What horn are you going with now that the Emilars are up for sale?Originally Posted by Ian Mackenzie
What horn are you going with now that the Emilars are up for sale?
Tad
Gotta love Sid Haig.
Wow! After looking that over I just realized that I'm really not all that into the whole "audio" thing.
I guess my rather conventional, E2-inspired JBL loudspeakers will just have to do.
.Well looking Greg.s most successful designs they been anything but conventional.Wow! After looking that over I just realized that I'm really not all that into the whole "audio" thing.
I guess my rather conventional, E2-inspired JBL loudspeakers will just have to do.
My interpretation is this field belongs to esentric self obsessed artisans.
Still a work in progress.
I will post a pic in the near future.
It will go in the living room that spans 4 x 8 metres overall.
I'm excited to see some updates. This thread is one of my favorites and the amount of information you gathered is much appreciated. Thanks for your hard work.
Al
Hi Allen,
Yes it uesful as its eary to forget a lot of this stuff over time.
I am in Bali at the moment but a few words are in order.
Like all diy projects it can take a while and sometimes you end up with a pile of drivers that need a home!
So I need to get this sorted out.
After you build a new home and everything is beautiful someone like Dan Ashcroft is in high demand to make something that looks good on paper fit the decor of the home.
The living room is vast and open plan so a system with some controlled directivity might be useful.
Coming up with something that will work is mean feat and I can now see what JBL crapple with in coming up with suitable systems.
There are several approaches but the ancher (for now ) is available components , the Tad 4003, Tad 703, AE 15M, Emilar horn and possibly a Tad horn.
I also have a pair Century Gold 1200H 12 inch woofers which will be the fall back plan if the above AE15M woofer box does not work out..its needs to be 7 cu ft 3. The 1200H and a helper sub would no doubt be more WAF friendly but not nearly as much fun as a large format system.
Another option at a pinch is I could turn out a 4344 for the living area and use the project system for the home theatre.(at least I know it works
As a reality check I plan to first trial existing boxes in the living room to assess the bass response.
This will be the get-go on progressing with the AE15M woofers. They are much like the Altec 416. A big box is unfortunately a must
If this looks promising I propose to look at a direct radiator (woofer ) system and compare with a Tractrix front loaded system.
If not I will post WTB a pair of LE14's
I was reading Mr Widgets long thread on curves mdf panels and that looks scary.
So the cosmetics will be added after the working systems is prepared. ie a curved front facia and sides as require.
A bit like trowling on makeup but there you are.
I have conducted an inital trial.
Using my reference monitors manufactured by ex Linn designer Rod Crawford some insights into the new listening environment were obtained.
The idea behind this approach is that as a known base line I can gauge what the room is doing as this is as much an unknown as the new proposed system
The overall presentation was spacious, open and punchy with surprising sound stage depth.
There are not really obvious room modes room modes but the anticipated room gain below 100 hz was not overly apparent.
Some rear field effect is noticable towards the rear of the room some 25 feet back
The step will be to trial the significantly larger project boxes with the AE15M woofers and see how they fair.
I recently restored the diy 4345 in the listening space, just need the correct JBL blue paint and I will bark them up. Unfortunately they are a bit too big.
The earlier post on décor still applies......
My latest idea is a wedge shaped box not unlike a pa monitor with a curved front and sides.
Currently exploring horn and wave guide options
As far as box style is concerned i really like the L200 low boy classic design not withstand its dimensions were acoutically sub optimal.
While mostly listen in the setted position i am mindful of stand up listening so i have been pre planning with a string line to obtain the best compromise in terns of horn mouth height and baffle angle.
The ear height seated is 95cm. So i have gauged the tlit angle for centre horn height of 85cm. This gives a modest but effective lift in the horns. 30-40 degree vertical window in the far field .
The overall enclosure height will around. 95cm. This was derived emperically after accounting for hornheight driver spacing. The L200 was about 80cm in height dimension . This would have placed the botton of the woofer only 15cm off the floor and possible caused booming ro mask the midrange and a suck out in midrange due to floor bounce cancellation.
I have not yet decided if i will opt for straight side walls or truncated with a narrow rear wall . The overall width is about 60 cm. The volume of the side wall would all provsion for front horn loading of the woofer anf would break up the box modes with staight walls.
The plan would be a tractric horn or Altec from 120 hz to 600 hz.
The construction would be more challenging but woulf provide time alignment with the Tad 400 driver and the woofe.
Horn loadin the woofer also sorts out the difficult match up of horn and wooder coverage angles and defines the direct sound patten.
On reading Tom Hidley's'early studio design ideas in a aper by Philip Newell the choices are simple in terms of how to manage the amount of reflected sound wih either room treatment or controlled coverage. I dont want to intall deliberate room treatments
So control of coverage angle is useful where the system is targettinf more of a near field listening pattern.
If you look at the attachment Blue Thunder that is the idea.
It's been a while since my last post and like other members projects....sometimes life gets in the way.
Over the past 6months l have been evaluating system configuration that would perform appropriately but also blend cosmetically in the listening room.
This is a very challenging quest and is a labour of love no doubt for the industrial design team at JBL.
I have looked at Tad horn clones, SEOS wave guides.and while these horns have some good attributes the wide of the horn make the enclosure look too much like a big box.
So, l have tacked the box dimension on two fronts, sourcing of a different driver and a new norm/waveguide.
The selection has come down to the JBL 2216 Neo driver that requires 60 litres less box volume than the AE15M @130 litres and the AH700 conical horn (no longer in production) that has a 40cm diameter and is some 20cm deep.
The room classic styling with leather furnishings and oak floor lend to box design not unlike the Altec Iconic but scaled for the 130 litre enclosure size.
The horn will be mounted on an adjustable pivot to enable critical adjustment of the driver angle and displacement relative to the woofer.
Some images below of raw horn images and FR curves.
It's not a text book perfect world but with some careful eq of the off axis power response excellent subjective performance rivalling most two way systems can be obtained.
The actual test curves will be published soon enough.
Okay,
Over the holidays l have been able to finalise the prototype crossover.
The woofer is a piece of cake and there is box diffraction issue in the test box that l will ignore for now.
The conical horn took some understanding. It is true CD and when equalised extends to about 18 Khetz.
However the power response at 0, 5, 10, 15, 20 and 25 degrees varies somewhat and is hot in the 2-3 k hertz area on axis until about 15degrees off axis.
After a lengthy read of all the Harman papers by Floyd l reasoned the controlled directivity and off axis power response was more important than the 0-10 on axis response from the horn. Floyd says only equalise resonances, not diffraction and shoot for as smooth as possible response from 700-7000 hertz.
So the next step was to find the best fit for the off axis power response of the horn into the listening room.
Reviewing the averaged equalised power response 15 degrees off axis was the best overall compromise.
Based on this decision l set about to design for a flat response at 15 degrees off axis to the centre axis of the horn.
This will require the listening height to be 320 mm below the centre axis in the vertical plane of the horn at a distance of 2 metres. The seating is a bit more like 3 metres but the variation in the angle along this long axis is not significant.
I ended up with very little passive equalisation of the 2-3 khertz hump and rather than fight the horn on axis response with EQ l am using the off axis response to the advantage of the listening window.
Physically the driver layout will have the horn elevated above the seated listening height with the on axis response going over the head of the listener into the far field of the room.
I figure the far field some 5-8 metres back will have so much reflected sound the on axis hump at 2-3 khertz will not be an issue.
There were 3 areas of passive equalisation l decided to focus on in the end :2.5 k hertz, 4.5 khertz and 9 k hertz. This was to level the response overall.
I used a series LCR circuit to equalise the horn response overall and borrowed the topology from some of JBL recent designs. The low value of the Fixed LPad shunt resister provides low voltage drive ripple at the low end of the passband (unlike the 4430/4435 networks).
I will incorporate a variable L pad into the LCR to provide an HF adjustment. An LPad will also be inserted in the equalisation for fine tuning presence of the 2.5 Khetz region.
Structurally the horn will elevated some 320 mm above the listening height and the horn tlitled at 5 degrees back.
The bottom of the horn will be some 100 mm proud of the top surface for cosmetic reasons and to minimise enclosure diffraction.
The horn and woofer centre is 500 mm apart.
The bass reflex enclosure is conventional and 130 litres in volume and 102 cm in height.
The bass response can be boosted per DB Keele's alignment to improve low end extension and displacement limited output.
The overall approach of a smooth off axis response may be against conventional wisdom but if it's assumed the listener will be at about 2-3 meters from the loudspeaker the above approach will yield the best power response tracking on axis to the listening window per the Harman paper.
The curves are self explanatory but the crossover requires some description.
The crossover is an hybrid individual active and passive cascaded 3 db networks that when combined are LR12 CB network on the high pass section The woofer is active 12 db LR.
I spent a lot of time getting the phase in the crossover right and with careful adjustment the notch is some 30 db below the reference level.
The lateral displacement of the driver acoustic centres, the driver phase response and crossover phase response all impact on the phase optimisation in the crossover region.
I was able to optimise the phase null at the crossover point using LEAP.
The system could be described as phase compensated in the crossover design.
Fine adjustment of the crossover point spreads of the woofer and horn and the lateral displacement can be evaluated with LEAP.
Woofer acoustic voice coil= 115mm
Horn acoustic voice coil = 280mm
(Referenced the front baffle)
The distance from the horn mouth flange to the driver vc is about 275mm so the horn assembly can be conveniently located above the woofer enclosure.
As a result the summation is in fact dead flat at 0 degrees (320 mm below the horn) through the crossover region.
I also spent a lot of time investigating the vertical polar nulls.
The nulls above and below the central lobe are normal (in the crossover region).
I don't think this aspect of the design has precedence over other aspects including flat response and a reasonable match of the woofer and horn polar response at the crossover point. The horn polar response is 60 degrees and unloads at the lower end. The crossover point is 800 hertz. This will be evaluated with pink noise once the system is assembled.
After a while l was able to come up with a polar response lobe that favours the bottom half of the horn axis through the crossover region. When the driver phase is reversed the true frontal lobe is focused on axis.
Of course it remains to be seen what happens in practise but l am confident that the LEAP sims and LMS ground plane measurement are good.
The drivers are Tad4003! AE 15M (JBL 2216 to replace the AE15M)
From top to bottom
Vertical lobe in phase
Vertical lobe out of phase
On axis equalised response (at listening position 2-3 metres)
Network schematic
Voltage drives
These curves are 5 db per division. I found 5 DB divisions useful for ironing out the bumps but l prefer 10 divisions for looking at the flatness and levelling of the overall response.
I don't think the notch at 2k hertz is a major issue, it's not the driver but the crossover artefact. Much steeper slopes would be required to minimise this.
I have combined the driver guide curves and the system curves below to illustrate the deviation in the power response.
All ground plane measurements were done in my drive way.
From my recollection using the 4430 crossover it had a shelf L pad around 2 Khetz and the effect is to warm brass instruments. I generally had the LPad up full do its not an unpleasant effect. I may create a adjustment for this once the system is up and going.
I have not listened to the most recent iteration of the crossover but using a simplified all passive crossover it was quite promising. Low level details are early discernible and you could really hear the recording environment.
These images are all via the I Pad Mini camera as l did not have access to upload via the desktop server.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Reply With Quote