/‎2019 at 6:22 PM, etc6849 said:

measurements of the hypex ncore modules.  

https://www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-hypex-nc400-diy-amp.5907/

 

 

 

Thanks for the link.

[parlophone1]

I would appreciate if someone could direct me how to put up an active 2-way system that includes turntable.

I am aware of AVR receiver which could provide for a preamp section.

 

However, recently I had the opportunity to get my hands on Crown SL2 preamp, which to my knowledge would be suitable to provide for an active two-way system that would include a turntable preamp.

 

What would be used as a preamp in this configuration?

 

In fact, to my opinion it would be very useful to do instructions on how to assemble:

1. An all digital system,

2. A system that include an analog component(s).

 

Partly, the reason for posting this question lies in the fact that I do not want to part with my LPs as a source of music. I am sentimentally attached to it.

 

Thanks.

[Chris A]

Just insert the output of your analog phono preamp into the tape or aux inputs on your Crown preamp.  The Crown preamp output would be connected to the input of (at least) a 2-in, 4-out DSP crossover.  Remember that the default input to a DSP crossover is analog... 

 

 

The above connection diagram is taken from here.

 

To do an all-digital source to the output of the DSP crossover, you will need a preamp (or source with volume control) with digital output, such as AES3 or S/PDIF, and a DSP crossover with an AES3 or S/PDIF input.  User etc6849 (above...his name is Ellery) uses JRiver running on a PC out through an AES bus (RME AES32 PCI-E 16 channel input/16 channel out card) to a Xilica XD-4080 (note the "XD" instead of a "XP" in the name) to do multichannel all-digital connections through to the amplifier inputs.  This replaces the DAC and digital server in the diagram above, and adds a digital line between the preamp (or source PC, eliminating the preamp) and the DSP crossover.

 

Chris

[etc6849]

 

On 9/30/2019 at 7:49 AM, Chris A said:

This replaces the DAC and digital server in the diagram above, and adds a digital line between the preamp (or source PC, eliminating the preamp) and the DSP crossover.

 

What I found was that eliminating the extra unnecessary DAC improved performance.  I'm not sure if this is because the ADC was the limiting factor (which also gets eliminated when you send digital straight to your DSP), but likely.
 

Just in case anyone says all DACs sound the same: it depends on your entire signal chain and even the noise in your room.  When your amps are no longer the limiting component affecting overall dynamic range, the extra DAC and ADC can be the remaining factor needing improvement.  I bet most of the blue DAC measurements below are from DACs that will sound the same (all are very well engineered), but many of Amir's SINAD measurements of AVRs and even expensive DACs look so horrible, that for sure those using sensitive speakers with a consumer grade AVR/AVP or poorly designed DACs are missing dynamic range if they use a leading amp like the Hypex nCore NC400 or Benchmark AHB2.

 

I wouldn't have believed how much dynamic range people are missing, but the measurements don't lie:

https://www.audiosciencereview.com/forum/index.php?threads/master-sinad-distortion-comparison-graph-for-dacs.4814/

 

 

Amps are even worse too. 

 

 

Compared to the best known measurements of the Benchmark AHB2:

 

 

 

 

 

[PrestonTom]

I fully agree that eliminating an extra A-D & D-A conversion is always a good idea. BTW, those plots showing 70 dB or 140 dB down points on distortion are both impressive (even -70 dB is quite a bit actually).

 

However I am suspect of the -140 dB measure, since it is very difficult to measure that low of a noise floor. Even if it was true, it may not be terribly meaningful since the "true floor" will be determined by a host of other contributors in your audio chain (including the compressor on the refrigerator down the hallway etc).

 

Good luck,

-Tom

[etc6849]

1 minute ago, PrestonTom said:

However I am suspect of the -140 dB measure, since it is very difficult to measure that low of a noise floor. Even if it was true, it may not be terribly meaningful since the "true floor" will be determined by a host of other contributors in your audio chain (including the compressor on the refrigerator down the hallway etc).

 

Indeed.  It is very hard to measure that low from what I understand.  Amir has at least $50k in test equipment (his Audio Precision APx555 alone is over $25k?).  I definitely don't think most reviewers (unless they are technicians or test engineers) and many manufacturers (sad but true) have the necessary test equipment to measure that low.

[etc6849]

Here are some more measurements to consider.  I've owned several of these in the past before I found the path to audio nirvana 😀

 

It seems one can gain 30dB of dynamic range or more by using digital out from a PC and using a well measuring DAC and feeding an amp (even without tri or bi-amping).  Combining everything possible to reduce overall distortion (room treatments, signal chain, multiple subs, actively tri-amping) has created a system that most will never hear the clarity of.  Well worth the expense for the performance I'm getting.  I know I could be driving a nice car instead, but most of my gear will hold it's value better than a car.

 

Some AVR measurements:

 

Some AVP measurements:

 

 

 

[Chris A]

On 10/6/2019 at 8:35 AM, etc6849 said:

It seems one can gain 30dB of dynamic range or more by using digital out from a PC and using a well measuring DAC and feeding an amp (even without tri or bi-amping).  Combining everything possible to reduce overall distortion (room treatments, signal chain, multiple subs, actively tri-amping) has created a system that most will never hear the clarity of.  Well worth the expense for the performance I'm getting.  I know I could be driving a nice car instead, but most of my gear will hold it's value better than a car.

I believe there needs to be some discussion on the noise and harmonic levels here--and understand that it's not really harmonic distortion which is audible/objectionable, but rather the lower sidebands of modulation distortion.  In my experience, modulation distortion levels at -50 dB at the maximum listening level is quite good performance for the system, which is typically dominated by direct radiating woofer sidebands. 

 

The -116 dB figure that "Amir" on the other website you show is quoting ("Our best case hearing threshold is -116 dB SPL") is at least 50 dB below the best quality phono system ever created, and probably 40 dB below analog tape performance.  So in order to approach within 20-30 of the noise floor audibility of these DACs and amps, you would have to be cranked up to well above 100 dB in-room, with an extremely quiet listening room, i.e., much quieter than my listening room (note: outside noise levels are generally much higher):

 

 

...and have a music recording and source that's good to -100 dB of as-produced noise floor/dynamic range (of which I presently know of none of those). 

 

So, to put a little perspective on this, like many digital and electronic performance discussions nowadays that seem to continue to go off into infinity...the THD and noise numbers can keep getting lower and lower (and the wallet thinner and thinner), but the threshold of audibility has probably already long since been left behind.  If you put your ear inside the horn and can hear little or no noise when the preamp is cranked up for 95-100 dB in-room playback (which I almost never experience nowadays in an effort to retain as much hearing as I can for as long as I can), then I think that you've reached a reasonable point of diminishing returns.  I'd instead recommend putting your money into better loudspeakers (i.e., those without direct radiating woofers) and better quality multichannel recordings, once the noise floor threshold of audibility has been reached in-room.

 

Chris

[etc6849]

I guess it depends on space.  To be totally honest, if I can fit two horn loaded subs or 5 well engineered direct subs like I have now (9 sub drivers in total), I think my current setup would win in terms of measurable distortion at the MLP as my room is on the small side.  

 

What really surprised me was the performance I gained by going with 5 subs, especially considering I have three woofers in each speaker crossing to the 9 sub drivers (for example, there are 12 drivers working for a 100Hz tone).  This has led to a setup with measurements like .363% THD at 29 Hz at 90dB and .239% at 80 Hz at 92dB:

 

 

I did measure my room as you rightly point out.  With my lights off (LED lights buzz just a tiny bit), I get this RTA result:

 

I see plenty of places where less capable amps would make hiss depending on the spectrum of the noise content (and they did, which is why I spent so much to replace them).  This was the case with the old Emotiva XPR amps I had, and I think their SNR was around 92dB at 1 watt, but they had a lot of hiss when I tri-amped.

 

5 hours ago, Chris A said:

If you put your ear inside the horn and can hear little or no noise when the preamp is cranked up for 95-100 dB in-room playback (which I almost never experience nowadays in an effort to retain as much hearing as I can for as long as I can), then I think that you've reached a reasonable point of diminishing returns.

 

Yep.  I have reached that.  I do hear a very faint hiss when my head in the the horn.  However, I still may buy several 8 channel DACs at some point and do all the DSP work on the PC.  I haven't yet, mainly because I'm not sure I'd hear a difference.

 

 

 

[Chris A]

Small rooms have significant issues activating/filling up all the below-Schroeder-frequency room modes to even out the infrasonic and deep bass response due to the relative density of room modes as room dimensions decrease.  So having multiple subwoofers will definitely allow the activation of all the deep bass room modes via placement of the subwoofers at the nodes and anti-nodes of those room modes.  It also significantly decreases the required motion of the woofer diaphragms (square root of 2 reduction in maximum displacement for every doubling of the diaphragm area or horn mouth area of the subwoofers in-room).  Geddes' and Toole's method of increasing the number of subwoofers in-room is a good approach.  

 

Your ears are younger than mine, and your listening room is quite a bit quieter (and no doubt due to your extensive acoustic treatments) so a lower noise floor is probably desirable in your case. 

 

My room, perhaps not so much, and only if my middle ears happen to be clearer than usual because there aren't threshold levels of offending pollen/mold spores in the air.

 

Chris

[etc6849]

@Chris A

 

Music itself can mask THD too.  However, it seems some instruments are better than this than others such as pianos that don't mask amp distortion as well (since they produce overtones that are not perfectly harmonic per this link).

 

Here are some calculations I did for matching an amp to my bookshelf speakers.  I should have probably done this prior to tri-amping my main speakers, but the driver data was not readily available.  For most though, driver with horn sensitivity is likely available, so they can follow the same methods as below provided they have data for their amp (I used actual measurement data from the links below to ensure accuracy instead of the manufacturer's data).

 

Klipsch P-17b speakers:  http://e994010f48279d85b5d7-a0bc3fbf1884fc0965506ae2b946e1cd.r57.cf2.rackcdn.com/legacy-files/p17b_-_Spec_Sheet_635738650125214000.pdf

Emotiva XPR-1 amp measurements:  https://hometheaterhifi.com/reviews/amplifier/power-amplifier/emotiva-xpr-1-monoblock-power-amplifier/

Benchmark AHB-2 amp measurements: https://benchmarkmedia.com/blogs/application_notes/interpreting-thd-measurements-think-db-not-percent

 

For the XPR-1 amp, I calculate that it will have a worst case 34 dB SPL of distortion at a 80dB SPL listening level (assuming 20 dB peak, giving a 100dB SPL peak).  Also, using the XPR-1, my speakers would reach an SPL of 121dB, and have 54dB SPL coming from the amps distortion.  Since this exceeds the speakers max recommended output, likely the distortion at 121dB SPL would be greater than 54dB (amps + speakers)...  Don't worry, I don't plan to ever have them that loud.

 

If I use an AHB-2 amp with the bookshelf speakers, I'd see a peak level of 113.6dB (fine for any volume I'm going to use them at), and the distortion would be below 0dB SPL at -6.4dB SPL, so it would be impossible for it to ever be audible.

 

I'm ignoring the Benchmark DAC 3 DX (DAC I'm using), but it's not the limiting component for the XPR-1 amp case.  If both amps were played back at 80dB SPL with a 100dB SPL peak level, we'd see the AHB-2 contributes -20dB SPL (not audible) while the Emotiva XPR-1 contributes 34dB SPL (might be audible depending on room).  This surprises me some as the Emotiva XPR-1 measures as good as amps costing a lot more.  Based on the used market though, AHB-2 versus two XPR-1's seems fair based on used prices.

 

Given that the bookshelf speakers are in my living room, I am fine with 34dB SPL of noise from the amps as it is a noisier room and I doubt I will hear a difference.  I have less in the XPR-1's anyways as I bought them off craigslist

 

 

Bechmark AHB-2 case:

(I only calculated noise from the amp for max level below, but obviously any level below that will be even lower)

 

For fun, I re-did the max SPL from the AHB-2 amp using Pascals, then converted to SPL after cross-multiplying to obtain the new Pascal value...  Of course it is the same:

 

Application Notes I studied:

https://benchmarkmedia.com/blogs/application_notes/interpreting-thd-measurements-think-db-not-percent

https://benchmarkmedia.com/blogs/application_notes/speaker-efficiency-and-amplifier-power

https://benchmarkmedia.com/blogs/application_notes/audio-rules-of-thumb

[robi2]

Hello guys,

 

I’d like to know if exist a sort of pro amplifier as follows:

 

1)Coax or xlr digital input.

2)Optical digital input.

3)Input switch and master volume control (Could be by IPad app also).

4)six output channels amplifier.

5)Programmable filters and dsp for eq by external pc.

 

Thank you!!!

 

[Chris A]

None that I'm aware of accept optical inputs.  The requirement for 2-in (assumed), 6-out is also difficult to find.  I see amplifiers like the following 8-channel I/O DSP amplifiers (which I know nothing about):

 

https://www.qsc.com/resource-files/productresources/amp/cxd-q/q_amp_cxdq_8ch_specs.pdf

 

In short, I think that a separate DSP crossover box with S/PDIF (TOSLINK or coaxial) connected to a 6 channel amplifier box is a much more likely configuration--like a miniDSP 4x10 HD with your choice of multi-channel amplifier (6 channels in your case, leaving two unused output channels from the 4x10 HD).  I'd recommend balanced connections (XLR) to reject common mode noise.

 

Chris

[robi2]

23 hours ago, Chris A said:

None that I'm aware of accept optical inputs.  The requirement for 2-in (assumed), 6-out is also difficult to find.  I see amplifiers like the following 8-channel I/O DSP amplifiers (which I know nothing about):

 

https://www.qsc.com/resource-files/productresources/amp/cxd-q/q_amp_cxdq_8ch_specs.pdf

 

In short, I think that a separate DSP crossover box with S/PDIF (TOSLINK or coaxial) connected to a 6 channel amplifier box is a much more likely configuration--like a miniDSP 4x10 HD with your choice of multi-channel amplifier (6 channels in your case, leaving two unused output channels from the 4x10 HD).  I'd recommend balanced connections (XLR) to reject common mode noise.

 

Chris

 

Thank you Cris!!!

 

I kindly ask:

 

1)Do you think exist in miniDSP lineup a similar product that could provide 6 (or more) digital outputs too ???

 

2)Could you suggest some good sounding pro amplifiers (even without digital input, in case of use of the the miniDSP 4X10 above. 

 

Thanks!!!

Edited November 8, 2019 by robi2