Page 55 of 58 FirstFirst ... 5455354555657 ... LastLast
Results 811 to 825 of 863

Thread: Project M2 DIY Thread

  1. #811
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    7,822
    Hello Ian

    You didn't steal it I posted it. I don't have a problem with you building from it and suggesting changes. That's fine that is what this forum is about sharing where we can to help each other out. That and did say where you got it from.

    Thank You

    I did a couple of Leap simulations on changes to R2 It's easier to see what actually happens as opposed to just looking at the voltage drives. First up is the response with no notch. That is with the fixed attenuator in place.

    If you change the driver attenuation that entire curve position/level above will change based on the attenuation so using a variable attenuator there is going to effect the R value in the series notch.

    I went for maximally flat on the M2 horn I have a 1db step starting at about 1K. That said I set the attenuation and then tweaked the R series value.

    The effect is larger bandwidth wise than it appears looking at just the voltage drives. Not that it's not there just easier to see looking the the FR simulation.

    So no notch, flat at 6.8, 2 ohms and 15 ohms Hinge starts at about 6K but effects the curve all he way back to 3-4K. Has like a see/saw effect.

    I didn't look at changes to the parallel notch vs attenuation level. Keep in mind you may have to tweak that as well and possibly the band pass as the notch and the band pass combined form the roll off.

    Rob
    Attached Images Attached Images     
    "I could be arguing in my spare time"

  2. #812
    Senior Member srm51555's Avatar
    Join Date
    May 2013
    Location
    Wisconsin
    Posts
    600
    Ian and Robh,

    Thank you for taking the time and posting this, it is much appreciated.
    Quote Originally Posted by Ian Mackenzie View Post
    Hi Scott,

    Got ya.

    Just looking at your recent posts. What drivers are you currently using?

    It seems you have tried the stock M2 drivers and moved on. I haven’t personally set my M2 horns up yet as l am buried in some other work right now. You seemed stuck. But l note that due to the wide dispersion on the horn you “may “ need to globally attenuate the horn response down overall by about 1.5 db or more ref the 2216nd. Or taper the response down to suite your conditions. This may defy convention.

    Why?

    Due to the very wide power response of this particular wave guide the on actual on axis response while useful for crossover design is not necessary appropriate for listening in a domestic situation where you have six flat un treated surfaces.

    Trust your ears not the measurements because your ears are hearing the whole room. Not just the direct on axis response if that makes sense. The D2 driver is possibly more surgical and analytical than what your used to and that is why l posted about analogue EQ. DSP EQ can also cold, analytical or sterile and that is also part of the story. Not really appropriate in the hifi world. So you might be having difficulty acclimatise to the presentation.

    Back to the network what 4313 alluded to was a network that allows you to manually adjust an L pad back in the 2-3 khertz area just like the 4430 crossover schematic. It also has a HF brilliance LPad. These L pads enable easy adjustment so you can make adjustments and listen. That is the key.

    It’s a very simple network that you can play with. It won’t be technically dead flat. But that l don’t feel is what’s important in your space at the moment.

    Trust your ears not your on axis measurements at the moment. Ruler flat response on paper is the icing on the cake. But not what’s in the cake. You need to get the mix right. If you look at the M2 overall EQ JBL significantly doctored the 2216 bottom end to give the presentation balance. It’s like a see saw. Too much mids and highs means your ears need to turn up the bass tone control to make it sound right. First world problem.

    The whole deal on it must be flat goes right out the window because your ears are receiving the total sound power and they are saying it’s way too hot. Simply dial back the treble and dial up the bass like a tone control so to speak. Don’t be locked into the rigid regime of obtaining a ruler flat response. That’s where perhaps the confusion is.

    I hope this brings you some insight.

    Ian

    Edit Below is Robert H M2 passive network for the 476Mg driver. It is close enough for any conventional JBL driver. If you have a 16 ohm drive simply put a 20R resister in parallel with the drive and the network won't know the difference.

    I have added 3 Lpads labelled in the attached schematic.
    These L pads allow independent adjustment of 3 hertz Presence, 10K hertz brightness, and overall Horn level.

    Technically an 8 Ohm L pad can be wired up for an 8 ohm trimmer or a 33 ohm trimmer as it has two independent wire wound tracks.
    This embodiment will enable on the fly adjustments I mentioned above so you can listen and tweak just the way Greg Timbers does.

    As I said earlier the ruler flat stuff is really just finessing the optimisation once you have a ball park listenable system which is the whole idea of using classy drivers.

    Notice I actually made the effort to upload nice clean pdfs...LOL. Enjoy your weekend where ever you are.

    Ian,

    Currently I'm using a 2451 with D8R2450SL diaphragms. The M2 waveguides are attenuated through the DSP, but not sure how much at the moment. A member noted on ASR that JBL preset a High Shelf at 8khz (-3dB) when using it's SDEC DSP as a crossover which I have not been able to try yet. I agree with the ears being the final test. Below is a picture of two measurements (~1 meter away) I took in the beginning the year. As a disclaimer, it's a poorly taken picture and I usually only like to use screenshots but I didn't save the measurements and this is all I have. From this I feel the speakers measure similar (outside of the hf loss at 11k for the 2405) but sound completely different. Both used the same amp.

    Red is a 2235/2379/2440be/2405 in 5cf cabinets
    Green is the DIY M2
    Name:  plots.JPG
Views: 280
Size:  51.3 KB

    I think the passive with the L-pads my be the way for me to do due to the ability to adjust on the fly as you noted. Funny thing is that these speaker sound better than most have heard but there is always that 1% we are chasing, plus what fun is it to leave it alone.

    Have a great weekend and thanks again,
    Scott

  3. #813
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    7,822
    Hello Scott

    Your Welcome

    Thank You for posting your measurements. You have the same on axis 3K peak I have in my measurements. I was wondering if it was me or real as in an on axis bump the averages out as you go off axis. Glad you are enjoying them. I really like the way they sound and image!

    Also there is a hidden R value in L1. When I start the sims I use perfect inductors and add a resistor in series. I change the resistor value to see how the DCR effects things and then just find an inductor that has the same or similar DCR value and leave the R out of the actual build.

    Have fun!

    Rob
    "I could be arguing in my spare time"

  4. #814
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    Hi Scott,


    Thankyou for posting your measurements.

    I hauled my M2 horns out tonight and plan to try and get them ready with the passive crossover as posted and shoot some measurements while the weather holds over the weekend.

  5. #815
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    7,822
    Quote Originally Posted by Ian Mackenzie View Post
    Hi Scott,


    Thankyou for posting your measurements.

    I hauled my M2 horns out tonight and plan to try and get them ready with the passive crossover as posted and shoot some measurements while the weather holds over the weekend.

    Great! Looking forward to see what you get. What are your compression drivers??

    Rob
    "I could be arguing in my spare time"

  6. #816
    Senior Member srm51555's Avatar
    Join Date
    May 2013
    Location
    Wisconsin
    Posts
    600
    Thanks for the kind words. I also look forward to what Ian's comes up with, weather permitting.

    Thanks,
    Scott

  7. #817
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    Quote Originally Posted by Robh3606 View Post
    Great! Looking forward to see what you get. What are your compression drivers??

    Rob
    D2, 4003, 2447.

    Not quite ready. A lot of preparation to do. Trying to sneak this in before my reno's. Got the adapters and the D2 set up on an M2 wave guide today. Decided on M5 taped holes into the shell. Seems okay.

    For those reading this post the phase response of a compression driver and horn/wave guide is not linear. A departure from the phase angle at the lower end of the pass band occurs. It's not the response but the phase response that is important in the HP filter with a wave guide or a horn. It varies depending on the wave guide flare rate and the driver resonance. With a tight crossover region like in a 6th order slope and natural alignment of the drivers it can work out nicely in analogue. If the driver phase is shifting too much its simply a case of moving the crossover point to a better spot.

    Below are two test measurements. BTW the date is wrong on this very old XT computer running Linear X LMS. I don't feel like fixing it just to see who's awake.

    (1) The M2 wave guide response at 1 metre ground plane (indoors)
    (2) The impedance.

    These are NOT reference test results. Alright. Just a trial measurement.
    The wave guide was off the vertical axis with the mic so don't panic.

    Notes
    You can see the minimum phase transform on both tests that relate to my comments above.
    The divisions are 10db per division. Its actually fairly smooth and should not present any problems for a diy attempt at a passive crossover / EQ.
    The M2 wave guide phase has some straight lines which is good news. Just proves I do know something about this stuff.

    I may open another thread with detailed measurements of the M2 wave guide focusing on a passive / active analogue crossover filter and EQ. This will avoid any confusion / clash of the wills with those who have their hearts and minds set on a negative engineered JBL M2 digital system using JBL calibration files.

    In my experience anything to do with a stock JBL system crossover of any sort is never diy friendly. That's one of the reasons I am focusing on the alternative approach and just to be a pain the ass because I love my job...LOL.
    Attached Images Attached Images   

  8. #818
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    Hi Rob,

    Based on my preliminary tests l think that 3 khertz peak is due to proximity to the mouth. Possibly a phasing thing with the geometry of the wave guide and those fingers in the throat. I did not get that at 2m. Unfortunately my rear neighbour raised a concern for her cats mental health so l had to stop my mad professor experiments.

    I did try some measurements in doors and it quite tricky.

    Will try some measurements in my driveway tomorrow.

    Ian

  9. #819
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    7,822
    Quote Originally Posted by Ian Mackenzie View Post
    Hi Rob,

    Based on my preliminary tests l think that 3 khertz peak is due to proximity to the mouth. Possibly a phasing thing with the geometry of the wave guide and those fingers in the throat. I did not get that at 2m. Unfortunately my rear neighbour raised a concern for her cats mental health so l had to stop my mad professor experiments.

    I did try some measurements in doors and it quite tricky.

    Will try some measurements in my driveway tomorrow.

    Ian

    Ok thanks! I had to do my measurements inside, pandemic no vax quarantine, I also had a tuff time finding a spot where I could get reliable repeatable measurements took me a couple of try's to get it right. Looking forward to your measurements. I was about 1 meter + oriented corner to corner 4 ft out.

    Scarring the cats! How loud were you!

    Rob
    "I could be arguing in my spare time"

  10. #820
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    I'm going to compare a ground plane measurement with a free standing measurement with the M2 free standing outside today.

  11. #821
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    Quote Originally Posted by Robh3606 View Post

    Scarring the cats! How loud were you!

    Rob
    Only 0.7 volts, -12db on the 2.83 volt level. I did a number of sweeps and had a feeling the neighbors might crack it…Lol. She was not in the mood for small talk.

    The old 486 Pentium had a murmur yesterday. Would not start. All my measurements at risk. I had it on all night. Must have disappeared up its own clacka…! Despite being old school LMS has some really cool data functionality so am going to get an industrial pc with a Isa slot built. Plan to get pocket Clio.

  12. #822
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    7,822
    Quote Originally Posted by Ian Mackenzie View Post
    Only 0.7 volts, -12db on the 2.83 volt level. I did a number of sweeps and had a feeling the neighbors might crack it…Lol. She was not in the mood for small talk.

    The old 486 Pentium had a murmur yesterday. Would not start. All my measurements at risk. I had it on all night. Must have disappeared up its own clacka…! Despite being old school LMS has some really cool data functionality so am going to get an industrial pc with a Isa slot built. Plan to get pocket Clio.

    Clio Pocket works just fine all my "M2" measurements were done using it. They have updated it 3 time since I have owned it. It's self contained, self calibrating and you just need a USB port. It's a really nice portable measurement system. First time you calibrate wait about a 1/2 hour for the control box to reach it's temp equilibrium.

    Yeah I hear you on the computer my Clio 7 and Leap are on an old Pentium running XP so it's only a matter of time! Hope you get it back! I had the same thing happen and switched out the power supply and it came back to life.

    Rob
    "I could be arguing in my spare time"

  13. #823
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    Yes its annoying. It froze today. I finally did some indoor measurements tonight with the wave guide elevated 1.6m off the floor and gated it at 1m and 3.4m for the bounce. Seemed to work out.

    Obviously the measurements need further work.

    Attached is a simple M2 diy passive high pass network based on the JBL 4430 style of network developed by D.B Keele and Don Smith.

    Hammered this out in about an hour.
    My focus was simplicity for easy of construction and cost. The 5.2mH inductor can be a 2-3 ohm air core or a descent ferrite core. Its simple so it's not perfect. (No gripping) It has two controls as labeled and set up for a D2 driver wired in series. The input impedance is about 10 ohms and has a minimum impedance of about 6.6 ohms at 20 Khertz. The acoustic slope is 36 db. This can be revised if required.

    When more time permits I may embellish it further. This is a startup passive high pass network for those launching into an M2 clone.

    Edit 41/05/22
    The acoustic slope of the D2 at the edge of its passband is between 18-24 db / octave. So I have used an 18 db electrical network. The crossover summation needs to be confirmed. The two LCR Series Traps help damp the impedance peaks due to the horn and a bump at 3 K hertz and 700 hertz. Unlike the 2344A the M2 horn loads the driver nicely so there's no voltage ripple across the network.

    2216nd passive low pass network to follow.

    Edit 02/06/22 The schematic is being updated.

    Edit 03/06/22 File data being updated.

    Will be be quite busy over the next few weeks so l will see how far l get with this little project.

    For those who are like hello? The M2 is in current production. I do have affiliations and connections in the industry and itís not welcome to have un controlled technical information floating around the Internet.

    Ian

  14. #824
    Administrator Robh3606's Avatar
    Join Date
    Apr 2003
    Location
    Rocinante
    Posts
    7,822
    Nice Ian

    One suggestion change your graph scales to cover the area of concern. You can also export in .jpg so you don't have to download to view. Reminds me of the 4367 where they use fixed pads instead of pots. Check out the schematic when you get a chance.

    Rob
    "I could be arguing in my spare time"

  15. #825
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,455
    This was done at 2.30 am in the morning down under time.

    Your reading my mind again. Stop it. The first thing l looked at was the 4367 schematic.
    Iím planning on talking with Greg about the 4367 passive network. The 4367 has different EQ. It didnít sim well. At least with series voice coils.

    Itís a work in progress. The measurements are not optimal. When l have time l will re publish the curves as Rev2. They can open it if interested and download as it evolves. Pdf are much better to print and email for over the over 50ís with reading glasses. I prefer not manipulate the scales. Whatís important is a consistent perspective with log scales. I find magazine curves misleading.

Thread Information

Users Browsing this Thread

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

Similar Threads

  1. JBL LN3 Project, *Official Thread
    By Nightbrace in forum Lansing Product DIY Forum
    Replies: 58
    Last Post: 07-13-2016, 03:08 PM
  2. Fundraising request; donations to Lansing Heritage Project May loudspeaker project
    By mikebake in forum Lansing Product General Information
    Replies: 7
    Last Post: 04-21-2011, 12:37 PM
  3. Project May, similar privat project?
    By Flodstroem in forum Lansing Product DIY Forum
    Replies: 14
    Last Post: 02-21-2007, 03:42 PM

Posting Permissions

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