Page 1 of 2 12 LastLast
Results 1 to 15 of 19

Thread: Biggest JBL midrange-driver ?

  1. #1
    Senior Member
    Join Date
    Sep 2008
    Location
    Germany / Hamburg
    Posts
    659

    Biggest JBL midrange-driver ?

    Hey,

    I`d like to build a pair of speakers with W15Gti-carwoofers as the lowrange... As these drivers don`t behave to well in the upper reagion, I`m looking for a midrange/midbass driver that can be crossed at 80-100hz and go all the way up to 800hz to mate some 2445 drivers..

    The JBL 2123 unfortunatelly starts playing at 250hz I guess...

    Any recommendations ?

  2. #2
    Senior Member grumpy's Avatar
    Join Date
    Oct 2003
    Location
    SoCal
    Posts
    5,742
    2206? Big party speakers?

  3. #3
    Senior Member 1audiohack's Avatar
    Join Date
    Jul 2007
    Location
    Las Vegas Nevada
    Posts
    3,092
    Or 2220, if you want a 15".
    If we knew what the hell we were doing, we wouldn't call it research would we.

  4. #4
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    8,170
    Or E-145's

    Rob
    "I could be arguing in my spare time"

  5. #5
    Senior Member
    Join Date
    Sep 2008
    Location
    Germany / Hamburg
    Posts
    659
    Thanks guys
    Nice recommondations!

    I got a pair of E-145 myself and believe they`re awesome driver! They cover 95% percent of most tracks, but sometimes I miss the very low end. A 2235h is not my choice, as the midbass is way worse in my oppion...

    So I thought of adding some subwoofers to the E-145. I`ve tried some JBL`s and JL`s but they don`t seem to blend in very well. Even though we used one sub each side closely to the mains!
    Someone come up and told me, the very low crossover-points of 50-60hz is the main cause of "slow bass"... It causes phaseshift and big groupdelay, when filters are at low frequencies!
    The guy advised me to use a higher crossover-point somewhat near to 80-100hz, so groupdelay of the crossover will decrease and the bass will sound more time coherent and not "slow"...
    But when I crossover the E-145 that high, I believe the cone won`t be moving much at high levels and the voicecoil will leak cooling-air caused by a moving cone. So the E-145`s coil could suffer a heat-damage due to no conemotion/cooling-air
    So I thought, I should use a driver instead which operation-range and crossover-point should be higher by it`s natural tuning. A JBL E-110 couldn`t play any lower and should be crossovered at 80hz anyway, so there wouldn`t be any heat-problems for the coils, because conemotion is in the normal range as it was designed from scrach!
    This is why I´m looking for drivers which perectly match a crossover point of around 100hz....

    But on the other way, most professional brands and speakerdesigners tend to pic a crossoverpoint of 250-300hz, not as low as 100hz... So I might be on the wrong way...
    I thought 100hz crossover would be great, because less filter-groupdelay as 50hz crossover, put still the midbass-punch 100hz and up from a decent driver like 2220, 2206 etc...

  6. #6
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    8,170
    I believe the cone won`t be moving much at high levels and the voicecoil will leak cooling-air caused by a moving cone.
    It's an underhung coil. You couldn't get better heatsinking with the mass of the magnetic structure.
    What kind of power levels?? I run mine with 100 watts and it's more than enough and with those power levels you could run them all day long.

    I have used Le-14 and 2235 based subs with my E-145's and had no issues. What are you using to crossover?? I am using 24Db L/R slopes in an active crossover @ 80hz.

    Rob
    "I could be arguing in my spare time"

  7. #7
    Senior Member pos's Avatar
    Join Date
    May 2007
    Location
    France
    Posts
    2,628
    Quote Originally Posted by Dr.db View Post
    So I thought of adding some subwoofers to the E-145. I`ve tried some JBL`s and JL`s but they don`t seem to blend in very well. Even though we used one sub each side closely to the mains!
    Someone come up and told me, the very low crossover-points of 50-60hz is the main cause of "slow bass"... It causes phaseshift and big groupdelay, when filters are at low frequencies!
    What crossover unit are you using?
    In order to properly cross at 50Hz (ie with phase coherent and amplitude complementary acoustic responses from the two drivers) you absolutely need to take the natural rolloff of the E145 in its (closed?) box into account. You can measure this with a mic very close to the cone.
    If you are using a closed box then you already have a 12dB/oct HP filter somewhere were you need it. You will need asymmetrical slopes, and probably a bit of EQ.


    Regarding the phase shift issue, it is real and will cause audible delays (but might not be responsible for the problems you hear, as the room interaction is far more audible). The steeper the acoustical crossover slopes the more phase shift you will get.
    That said, these phase shift can easily be corrected nowadays using convolution (FIR) in front of the actual crossover, without any drawback other than an additional global delay (the highs will have to "wait" for the lows so that everything in temporally aligned).

    What is your source and signal path?

  8. #8
    Senior Member
    Join Date
    Sep 2008
    Location
    Germany / Hamburg
    Posts
    659
    Thanks a lot for your qualified answers!!


    @ Robh3606:

    I run them with a Harman/Kardon Citation 200w rms/channel... Sometime`s I drive them to their limits for an hour or two....
    I use a dbx223-crossover, but its not the best I know.... We tried 50 and 60hz crossovers with 18db of that dbx unit.

    As I said before, we were afraid 80hz or 100hz highpass for the E-145 would be too high and it could suffer a heatproblem due to less cone-motion when driving with full 150-200watts...
    You don`t seem to share my concerns, calming to hear that. Would you say, I could even crossover the E-145 higher than that, lets say 120hz...!?



    @ pos:

    I use an old Harman/Kardon 402 as pre- and tweeter-amp. Crossover is a simple, analog DBX-223.... EQ is some old Yamaha 31band to keep the most extreme peaks e.g. room-mode´s down...
    E-145 is in a 4cubic vented enclosure tuned to 50hz...
    In which way should I consider the natural roll-off; do you have any practical advices?
    BTW; what are asymmetrical filters!?

    You said, the steeper the filter, the bigger the phase-shift issue...
    This also applies to lower cuttof-freq., right!? So 50hz has more groupdelay/phaseshift than 80hz ?

    FIR`s don`t mess with the phase, that is correct.
    But at this moment, I`m trying to build my speakersystem without a DSP, so I am depending on "traditional" groupdelay-comensation...

  9. #9
    Senior Member pos's Avatar
    Join Date
    May 2007
    Location
    France
    Posts
    2,628
    Quote Originally Posted by Dr.db View Post
    E-145 is in a 4cubic vented enclosure tuned to 50hz...
    In which way should I consider the natural roll-off; do you have any practical advices?
    a vented box will give you an acoustical high-pass filter with a 24dB/oct slope. You will want to add an additional high-pass filter to reduce excursion under the tuning frequency.
    Going sealed here would probably be a good idea: it would simplify measurement and filtering, give you a more predictable behavior, and alos let you target a lower slope for your crossover.

    So first thing to do is to plug the bass reflex port (Zilch plugs!!).
    In order to evaluate your box rolloff (ie the acoustical high-pass filter that is already there) you will need to take a measurement with a mic very close to the cone (let's say 1cm, near the dustcap).
    If you send that measurement to me (impulse response as a text or wav file) I should be able to tell you what electrical high-pass filter you will have to add to obtain a given acoustical target (LR24 for example).

    BTW; what are asymmetrical filters!?
    Your W15GTI will probably not have any significant natural low-pass filter in the 50Hz range, so any electrical low-pass filter you will apply with result to the same acoustical filter.
    On the other hand as your E145 will have a naturall rolloff in this range you will have to take it into account when designing your filter: acoutical filter = eletrcical filter + "natural" roll.

    What you absoluterly want is a phase-coherent and amplitude-complementary crossover. The simplest target is a symmetrical LR (same slope and fc for the low-apss and high-pass).
    So for example if you are targeting a LR 24dB/oct crossover, you will probably need to use a LR 24dB/oct low-pass filter for the W15GTI, whereas the E145 in a sealed box will typically require a
    12dB/oct high pass filter with a specific Q (and additional EQ...).
    The close range measurement of the E145 will tell what Q and EQ need to be used.


    You said, the steeper the filter, the bigger the phase-shift issue...
    This also applies to lower cuttof-freq., right!? So 50hz has more groupdelay/phaseshift than 80hz ?
    Yes, but 50Hz is probably a better target.

    FIR`s don`t mess with the phase, that is correct.
    But at this moment, I`m trying to build my speakersystem without a DSP, so I am depending on "traditional" groupdelay-comensation...
    FIR can correct the phase distortion of an analog (or digital IIR) crossover.
    If your source is a PC this can easily be done inside your media player software (with a convolution engine running the proper FIR correction).

    All in all, even if you do not want to correct phase, going for a properly summing crossover (coherent and complementary) is very important.

  10. #10
    Senior Member
    Join Date
    Sep 2008
    Location
    Germany / Hamburg
    Posts
    659
    Thanks a lot for your descriptions!!
    I think I got the idea of coherent phase / crossover-slopes


    If I will achieve identical acoustical-roll-off for midbass(12db acoustical + 12db electr. =24db acoust.) and subwoofer(24db electr.), they should be phase coherent again
    Both drivers have the same phase-turn (360°) at the crossoverpoint, so they would match perfectly at this point, right!?
    But the phase will be turned about 360° at the crossoverpoint only, the frequencies below and above would still behave in the "normal/correct" phase (0°turned) and would not match the turned phase at the crossoverpoint...!?
    So there wouldn`t be cancellations at the crossoverpoint, but still a groupdelay due to the partially turned phase at the crossoverpoint



    What will happen, if I run a bassreflexbox with 24db/oct. roll-off at 50hz tuning, but use a 12db-highpassfilter at 80hz ?
    The acoustical roll-off should be 12db, because the electrical roll-off "kicks in" way before the natural roll-off blends in...!?
    At 50hz it would summ to something like 36/oct. but should be muted enough not to be heard any more, right!?



    Ps. I actually don`t have any measurement equipment right now, sorry.
    But thanks for your offer to calculate the correct highpass!

  11. #11
    Senior Member pos's Avatar
    Join Date
    May 2007
    Location
    France
    Posts
    2,628
    Quote Originally Posted by Dr.db View Post
    Thanks a lot for your descriptions!!
    I think I got the idea of coherent phase / crossover-slopes


    If I will achieve identical acoustical-roll-off for midbass(12db acoustical + 12db electr. =24db acoust.) and subwoofer(24db electr.), they should be phase coherent again
    That is correct.

    Both drivers have the same phase-turn (360°) at the crossoverpoint, so they would match perfectly at this point, right!?
    But the phase will be turned about 360° at the crossoverpoint only, the frequencies below and above would still behave in the "normal/correct" phase (0°turned) and would not match the turned phase at the crossoverpoint...!?
    Phase response of a LR 4th order filter, be it a high or a low pass, will be the same: it will be 180° at the crossover point, and asymptotically reaching 0° above that point and 360° under that point.

    Something like this:
    Name:  TextbbokLR4.gif
Views: 893
Size:  12.6 KB

    Both the high and the low pass will be phase coherent because they both have the same phase shift throughout the crossover.

    So there wouldn`t be cancellations at the crossoverpoint, but still a groupdelay due to the partially turned phase at the crossoverpoint
    Yes, but group delay is not that audible. What matter most is the phase coherency between the two drivers.
    And then the global phase shift/group delay can still be corrected using FIR before your actual crossover, buy "pre shifting" the phase the opposite way.

    What will happen, if I run a bassreflexbox with 24db/oct. roll-off at 50hz tuning, but use a 12db-highpassfilter at 80hz ?
    The acoustical roll-off should be 12db, because the electrical roll-off "kicks in" way before the natural roll-off blends in...!?
    At 50hz it would summ to something like 36/oct. but should be muted enough not to be heard any more, right!?
    Phase shift will caused by a 24dB/oct filter will already be apparent several octaves above the actual crossover point. You will have to take into account anyway.
    If you look at the example image above, you will see that the phase is still something like +90° above active above the crossover point. The exact phase shift "shape" depends on your actual BR Q, but you get the idea.

  12. #12
    Senior Member
    Join Date
    Jul 2003
    Location
    Orange County, CA
    Posts
    3,597
    Quote Originally Posted by Dr.db View Post
    Hey,

    I`d like to build a pair of speakers with W15Gti-carwoofers as the lowrange... As these drivers don`t behave to well in the upper reagion, I`m looking for a midrange/midbass driver that can be crossed at 80-100hz and go all the way up to 800hz to mate some 2445 drivers..

    The JBL 2123 unfortunatelly starts playing at 250hz I guess...

    Any recommendations ?
    Not true. The W15GTi is as smooth or smoother to 1kHz as a 2235 and extends lower. At one time I did the testing and printed my results here. I use the W10GTis crossed over at 800 Hz with the 2425 on butt cheeks for my four surround speakers, but you've got to pad the horn way down.

  13. #13
    Senior Member
    Join Date
    Sep 2008
    Location
    Germany / Hamburg
    Posts
    659
    @pos:

    Thanks a lot for all your specific answers, you`re a big help!!


    Quote Originally Posted by pos View Post
    Phase response of a LR 4th order filter, be it a high or a low pass, will be the same: it will be 180° at the crossover point, and asymptotically reaching 0° above that point and 360° under that point.

    Something like this:
    Name:  TextbbokLR4.gif
Views: 893
Size:  12.6 KB
    Did I get it right, when I`m saying the phase will be the same(0°) little above HPF & LPF, and is progressivly turned at the crossoverpoint and finally is 360° turned all below the crossoverpoint ?
    So I would have to reversal the voltage on the lowfreq.-driver to get it to 0° phase again!?



    Quote Originally Posted by pos View Post
    Yes, but group delay is not that audible.
    In the upper frequencies this is true. But I read several times that it becomes a big issue in low frequencies and I shouldn`t use a crossover below 150-200hz, because of major group delay that would cause a "slow bass"...
    Again, I don`t have FIR, just a simple analog crossover...



    @ toddalin:


    That sounds great.
    But does the W15gti just run without any dips and peaks that high, or is it actually still sounding very dynamic and punchy above 100hz ?

  14. #14
    Senior Member pos's Avatar
    Join Date
    May 2007
    Location
    France
    Posts
    2,628
    Quote Originally Posted by Dr.db View Post
    Did I get it right, when I`m saying the phase will be the same(0°) little above HPF & LPF, and is progressivly turned at the crossoverpoint and finally is 360° turned all below the crossoverpoint ?
    So I would have to reversal the voltage on the lowfreq.-driver to get it to 0° phase again!?
    The phase will have turned 360° in the LF, but you don't need to reverse polarity.
    You would need to do that if you were targetting a 2nd order crossover (12dB/oct), which would be 180° toward DC (ie polarity inversion).

    In the upper frequencies this is true. But I read several times that it becomes a big issue in low frequencies and I shouldn`t use a crossover below 150-200hz, because of major group delay that would cause a "slow bass"...
    Again, I don`t have FIR, just a simple analog crossover...
    Yes it can be audible in the LF, but if your crossover is phase-coherent then you will be able to *compensate* for it in front of your crossover.
    For example if your source is a PC then you can load the appropriate FIR correction inside your mediaplayer (using a stereo convolution engine, usually free), or you can insert an hardware convolution engine (such as the openDRC) between your source and crossover

  15. #15
    Senior Member
    Join Date
    Jul 2003
    Location
    Orange County, CA
    Posts
    3,597
    Quote Originally Posted by Dr.db View Post
    @pos:


    @ toddalin:


    That sounds great.
    But does the W15gti just run without any dips and peaks that high, or is it actually still sounding very dynamic and punchy above 100hz ?
    At least as smooth as a 2235 to that frequency and still very dynamic and punchy. If you have one, you can judge for yourself. Maybe the Moderator and find the old thread that had the comparison data.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Biggest Fullrange driver?!
    By Dragan666 in forum Lansing Product DIY Forum
    Replies: 25
    Last Post: 04-23-2009, 08:17 PM
  2. Midrange driver
    By bulgroz in forum Lansing Product DIY Forum
    Replies: 5
    Last Post: 07-17-2007, 05:18 PM
  3. LE5 Midrange driver question
    By kworrilow in forum Lansing Product Technical Help
    Replies: 12
    Last Post: 12-20-2006, 10:35 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •