Page 16 of 21 FirstFirst ... 61415161718 ... LastLast
Results 226 to 240 of 303

Thread: Discussion Thread JBL 4343 to 4344 upgrade

  1. #226
    jandregg
    Guest

    4345 crossover

    Rick

    I am 30 miles south of atlanta, gainsville is a bit north of atlanta. Probably an hour or so trip between the two. I grew up in KC and still have family there. Next time your in the area drop in for a listen. I am always glad to show off my jbls.

    I hope to get the second crossover built this weekend. Then do some serious listening.

    Cannot afford any really expensive caps, but will try mixing and bypassing on the charge coupled. I mixed some auricaps in with the daytons on my old bypassed crossover and found it made a huge difference. When people on this site first started talking about charge couple they were saying that bypass was not needed. In later post around the first of the year some ( Ian ) were reporting good results from bypassing in some instances. What a happly conundrum.

    John

  2. #227
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,939
    It depends how far gone you already are..in other words if your amps are AAA with AAA prices spare no expense...so scale you outlay accordingly..its all relative.

  3. #228
    Senior Member porschedpm's Avatar
    Join Date
    May 2004
    Location
    Reno/SF Bay Area
    Posts
    483
    I’ve been invited by Ian Mackenzie to publish an outline of the upgrades that I’ve made to convert my 4343’s to 4344 spec. And give my impressions of the resulting sound. This 2 ½ year upgrade process has culminated with the recent addition of custom built active and passive crossovers built specifically for my application by Ian. A summary of these upgrades are as follows. For more description and detail of these upgrades you’ll want to drill back through the 4343-to-4344 Upgrade Discussion thread.

    Upgrade 4343B drivers to 4344 spec.
    1. Replace LF 2231H drivers with 2235H
    2. Replace MF 2121H drivers with 2122H
    3. Replace HF 2420 compression driver with 2425J.

    Bi-Amp speakers:
    1. Pass Labs 250w/ch Class A amp for LF drivers
    2. Pass Labs 150w/ch Class A amp for MF-HF-UHF drivers
    3. Ashly XR-1001 2-way active crossover

    Upgrade stock passive crossovers:
    1. Outboard crossovers custom built specifically for 4344 application by Ian Mackenzie.
    2. Outboard crossover design completely bypasses the 4343’s Bi-Amp switch
    3. Outboard crossover design completely bypasses the 4343’s L-Pads

    Added internal Brace

    Upgraded active crossover:
    1. 2-way active crossover custom built specifically for 4344 application by Ian Mackenzie

    Evaluation of upgrades.
    1. When I upgraded from the 4343B to the 4344 drivers the changes were subtle. The low frequencies now seemed a little more detailed. The mids seemed to sound more natural and the highs seemed to be much more extended and natural (although I should add that some readers have said they prefer the aluminum diaphragm in the 2425H).

    2. Bi-amping the speakers added quite a bit of detail and authority to the low end. I had issues though that I attributed to the Ashly active crossover (or possibly the internal Bi-Amp switch or a combination of the two). What I gained in detail and authority on the low end, I gave up in terms of transparency and fullness of the music from the mids on up. The music sounded somewhat thin and undynamic. It seemed like the music was veiled. I described the sound to be as if there was something in between me and the speaker. I could still hear the music but it just wasn’t alive. I went back to a single amp set-up to verify it was the Ashly crossover/bi-amp switch causing this. I eventually decided that bi-amping was still the way to go and so went back to the bi-amped set up and started looking for a better quality active crossover.

    3. Adding in the custom built outboard passive crossovers, eliminating the internal bi-amp switch and L-pads, and adding the internal bracing was all done at the same time. While it would have been ideal to listen to the differences after each of these upgrades separately, it wasn’t practical to do so. But taken together, and probably due in most part to the new passive crossovers, they allowed the music to open up and be more full and transparent. The speakers sounded better than they ever had before. Even though there was still some slight veiling effect from the Ashly crossovers, it appeared to be minimized.

    4. It’s been a little over a month since I upgraded the active crossover to the custom active crossover Ian built specifically for my application. See pictures below active crossover. What you don't see is that the power supply is oused in a complete differen box and is hidden behind the equipment cabinet. Adding in this active crossover was another quantum leap forward. The veiling caused by the Ashly switch was now completely gone. And there was huge increase in detail from top to bottom. There was an overall new transparency and fullness to the music. The low end had a huge increase in control and authority of the bass. The mid through highs were now so crystal clear that any imbalance between drivers immediately stood out. It took me several attempts to get the balance amongst the drivers correct but once I did, the music became jaw-dropping good. It was detailed and dynamic. In fact it became hard to stop listening because it made all the old music I’d heard hundreds of times before, fresh and exciting again. I found subtle nuances I hadn’t noticed before because the detail just wasn’t there before. I still am amazed at how good it sounds when I turn on the music after being away for awhile. It’s nothing short of awesome. The music is now full, dynamic and detailed.

    It's taken over two years to get to this point but this latest upgrade has made the whole upgrade journey worth it. I can't over emphasize the importance using good quality crossovers. The quality of your crossovers should be at least equal to the quality of your other components. Really the easiest part was sourcing and installing the 4344 spec drivers. The most challenging part was upgrading the crossovers. You can use the stock 4343 crossover temporarily, but it was never designed for the 4344 drivers. And if you can find a 3145 network out there chances are it's time for it to be upgraded anyway. So invariably you'll be faced with having to custom make or if you have the time and expertise to do so. I had neither, so I was fortunate to have Ian custom make the crossovers for me. I asked Ian for his help because he understood the amps I was using and could match the quality of the crossover components to them. But there have been several LH threads written on building crossovers and there are several Forum members that have the ability to custom build networks. My deepest thanks go out to Ian for his patience, persistence and technical wizardry in building these phenomenal crossovers. It's hard for me to imagine how the music can sound any better. Although I have heard Ian's currently toying with the idea working on incorporating a small Pass DIY amp into an active crossover chassis to run the UHF-2405's. Sounds interesting.
    Attached Images Attached Images   

  4. #229
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,939
    Ed,

    Thanks for the thoroughly informative post.

    Its nice to see a challenging project like this turn out the way it has.

    I must say this would never have happened had it not been for Ed's passion, persistence and devotion that took this project to where it is today.

    It was a lot of fun and I am so pleased we meet and worked on this project together.

    Enjoy.

    Ian

  5. #230
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,939
    Okay,

    About the self powering of the 2405 it looks like Ed wants to go top shelf (again) so it might have to be one of these.

    This particular amp is a mini class A power amp with some similarities to the amps Ed is already using in his system. Ideally, all amps in the mid and high registers should be of identical design/type.

    Below are some old but timeless links to the above mentioned active crossover that has been the result of numerous improvements. There is a sister unit that is embedded permanently in my current system

    http://www.passdiy.com/gallery/hi-lo-xover-p1.htm

    The little amp also deserves some mention as it was reviewed some time ago now..excuse my enthusiasm at the time.

    http://www.passdiy.com/gallery/alephx-p1.htm



    Ian
    Attached Images Attached Images    

  6. #231
    Senior Member JBLnsince1959's Avatar
    Join Date
    Dec 2003
    Location
    KC - land of ahhhhs
    Posts
    1,795
    porschedpm

    Great post and congrads on your upgrade. Glad to hear it has gone well for you. I'm waiting to hear how the Highs are affected with the new little amp..

    Ian...Fantastic job...You're a real asset to this forum more than you know. I know you were there for me last year when I had questions and I really appreciate it....way to go buddy..

    Just to let you know, this is one of my favorite threads..while I don't have these speakers the principals can easily be applied to other speakers

    thanks guys....

  7. #232
    Steve Gonzales
    Guest

    Outstanding!

    Thanks Porschedmp and Ian Mackenzie!

  8. #233
    Senior Seņor boputnam's Avatar
    Join Date
    Apr 2003
    Location
    northern california
    Posts
    6,142
    Quote Originally Posted by porschedpm
    The veiling caused by the Ashly switch was now completely gone.
    Doods...

    First, thanks for the detailed postings and even more congrats for the 2-yr long exercise. That is devotion. You and Ian look to have done one hell of a job. I wish I had reason to travel to Reno...!!

    The quote confuses me. I've "heard" that from but a few members here, and don't get it. Were you able to collect any measurements of the response, before and after? Could you post some plots? Also, you didn't post whether there is any EQ in the signal path, and if so what it is, and what corrections you might be imparting.

    I ask because I just inserted a Bryston 10B in-place of the Ashly - but not for the reasons you did, nor with the results you got. I am trying to achieve what I understand is Greg Timbers' ideal active crossover application with a 4345 - use the active crossover in a biamp set-up to replicate the voltage drive of the original HPF (-12dB, 290Hz).

    I have not yet Smaart'ed this - I am waiting Widget's visit this weekend for some A/B, and don't want to corrupt the current set-up before I go fully into the Bryston config - but what I did get (admittedly NOT properly EQ'd and therefore results/opinions are not relevant), is seemingly improved tonality, and better channel separation. I say not relevant because in my soundguy work, I typically am confronted with a system that sounds either muffled or bright, or "veiled" or boomy, and have found time-after-time that proper measurements of room response and adjustments for that response vastly improve the character of the sound, and the perception of the system (and audience response and appreciation!). This is experience based upon Smaart'ing dozens and dozens of systems, often when I felt there was either no hope, or it was already so good there was little improvement that could be made. Wrong on both accounts.

    The "brightness" many attribute to Bryston, IMO, relates to EQ. Bryston may be more efficient in their reproduction of the HF part of the spectrum - therefore, systems not "aware" of that will certainly sound brighter.

    I've got measurements before this switch, and soon will Smaart with Bryston and share those.
    bo

    "Indeed, not!!"

  9. #234
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,939
    Hi Bo,

    Your post poses several questions so it may take several posts to reply.

    Firstly,

    The transparency is not related to amplitude response measurments but the E 2 E end distortion and other losses that are reduced with the active crossover and external crossover.

    I was not able to discern any change in response from new or old crossovers (the voltage drives being almost identical)..calibration of the system is the Key.

    We found the final L Pad setting more critical with the improved crossovers, I found about -1 db on the mid range and HF and UHF 0 postion using my net spl calibration method best with active using the ststem design 18 db slopes.

    Hope this answers some of your questions, no Doubt Ed will add some comments.

    regards

    Ian

  10. #235
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,939
    Quote Originally Posted by boputnam
    Doods...

    First, thanks for the detailed postings and even more congrats for the 2-yr long exercise. That is devotion. You and Ian look to have done one hell of a job. I wish I had reason to travel to Reno...!!

    The quote confuses me. I've "heard" that from but a few members here, and don't get it. Were you able to collect any measurements of the response, before and after?

    No, I don't think Ed has an RTA or a Clio for pre 4344 post 4344 response.

    Could you post some plots? Also, you didn't post whether there is any EQ in the signal path, and if so what it is, and what corrections you might be imparting.

    There is no graphic equ I am aware of, however the Mac preamp has facility for eq via tone controls

    I ask because I just inserted a Bryston 10B in-place of the Ashly - but not for the reasons you did, nor with the results you got. I am trying to achieve what I understand is Greg Timbers' ideal active crossover application with a 4345 - use the active crossover in a biamp set-up to replicate the voltage drive of the original HPF (-12dB, 290Hz).

    This would be much easier to discuss verbally.

    Question: What was your reason for installing it? (Are the voltage drives from the Bryston tailered to those of the passive 3145 crossover for the woofer to mid range driver?)

    My understanding you were happy with the Ashly.

    The passive filter voltage drive maybe difficult to emulate. The actual active slope is 18 db as you maybe aware prescribed by JBL. Perhaps if someone fluent with Soundeasy could take those passive voltage drives and compute an active eq simulation and schematic ..Tim G comes to mind.

    We could attempt to modify the 12 db slope with the new crossover to match the current JBL passive voltage drive but this has not been attempted. I will look at it but it will take some time.

    At any rate the 18 db slope in conjuntion with the active crossover is subjectively superior.

    I have not yet Smaart'ed this - I am waiting Widget's visit this weekend for some A/B, and don't want to corrupt the current set-up before I go fully into the Bryston config - but what I did get (admittedly NOT properly EQ'd and therefore results/opinions are not relevant), is seemingly improved tonality, and better channel separation. I say not relevant because in my soundguy work, I typically am confronted with a system that sounds either muffled or bright, or "veiled" or boomy, and have found time-after-time that proper measurements of room response and adjustments for that response vastly improve the character of the sound, and the perception of the system (and audience response and appreciation!). This is experience based upon Smaart'ing dozens and dozens of systems, often when I felt there was either no hope, or it was already so good there was little improvement that could be made. Wrong on both accounts.

    Interesting observations;On the basis that the L pads are setup correctly the distinction in Ed's system (or the reference system) is not relevent to Eq flatness. The veiling (we both observed) refers to loss of information and Hf brightness. This is attributable to what happens in the signal processing. This being insertion losses from cables and terminations, transparency of signal processing electronics and the power amplifiers.

    The "brightness" many attribute to Bryston, IMO, relates to EQ. Bryston may be more efficient in their reproduction of the HF part of the spectrum - therefore, systems not "aware" of that will certainly sound brighter.

    Could be the case! We found ther setting up of all 4 drviers far more critical in terms of level matching.

    Perhaps a process of elimination may assist here. Temporarily bypass the Eq 1/3 direct from the preamp to the crossover=>and power amps. Then switch in the EQ in flat mode but not bypassed and then alternate with the Ashly and Bryston.

    The brightness could also be attributable to the source or even the power amps. Also look at the net spl method of setting up the Lpads, it works.

    I've got measurements before this switch, and soon will Smaart with Bryston and share those.
    As a post script, I spent considerable time (months) evaluating the active crossover using similar amps and obviously drivers and passive crossovers identical to Ed's.

    Very subtle variations is the final part selection and usage in the active crossover provided a presentation that was both transparent and detailed without any sonic glare.

    The Jeft differental pair & bjt discrete class A output used in the opamps in the crossover have characteristicly somewhat higher THD than the best chip opamps, typcially 0.05 % verus 0.005 distortion. (I hate bjt front ends in any amplifier, they lack detail and sound un natural to me imho). But they have other special qualities that are particularly attractive for audio.

    Jfets are analogous to Valves and are known to be more transparent and offer finer resolution than equivalent BJT eqivalent designs. They also have low feedback resulting from low open loop gain


    The final melt of the active crossover and the power amps is the essence of what happens, and of course the passive crossover parts and terminations. In this project we addressed all these areas to deliver a final result which I would consider neither warm nor dry, bright or soft sounding.

    Because the various aspects of the signal chain are cascaded, its is often rather confusing to determine where the audible traits are attributable and for this reason changing out one component may or may not give the desired result.

    Of course it depends on what you want and like. Some people like and prefer a bright detailed sound and that is fine if that is what you are looking for.

    Ian

  11. #236
    Tom Loizeaux
    Guest

    Great thread!

    I think this thread is an example of one of the best things about this Forum. Many of us have, or want to aquire some classic JBLs - so this kind of give and take information, along with lots of sharing of real experience on very specific issues makes the prospect of restoring and even upgrading these classics very attractive.
    My thanks to all of those involved. Your time and work has elevated the value of this Forum.

    Tom

  12. #237
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,939
    Tom,

    Glad you enjoyed the read. Unfortunately none of this is a overnight exercise and it can be tedious.

    But if you follow a logical process and execute each phase correctly the results can be quite rewarding..and its fun!

    The next phase we propose is to consider fixed resisters to replace the variable Lpads once their settings have been verified. The Lpads are not a very precise way of setting up the drivers using the db markings on tnhe foil cals as Ed will attest to. Hence we measure and calibrate each drive one at a time

    Incidentally the 4344 requires the mid to be attenuated -5 db from maximum, the horn and slot -6 db from maximum for proper alignment with the 2235H a best we can tell.

    The rta is probably not accurate enough to compare individual frequencies. Hence the net method of maximium less a known amount is best for reading exact attenuation of selected spot frequencies. However if you did use an analyer you would find the overall response remarkably smooth! Using broad divisions like 12 db..I did and it sounded perfect. I find I can get a really smooth response with the net method far more so than using an analyser or RTA particularly when the system has a rather wiggely response like the 43XX systems. How do you find the mean response with a driver that varies +- 2-3 db with all sorts of peaks and bumps and dips?

    Even with smoothing its an approximation and I am looking for 1/4 db or better precision matching of the actual sensitivities (levels) of all 4 drivers.

    If you can get a mean value of SPL b/n left and right UHF, HF , and Mid then so both left and right are as close as manual measurement allows then we can consider measurement of the L pads to arrive at some fixed resisters values.

    Note this may require 3 or 4 attempted readings to get a mean reading on the meter because of the sloppy nature of the Lpads.

    The key is to get both Left and Right channels the absolutely same. You will just know when its right..the speakers drivers outputs will no be distinguishable from each other and the imaging will drag you in and engage you.

    Ian

  13. #238
    Senior Member porschedpm's Avatar
    Join Date
    May 2004
    Location
    Reno/SF Bay Area
    Posts
    483
    Quote Originally Posted by boputnam
    Were you able to collect any measurements of the response, before and after? Could you post some plots? Also, you didn't post whether there is any EQ in the signal path, and if so what it is, and what corrections you might be imparting...
    Hi, Bo. Sorry, beyond my trusty little SPL meter I didn't use, and actually don't have, any measurement equipment. (I'm open to suggestions on what would be good for the amateur, slightly more than casual home user, though) There are 8 bands of tone controls on my preamp but they're always set to flat.

    I'm interested to hear how you like the Bryston. I realize Reno's a bit out of the way. I'm still loving my new set up, like a kid with a new toy. But later on, say in a few months, or if we ever have a Bay Area LH gathering, I might be able to be talked into bringing Ian's active crossover down for your evaluation and listening pleasure.

    Ed

  14. #239
    Senior Seņor boputnam's Avatar
    Join Date
    Apr 2003
    Location
    northern california
    Posts
    6,142
    Quote Originally Posted by Ian, in Tardis
    The passive filter voltage drive maybe difficult to emulate. The actual active slope is 18 db as you maybe aware prescribed by JBL.
    True, and correct. The specific 4345 cards for the 5234a have a -18dB slope. I talked to Giskard extensively about this aspect of the active 5234a. He reiterated a comment from Greg Timbers, leading me to elect the -12dB slope (as per my post). There is a risk of some loss of definition, and of overdriving the 2122H (due to the shallower slope of the HPF). Of the former - I am listening hard for this, but thus far have not discerned it after dozens of A/B's with the Ashly. To the latter - I sensed some overdriving of the 2122H (even at modest gain) and reduced the gain on the MF-HF amp, as Giskard suggested might be needed. He predicted that correctly.

    Quote Originally Posted by Ian, in orange
    We could attempt to modify the 12 db slope with the new crossover to match the current JBL passive voltage drive but this has not been attempted. I will look at it but it will take some time.
    I'd be most interested in your findings, if you ever do this.

    Quote Originally Posted by Ian, continued
    At any rate the 18 db slope in conjuntion with the active crossover is subjectively superior.
    Yes, I agree. However, I here elected to follow the guidance of the 4345 designer, and try and emulate the voltage drive of the HPF in the original 4-way passive network. This -12dB slope was chosen by them after hours and hours of listening and tweaking. I
    n "purist" approach, this Bryston has no switches - it is fixed at -12dB, 290Hz. It is a big change coming from the -24dB of the Ashly, but I cannot yet say definitively there is any loss of definition using the -12dB slope.

    I've not yet tweaked anything. Widget plans to come 'round this weekend for some hearing sessions, and I'm interested in his appraisal.

    Thanks for the comprehensive reply(s).

    One last thing:
    Quote Originally Posted by Ian →
    The veiling (we both observed) refers to loss of information and Hf brightness.
    This intrigues me, still. You've made so many improvements I wonder how much of this can be attributed to removing the Ashly, per se. This characteristic is something I am listening hard for, but can't verify this "veiling" in my set-up. I think our systems were quite different at the start of this Ashly comparison (mine is all balanced, XLR interconnects with very low noise floor). I think there is a bit more definition with the Bryston (at 3x the Ashly's cost, there should be...) but there is no loss of information in using the Ashly.


    bo

    "Indeed, not!!"

  15. #240
    Senior Member
    Join Date
    Nov 2003
    Location
    Toronto
    Posts
    1,162
    Quote Originally Posted by [COLOR=black
    One last thing:
    This intrigues me, still. You've made so many improvements I wonder how much of this can be attributed to removing the Ashly, per se. This characteristic is something I am listening hard for, but can't verify this "veiling" in my set-up. I think our systems were quite different at the start of this Ashly comparison (mine is all balanced, XLR interconnects with very low noise floor). I think there is a bit more definition with the Bryston (at 3x the Ashly's cost, there should be...) but there is no loss of information in using the Ashly.


    [/COLOR]
    Bo, not meaning to muddy the waters here, but in my triamp set up with Ashly's doing the 50 Hz L/P and 200 Hz B/P with a high quality passive network in front of my H/P amp instead of using active at 200 Hz (at only a 6 dB slope), there is a noticeable diffrence if I sub the Ashly in temporarily.

    The same is true for my Ashly 1/3 octave EQ, it is relegated to operation below 200 Hz (I know, - what a waste!). It's audible when used full range in my system.

    I have to agree with Ian too, on the BJT/JFet issue, my experience has been the same.

Thread Information

Users Browsing this Thread

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

Similar Threads

  1. How to Upgrade your JBL 4343 to a 4344
    By Ian Mackenzie in forum Lansing Product DIY Forum
    Replies: 3
    Last Post: 09-08-2005, 10:33 PM
  2. 4350 Monitor Plans - Discussion Thread
    By boputnam in forum Lansing Product General Information
    Replies: 6
    Last Post: 05-28-2005, 09:54 AM
  3. Upgrading a 4343 to 4344 components
    By porschedpm in forum Lansing Product Technical Help
    Replies: 9
    Last Post: 10-29-2004, 10:45 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
  •