153 replies to this topic

[Paul Spencer]

My next mains still in design stages will be a coaxial hybrid horn/waveguide arrangement. It is based on the Danely "Synergy horn" which is actually a range of speakers, they are all something like this:

[IMG]http://www.google.com.au/imgres?q=synergy+horn&hl=en&biw=1280&bih=582&gbv=2&tbm=isch&tbnid=G082h3WQxeOcuM:&imgrefurl=http://www.diyaudio.com/forums/multi-way/156988-gedlee-summa-vs-lambda-unity-horn.html&docid=Gc63eWxwiBFL1M&w=369&h=371&ei=ZB1KTqr5AobgiALNw5y4Bw&zoom=1[/IMG][img]http://2.bp.blogspot.com/-fuzjGZgsbAk/TZVR4YC0zfI/AAAAAAAAAc8/YbX3U9M-qos/s1600/SPL-td1_rear_Fullsize.jpg[/img]

I've written about it on the blog:
http://redspade-audio.blogspot.com/2011/03/synergy-horn.html

Why this one? I'm after a couple of things that don't easily come together in the same speaker:

• suits a built in studio soffit style install
  
• very high efficiency ~100 db 1w1m
  
• controlled directivity
  
• low distortion
  
• extreme dynamics
  
• phase coherent
  
• point source with sharp image focus and a big sweet spot
  
• drivers operated well away from breakup range

No other speaker can do all that. Direct radiators use large drivers to get efficiency up and they are all offset a fair amount in their vertical spacings - they fall short on most of those goals. Horns lack the coherence but do some of those things very well. FR drivers have the coherence but have issues with directivity, breakup. Don't get me wrong, I'm not saying the alternatives aren't valid, or that this is the best speaker in the world, just that it seems to meet all my goals without compromising any of them. I had intended to build something like a Gedlee Summa which uses a 15" pricey pro driver up to 1k. Instead I'll be using 4 x 5" drivers per side - much easier to get it to work.

I was stalling on this project as I couldn't figure out how to simulate it. Finally I realised I could do it with Hornresp quite easily:

[img]http://4.bp.blogspot.com/-xaco-dJ3V2s/Tj9eAw86vlI/AAAAAAAAAnE/RRdTSX1Lltw/s1600/hornresp-misco-vs-pyle.gif[/img]

So it will feature:

B&C DE250 compression driver at the apex of the horn/waveguide
4 x 5" closed back mids (250 - 1k)

Then it will cross to my "Black widow" active woofers (current project) and dual Rythmik subs under floor tapped horns (future project).

So lots of audio projects happening at the moment! I'll be having some GTGs to show them off when show worthy, hopefully when the weather warms up and we hit the season for BBQs and GTGs.

Edited by Paul Spencer, 16 August 2011 - 05:56 PM.

[mondie]

Wow. That is insane, looking forward to following your progress on this one Paul.

[Drizt]

Wow. That is insane, looking forward to following your progress on this one Paul.

Agreed, on both accounts.

[Frizzo]

I like it!

[caddisgeek]

Agreed, on both accounts.

I agree with your argreement

[Antripodean]

Paul, great to see it is taking shape. Seems like you have decided on the Pyle drivers. My understanding from the DSL team is that the mids need to be damped down more for home use to achieve a flatter FR across a broader range. Hence, I think you will need all the midrange sensitivity you have. I will be good to see your akabak script for this one
I'm just wondering why you say "coaxial" as there are no coaxial drivers used in the build?

[Paul Spencer]

The Synergy is a "coax horn" - I wanted a more generic term as not everyone will know what the Synergy is. Coax could refer to the drivers it's true, but in this case it's about the horn itself.

Yep, decided on the Pyle drivers. Max predicted SPL is 120 dB, that is 10 db headroom at THX reference.

They are slightly less effective based on lower BL and higher mms. If modelled with low mms and higher BL the efficiency at the top comes up, but I will just use EQ, I'd rather it droop at the top end rather than the bottom. I've modelled in hornresp only, too impatient to learn Akabak just yet. I could spend more time coming up with an Akabak sim than it would take me to build and measure the real thing. I can also try in a prototype some different options. It's been enough waiting already!

Last night I altered the black widow port to try to tune it higher and it resisted my efforts. You find out that stuff when you build and measure it and I don't always trust the model too much.

So there might not ever be an Akabak model from me! We'll see ...

Now do you remember our nearfield measurments of your Unity? The response was dead flat, almost perfect. I don't know if that means the crossover is compensating, or that the model isn't quite right.

[mr-happy-pants]

Your endeavors never fail to impress me Paul

It will be great to catch up chew the fat at Bathurst re all your little (?) projects

[Paul Spencer]

Looking forward to it ... wow, we're almost there!

[Antripodean]

The Synergy is a "coax horn" - I wanted a more generic term as not everyone will know what the Synergy is. Coax could refer to the drivers it's true, but in this case it's about the horn itself.

Now do you remember our nearfield measurments of your Unity? The response was dead flat, almost perfect. I don't know if that means the crossover is compensating, or that the model isn't quite right.

I have to agree that there is a lot of benefit in building to see what you can achieve. Nothing like measuring the actual. As always, if you need help building just let me know

I always think of coaxial on the same axis hence why I don't see the synergy horns as coaxial, except a couple of the mid models like the SM60M. For me the idea of drivers acting as a point source due to time alignment and ensuring the driver overlap on the horn is within 1/4 wavelength is how I picture it. No matter, it's the results that count

I think the crossover on the Unity Horns, as used by William Cowan and myself, is pretty darn good Here's Cowan's in room FR unsmoothed. Excellent! [ATTACH=CONFIG]35224[/ATTACH]

[Paul Spencer]

Coaxial - depends if you are looking at the orientation of the drivers themselves or the actual outputs of the drivers. They do share a common axis in terms of their radiation. I think Danley has called them a coaxial horn somewhere.

Considering the woofers don't want to run up to 300 Hz without a fight, I might have to run the horn down lower. I can supersize it to reach 250 Hz.

The surround woofers actually sim quite well as mids in the horn as well.

[mr-happy-pants]

Damn this is getting exciting.

[56oval]

Damn this is getting exciting.

+1 ,will be very interesting .But Paul really needs to pull his finger out .

Cheers

[jaspert]

You have been talking about this for a while. I hope it won't be completed at "56oval" diy project time frame as I hope to have a listen in the near future.

[Paul Spencer]

I've been stalling because I hadn't figured out how to simulate it. Now I've sussed it out well enough to knock up a prototype, I should have a first crack working some time in the next few weeks.



I have a friend helping out with some CNC work. A Synergy horn can be relatively simple, but I'm making the throat a bit more elaborate. It starts out as a 1" circle for the compression driver at the apex. Normally it's a conical pyramid horn shape, but I'm wanting to do an oblate spheroid profile. Basically it's a conical with a radius at the start as seen in profile. With a conical horn you have a sudden transition from the narrow angle in the compression driver to the wider angle in the horn. Discontinuities are never a good thing in horns. The oblate spheroidal profile that Earl Geddes has championed seems the ideal profile out of the few constant directivity options, but it's tricky to transition from circle to square while at the same time doing an oblate profile! I'm hoping that the CNC guy can figure out the transition.

In theory I can make the mouth as wide as 1.2m if needed. Bigger horn means the mids will be loaded to a lower point, otherwise I might have a hard time getting mids to meet the woofers. I want to avoid more woofers to bridge the gap - things are already complicated enough as 3 way + subs.

[Paul Spencer]

Drivers arrived today! All in little boxes. 8 x 5" closed back mids and also 2 x 8" midwoofers for the surrounds. Ahhh the temptation to test and fire them up right now! Instead they sit by my PC to distract me all day.

[ehtcom]

Well, where's the picture?

[Paul Spencer]

This is all of them ...



... minus the other 2 compression drivers and the other 18 that are in the system. Looking forward to the weekend!

[Drizt]

Very cool. I look forward to seeing / hearing the end result.

[ehtcom]

Nice, thanks Paul

[mr-happy-pants]

Holy snapping ar$eholes Batman!

[Paul Spencer]

Looking forward to the weekend! Next step is a test box for the surrounds (the other project) and my first crack at a Synergy.

I've been learning a few new tricks. The Synergy likes to have narrow dispersion or the bottom end loses efficiency and drops off. I can compensate for that with more drivers, ie step up to 6 per side. 90 degrees seems to be a challenge, it seems to want to be a 60 degree horn. Extension is another issue. They will go to 300 Hz easily, but I'd prefer more like 200 Hz which makes the crossover to the woofers so much easier. The woofers don't really want to run up to 300 Hz, but 200 - 250 Hz is much easier. Lots of juggling going on.

[Paul Spencer]

Ok I'm no longer looking forward to the weekend. Trying to figure out the angles of this darn horn is killing me! It's a weird 3D mitre that is doing my head in! Oh the joys of geometry!

[Paul Spencer]

Finally managed to work out the mitres today and started cutting. No pics this time, but they will come when all the bits are cut.

My dad figured out the mitres by trial and error, cutting little samples and adjusting til they fitted together. I figured them out by building it in CAD, but we both came up with different angles. I came up with 14 degrees, he came up with 33! We couldn't make sense of it all, but we decided to go with what he worked out as we had the physical example that it actually worked. Well, today I figured out the answer. Dad mitred all pieces equally, but on CAD I had done it with only half of them. The result either way is the same, but the CAD version is quicker to build. Fortunately we worked that out before wasting any time. Now it will be some time before I can touch this project again, I think it's going to drive me nuts!

[Paul Spencer]

What day is today? Speaker Saturday ... let the horn building continue! Hoping to have something that makes some noise today. We'll see ...

[Paul Spencer]

Quite a bit done today.

Mounting plate for the compression driver:


All the bits cut:


Including holes, which were quite awkward!


Next up is making up a frame to help with assembling and a bit of a check to make sure everything is right. While I plan on getting the final thing done on CNC, physically building one from scratch is really good to get a feel for how it works and goes together. There are all kinds of details I didn't anticipate ahead of time. And of course, every dimension, every angle is different to what you might expect - all has to be worked out with trig. The side cuts aren't at 60 degrees, even though it's a 60 degree horn. Every edge has a specific angle, nothing is at right angles, except the ports and they had their issues as well. It all works out to be so much more complicated than expected, even though what you see is a dumbed down version!

More on the blog:
http://redspade-audi...orn-update.html

Also managed to do a few things on the surround speakers.

[56oval]

Bloody work blocks your blog & pic's .

Cheers

[Paul Spencer]

Sorry Mal, I'll have to take down the topless pics so you can view!

[zog]

good to see the project is moving along, very impressive!!

[Paul Spencer]

I just need one more full day to work on this and I should start making some noise! Didn't get to touch it today

[gainphile]

The buzzer like all power tools is a very dangerous piece of equipment. My poppa chopped off part of his index finger many years ago. When I heard about his accident as a little boy, it inspired caution with power tools.

Believe it or not I was ready to make a DIY table saw !! Lucky a quick google/youtube put me to my senses. Apparently anything with high circular motion are VERY dangerous and kickback/binding are real problems. Some 60,000 injuries in USA/year, they even have lawyers specialising in table saw!!

I'm now looking for a good quality jigsaw. Apparently they have good safety record (and hence why I still have all my fingers despite making speakers these few years) lol..

[DQ828]

Gainphile

You need to buy yourself one of these , I would if I could afford it.

Festool make the best jig saws I've ever used but once again you pay for it. Go the Bandsaw

Paul

Your a very adventurous man, I hope they sound good, bloody good. It's VERY handy having access to a CNC router. Looking forward to seeing the result.

Edited by DQ828, 04 September 2011 - 08:12 PM.

[Paul Spencer]

CNC routers are all over the place now. I've usually done things the slow way in the past. Then one day Roger came over with CNC cut bits and it all came together so fast it was a bit of an eye opener. You can have a working speaker in a weekend!

It doesn't cost much to buy a table saw, if I didn't have access to one I'd buy one. I make sure my hands never come anywhere near the saw. Only thing is if you buy large sheets then you need two people to cut them up. If you always work as a one man band then you can just get a portable circular saw. With a bit more time and effort one person can cut up a big sheet. Most of the cuts for the Synergy horn had to be done that way, except at weird angles.

Your a very adventurous man

I wanted a challenge! And I couldn't think of anything more difficult than this. The Synergy also has a few traits that suit me right now.

Believe it or not I was ready to make a DIY table saw

That's one thing I wouldn't DIY. It's a lot of work to make something decent, and you can buy them so cheap. You could spend two whole weeks on a project like that, or a whole lot of weekends! Most guys wouldn't find enough spare time in a year to do it.

[Paul Spencer]

Almost there ... shortly I'll be firing them up for the first time ... they are a bit bigger than expected!

[Drizt]

Almost there ... shortly I'll be firing them up for the first time ... they are a bit bigger than expected!

lol, they were always going to be monsters Can't wait to see the pics.

[Paul Spencer]

almost there .......

[Paul Spencer]

Everything takes ages when there are 8 midrange drivers! ... still hooking up cables

[Paul Spencer]

Now they are up and running. A few shots taken just before I set them up:

[Drizt]

How do they sound?

[Paul Spencer]

Without a proper crossover, ie just a guessed one ... LOUD! I'm working on a crossover now, the mids extend lower than expected, but only require just one broad notch filter to smooth things out. The compression driver is reasonably smooth in the horn. The black widows now have to come back in but they only have to get up to about 230 Hz, a good thing because they have early roll off and by 300 Hz lose their high efficiency. First impression, without doing it properly they don't sound as good as the surrounds, but that's no surprise. These are a lot more tricky.

[gainphile]

Very nice...!

Is it the same size as Roger's Unities? Hard to tell without reference scale.

[Paul Spencer]

These are super sized. Roger's are 465mm external, mine are 540mm so they can run a little lower. There is some room for improvement but they are performing really well. They image differently to waveguide speakers and they are stupidly sensitive. When the mids see 0.01w they kick serious butt! The Black Widow woofers need some work now ...

I'm waiting for some pro speaker stands to show, I'll build a really nice polar measurement rig. A bit like yours but mine won't double as an ironing board. I'm really curious to see how this thing measures off axis.

[gainphile]

It would be interesting how you'd rotate that thing at the mouth!

Perhaps move the mic circularly instead?

[fury]

They look fantastic Paul!
Can't wait to see measurements of these things!

[DQ828]

How's the WAF?