After 5 years it finally begins

[sebackman]

Some of you may know that the M2 hysteria was started back in 2015 when I ordered the first 2pcs JBL M2 WG's by their spare part number trough a very friendly speaker shop in Florida (they still are!) and actually got them to my very big surprise. They were the first two to be sold separately. I quickly ordered 6 more and got them too. Pure joy.

I posted data here on LH and I think some jaws dropped in that they were actually obtainable at that point. Many have since successfully built M2’s or clones in various forms. The rest is history.

My own WG’s have been sitting idle while seeing others build nice speakers over the years. I have supplied M2 adapters to many and assisted many builds in any way I could.

Now its time to get my dusty WG’s in service.

This tread will cover the original idea with replacing my current LCR system with DIY M2 Rev RoB WG’s, 1400PRO, 2451Be and 045’s.

Just to be certain I’m also building a pair of faithful M2 clones at the same time. Maybe not that faithful, I’m going to use the -1 woofer and 4” drivers (also have D2's). More of that in a separate thread if anyone would be interested.

Attached are some pictures and more to follow.

Kind regards
//RoB


85331
85329
85330

[jmpsmash]

nice!

what's happening at the last photo?

[srm51555]

Your threads are always a good read, cant wait to see the progress. Is there a reason you chose the 2216-1’s over the nondash?

Thanks,
Scott

[pos]

Yes RoB, I think all of us M2 DIYers can be grateful to you :)


m going to use the -1 woofer and 4” driversSL diaphragm I presume. Hmmmm, passive crossover maybe?

[srm51555]

Yes RoB, I think all of us M2 DIYers can be grateful to you

Yup, both of you:)

[more10]

nice!

what's happening at the last photo?

Looks like he is scanning them in a CNC machine

[1audiohack]

Didn’t he cut them shorter and port two of them for super tweeters? Looks like a mill op to me.

Barry.

[Robh3606]

Looking forward to see you complete them! I have been sitting on mine as well. One of these days

Rob:)

[sebackman]

Thank you for the positive comments.

I wanted to keep the 2451Be’s as the Be dias do have a nice tone but at the cost of sharp roll off in higher frequency. I would have preferred to build a 2-way system but the tests I did using the BSS DSP to flatten the curve felt like it lost the crisp in the higher segment. I can get the curve pretty good with a lot of EQ but when adding a tweeter, the treble got more clear and “airy”. Even if the curves look pretty much the same. I did not experience the same with the SL dia’s using required EQ for them. Odd

I typically hate when science (measurements) and perception deviate like that. I don’t believe in psychoacoustics. Be as it may, 3-way with sub it is. :-)

Anyway, as I have the 045’s with the K2 UHF horns they will have to be used here too. Many new speakers strive after providing a point source and JBL has over the years integrated UHF units in their horns to reduce the acoustic centre distance. I decided to do the same and integrate the 045’s in the M2 WG’s. Hence the CNC milling.

This system is also wall hanged to mitigate animated discussions with other family members, so the requirement was that these are keep rather narrow. The M2 WG’s are quite wide so the only way was to put the blade in them.

Attached are some pictures. I will revert with more info over the coming days.

85364
85365
85366

[sebackman]

In the CNC job we also took off the upper edge to minimize the baffle space. The 045 horn’s was recessed on the side were the logo normally should go. I know they are JBL’s so I don’t need that 😊 .

The next job was to make them narrower. That was a nightmare as the material these are filled with is some kind of glas/plastic mix that is a pain to saw in. I had to change the blade after each cut. And the smell is not for the fainthearted.

I made a jig to be able to cut straight as these WG’s are curved in many directions. A bit crude but it worked out just fine. They will be recessed in the new cabinets and after repainting I hope they will look factory to all but the trained eye.

85367
85368
85369
85370

[macaroonie]

Our Swedish McGuyver , Great work Rob.

Prob some mineral fill in that plastic , did you try pure carbide , 3 flute spiral downcut at a lower speed..

M

[sebackman]

Hi Macaroonie,

I used a hard metal blade and that just died after each cut. I had three blades but had to file them for the second cut. Carbide would have been perfect but probably at a price. Anyway, now they are in a nice "wall-hanging-friendly" size.

I have done some preliminary measurements and the difference from a "non-molested" WG is not that big. I will have to do better measurements soon. My workshop is rather small so I will have to figure out a way to get some decent flight time before the first echo.

I want them as compact as possible so using the squarcle 1400PRO with the M2 WG is very area effective.

The centre is mounted above an opening in wall into the next room and my screen is roll-down with speaker cloth at the top black-drop. It is angled 15 degrees down so the new one will have to be similar.

The new system wil be LCR with the new cabinets described here and the old fronts will be used as L+R rear. My faithful SUB1500 subs will stay as they are still very good compared to many alternatives. All will be finnished in light Oak, as today. Everything for the family peace.

Come to think of it, I have no idea were to put the M2 clones we are building at the same time. As there is no chance to hang them on the wall, I guess there is some "compensation" needed to keep the family peace.


I have a professional furniture builder do them as I’m all to familiar with my limited carpenting skills and I have no CNC ´machine :-) . I’m supposed to get them within a few weeks.

85376
85377
85378
85379

[sebackman]

Here are the ideas for the new center.

85380
85381
85382

[bubbleboy76]

Very fun to follow this project, subscribing to this thread now! I wish you good luck.

What is the rough timetable, will it be ready for Christmas?

Please tell us about all family compensations you will have to do! I guess they are not fixable with DSP!

Are you using the same carpenter as I did (Södertälje Specialsnickeri)?

[sebackman]

Hi bubbleboy76,

I guess a "DSP compensation" would have limited IRL compensation impact .... :-) . Probably healthy to stick to other options.

We are not using the same carpenter. The one we are using is located 300km south of me. They are also building the monitor cabinets for us.

The left and right speakers will be rather shallow to look "smallish" on the wall. I know that will impact low end but with DSP and reasonable amount of power in a normal living room I think I can compensate for the fall off below 200hz. I will XO @ 70Hz towards the subs. I hope that will be fine.

Right now I have two SUB1500 subs in closed 70L cabinets and they sound just fine. However I do have two more almost brand new SUB1500 drivers so maybe I should make a second pair. I don’t need more bass but maybe 4 subs can reduce the nodes a little. That is really not a problem in my room either but I need a reason to do something with them. They are not making anyone happy in the cardboard boxes in the shop. Or just stop the madness and sell them.

Right now my favorite subs are the 12” active LSR4312P, they are just wonderful under a pair of 305P active monitors for office duty.


I hope to get the new cabinets before X-mas. Maybe the M2’s at the same time. That would be nice.

Electronics will the same (now almost ready for the museum) Meridian G68 pre with digital 7.1 (only using 5.1) out feeding SPDIF (24/96) to a BSS BLU800 with 8 analogue out and a separate BSS BLU-BOD DAC that adds another 8 analogue channels out. The BSS feeds two Crown CTS8200 (8-channel amps) and a MC2 (class B sub amp) amps via two 8 channel Burr&Brown PGA2310 based VCA’s.

I also use a BSS BLU-USB unit to feed digital sound into the BSS DSP. That is a pretty neat unit and not expensive. You can actually run 7.1 through them as they are 8 channel (in and out) devices by using a program called Virtual Audio Cable. With that you can assign an output from a SW player to any output and feed asynchronous USB audio direct digital to the world of BSS DSP as long as they have a BLU-LINK connector. Brilliant !!

Fronts are 3-way and rear will be only 2-way. The rears will use the old front cabinets with dual 8” drivers and a 2451SL on the VTF F12 WG (as today). The SL’s can go all the way which the Be’s cannot (in my opinion) , hence the 045’s in the new fronts.

Enough ranting today, attached are some drawings of the left and right cabinets.

85384
85385
85386

[Mr. Widget]

The left and right speakers will be rather shallow to look "smallish" on the wall. I know that will impact low end but with DSP and reasonable amount of power in a normal living room I think I can compensate for the fall off below 200hz. I will XO @ 70Hz towards the subs. I hope that will be fine.
I have found shallow boxes to allow more midrange coloration to reflect back through the LF driver's cone. You may want to explore internal diffusers or Sonex foam on the back wall of the enclosure to help mitigate this.

Have you mocked them up?


Widget

[sebackman]

Hi Widget,

Thank you for the input.

I did experiment a little with current cabinets that are also shallow. However, in the end I just settled for a layer of wool felt (about 1/2") applied with excess material creating an un-even surface on the rear wall and some more damping in the box. There is also some bracing on the back wall that limits the flat surface.

Once I get these new cabinets, I will have a go at checking if such addition of a diffusor foam at the back is audible. I do have some triangular shaped diffusor foam in the shop but I have no idea how it react over frequency. Good ideas must be tested.

At reasonable levels in a living room I guess the potential distortion is limited. At higher volumes it may prove more problematic. My system rarely sees concert levels unfortunately. :-)

The wool felt is actually the same wool felt I use in the drivers when removing the often-aged foam in the rear cover. I typically just remove the old foam and cut a circle of the wool mat and glue in. I cut the circle the same size as the outer diameter of the back cap and when I glue it in it will nicely cover the sides a little like in the 475Nd/476Nd. I recognize that they a have larger back caps but this works just fine. I also added a picture of my favourite diaphragm in the 2451 cores, now difficult to find.




85404
85405

[sebackman]

85448

[sebackman]

Christmas comes early this year.

Sorry for picture being upside down. No idea how to fix that. Normally ok, but this was uploaded from an iPad.

[sebackman]

85483
85484

[Mr. Widget]

Looking great!
Oh, and I fixed your previous image... much easier to see with it right side up. :D


Widget

[srm51555]

Yes looking great, progress is moving along pretty nicely.

[sebackman]

Still some work needed on the WG to make room for the 045. Slowly getting there.

85489

[pos]

85490
"RoB 1" :D

Very cool looking Robert!
That is a one of a kind and very refined system for sure, and with these top components and well executed integration it is going to be top notch for sure :)

[sebackman]

Hi POS,

Rob1 was the first M2 WG sold as a spare part back in 2014. That is where it started.

Now butchered to form something else.

More info to follow.

[sebackman]

Saturday fun in the shop.

2430k, 2451/475Nd, 2451Be & 2453SL

Impedance sweep naked, with VTX horn and M2 horns.

Tomorrow frequency measurements on all combinations.

Slowly getting there. :-)

85534
85535
85536

[ivica]

Saturday fun in the shop.

2430k, 2451/475Nd, 2451Be & 2453SL

Impedance sweep naked, with VTX horn and M2 horns.

Tomorrow frequency measurements on all combinations.

Slowly getting there. :-)



Hi sebackman,

Only JBL Be diaphragm the (476) is "missing"....
Anxiously waiting the measurements...

regards
Ivica

[sebackman]

2430K and 2451Be on M2 WG. Be is smoother but falls off sharpely after 15k.

85539

[sebackman]

Huge plans to cover some ground during the holidays was abandoned in the same pace they were made.

At least the cabinets are now reinforced for wall hanging and wall mount cups for left and right installed. The baffle is cut out to make way for the 045’s.

045 horns are attached to M2 wave guide and 045 drivers are attached to the horns.

Later this week I will remove the remaining two 2451Be’s from the old system replacing them with 2453SL’s and changing them from 3-way way to 2-way.

The measurements I did on D2’s with and without the passive JBL M2 filter show that the capacitor is mainly a protection cap and replacing it with a 6db filter in the BSS/DSP achieves nearly the same curve. The cap in the M2 filter is not a Butter slope, the D2 extends a little further down with the cap compared to 6db Butter I used in the DSP. Please see picture.

The XO used in my measurement corresponds to HP 6db filter at 957Hz. The frequency is the result of running a backwards calculation from a calculated 20,8 ohms load that the 8mf in series in the original JBL filter would be seeing. The 20,8 ohms load seen from the cap is a result of the average impedance between 800Hz and 20kHz (26,3 ohms) of the D2 driver on the M2 and 13 ohms in series and the 11ohms shunt in the filter. Lowering the XO can probably mitigate some of the difference, but the passive cap is probably not a Butter curve with those resistor values.

If you are building an exact M2 clone (this build are rather ClownsJ ) I would recommend keeping the passive filter to mimic the exact JBL curve. However, I do think it is possible to change the slope and/or the cut off frequency (lower) of the 6db filter in the DSP do achieve an identical curve without the cap’s, for the one willing to invest some time.

In my build I will use 4” drivers with Be diaphragms and 045 UHF’s so I have no curve to adhere to, except for the almighty “straight line”, so I will not use capacitors for the 4” drivers, only an attenuator with -8db. They are sturdy enough.

For the 045 UHF’s I will use a cap for protection as they are much more fragile than the 4” and I cannot use an attenuator for protection as the sensitivity is low, 87db @ 4ohms. That gives me about 90db when compared with the other drivers that are 8ohms. That will imply a real boost in the DSP anyway.


85614
85612
85613

[ivica]

....

Later this week I will remove the remaining two 2451Be’s from the old system replacing them with 2453SL’s and changing them from 3-way way to 2-way.

The measurements I did on D2’s with and without the passive JBL M2 filter show that the capacitor is mainly a protection cap and replacing it with a 6db filter in the BSS/DSP achieves nearly the same curve. The cap in the M2 filter is not a Butter slope, the D2 extends a little further down with the cap compared to 6db Butter I used in the DSP. Please see picture.

The XO used in my measurement corresponds to HP 6db filter at 957Hz. The frequency is the result of running a backwards calculation from a calculated 20,8 ohms load that the 8mf in series in the original JBL filter would be seeing. The 20,8 ohms load seen from the cap is a result of the average impedance between 800Hz and 20kHz (26,3 ohms) of the D2 driver on the M2 and 13 ohms in series and the 11ohms shunt in the filter. Lowering the XO can probably mitigate some of the difference, but the passive cap is probably not a Butter curve with those resistor values.

If you are building an exact M2 clone (this build are rather ClownsJ ) I would recommend keeping the passive filter to mimic the exact JBL curve. However, I do think it is possible to change the slope and/or the cut off frequency (lower) of the 6db filter in the DSP do achieve an identical curve without the cap’s, for the one willing to invest some time.

.....

Hi sebackman,

Nice work and very interesting.
Talking about the VHF driver response round 1kHz +/- I have to say that I can imagine the driver+M2Horn impedance would be a bit higher and more complex then expected in the calculation, so total response would be influenced, but I believe that You would find quite good solution.

For me D2 vs 2451/Be_Trx responses are much more unexpected. I am not sure, but I can imagine that high-frequency dispersion of D2 vs 2451&M2 would be different, as their responses differs more then 6dB over 10kHz. May be their phase plugs 'produce' different "wave front" at the drivers mouths, I would expect quite wider HF dispersion D2&M2 combo then 2451&M2 combo, but that is only my personal expectation looking at the drivers&horn F/R.
"Better" LF response of 2451/Be_Trx is something quite expectable for me, as for the 'ring-radiator type diaphragm' is not the best solution for the larger diaphragm excursion. I can imagine THD would confirm that.

Regards
ivica

[sebackman]

Hi Ivica,

The impedance of the “naked” D2 does indeed shift quite a lot over the range. The passive filter does really matter as can be seen in the plots below.

All is D2/2430K marked “M2” (brand new and I measured two different samples) on a new M2 wave guide, no baffle.

First plot is swept impedance taken with DATS II (calibrated)

Second plot is with the JBL M2 passive filter

Third plot is frequency with the correct BSS DSP correction and JBL M2 passive filter

Not too shabby! Please bear in mind that there is no smoothing at all applied here.

Best regards

85619
85620
85621

[sebackman]

Beryllium drivers installed. UHF will get protection caps but drivers only an attenuator.

Tomorrow they are going into the cabinets.

85671
85672

[sebackman]

Just got my hands on two "Holy Grale" diaphragms in the form of 2pcs NOS original "Northridge" D16R475Nd. Great. They are not for this build but in my opinion the best dias for the large format 4" drivers. To the left new SL dia’s from Mexico, to the left and Northridge NOS. I hope they are ok. If so they are going into the M2 build.

Filter done and cabinets done. Bolts turned out to be too long for the M2 horns so that will have to wait for tomorrow.

I did check the impedance curves for the left and right 2451Be drivers on their respective M2 horns and it turned out to be a pretty good match. Please attached chart.

85682
85683
85684
85685

[pos]

Good find on the SL diaphragms!

I cannot read the impedance measurement, the image got resized too small by the forum engine.
Are your Be diaphragms 8 ohm versions?
What MC2 models are you using?

The cabinets look really good too! :)

[sebackman]

Hi Pos.

I have no idea why the 475Nd dia's do measure "better" than other SL diaphragms in the 2451 cores (in my opinion) They are from the older K2 475Nd 2” driver, but the SL diaphragms should really all be the same.

These are similarly smooth to the Be's but will go all the way to 20k. Sound wise the Be’s are slightly more detailed but at the cost of needing a UHF unit. With this new build I will have both, 2451Be + UHF in one system and 2451/475Nd (no UHF) in the other. Both on M2 WG’s.

There is also a measuring difference in 2450SL and 2451 cores even if they look the same. I have no idea why that is, but it is consistent in my measures. Maybe the phasing plug is different or mounted differently. Maybe someone else here do know if and what the physical difference might be.

Impedance chart just shows that the pair (L+R) are very well matched indeed. Just sheer luck I suppose. Cores with changed/upgraded diaphragms are normally not that close over the entire spectrum, regardless of diaphragm. You typically never get them 100% equally mounted and seldom the drivers in the exact same position on a horn/waveguide. There are also different cores and sometimes this is marked by handwriting under the back cap with plus or minus measures, showing the phasing plug position, as I understand.

Maybe JBL matched cores and diaphragms at the factory but there is no way, that I know of, to specify a replacement diaphragm in relations to such markings. If they used shims to adjust the diaphragms height at the factory, I have at least never seen that in a 2451 core. The deviations from core to core may be small to the eye but do mercy less show up on the measures. This took some fiddling.

My Be's are 8ohms. Today I would have gotten 16ohms, but they are what they are.

I'm using a MC2 T2000 and a T3000 for driving subs.

I tried, on a suggestion from Widget regarding shallow cabinets, to use acoustic foam instead of fiberglass and there were neglectable measuring differences but since I had already bought it I might as well use it. Maybe it will be better in s shallow cabinets as Widget indicated, and I hope it will not fare worse than the fibreglass stuff.

Today I will give all the drivers a measure to make certain they are ok and then mount the assembly in the cabinets. As they are shallow it will be a challenge to get all 10 bolts in place on the back of the waveguide.

[pos]

Hi Rob,


There is also a measuring difference in 2450SL and 2451 cores even if the look the same. I have no idea why that is, but it is consistent in my measures. Maybe the phasing plug is different or mounted differently. Maybe someone else here do know if and what the physical difference might be.That is pretty strange indeed, one would expect an exact same internal design between the two. Might be something related to an evolution of the build process that could also appear between two 2450SL or 2451 from different eras?
Do you have measurements of that difference to share, or maybe a description of what was the most striking difference you consistently saw?


I'm using a MC2 T2000 and a T3000 for driving subs.Nice! What are you using for HF and UHF?

[sebackman]

Hi Pos,

I can post some comps next week. It is odd indeed.

I'm stuck on the old Crows CTS series class AB amps, fed from my BSS BLU800. They are as good as any with an odd topology used in several Mark Levinson amps as far as understand (I have been told they were designed by tech's on loan from ML).

I have tried many other amps and some are no doubt better and more modern but these old workhorses sound good and are available to reasonable prices on the second hand market. They were expensive new. I talked to a Crown tech a few years ago and the guy claimed that they were going EOL as Crown did not earn enough on them. Expensive to build, expensive components and over engineered for the intended use/price point. "More HiFi than installation".... I have no clue if that is correct but I like them.

He turned out to be right, a year later they were EOL (2018). I have many of them and use the 4 and 8 channel units for all active DSP systems. I have two 8-channels (CTS-8200) in the HT rack driving 3-way fonts and 2-way rears.

It is important to have the attenuators at full when using them. That is a weak link.

[sebackman]

Did some sanity measures today before the drivers goes into the cabinets.

Everything calibrated and checked against known ref.

I measured 3pcs of each and will use average for DSP filter design.

2451Be is measured on my butchered M2 waveguides after a BSS BLU50 DSP XO with LR48 @ 400Hz. MLS window is 5ms no visible reflections in window.

045's is measured in JBL K2 horns inserted into my butchered M2 waveguides after a BSS BLU50 DSP XO with LR48 @ 6kHz. MLS window is 3ms no visible reflections in window.

Not too shabby. In fact pretty amazing how good the M2 waveguide is. Very little correction needed here and that is good news.

Please bear inmind that there is no filtering or smoothing, just raw mic data.

I will do new measures in the cabinets but this is a good start.

85691
85692

[ivica]

Did some sanity measures today before the drivers goes into the cabinets.

Everything calibrated and checked against known ref.

I measured 3pcs of each and will use average for DSP filter design.

2451Be is measured on my butchered M2 waveguides after a BSS BLU50 DSP XO with LR48 @ 400Hz. MLS window is 5ms no visible reflections in window.

045's is measured in JBL K2 horns inserted into my butchered M2 waveguides after a BSS BLU50 DSP XO with LR48 @ 6kHz. MLS window is 3ms no visible reflections in window.



Hi sebackman,

Very nice drivers as can be seen.
I have not get anything like this on MY measurements, if not a kind of smoothing have not been applied.

Regards
Ivica

[engineerjoe]

Very very nice work so far. I wish I could hear them in person. Thank you for sharing all this. I will reread thru all this again and see that I didn't miss anything!

[sebackman]

Thank you for the kind words.

All diaphragms centered and measured.

A bit of a setback today. The two 1400PRO I have had on the shelf for many years waiting for this turned out to rub. They are fresh reconed so I have no idea how that is possible, but rub they do. Both of them...

I have a hard time to beleive that I put them away when I bought them without checking cone travel. Well, they can't be used as is. I have to figure out tomorrow what to do. I cannot go back to the seller after so many years.

At least the center is done. Only cloth on the front frame remains there.

The cabinet is angled 15 degrees as the center is located under the ceiling pointing down towards listening position and the screen has speaker cloth in the upper blackdrop.


85713
85711
85712

[sebackman]

A good friend came to assistance. A brand new set of 1400Pro’s. Nice. Big thank you to Mr J!

85716
85717

[srm51555]

Thats a good friend, everything is looking great.

[sebackman]

Good friends allways help :-)

Today was a good day. After +5 years the fronts are finally done, except for the cloth on the fronts.


Thank you Mr. J for the 1400PRO’s. And big thank you LH members 4313b, Widget and Guido for Be’s & UHF horn help. Also a thank you to Kalle and Johan1 for help with finding a good carpenter.

Not too shabby with the butchered M2 horns. If one did not know the originals they may have passed unnoticed. Even the 045’s (K2 9900 horns) fits reasonable good. The goal was a very compact front.

I will try to measure over the weekend. 4-way active with SUB1500 subs XO @ 80Hz, 1400PRO from 80-800Hz, 2451BE from 800-10k and 045 above. I have no idea if it works or if it will sound disaster. The M2's will have to act reference.

This is fun!

85722
85723

[Mr. Widget]

Looking really good! Very cool!!!

I assume you can flip the 045 horn and M2 horn and compare that with this configuration. I am not suggesting that you should change your orientation, but I would be curious to see how flipping it would affect the sound.


Widget

[Robh3606]

Those look great!

Rob:)

[srm51555]

They look awesome!!

[ivica]

Good friends allways help :-)

Today was a good day. After +5 years the fronts are finally done, except for the cloth on the fronts........

I will try to measure over the weekend. 4-way active with SUB1500 subs XO @ 80Hz, 1400PRO from 80-800Hz, 2451BE from 800-10k and 045 above. I have no idea if it works or if it will sound disaster. The M2's will have to act reference.

This is fun!



Hi sebackman,

Very nice work, and the results.
I wonder 800Hz VHF crossover is meay too low for M2 horn. I would suggest 1kHz or even 1.2kHz as can be seen from the 2451Be/Trx measurements.

Regards,
Ivica

[bubbleboy76]

This is so cool :)

If you get any 045s over when you are finished, I might be interested. One of mine 045Be are faulty.

[sebackman]

Short update.

Sorry bubbleboy76, no spare 045's. Check with LN.

Ivica, I have not gotten so far to figure out XO yet. The 1400PRO is clean to about +1200Hz so I might potentially increase to 1kHz XO as I have no bass in these, so very limited cone travel. I don't know if that made any difference when designing M2 but 2216 will have to generate some real cone movement to produce bass.

I have finally started to prepare for the XO and DSP correction file. The JBL M2 file does not seem complicated and in the BSS version, there are 17 PEQ points for each driver but no FIR or phase corrections. That means that the chosen XO points and the cabinet are very well integrated indeed. The delay in the M2 is somewhat surprising long at 271ms. To determine the acoustic delay in mine I just shot an MLS puls on each driver and note the flight time/distance the system gives me to a fixed point (mic) about 2 meters out and use the difference as a basis. Some very advanced tools in the form of a "setsquare" direct on the monitor was used to determine that the peaks matched. Delay was calculated at 125ms for the woofer and 427ms for the 045.

85866

A first set of data wa collected out doors with no flat surfaces within 5m, except for the ground. Hight of pole mount was just above 2m. A bit chilly at 10 degrees C, but with a lot of coffee it was fine.

85867

I order to get a reference for timbre a very competent monitor in the form of an active Genelec DSP studio monitor was collected. Very valuable to have a good ref. The Genelec is within +/- 1,5db 58Hz to 20kHz with clean phase all the way. That is not too shabby.

85868

85869

Best reagards
//Rob

[ivica]

Short update.

Sorry bubbleboy76, no spare 045's. Check with LN.

Ivica, I have not gotten so far to figure out XO yet. The 1400PRO is clean to about +1200Hz so I might potentially increase to 1kHz XO as I have no bass in these, so very limited cone travel. I don't know if that made any difference when designing M2 but 2216 will have to generate some real cone movement to produce bass.

I have finally started to prepare for the XO and DSP correction file. The JBL M2 file does not seem complicated and in the BSS version, there are 17 PEQ points for each driver but no FIR or phase corrections. That means that the chosen XO points and the cabinet are very well integrated indeed. The delay in the M2 is somewhat surprising long at 271ms. To determine the acoustic delay in mine I just shot an MLS puls on each driver and note the flight time/distance the system gives me to a fixed point (mic) about 2 meters out and use the difference as a basis. Some very advanced tools in the form of a "setsquare" direct on the monitor was used to determine that the peaks matched. Delay was calculated at 125ms for the woofer and 427ms for the 045.


A first set of data wa collected out doors with no flat surfaces within 5m, except for the ground. Hight of pole mount was just above 2m. A bit chilly at 10 degrees C, but with a lot of coffee it was fine.

I order to get a reference for timbre a very competent monitor in the form of an active Genelec DSP studio monitor was collected. Very valuable to have a good ref. The Genelec is within +/- 1,5db 58Hz to 20kHz with clean phase all the way. That is not too shabby.

Best regards
//Rob

Hi Rob,

Nice to be seen that THE PROJECT advanced.
From the point of LF bass response I would not expect any spectacular results from the mentioned drivers even with the larger EQ applied, especially in such small box. From the simulation point using over 150Lit bass-reflex box with 2216nd drivern can reproduce a kind of "usable bass response".
But using 15-inch driver over 500~600 Hz may produce a "directivity" problem, especially if such "wide dispersion" horn as M2, is applied. Using 'smaller diameter driver' would improve such "problem".
Using here 14-inch driver over 1kHz may would introduce , such mentioned "directivity" issue. But as can be seen on JBL EON 615 solution a kind of "slot dispersion bass cover" would be helpful, but 'scalled' to the bass driver size

https://reconingspeakers.com/product/jbl-eon-615-woofer-cover-plate-5039904/

I believe that 'pure rectanglar' cover round 8-inch width can sovle "the problem".

Regards
Ivica

[sebackman]

The curve for the 1400PRO, even in such small cabinet, is pretty encouraging. Not much bass however. I do XO at 80Hz to the subs.

Here are some measurements

1400PRO in cabinet

1. Nearfield 170ms 3mm
2. Far field 30ms 146cm outdoor
3. Far field 30ms 146cm with EQ outdoor

86215
86216
86217

[sebackman]

Here are some measurements for the 2451Be's in cabinet on my butchered M2 horns.

No smoothing and please check scale change.


1. Far field 6ms 154cm outdoor
2. Far field 6ms 154cm with EQ outdoor


86218
86219

[sebackman]

Here are some measurements of the 045 on the K2 S9900 horn incorporated into my butchered M2 horns.

Also attached is a system plot on the full system.

All these curves are taken outdoors on a 200cm speaker stand with no reflective surfaces within 5m except a sloping grass area 200cm below.


1. Far field 3ms 145cm outdoor
2. Far field 3ms 145cm with minimal EQ outdoor
3. Far field full system 10ms 157cm outdoor with EQ & LR24 @ 800Hz and 13kHz

86220
86221
86222

[sebackman]

Here are a comparison between the old and the new. Slightly bigger but still acceptable by the Government.

86223
86224

[sebackman]

The room is about 5x8m and pretty decent when it coems to modes. 2x JBL SUB1500 subs do low end duty, one at the font of the room and one at the back.

Electronics is still same old fully digital rig with Meridian pre, BSS DSP and Crown CTS amps. Soon to be turned into a museum piece….

Drivers has no protection cap but a 8db attenuator. 045's are rather low sensitivity so they have no attenuator but are protected by a cap as they probably would react negatively on any voltage spike.

This summer I will devote a few days to re-measure and se If I can get the XO's more seamless by playing with the delays and potentially fiddle little with the EQ points.

86225
86226

[bubbleboy76]

Really cool. Nice work.
Nice measurements as well.
I would angle the L and R more inwards to reduce SBIR from sidewalls a bit, but that's just me.

[sebackman]

Hi bubbleboy76,

Thank you.

I agree, I would turn them in more if possible.

The sides are both angled both inward and downward as much as the wall mount allows, please see better picture attached.

The angles (down and in) are pretty spot on to where we sit.

Wall reflektions is far from optimal but the M2 WG is pretty nice off axis, better then most alternatives so it is not perceived very sonically disturbing. Maybe I will add some some kind of foam there later.

I can't really move them more inwards either, with this size screen.

Compromises, compromises.

The 045's blend very well with the Be's and allows for limited EQ on the Be drivers. I have no idea if it would have been better to put the 045 horns at the top as in the K2's, or if the centre position is equally good.

I have not seen any documentation outlining which placement is preferred. Probably the top mount as this is what JBL does in the K2’s. But that could also be driven by aesthetics. For me it was a matter fitting into a small cabinet.

Fronts are already made and hopefully I can get the cloth on this weekend. They attach with resessed magnets.


I hope to be able to place the M2's better in the 2-channel room.


86227

[bubbleboy76]

Maybe you can try some temporary thick damping on the walls close to the speaker, just to see what happens.
I would gain from it myself I think, but I can not do it due to WAF, so I don´t want to try it. I don´t want to know what I am missing...

I think you placed the UHF-horns perfectly.

[srm51555]

They look great and thank you for sharing all of your work. They make for great reference as well.

[Mr. Widget]

Yes, very nicely done! Thanks for sharing!


Widget

[Titanium Dome]

The 045's blend very well with the Be's and allows for limited EQ on the Be drivers. I have no idea if it would have been better to put the 045 horns at the top as in the K2's, or if the centre position is equally good.

I have not seen any documentation outlining which placement is preferred. Probably the top mount as this is what JBL does in the K2’s. But that could also be driven by aesthetics. For me it was a matter fitting into a small cabinet.



I think your location of the 045s is fine given your mounting position. Traditionally, the design aesthetic is to put the 045 on top, as you noted. Greg Timbers did the same thing on his personal, custom built Arrays which I've dubbed the Timbers Arrays since he sold them to me. Our discussion of the Be HF and UHF drivers lasted about 45 seconds.

Someone (4313B I believe) posted a photo of the speakers in Greg's home with the 045 Be drivers disconnected. When I picked them up, the 045s were solidly connected and sported some new networks. When I asked if they were really needed, he commented that he felt "a little something" was missing and the new boards and 045s "gave it back" as long as the UHF drivers were within +/-30 horizontal degrees of the seated listener's ear height.

It seems like your decision is a fortuitous one, since having the 045 driver at the top of the cabinet could be significantly outside the 30 degree range.

Your work looks great!

[sebackman]

Hi all,

I know this is an old thread but did realized that I have not updated this for some time and some things have changed.

Thank you for all the nice and supportive comments.

The system is fine, however I never got the PRO1400 to be happy in the closed cabinets so even with DSP and ample of power they never really produced as good as they do in other cabinets.

What to do? Bring out the router again!!

I ported the cabinets on the rear side, no room on the front, with two 3" ports. Even if the cabinets are really too small it still works surprisingly good with porting. I also could take down the sub XO from 85-ish to 65-ish which made the blend better.

Lesson learned, PRO1400 does not like small closed cabinets and DSP/power can't fix it. I should have known...

90090
90091

There was no space for two reasonable closely placed ports on the center so it only got one port on the side. XO 75-ish. Works decent for dialog on films

The PRO1400 is really a nice driver and the sound is very much like the 2216Nd (I have both), apart from the lack of low bass. A bit odd that not more DIY'ers use them more often.

The Be diaphragms are really sweet but probably does not warrant the additional cost, including 045's, compared to ordinary SL diaphragms and a good DSP curve.

The placing of the 045's was probably sheer luck. They did not fit on the top side but in reality the listener cannot distinguish between the 045's and the 4" Be's. They are one. Even if Xo is lowered.

I just did a pair of Array 4890 for a friend and swapped the old alu horn for a modern waveguide (VTX). They sound wonderful with a good DSP curve. When compared to my Genelec refs they have the same sound picture but more dynamics, more detail and can play stupid loud. We added some 2268H based subs to them. He was happy!! Not sure he will ever return to his day job, he may be stuck in his couch with Tidal as the only app on his iPad :-) .

Stay safe everyone.

Have a nice weekend, when it arrives

//Rob

[srm51555]

Thanks for sharing Rob, we always like to follow your projects.