Finally, my 7.1 full bi-amp experiement

Discussion in 'AV Receivers' started by John S, May 11, 2005.

  1. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Ok, I got Qty.7 Audiosource AMP/1A amps comming in.

    I got Qty.4 Gemini CX1000 stereo electronic cross overs.


    I guess my 1st try at this, will be a simple passive Capacitor type xover from my mids to my tweets in my factory non-bi-ampable 3 way JBL s38 speakers, just because it is an easy way to try it more than anything.


    I figure at this point, I'll at a minimum be able to use the amps as mono blocks with the factory 3 way passive xovers in a more separates config using my AVR as just a pre/pro if I cannot get the speakers to cross as smooth as the factory integrated passive xovers.


    Any last minute suggestions with the experiement?
    I should be posting back in a week or so with my results.

    So I'll be using the AudioSource amps bridged, one to each 8" driver in my s38's, and I'll be using one internal Denon amp to each sets of mids and tweets. CHT-15 Velodyne Sub.

    Wish me luck, I'm going to need it. [​IMG]


    In thinking some more on this.

    I guess the first step will be to use the system for a few days with just the new amps and the passive xovers in the speakers. Then go ahead and do the true bi-amp'd test.

    So really three compairsons.
    Denon 4802 AVR alone with the passive xovers in the speakers.
    Denon 4802 AVR as the pre/pro using the new amps with the passive xovers in the speakers
    Denon 4802 as pre/pro with the electronic xovers before the amps, using the new amps for the low end drivers and using the Denon amps for the mids and tweets using the simple capacitor xover to the tweets.

    I wonder if my sub will be able to keep up with that sort of power though. May have to improve it somehow, if so, just another example of one upgrade leading to the need for another.
     
  2. Chu Gai

    Chu Gai Lead Actor

    Joined:
    Jun 29, 2001
    Messages:
    7,270
    Likes Received:
    1
    Trophy Points:
    0
    Well, the power you need is the power you need. While it's nice to have reserves, and I happen to be a fan of that, if you don't use them, then it can become a moot point. Depending upon your room though, there might not be any limiting to speak of.

    John, let me give you some advice. First of all, I don't mean for anything I've said to be taken that I'm raining on your parade. In fact, I applaud your desire to take what you've got and somehow make it better.

    First of all, for all you know, and I'm sure many will agree, I'm just some kind of schmuck giving you bad advice. Please consider making a post in USENET, aka Google Groups to rec.audio.tech and direct some questions about crossovers to a Richard Pierce. Likely you'll get some responses and hopefully from him. He's a pretty respected speaker designer and scientist and he won't steer you wrong.

    Second, if you consider this StereoPhile link, you'll find on of their observed shortcomings of the S38's were cabinet resonances. You might want to consider asking Pierce about what you could do in the way of bracing to mitigate those as that would likely add to a subtle but real improvement in the speaker's performance.

    Third, also in the Stereophile review, you'll find that they observed that various suckouts were found depending upon how the speakers were placed. To me, that would suggest that you spend some serious time investigating this maybe with the aid of a soundcard and a microphone in order to get a good, uniform frequency response at your listening position.

    Fourth, I think you're barking up the wrong tree with the crossovers. Maybe what makes more sense is to use your new amps to drive the front three speakers while your receiver picks up the slack for the remaining ones which generally don't require anywhere near as much power. That means you could return the amps you really don't need as well as the crossovers. Yeah, you'd eat some s/h but you could take the money saved and then look into my last comment.

    Lastly, I suggest you pick the brain of an HTF member by the name of BruceD. He's using crossovers, not the ones you've got, but with a steeper slope, to better integrate the sub with the other speakers. Try making a separate post or even PM'ing him. He's a hell of a nice and knowledgeable guy who writes very lucidly.
     
  3. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Alot of that advice is already headed. Keep in mind, I was an audio and electronics engineer for nearly 2 decades. I'm the guy that most professionals I know call with these sorts of questions.

    Thanks for the involved post. Interesting in that the slopes are fairly gradual in the S38's passive cross and while the woof to mid seems to have some additional xover circutry, it really looks like the cross to the tweet is as simple as a cap. I have already taken a stock passive xover out of one speaker and sort of broken it down.

    I ended up goin with a 12db xover slope, which is just a little just steeper than the passive xover to the mid. The tweet should end up basically identical.

    I have actually went and auditioned in person a few very high end setups that use all channel bi-amp setups. It sure seemed to have the desired effect on additional headroom.

    I actually do have an Oscope with a Calibrated Mic., Noise, Tone, and Sweep generators left over from the days when I used to do live and studio work. So I should be able to chart what is going on. So far though, I have only tested at 1 meter from an S38 though, one of the reasons I like them even more than when I decided to buy them. My current room is NOT a good room as no living room environments are. [​IMG]


    If what I do does not equal what it was, simple enough, I will just use one monoblock per channel to each speaker with the passive xover put back in them.

    I actually think I like the cabinet resonances and probably want to maintain that characteristic.


    The AVR will be powering the mids and tweets, but your right, that will only take a fraction of the power and the internal amps of the Denon will hardly be breaking a sweat, which is good, and adds even more to the headroom increase.

    As stated in the post, this is experiemental. I am hopefull to gain performance more on par with a $20K system with it. Most of the people that have done this tell me I will achieve it.

    I think also, once I do this, I am going try running large with only LFE going to my sub.

    Will be a difficult HT testing next couple of weeks for sure.

    I actually called Audiosource and the amp is stable at 2 ohms per channel, or 4 ohms bridged, the AMP produces a whopping 300+ Watts bridged at 4 ohms according to Audiosource.
     
  4. Matt Bahnson

    Matt Bahnson Auditioning

    Joined:
    Apr 26, 2005
    Messages:
    14
    Likes Received:
    0
    Trophy Points:
    0
    John,
    I'm right there with you man. I too was an audio engineer for some time and it just so happened that I got out of the recording engineering business right as the JBL S38 speakers came out and the first time I heard them, the one thing that came to mind was "Genelec." I could have been easily fooled into thinking that the speakers I was listening to in the store were most definately modeled after the Genelec powered studio monitors. I love those speakers. and as for the cabinet resonance...those things resonate less than any other $129.99 speaker I've ever heard. Keep up the good work and that's so great that you've adapted room analization into this experiment. I feel your results will be very pleasing! I can't wait to do a similar project in my home. I'm not gonna go active xovers on all channels, but I am definately gonna go with separate amplification.
    Good luck!

    Matt
     
  5. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Thanks for then encouragement Matt. I am constantly taken back by the magnitude of this undertaking.

    People are absolutely encouraged to raise questions about it, that is why I posted. Just to make sure I have really tried to cover all the bases here.

    I just figure the more input the better. I guess the only posts that would really bother me, would be those saying not to bother to try it, or change your speakers instead, stuff like that.

    I like everything about my current speakers, I doubt I will ever change from them.

    You know, I had to install more outlets back to my breaker box at the Components location for all this even, so I am beyond committed to try it.
     
  6. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
  7. VinhT

    VinhT Second Unit

    Joined:
    Feb 14, 2002
    Messages:
    357
    Likes Received:
    0
    Trophy Points:
    0
    A critical aspect of using active crossovers that the article barely touches on is frequency response. Many drivers do not exhibit ideal (flat) frequency response above and below the typical crossover region. A passive crossover is a unique, versatile solution to a particular combination of drivers that takes into account many different factors, including any anomalies that the drivers may exhibit. You cannot just switch to an off-the-shelf active crossover using generic slopes and expect good results.

    Many bi-amping tweakers, especially those in the automotive arena, fail to properly address this issue (if they are even aware of it to begin with). This does not mean that a DIYer cannot concoct an acceptable active speaker using generic components. It's just highly improbable. I suppose that with a high resolution equalizer for each driver and proper measurement tools, an amateur could produce something that has potential. But, at that level, the cumulative expense might warrant a look at commercial offerings instead. With custom-tailored solutions, companies such as Genelec and the like can use active equalization and other integrated circuits to facilitate better integration of the drivers.

    I noticed that no one else seems to have the heart to rain on your adorable quest to achieve $20,000 performance with limited resources. Chu was much too subtle. I will go ahead and say it. Even worse than a waste of time, the experiment may result in significantly reduced performance. The money was not spent in complete vain however, as having separate amplifiers for each surround channel is a nice consolation.
     
  8. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Interesting.. Well I am close. I would like to switch the pre/pro to something that offered eq on each channel. But that change will be down the road a ways. My sweep and oscope measurements will be at 1 meter as my current room stinks. But not to long in the future I will have a dedicated room, that will be most excellent.

    I still think quite the opposite though. I have expeince with this in the Studio control room environments and have found it to make the most difference of anything period.


    But thanks for the opinion. Bi-amp'd much in your day, I mean how much experience do you have with bi-amp systems in general? I at least do have lots of experience with it.

    The only cross to figure out here is going to be the low to mid. I have decided per the article to use the stock mid to tweet crossover circuitry. I looked more in dept at the stock cross over. Should be very easy to still use it.

    The slopes and phase are pretty simple on the low to mid cross in it. Should be easy to simulate.

    I may end up getting better crossovers in the future. These I bought are really just to see if it will be worth it.

    Of course I attached an article from a high level person, that really dispells every fear you just posted about it.

    So once again, I will decide for myself. I am amazed at how easily the concept is dissmissed by those without a lot of experience with it.

    I do agree, you can't really trust your ears though, as that would / could most likely lead to just trade offs and the actual flatness may be hurt, while the better performance in raw output may be increased causing it to be actually be about even.

    It is at worst an interesting experiement, at best a quantum leap in performance.
     
  9. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    PS: The real problem with the next step better crossovers is that the slope is way to steep to achieve anything like the stock passive crossovers. Fully adjustable slope would be way best. I have not looked into it, but I'll bet anything that does that costs a mint!!!!

    I think the cheap crossovers I have will be good enough to at least tell me if it is worth persuit, if not even just about perfect. The cross in the speakers I am using is so simple, I can't believe that I can't improve upon it some.


    But alas, those speakers are very flat, remarkably flat for what they cost. So it will be a challenge to get them to be that flat. I figur if I can get +/- 3db flatness, even if it doesn't match their ultra stock flatness, it will be good enough.


    You make an excellent point though, the only thing I have really wasted is a couple of hundred on the cheaper crossovers. Of which I could probably sell them off and minimize that loss quite a bit even.

    The time involved will be chalked up to the learning experience.
     
  10. Wayne A. Pflughaupt

    Moderator

    Joined:
    Aug 5, 1999
    Messages:
    6,208
    Likes Received:
    56
    Trophy Points:
    9,110
    Location:
    Katy, TX
    Real Name:
    Wayne
    This guy had pretty good luck doing it – to say the least.

    John, it’ll be interesting to see how it turns out. I only have a couple of concerns with your project, first being the low-end crossovers. You might well end up loosing any performance gain right there.

    I used a Carvin crossover in my system for a short while between my mains and subs – certainly not the best crossover out there, but I think you’d probably agree its better than a Gemini. It noticeably colored the sound of my mains, making the highs sound raspy and added some exaggeration in the mid bass.

    Bottom line, I’d go with no less than an Ashly, or Rane as a last resort – certainly nothing below that.

    The second concern in the SQ department is that you’re bridging the amps. It’s fairly well agreed upon in audiophile circles that amps just don’t sound as good when they’re bridged. Besides, one of the fringe benefits of active bi-amping is that you get loads of “free” wattage once the power-robbing passive elements are out of the signal chain. So in light of that it’s rather senseless to bridge the amps. You could better use the extra channels to do a three-way system.

    Regards,
    Wayne A. Pflughaupt
     
  11. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    3 way systems compound phase issues. The cross overs have decent enough specs, I don't think they will be an issue, but as said, better crossovers may be in order, just not for the initial test. As said, if the idea is to mimic the passive crossovers, none of the better crossovers will work as well with their 24db slopes. If I can't see improvement on 2 channel with these crossovers, then I will probably just abandon the idea.

    I sort of disagree with the bridge thing. I have not found an amp yet that didn't perform better bridged. The somtimes higher distortion rating, is more than made up for by the additional head room provided by doubling the power. Maybe a full range bridge, but not a pre-corssed bridge at least.



    Thanks for your post, as said the more the idea is questioned, the better my experiement will be.
     
  12. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    PS: Keep in mind the bridged amps will only handling up to 800hz...

    One other reason, is it takes so much less power to drive only mid and high frequncies. That if I were not bridging, I'd have to use the AVR amps for the lows and the single channels of the amps for the mids/highs.


    Correct gain staging is important to me. I hear systems all the time, that are only proper at reference level where they calibrated, go below that and the levels just aren't correct.

    I would persojnally would never just put better amps on my fronts for example.
     
  13. VinhT

    VinhT Second Unit

    Joined:
    Feb 14, 2002
    Messages:
    357
    Likes Received:
    0
    Trophy Points:
    0
    Mr. Pflughaupt,

    There is a rather significant difference between the endeavor discussed on Home Theater Talk and the situation presented here. Michael P. obviously has several advantages. He designed and built his own active crossovers, demonstrating a level of competency far greater than that of the average DIYer. He actually has measurement software. He also understands the limitations of his measurements (uncalibrated mic, reflections, etc). As a result, his project had more potential to begin with.

    I will take a moment to touch on a topic that DIYers may be unfamiliar with. Loudspeaker design is not nearly as simple as merely achieving flat on-axis frequency response. Off-axis performance is extremely influential in determining how the speaker sounds in the listening environment. Professional speaker designers will take extensive off-axis measurements and use a complex weighted average formula to derive a speaker's power response. The power response is a good indicator of how the speaker will actually sound in-room, and the designers will usually have a target power response in mind. Unfortunately, many DIYers neglect off-axis performance, and even if they are aware of the issue, often lack the means to adequately quantify it. Instead, many DIYers simply ignore the fundamentals of advanced loudspeaker design and merely assume that what they come up with is superior to professional solutions.

    I noticed that Rod Elliott's website is frequently referenced by proponents of active crossovers. It certainly has been instrumental in inciting DIYers to abandon passive crossovers in hopes of an active revolution. I feel it is irresponsible to lead folks to believe that with a few amplifiers and active crossovers they can easily produce better speakers than those designed by professionals.

    Lastly, I would like to address your disdain for passive crossovers.Respectfully,
     
  14. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Thanks for the additional information Vince.
    Some good stuff in all these posts.


    So what are some of the least expensive electronic cross overs that would at a minimum allow me to decide on the slope?
     
  15. Wayne A. Pflughaupt

    Moderator

    Joined:
    Aug 5, 1999
    Messages:
    6,208
    Likes Received:
    56
    Trophy Points:
    9,110
    Location:
    Katy, TX
    Real Name:
    Wayne
    Vinh,

    Interesting and highly useful information, especially your second paragraph.
    I have no disdain for passive crossovers. If you read my last post on the thread referred to, you’d know that.

    John,

    I don’t know of any budget crossovers with selectable slopes. However, over the years Ashly has produced crossovers with 12, 18 and 24 dB/octave slopes. You can get them cheap enough on eBay that it might be worth acquiring some for experimental purposes. You would have no problem selling them when you’re finished with little or know money lost.

    Short of that, you might look into some digital crossovers like the dbx Driverack or some Behringer models to see of they have selectable slopes – provided you don’t mind the extra AD/DA conversions.

    Regards,
    Wayne A. Pflughaupt
     
  16. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Man, does it take time to test.....

    So far, I have only got testing of some differences between Denon amp with passive cross and the Audio Source Amp's, using the Denon for the Pre/Pro with the passive cross.

    On a positive note, the AudioSource Amps seems extremely capable bridged driving the full range. It got to reference really quick on the Denon Volume control anyways. [​IMG]

    I am going on a pretty long vacation, I leave Friday, I am questioning how much bi-amp testing I can get done. It would be easier to just go a head and get an H/K 7300 AVR and actually tri-amp it. It will be interesting to see how I feel about it all after I get back from vacation.
     
  17. VinhT

    VinhT Second Unit

    Joined:
    Feb 14, 2002
    Messages:
    357
    Likes Received:
    0
    Trophy Points:
    0
    Mr. Pflughaupt,

    Unless my eyesight (or web browser) is failing me, you only made one post in the referenced Home Theater Talk thread. You recalled a bad experience with a modified passive crossover from Genesis Physics, and read about the efficiency advantage of active speakers. You then proceeded to make a statement which I chose to omit last time.Unfortunately, his tools will not be particularly useful for his experiment. He lacks the means to adequately visualize loudspeaker frequency response, which means he will be doing blind crossover work. But what about if he just duplicates the crossover from the product literature? No good, because the specified crossover is the acoustic crossover, the combination of the natural roll-off of the transducers and the electrical filters applied. So he knows the end result, but cannot "see" how they got there. I suppose that given enough experience designing passive crossovers, he could theoretically reverse engineer the JBL one (which he claims to have done), but I do not get that impression here. Many designers do not use textbook filters, and the crossover may include optimizations, causing great difficulty when attempting to derive the transfer functions by sight alone.


    John,

    From an unrelated thread, you expressed some interest in purchasing a receiver with all-channel EQ. I do not know if your intention is to use it as the equalization suggested by Mr. Wiggins, but as a preemptive word of caution, the EQ offered by receivers is often far too low-resolution to be of much use in an active crossover application.
     
  18. Michael R Price

    Michael R Price Screenwriter

    Joined:
    Jul 22, 2001
    Messages:
    1,591
    Likes Received:
    0
    Trophy Points:
    0
    John,

    If you have a calibrated microphone, then just load up Speaker workshop or some other MLS measurement software into your computer and you're golden.

    I wouldn't try designing active crossovers by ear or even with something like a Radio Shack meter or scope, but once you get a good handle on the sound of the system and your drivers (by listening a lot) it's a good idea to adjust that way. I ended up with something a little different from the "ideal flat frequency response" when I tweaked my crossover by ear.


    Vinh,

    A lot of speakers rely on their passive crossovers to equalize out anomalies in the drivers and crossovers, but you don't always need to do anything besides the driver rolloffs and maybe a shelving filter for the baffle step. I have enjoyed the results of speakers made without the crossover "equalizing" pass band responses, which is usually unnecessary with hi-fi drivers and good selection of crossover points. I just got to listen to a speaker which was made with an Alesis plate amplifier - completely textbook, 8th order active filters at 1.8 KHz - and some very nice Parts Express drivers. Except for a tweeter-woofer level mismatch which I think the owner just corrected, they sounded really good. It is possible to avoid an involved design process, with the right components.

    Wayne might have meant that as the driver impedance changes under load (from voice coil heating and the large signal movement of the cone) the transfer function of a passive crossover changes. That may be one reason why active crossovers sound more dynamic.
     
  19. John S

    John S Producer

    Joined:
    Nov 4, 2003
    Messages:
    5,460
    Likes Received:
    0
    Trophy Points:
    0
    Thanks for the additional informations. I still hope to get a limited test on the bi-amp in 2 channel.

    I'll do my own graphs manually with the O-scope as I am most familure with the method.


    My crossovers are three way.. Who knows when I get back from vacation I may just decide to go a head and get the H/K 7300 and add tri-amp to the test mix. [​IMG]


    Maybe with more expensive speakers. But I have yet to see any complexity in passive crossovers. The s38's really seem to have the most basic of crossover technology in them.


    And yes the H/K EQ is what it is... But I do not see a nice out board eq on each channel in my future.
     
  20. Wayne A. Pflughaupt

    Moderator

    Joined:
    Aug 5, 1999
    Messages:
    6,208
    Likes Received:
    56
    Trophy Points:
    9,110
    Location:
    Katy, TX
    Real Name:
    Wayne
    Vinh,
    My apologies – I though you were referring to a different thread about Michael’s project that ran here on HTF at about the same time.

    Still, you’re blowing this way out of proportion. If I really “disdained” passive crossovers as much as you imagine, I’d have ripped them out of all my speakers. I haven’t. So please, give it a rest; it’s contributing nothing constructive to this discussion.

    Regards,
    Wayne A. Pflughaupt
     

Share This Page