Print Page | Close Window

BMS 4592ND performance on a 250Hz tractrix

Printed From: Speakerplans.com
Category: General
Forum Name: Feedback Forum
Forum Description: How Your Projects Perform
URL: https://forum.speakerplans.com/forum_posts.asp?TID=85541
Printed Date: 28 March 2024 at 9:54pm
Software Version: Web Wiz Forums 12.06 - https://www.webwizforums.com


Topic: BMS 4592ND performance on a 250Hz tractrix
Posted By: csg
Subject: BMS 4592ND performance on a 250Hz tractrix
Date Posted: 28 February 2014 at 9:58pm
Ive been messing around with several new flares recently with a BMS 4592ND as the mid / high end of my new large format mid high box.

I have never been convinced by the sound of constant directivity flares. Radials / bi-radials better on axis, but worse off.

i took a bit of a punt and ordered a couple of 250Hz flare rate Tractrix flares from Stereolab in Germany with a view that a 250Hz tractix should be able to support a 400 Hz isn high pass, LR 4th order.

straight out of the box, no EQ, just the described high pass gave the response below, taken on a type 1 spectrum analyser. very impressed indeed.

http://s264.photobucket.com/user/csgal14tp/media/IMG_4319_zps7706e4cc.jpg.html" rel="nofollow">


the off axis response i heard this afternoon surprised me though - i thought these would beam like hell high up, but they seem to give a fairly satisfactory 80 degrees or so.

http://s264.photobucket.com/user/csgal14tp/media/IMG_4322_zpse6c53191.jpg.html" rel="nofollow">
http://s264.photobucket.com/user/csgal14tp/media/IMG_4320_zps143e88b7.jpg.html" rel="nofollow">

the sound, compared with an electrovoice HP940 and a JBL 2385 is in another league. Early days in this design, but a very promising start.


-------------
“The fact is this is about identifying what we do best and finding more ways of doing less of it better”



Replies:
Posted By: lycantheleopard
Date Posted: 28 February 2014 at 10:37pm
thats pretty sweet! i'm glad its for a "large format" box though, i bet that horn is quite sizable.


Posted By: csg
Date Posted: 28 February 2014 at 11:32pm
18" / 450mm diameter at the front. You need a big horn to do this driver justice though i think, i want as much of the vocal range coming from the coax as possible.

-------------
“The fact is this is about identifying what we do best and finding more ways of doing less of it better”


Posted By: Keen
Date Posted: 01 March 2014 at 12:06am
awesome stuff csg. thanks for showing us. all the best with the project Thumbs Up


Posted By: y2stevo
Date Posted: 01 March 2014 at 1:57am
Nice,
What LF driver are you thinkg on pairing with?
A single or even dual 15 arrangement?


Posted By: matty w
Date Posted: 01 March 2014 at 4:16am
Everything about this combo just excites me ,just makes a lot of sense .
Response looks very promising , thanks for sharing

-------------
Black to black , red to red , blue to bits ....



Posted By: matty w
Date Posted: 01 March 2014 at 4:22am
How stiff is the horn ? With it being quite large plus the intention to use that bms driver nice and low ,as it should be ,you mite want to consider fibreglassing strengthening strips up the length of it !? Also maybe some tar damping around the outside ,all to help keep it sounding natural and not like a fiberglass/plastic ? Horn!

-------------
Black to black , red to red , blue to bits ....



Posted By: SMP
Date Posted: 01 March 2014 at 4:38am
I hope the updates will keep coming as the BMS is/are of particular interest at present. your results look very promising
nice work


Posted By: Teunos
Date Posted: 01 March 2014 at 8:36am
Is there any possibility you could do a directivity plot?

-------------
Best regards,
Teun.


Posted By: csg
Date Posted: 01 March 2014 at 9:53am
I will do some readings at angles once i have access to a suitable space. Anechoic spaces are rare round here, but i do have access to an unused airfield in a very quiet location, just need a dead calm day.

the horn itself being from hifi land is very dead, made from 1cm thick Corian. The makers pride themselves on the horns being non resonant and well damped. Stiffer than fibreglass, much more dead than anything metal.

the only re-enforcement i would in visage is support for the comp driver. Its not heavy, but the worst could happen going in or out of a truck. Someone managed to shear all 4 wheels of a flight case last year coming down a truck ramp.

there will be 2 12" drivers below this covering the 100 - 400 Hz band. I am hoping that i can avoid any significant horn loading here - 2 reasons being i want these to be reasonably wide dispersion, good for one box a side applications, and to avoid any colouration of the sound. The horn needed to give any decent gain in that passband would need to be fairly long too, and i want to keep the box to around 700mm deep. If the 12" drivers can cope in terms of output then great, and a reflex load just around the bottom cross freq. should help keep the efficiency up.


-------------
“The fact is this is about identifying what we do best and finding more ways of doing less of it better”


Posted By: login4
Date Posted: 01 March 2014 at 11:05am
just look at tannoy VQ range, huge horn Wink

nice project, look forward to the results



-------------
CELTIC SUBSONIC SOUND SYSTEMS


Posted By: Teunos
Date Posted: 03 June 2014 at 6:42pm
Any progress on these CSG? would love to the cab these ended up in Tongue

-------------
Best regards,
Teun.


Posted By: ShokZi
Date Posted: 17 June 2014 at 7:19pm
im also interested as i had my eye on those horns but couldnt find enough data on directivity to take such a leap of faith


Posted By: DJ-Dulux
Date Posted: 23 October 2015 at 2:02pm
Any updates, looks like a really interesting project....

Dupe...

-------------
Dupe...


Posted By: Sapro2
Date Posted: 23 October 2015 at 9:52pm
Agreed, i looked up Stereolab and all i found was a band. I would love more info, they look beautiful, good horns tend to lok right and hat is a corker. I am a fan of tractrix my lower mid horns are based on the tractrix curve.

Do they have a website?

-------------
Splat Soundsystem
Baby Sham pram Soundsystem
Sapro - SoundCloud
https://soundcloud.com/sapro
DJ Sapro. West country free party DJ and Producer.


Posted By: DJ-Dulux
Date Posted: 23 October 2015 at 10:24pm
Website is http://stereo-lab.de/2-inch-cf-190-Hz-horn-spherical-wave-tractrix-horn-wave-guide-lense-tromba-pavillon" rel="nofollow - here.

Dupe...

-------------
Dupe...


Posted By: odc04r
Date Posted: 24 October 2015 at 9:09am
Lovely horns, fairly pricey but I bet the quality is good. Bet they could be great for HiFi customs if you had a nice 1"/1.4" driver sitting around


Posted By: DJ-Dulux
Date Posted: 25 April 2016 at 4:26pm
Did you finish your work on these, I am wanting to build a bigger 12V rig over the next year and these may do the trick?

Dupe...


-------------
Dupe...


Posted By: SouthwestCNC
Date Posted: 05 March 2017 at 11:57pm
Wow Corian, nice inert material and impressive thermoforming. 10mm that is interesting it's only available in 9 or 12 which likely means that 2mm is stretched out in the thermoforming process. No easy task. Problem with it is its brittle, a good impact and it shatters, it also scratches very easily. Not sure how well it would fair in a touring application. Need to be boxed for certain.


Posted By: csg
Date Posted: 06 March 2017 at 7:53am
Unfortunately, whilst on axis sound was superb, these, and all other tractrix flares i have played with ( 18 sound  and a couple of unbranded ones i picked up somewhere) suffered from too much HF beaming, giving an inconsistent sound and making splaying multiple boxes impossible.

With flares, you cant beat the physics unfortunately - and i had to compromise and use large CD flares that whilst ultimately sounded poorer on axis were array-able and  far more consistent across their coverage area giving a solution which was workable.

For me, tractrix flares have one use - hifi where you have one focused listening position and need very high sensitivity through the use of very low powered valve amps, and thats about it.


-------------
“The fact is this is about identifying what we do best and finding more ways of doing less of it better”


Posted By: toastyghost
Date Posted: 06 March 2017 at 10:35am
Chris did you try the 4594Nd on the 18sound XT? With FIR filtering applied to correct some of the dispersion anomalies it's damn good - but as you say it's not ideal for more than two per side

Nor is it cheap!


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 2:32pm
18sound claims about the xt being a cd horn are bogus, Here is a sim we done of it.

Horizontal
https://postimg.org/image/bkg5g1djp/" rel="nofollow">

How did the fully round tractrix compare? id imagine it probably beamed more.


Posted By: toastyghost
Date Posted: 06 March 2017 at 3:22pm
Well there must be an issue with that sim because they provide their own isobars, polars, DI and so on here:
http://www.eighteensound.com/Portals/0/PDFs/XT1464.PDF" rel="nofollow - http://www.eighteensound.com/Portals/0/PDFs/XT1464.PDF


Posted By: DJ-Dulux
Date Posted: 06 March 2017 at 3:32pm
Really difficult to compare as the scales (dB and frequency) are so different?
 
Dupe...


-------------
Dupe...


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 5:00pm
The sim is slightly different, the 18sound sims look very smoothed, you will also notice my sim goes to 20khz the 18sound is zoomed in and starts at 500 and only goes to 10k. Stretched and zoomed in to that range they are very close. Outside of that range is where the issues lie.

There is also the fact that this sim is a scaled up 2" version, the 1.4" sim is almost identical tho.


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 5:03pm
Originally posted by DJ-Dulux DJ-Dulux wrote:


Really difficult to compare as the scales (dB and frequency) are so different?
 
Dupe...


The log10 frequency are exponents you just need to work them out.


Posted By: toastyghost
Date Posted: 06 March 2017 at 5:34pm
Originally posted by SouthwestCNC SouthwestCNC wrote:

The sim is slightly different, the 18sound sims look very smoothed, you will also notice my sim goes to 20khz the 18sound is zoomed in and starts at 500 and only goes to 10k. Stretched and zoomed in to that range they are very close. Outside of that range is where the issues lie.

There is also the fact that this sim is a scaled up 2" version, the 1.4" sim is almost identical tho.


They do only advertise constant coverage (not directivity) from 1.5kHz upwards but it is most common for isobars to be smoothed, often 1/3rd oct with colours stepped for 3dB divisions.

If you're comparing then it's best to keep the parameters the same.


Posted By: DJ-Dulux
Date Posted: 06 March 2017 at 6:41pm
Didn't notice the frequency scale, odd way of doing it? Teach me to read the axis label. Will have a better look later....

Dupe...


-------------
Dupe...


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 9:00pm
Takes a lot of processing power and memory (edit: ram) (60gb plus) to go up to 20k in comsol, may be that they didn't have it at the time.

What would your thoughts be on using a tractrix to say 7k below were beaming starts to occur with a non coaxial mf compression driver in addition with a cd flare to handle the high frequencies.


Posted By: toastyghost
Date Posted: 06 March 2017 at 9:26pm
Originally posted by SouthwestCNC SouthwestCNC wrote:

Takes a lot of processing power and memory (edit: ram) (60gb plus) to go up to 20k in comsol, may be that they didn't have it at the time.

What would your thoughts be on using a tractrix to say 7k below were beaming starts to occur with a non coaxial mf compression driver in addition with a cd flare to handle the high frequencies.


Those are measured isobars not simulations.

http://fouraudio.com/en/products/elf.html" rel="nofollow - http://fouraudio.com/en/products/elf.html most likely


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 9:33pm
Could well be. That's incouraging if so as comsol is very close at 500-10k. Then I ask why have they stopped at 10k? But proofs in the pudding we know it beams as csg experienced, if it's a measurement they have stopped at 10k for a reason.

But like you say they claim constant coverage not directivity. I was under the impression that was the same thing but obviously not, what does constant coverage mean?


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 9:41pm
Infact how can it be a measurement, a measurement is a driver and horn combined which would be invalid results for the pressure response of the horn alone.


Posted By: toastyghost
Date Posted: 06 March 2017 at 9:53pm
It suggests that measurements were made using their ND1480 driver on the horn in the PDF. Also as I read it the 10kHz mark on the isobars is not at the far right of the diagram? So it shows response above that - likely up to the 12.5kHz they claim directivity is maintained up to.

Look at isobars / directivity info for most manufacturers, you'll see they all tend to crop to show the 'relevant' parts. It's fairly rare to see data below 400Hz.


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 9:57pm
Ahh missed that bit, well spotted. Yes I understand there's no point in plotting outside of specified response but even at 12k they are 8k off the top end. Which to me says the recommended response is 500-12k


Posted By: toastyghost
Date Posted: 06 March 2017 at 10:00pm
Originally posted by SouthwestCNC SouthwestCNC wrote:

Ahh missed that bit, well spotted. Yes I understand there's no point in plotting outside of specified response but even at 12k they are 8k off the top end. Which to me says the recommended response is 500-12k


Well yeah, that's basically what they say…



Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 10:07pm
12k - 18k quite a large gap still in there dispersion plot. I think it starts beaming well before 18k Even so this a big implication as to the use of this horn. It really suggests it needs a hf horn on top.


Posted By: toastyghost
Date Posted: 06 March 2017 at 10:14pm
Here's a measurement somebody did of an Eminence N314T on the XT1464:
http://www.avsforum.com/forum/155-diy-speakers-subs/2008882-eminence-n314t-xt1464.html" rel="nofollow - http://www.avsforum.com/forum/155-diy-speakers-subs/2008882-eminence-n314t-xt1464.html

They do off axis too.

I think you'll find quite a lot of horns beam in the upper ranges. For example, VDOSC only claimed to maintain published 90 degree wide dispersion figures 630-12.5kHz


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 10:26pm
Yeah 10k being the point of no return. I'm just learning about speaker design so this is all a big learning curve for me, so big thanks for you input.

What would be the implications of crossing the xt at 10k say using a bms 4591 (2" version xt) To a hf 60/40 horn? TIA


Posted By: toastyghost
Date Posted: 06 March 2017 at 10:44pm
I think you're overestimating what is needed up that high, cross reference to many boxes if you can find any measure responses on or off axis that you think have nice HF. Perhaps go through Production Partner or download some CLF/GLL files. Or load some boxes into manufacturer simulation software such as MAPP.

Then also factor in air absorption and you might be surprised how little top end there really is on the average show.



Posted By: toastyghost
Date Posted: 06 March 2017 at 10:48pm


Here's a few to save you time. Note that the last few are array cabinets and have presets with boosted HF response in the associate preset to account for mid/lf summation of multiple boxes / get more HF out to the far field.


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 10:50pm
Ok that makes sense but having ascertained that anything over 10k has a 20/20 angle makes it pretty useless over 10k so 10k-20k required.


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 10:55pm
Yeah that is helpful advice , Ty.

Now need to find either a round or eliptical cd horn for 10k-20k. Square cd flare over an eliptical xt would look awful. Another hurdle.


Posted By: toastyghost
Date Posted: 06 March 2017 at 11:02pm
You might also have a lot of fun trying to get them to sum nicely without resorting to not so nice sounding super steep filters.

There's a lot of trade offs - the XT1464 might not look so great on paper but in practice it's probably the best off the shelf horn I've heard. And is right up there in the pro market when using the right driver too.

Here's a real world on axis REW measurement of BMS 4594Nd on one with a dual 12" LF section:


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 11:13pm
The intended application of all my research is for pretty large systems hence upscaling to 2". 8 hog stack for instance at the moment, The response is very good it's the dispersion over a large size system that I think it's going to get into trouble. If I can use it for the frequencies that have the 6040 pattern only with a separate horn For more cd and still get the xt sound quality even if it's only in the mf it would be a good trade off


Posted By: SouthwestCNC
Date Posted: 06 March 2017 at 11:16pm
Yeah the response is awesome, 12" not applicable for my purpose either as flh mid kicks go upto 370 just to complicates matters


Posted By: corell
Date Posted: 07 March 2017 at 12:06pm
Very detailed information about the XT 1464 can be found here :
http://lambda-labs.com/de/#/tx-3a" rel="nofollow - http://lambda-labs.com/de/#/tx-3a (tools4music test download)
and here :
hr-systems.biz/downloads/PAF-212-Dokumentation.pdf

both were done by the same person btw. the second one probably is the best documentation of any DIY top there is. Even if its not that fun to use google translator, i would suggest to read the whole article!


Posted By: bob4
Date Posted: 07 March 2017 at 1:46pm
Originally posted by corell corell wrote:


hr-systems.biz/downloads/PAF-212-Dokumentation.pdf


Any idea if cnc flat packs are available for this one? The online shop carries  a driver set, and protective plastic covers  but no flatpacks Ouch


Posted By: corell
Date Posted: 07 March 2017 at 2:20pm
look at h audio :)
oh i realise thats probably the online shop you meant, sorry. They had a CNC flat pack once, but no i cant find it anymore... i will ask fabian directly!


Posted By: bob4
Date Posted: 07 March 2017 at 3:33pm
Originally posted by corell corell wrote:

look at h audio :)
oh i realise thats probably the online shop you meant, sorry. They had a CNC flat pack once, but no i cant find it anymore... i will ask fabian directly!

thank you corell, that's very kind of you!

Clap

EDIT: can't find a crossover for the PAF212 either, neither pre-assembled nor in kit form Ouch

I suspect it might have not sold very well at the high (but justified!) price point. 


Posted By: corell
Date Posted: 08 March 2017 at 11:39am
Yeah, they didn't sell as much as you would have expected from such a professional design. It was the PA-Forum community projekt years ago, before Jobst took over the german DIY market.
Quite a few people disliked the very complicated passive x over (including allpass, notch filters etc) in these days of ease digital processing aswell. "Bauteilegrab" they'd call it ;) The passive alone is ~200€ per piece. Only sold preassembled.
I know Fabian personally, he also did the measurements of my own speakers, im 100% sure it is possible to get the x-over and building plans still, not so sure about CNC flatpacks if they took it of the shop.
Also, i have 2 new ND1460 lying around i might be willing to sell ;)


Posted By: bob4
Date Posted: 16 March 2017 at 2:38pm
thank you corell!! Thumbs Up



Print Page | Close Window

Forum Software by Web Wiz Forums® version 12.06 - https://www.webwizforums.com
Copyright ©2001-2023 Web Wiz Ltd. - https://www.webwiz.net