Bi-Amping with Active Crossovers - Pros and Cons

Discussion in 'Audio Hardware' started by Soundgarden, Jan 9, 2022.

  1. Davey

    Davey NP: Hania Rani/Dobrawa Czocher ~ Inner Symphonies

    Location:
    SF Bay Area, USA
    We don't really think of efficiency for line level crossovers, there is so little power dissipation that it isn't a consideration, so yes, they are much more efficient. The difference in impedance means that you can use much smaller, and possibly much better sounding capacitors than in a speaker crossover. The downside is the previously mentioned inductors which need to be much larger in value due to the high impedance. They don't have to handle any appreciable power though, so can be wound with smaller wire, but off the shelf values are limited. The preamp does need low output impedance for best results, or that value needs to be part of the filter equations.

    One with a good electronics understanding could actually build the passive crossovers into the amplifier circuits and avoid extra components since the amplifiers have at least one filter on the input, and possibly another in the feedback network, so they would really be active electronic crossovers then.
     
  2. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    Okay, this makes sense. Feel like I'm getting somewhere again.

    I don't have specs for the output impedance of my modified Marantz 7T pre-amp. I can ask. But the crossover will be built by the same builder. So that can easily be incorporated.

    This thought actually occurred to me. For example, the pre-amp could be wired to spit out two left channels (low and high) and two right channels. But this would have some impracticalities associated with it as well.
     
    Davey likes this.
  3. SandAndGlass

    SandAndGlass Twilight Forum Resident

    Questions that I will ask is this going to be a two or three way system and what crossover frequencies are you going to use?
     
  4. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    Three-way speakers. The line-level crossovers would just split the lows from the mids/highs. And I'd be using passive components at the speakers for the mids/highs. The low crossover is at 350 Hz.

    http://www.infinity-classics.de/technik/manuals/RS_1.5_technical_sheet.pdf
     
  5. House de Kris

    House de Kris VVell-known member

    Location:
    Texas
    I'm not Davey, but since he hasn't answered your question, I thought I'd offer my view on how removing the series inductor may change the box tuning. Simply speaking.

    The voice coil of a moving coil driver is indeed a coil of wire, and as such behaves like an inductor. This gives rise to the rising impedance with rising frequency that these drivers exhibit (and that is compensated for by zobel networks to make the electrical crossover behave as expected). But, this voice coil is more than just an inductor, it is an inductor that is intended to move based on the magnetic field it creates via the alternating current through it as it opposes a constant magnetic field provided by the magnet. To move freely in this constant magnetic field, it must be held in place by a compliant suspension. This suspension exherts forces on how this inductor wants to move in the permanent magnetic field and these forces are reflected in the impedance as deviations from what an ideal inductor would present. These forces are typically modeled as springs and masses to predict their behavior. But, this moving coil driver is more than just an inductor held in a magnetic field by suspension. It is also attached to a diaphram (cone) which moves air. The work of moving air is also reflected back to the impedance curve. Placing this moving cone driven by a suspended inductor in a box results in trapped air on one side of the cone. This trapped air will present a certain amount of springiness that then fights back to the free movement of the cone. These external brakes applied by the air can also be modeled by springs and masses. Part of this modeling is how strong the electric drive the moving inductor attached to the cone is supplied, in order to fight back on the forces provided by the trapped springy air. It becomes a complex set of weights and balances, but the delicate balance this system is altered slightly by making the driving force stronger or weaker. The series inductor of the passive low-pass filter makes the driving force weaker, thus resulting is a different drive strength, and resulting in a different balance, or equilibrium, than without said series inductor.

    Well, there you go, a single paragraph simple explanation. Some may claim technical hand-waving, but I did that in order to keep the concept simple.
     
    Soundgarden likes this.
  6. Davey

    Davey NP: Hania Rani/Dobrawa Czocher ~ Inner Symphonies

    Location:
    SF Bay Area, USA
    I had already mentioned how it affected it so didn't bother to elaborate, but the resistance changes the Qes parameter of the woofer, which will alter the system damping, and hence the bass alignment. The bass alignment Qtc is a factor of the woofer Qes, Qms and the box Q.

    Anyway, not much to worry about, just thought I'd mention it, if not using an inductor in the woofer circuit, then you don't have to alter the woofer parameter when doing the calculations.
     
  7. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    The technical-waving is essential and more appreciated than you know. The explanation to the layperson is what makes science accessible to those without mastery of the subject matter.

    And this explanation makes sense as to why removing the series inductor would have an impact. What the impact would be, and whether the change would be desirable, involves additional considerations. But I suppose, if you're starting with a speaker that is impeccably tuned, you run some risk of detuning.
     
  8. House de Kris

    House de Kris VVell-known member

    Location:
    Texas
    I'm going to be my typical contrarian, and say this speaker would be quite a challenge to change over to a bi-amp setup. If the goal is to just play around with bi-amping and learn how it's done, I think the AR-3a would be better. Here's the issue, the Infinity RS1.5 seems to be a three ana half way design. Have you looked at the schematic you linked to in the PDF? There's a lot of wackiness going on in the woofer section. They've got a second order low-pass on one of the coils (to match the second-order high-pass of the midrange), then the other coil is a single order way down lower. I suppose it is used to boost the low frequency capability. On top of that, they've got those huge capacitors in series with the woofers, giving it a hi-pass around 18Hz or so. If you were to make your own completely new box, that didn't need the second cone to act as a very low frequency helper, then you could just put both coils in parallel and treat it as a typical three-way. With your Phase Linear driving just the woofers (with nothing else in the way), and use another amp to drive the mids and tweets with their portion of the passive crossover.
     
  9. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    So the Dual Voice Coil Watkins Woofer, which is a part of the draw of getting this speaker, actually becomes a complication in this context.
     
  10. House de Kris

    House de Kris VVell-known member

    Location:
    Texas
    It's a draw because you can do cool stuff like using one coil to be a helper. Bet in your case, designing in a helper circuit might not be so fun.
     
  11. House de Kris

    House de Kris VVell-known member

    Location:
    Texas
    Oops, sorry. Didn't mean to step on any toes. I looked for Davey answers after the question was asked and didn't think I saw any. My bad for missing it.
     
  12. Davey

    Davey NP: Hania Rani/Dobrawa Czocher ~ Inner Symphonies

    Location:
    SF Bay Area, USA
    No problem, I didn't answer the question before so no toes stepped on, good discussion :)
     
  13. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    Well maybe then I’m back to using my Klipsch Heresy 1.25s for bi amp experimentation purposes.
     
  14. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    Honestly having both the longer layperson description and your shorter version using actual specs has a nice synergy to it.
     
    izeek likes this.
  15. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    Clear as the espresso I drink...

    The speaker and its voice coil inductance can certainly be viewed as a separate component from the external inductor.

    What might not be obvious in the passive crossover design schematic is that the exact frequency response and -3dB point will be different than simply plugging "8 ohms" into a crossover formula, because the nominal impedance is only a rating. One has to look at exactly what the speaker impedance is doing - even measuring in the enclosure - to determine the true -3 point and slope.

    [​IMG]
    (Tymphany 10")


    That's not a good place to start if you are wanting to replace passive with active. The woofer is dual-voice-coil, each with its own crossover network. Here's the output (guessing the individual coils are 8 ohm) for each:
    [​IMG]

    You can imagine that not only will you need two amps to drive this identically with active, but if instead you were to use one amp, combining the response into one line-level output for parallel coils, that will be another complex circuit.

    Some of the above might not be necessary - this may be a hack to increase efficiency by creating two resonators.

    The midrange also has sound tune-ups. You've got to remove only some of that circuit so it still does mid-to-high.
     
    Last edited: Jan 16, 2022
  16. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    Sigh.

    Well anyway for anyone who’s interested in a description of the DVC woofer, and who’s willing to wade through a fair amount of marketing speak, this document provides an interesting account starting on page 12.

    https://www.bayer-soundworks.de/app...+Creative+Technology+Of+Infinity+Speakers.pdf
     
  17. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    So perhaps what you’re saying is that yes I can separate the lows from the mids//highs via a simple line level crossover but I will still be heavily reliant the passive crossovers in the infinity 1.5 to deliver both the mids/highs from one amp properly to the mid and high coils and the lows from the other amp to the dual voice coils.

    The amp that would be driving the lows would send all the lows towards the woofer. But there would still be a fair amount of work from
    there to separate and prepare the signals for the DVCs. Is that right?

    If so this could still work with a simple line level crossover and basic speaker crossover mods but I wouldn’t really be “replacing” the speaker crossovers with line level crossovers since I’d still be relying on the speaker crossover pretty heavily.

    That said, I feel myself creeping back out over my skis again.
     
    izeek likes this.
  18. head_unit

    head_unit Senior Member

    Location:
    Los Angeles CA USA
    Here's my engineering take on it, probably some reinforcing the 11 previous pages:
    - Electronic crossovers are typically implementing ideal electric filters, with textbook looking slopes. Sophisticated passive crossovers won't have a filtering effect like that, they will be non-flat in order to make the summed response of the non-ideal drivers flat.
    - In those cases it's a crapshoot if active crossovers will end up sounding better.
    - Active crossovers can offer time alignment, which passives cannot. However this requires measurement and adjustment and fiddling around. As does the whole setting of crossover points and levels and so on. (It also requires either an additional A/D step, or digital signal input which is complex). The time alignment can be particularly helpful for horns is my anecdotal observation.
    --> Active crossovers are NOT "plug them in and set the frequency and level and done"-nope sorry, at least not for best results.
    If done well the sonic benefits can in theory outweight the one additional layer of electronics. However in one case we felt an active Dynaudio sounded no better than its passive counterpart. It just depends on the level of implementation and design skill. With nice sounding passives I would not tear them apart to actively biamp. I think of it more as a DIY thing for fairly elaborate systems.
     
    Soundgarden likes this.
  19. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    The dual voice coils are 2 and 4 ohms I believe.
     
  20. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    Not much has been said so far about phase coherence / time alignment in this thread. But I’ve been reading about it and how it can make a big difference. But has to be done via ACTIVE crossovers correct? Passive crossovers will not do this…
     
    Randoms likes this.
  21. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    And your point about ideal filters is new to this thread. Or at least put so simply/explicitly is new. Although I could have missed something earlier.
     
  22. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    One can measure the DC resistance to get an idea if that is true.

    If there is indeed a two-ohm coil, it simply may not be possible to extricate that from its impedance-tempering crossover, simply because of the extreme requirements on an amplifier and especially inability to parallel that with the other.

    One could simply use the 4 ohm coil, and take a completely different electronic approach to reproducing the original response actively. One would need to make mic measurements of the low frequency network and drivers as they are, and then the 4-ohm coil driver alone to come up with a plan.

    There are no resistive components in the low frequency network (besides the voice coil itself), so you might not be gaining efficiency in replacing it. The ultimate benefit would be the much sharper crossover slope that can be employed with an electronic re-design.

    One might instead ponder a mid-high split.
     
  23. Brother_Rael

    Brother_Rael Senior Member

    And I'll take active, anyday and every day these days. I spent years with passive speakers. There are some great models out there, but the compromises are too great for me. Acoustic energy produced an active for £900 that delivered a performance that speakers costing £5,000 would've wilted at. AVI produced an all in one system combining a preamp, active speakers and a DAC in a speaker housing long before most of the others started to and did it very well.

    So no, the subjective performance of active speakers likely aren't overestimated. For me, I'm usually surprised they don't feature more often and more highly as they deliver what audiophiles need. Great sound. There's the usual "oh they're too limiting, I can't use the amp I want" or "they're too clinical, I don't want neutral" and the like. That's uninformed (to some extent), or misled perhaps, comment. The flexibility is all there when you look at it.

    In terms of flexibility, I use a Linn preamp (although my AVIs have an onboard pre) for the extra inputs the speakers themselves don't offer me. I use the optical for TV direct to the speakers, the line level and phono amp all go through the Linn.

    The power amp being dictated by the speaker designer is no more a limitation than DAC chip in your standalone DAC or the driver being selected for you in the speakers you buy being preselected for you.

    The bottom line - the reduction in distortion (by orders of magnitude) offered by actives is reason enough to warrant inclusion in shortlists. The consumer is broadly led by marketing, reviews and dealer advice. I'm not so sure that's always in the best interests of the consumer - a trip to a good pro audio dealer is worth an hour of *any* audiophile's time. It was for me and it was a great experience. I recommend it, and actives, wholeheartedly.
     
    Randoms likes this.
  24. Soundgarden

    Soundgarden Forum Resident Thread Starter

    Location:
    Bend, Oregon
    How do you mean?
     
  25. harby

    harby Forum Resident

    Location:
    Portland, OR, USA
    A tweeter-only amplifier and active crossover. A very sharp 48dB crossover as with a miniDSP almost eliminates comb filtering.

    One would simply leave a much lower frequency capacitor on the tweeter for turn-on protection.
     

Share This Page

molar-endocrine