5 way horn speaker system project - tapped, bass, mids and tweeter passive active

[Rob Dingen]

The biggest problem is that it's not 12dB or 24dB acoustic only electric so you never get it spot on.
Maybe you need 6dB combined with the driver fall off to achieve 12dB.
You can do measurements so export them as a text file and measure the impedance and load them in Vituixcad and play around.

Rob

 

[Speedysteve7]

That's sounds interesting / is it another rabbit hole?
I won't know until I try 🙂

Next time I'm bored with things..

For now I'm very happy with the changes I've most recently made and am enjoying them.

 

[Rewind]

I always felt my RAAL needed steeper slopes and the midrange compression driver just needed 2nd order bandpass. My JA6681B drops like a stone below 450-500Hz in the Iwata 300, and above 12000Hz. But 12000 Hz is bit too high for it, if I am being critical, and there are better tweeters. The Raal is super clean, but lacks some qualities of the compression drivers, like «fullness». But no compression driver can match the airy highs of the Raal drivers.

 

[Speedysteve7]

The biggest problem is that it's not 12dB or 24dB acoustic only electric so you never get it spot on.
Maybe you need 6dB combined with the driver fall off to achieve 12dB.
You can do measurements so export them as a text file and measure the impedance and load them in Vituixcad and play around.

Rob

Thanks for this insight Rob.
I took my time to digest it.

Back in my time aligning only days, I optimised (hehe I thought), things using tried and tested, (by others) crossovers.
This meant a 3uF cap on the Vitavox S2 channel and a 1uF cap on the Raal Lazy Ribbon tweeters.
(I had tried other high pass X/O slopes on the S2s, reversing the polarity as required for 2nd order, or not for 4th).

Moreover, due to only having an 8 output Najda DSP X/O box, I left the high frequency open on the S2s, and as you see, they naturally die off - much faster than 1st order.
The tweeters were then piggy backed onto this open upper frequency output and crossed using a 1uF cap. I did try 2nd order briefly but could not say it was better sounding.

This does give rise to the phase alignment dialling in problems you pointed out.
The slopes are off.

I've recently started chaining a 2nd Najda DSP unit to do the tweeter X/O in DSP instead, and when measuring I found the slope that best matches and allows nice phase plot alignment is actually 8th order.
The S2s die off that quickly!

Why do I persist in using a 3uF cap for high pass on the S2s, even though the signal is going through DSP and is time delayed etc there?
Well, the S2s just sound better on the cap than the DSP doing the X/O. I can set the DSP X/O up as close as possible to copy, and it measures very similarly, but I always go back to the 3uF cap!

I also visited the high pass X/O point from S2s. The 3uF cap does it's thing and nicely takes care of the around 1000Hz or so hump the S2 driver has.
But looking at the slope and trying to align, found that a 4th order on the mid driver gives a nicer phase plot alignment opportunity than 2nd or 1st.

I've not yet looked in to exporting the frequency plots to Vituixcad.
Still on the Todo list.

 

[AllenB]

If your point is that different filter slopes applied to each of the mid /tweeter is needed to cross them and bring phase together, that's OK. To complement this point, if you find 4th and 8th works for this then 2nd and 6th should also work, since they have the same phase separation.

Aligning phase by trying to overlay the two, and choosing which order of filter to use, are blunt tools.. not that you can't tweak it by ear, sometimes. Using a simulator makes discovery easier, letting you find what extra little filters are needed.

 

[Speedysteve7]

After many weeks listening to all sorts of music on the Le Cléac'h 550Hz upper mid horns on the Vitaxox S2s - X/Os optimized and phase aligned -
I switched over to the Tractrix 400Hz horns.
Still keeping the horn mouths all flush and X/O optimized, SPL adjusted slightly and phase aligned.

Initial impressions are I like the T-400 horn more.
They are more horny🙂 (more direct and slightly more what I'll describe as horn sounding) Vs the Le Cléac'h.
The Le Cléac'h sound more natural in some ways, but the T-400s do something tonally that I like. More timbre perhaps?

I'd like to try a Le Cléac'h 400Hz horn or similar.
Having turned the Le Cléac'h 550s, I would struggle with the 480mm diameter required. Unless I make a horn lathe adapted drive off my existing lathe - too much work!?

Autotech in Poland is one option. $$$s.

I wonder if 3d printing (sand?) Might be an option?

I could perhaps make 12 or 9 sided curved petals but there is the throat adapting from sides to round to do, and I prefer round.

Any ideas?

 

[Rewind]

I made a home brew JMLC 500-550 Hz round horn and compared it to my JMLC Iwatas. I found the sound a bit intense right in front of it. I really appreciate the sound of the JMLC-Iwata 300. Mine is not Autotech, but also from Poland, and as wide but has a bigger vertical mouth, which is only an advantage. It partly 3D printed with a fiber glass throat. The filament is a selected to be a filament and is supposed to be hollow in some places. When comparing it to a vintage Yamaha metal horn (70cm wide) it sounds much better with this softer filament. Comparing it to my old solid wooden JMlC Iwata 220Hz horn with a curved throat, it sounds clearer. I had curve it to reduce depth, but it had a negative impact on sound quality. The winner so far is the JMLC Iwata 300. I wish it was a little bigger though. 250-270, I think is perfect.

 

[Speedysteve7]

That's very interesting Rewind.
Also interesting reading about how folk find their sound - very promising.
The Autotech Iwata-300 is too wide.
The 600 too small I fear. Nothing in-between!
But their eBay offering says they can make to any custom frequency..

First I'm going to try a hybrid or bastardisation🙂 of absorbent material on my T-400 horns mouths.
I have suitable for foam and felt, might try a mix of the two - can only but fail😂

Also recently I've had a revisit on acoustic room treatment placement.
I paid attention to where reflected sound could be absorbed.
Used the mirror technic to see where I could see the horns from listening area.

Along these lines for my loft room (note, this is not my room, something I found on the internet..)

 

[AllenB]

I'd have to ask what EQ was done before passing judgement. If harshness can be successfully removed with EQ, then you can hardly blame the horn. When a horn is EQed as best as it can, what remains should then define it's character.. and even then only if you can justify it's room interaction.

 

[John Sheerin]

Agree with Allen. If you didn't adjust the response between the two horns then you're just saying you like whatever the response shape produced by one horn is compared to another. Over a lot of listening tests, I've found that you can usually get two systems with similar driver configurations (not driven into distortion, etc.) to sound quite similar tonally if the in-room response is the same (kind of an obvious statement). If the two systems have differing directivity (which your horns might) then what remains after EQ will be an imaging difference. Lower DI (wider pattern) will be more spacious and farther away, higher DI (narrower pattern) will be less spacious and more in the room. The in-room response will also change based on your room and the horns' DI, thus the need to EQ in room. So if you like the response of one horn better, try correcting the other horn to have the same response and see if you can still tell any difference. Personally I always needed to be able to A/B these kinds of things with almost no delay between. The smaller the difference, the harder it is to pick it up across a long system change over time. Acoustic memory is not very long lasting for most people.

 

[Speedysteve7]

Thanks for the input.

When I did this

"Still keeping the horn mouths all flush and X/O optimized, SPL adjusted slightly and phase aligned."

That optimization did include minor EQs to ensure similar frequency response.
I'm not a fan of EQ'ing the hell out of things in pursuit of a totally flat line.
The two different types of horn, do not differ much at all in FR, distortion or phase actually.
The T-400 shows the classic S2 bump (largely tamed by the chosen crossover point), slightly lower in freq than the JMLC-550. Talking a couple of hundred Hz

What I hear, i put down to the different in horn profile characteristic - flare and mouth termination.
It's not a massive difference though.

Don't get me wrong, I like both.

Always curious about something different, esp if I can make it.

I do also like a change from time to time..🙂 keeps things alive and novel.

 

[Speedysteve7]

I got some time to have another go at phase alignment.

Note extra bass at 27Hz room mode.
I have 3 settings, from this, to flat. This one is 'shock and awe' setting when demoing bass. Sofa vibrates nicely to church and cathedral bass, and certain Electronica.
Flat is totally boring!

X/O 100Hz

The midbass to mid, X/O 300Hz.

Mid to upper mid. X/O 800Hz

I've matched the slopes as best I can so far. It certainly has helped align the phase plots better, I think?
Still not 100% I'm doing this right though, but it sure does sound nice!

 

[Speedysteve7]

Did the upper mid to tweeter 12600 X/O

These line up quite nicely too.

 

[Speedysteve7]

Here's the latest - room treatment.

I don't have room in the corners for big bass traps and don't really have an issue with the bass - once I've correct at 30Hz and 70Hz is applied in DSP.
No correction anywhere else.
It's not a clappy room by any means.

I did find that 1000Hz up treatment gave a much shaper image, esp vocals.

I tried thicker 500Hz and up panels but didn't like the effect.

To find the positioning I used the line of sight mirror trick.

It's not so obvious what is reflecting where, in a loft room and to the listening spot.

I have around 20% coverage excluding carpet etc.

Side ceiling, also end wall.

More behind the listening spot and to the sides (not in shot).

i revisited the SPLs and played around with the X/Os and phase checked to.

Really happy with the sound.

 

[Speedysteve7]

Having had great fun, and in fact continuing to have great fun with the SH50 alike synergy horns, I decided to try them as a 5 way!

3 long-standing components are still in the mix.

• Raal 9" Lazy ribbons.
• Vitavox S2s on 550Hz soft wood Le Cléac'h horns, I turned some 3 or 4 years ago.
• 2.4m tall/long Tapped horns (also a Danley Sound Labs design) I made some 12 years ago now!

What's new!
The Danley SH50 conical horn is now doing LF, using the trusty Eminence Kappa 12A.
Though playing a big deeper than trad mid bass straight horns - thanks to the horn loading being knitted with reflex port loading.
The SH50s are also being used Synergy wise, for the MF - from the 4 little 5" Celestion sealed back cone drivers.

Whilst 'lash up' can be used to describe the current physical assembly / look.
The acoustic / electrical setup is anything but.
I've spent lots of care and attention measuring, time spent on the frequency response, crossover slope choices fitting for each driver range, how they interface with the adjacent drivers, and phase alignment.

The resulting sound is I believe, the best I've ever had from my system!

Improvements over previous 5 way systems are most noticeable in the LF bass of SH50, more kick, texture, depth, plus more cohesion between the LF and MF from the Synergy design, and the more compact and vertical inline layout of the MF and HF drivers.

It's taken a good while to realise the potential.
I enjoy the journey and reaping the musical rewards.

Dynamically recorded live music really sounds live!
I guess thanks to all the drivers being the sort of thing you could hear live music over at a well prepped venue, but it's also super refined and accurate on studio recordings.

Another really cool thing is that within seconds I can configure it to be a pure Danley'esk 4 way - the tapped horns + 3 way synergy SH50.

I have 2 systems in one.
They sound quite different.

 

[Rewind]

Holy conical! My problem with Danley horns is that when simulated it is both a crappy midbass horn and a crappy midrange horn. Having everything centralized might help with certain things, and can give you the headphone effect, but having good quality horn profiles (in good material) is more important.

 

[Speedysteve7]

Yes, I was highly sceptical, but wanting a semi PA setup for a wedding venue, I built anyway.
For the wedding it will be the BMS comp drivers, perhaps 2 extra 5" MFs (4 in total), and another 12" LF driver (2 in total) per horn.

So the hifi set up sound..

With trad DSP crossovers on the SH50 the LF isn't too shabby out of the box.
Does need a PEQ cut at 120Hz.
Reaches down 60Hz and easily plays up to the MF at 325Hz cross.
The MF was a bit of a revelation measurement wise. Between 300Hz and about 1000Hz it was pretty smooth really. A little cut at a hump around 350Hz needed.

I am still dabbling with the "XO free" flatter phase integration that is reported to be so good. That uses mostly only PEQs to control things!

I can hear some potential there but integrating the tapped horns is not as easy.
A lot of PEQs are required - still work in progress!

Putting the Vitavox S2s on the Synergy horns has made the sound totally acceptable to me!
The HF shortcomings with the BMS 4550s is simply not there.
The S2 tone blends with the cone MF better too.
I do EQ cut the S2s at around 950Hz a few dB and still use the 3uF cap on them🙂
The S2s also play quite a bit higher in the SH50 than on my Tractrix or Le Cléac'h horns.
Why is that?
I'm getting the Vitaxox quoted "about 16KHz". Almost no need for a tweeter, for my ears.
I'm playing with the Raals coming in at 15KHz on various slopes, and orientations, to add air, but generally it's as good without.
If I were an 18 ye old I might say otherwise😃

I've yet to find the perfect room position for the SH50s -
Trying then out as wide as they will go in my room at the moment, and will also try them a bit further in.
Extra supports may be required for that though

The other thing is how well the SH50 LF integrates with the tapped horns.
Like they were made for each other - both from the Danley stable I guess🙂

I find listening much easier across a broad cross of music.
There's still the wow factor and grab you by the lapels shock and awe that I love.
Enhanced by the kick bass of the SH50 reflex LF..
They are just more even in their presentation across any type of music.
A neat trick!

I'll carry on reporting the evolution and even other Synergy horn builds, should they happen, on my Synergy specific thread.

 

[swak]

My JA6681B drops like a stone below 450-500Hz in the Iwata 300

The winner so far is the JMLC Iwata 300. I wish it was a little bigger though. 250-270, I think is perfect.

Rewind, I am about to start printing an JMLC Iwata and I also have the Yamahas. May I ask why do you feel that a slight larger horn would be better? I remember JMLC once writing that there would not be much difference by such a change, perhaps going down to 250 there would, but how much? Have you by chance measured GD and impedance on your horn?

I made a home brew JMLC 500-550 Hz round horn and compared it to my JMLC Iwatas.

Interesting, a bigger symmetrical (around 350) is on my list too, and on both I would like to try going down to a 25.4 mm diam. throat (into the driver). Other than the obvious difference arising from different operating ranges, I wonder if your comments come from hearing to a a pair towed as it was suggested by JMLC. Some prefer the narrower pattern of the symmetrical horns, others the wider of the E-JMLC (the JMLC Iwata being in the middle of both). I believe this is a matter of preferences and/or room.

I'd like to try a Le Cléac'h 400Hz horn or similar.

The Autotech Iwata-300 is too wide.
The 600 too small I fear. Nothing in-between!

You seem to be very happy with the Danley horn now, but if you do decide to try, it is possible to 3D print them or to get the AH425 (if axisymmetric), which is not expensive and can be shipped.

 

[Speedysteve7]

You seem to be very happy with the Danley horn now, but if you do decide to try, it is possible to 3D print them or to get the AH425 (if axisymmetric), which is not expensive and can be shipped.

True I am enjoying the SH-50 in room, much more than I thought I would.

It took a compression driver I really like to get it there sound wise.
I'm still not fully there with the phase alignment thing.
HF and MF is good, LF needs more work - trial and error as I'm not REW au fait.
To get phase control in having to advance the LF in time. Should not be like that?🥴
It still sounds better than anything else so far!

Never say never on the another horn set up though.
I have already tried the 5 way with SH-50 doing only LF and MF.
I will probably try the SH-50 as LF only next.
Bringing the Sea / Le Cléac'h (or other) horns? And the JBL2482s back as MF.
It's the only config I've not tried yet.

Once I've sorted the SH-50s out fully perhaps..

Asymmetric AH425 and even something similar for the mid frequency compression drivers could be something to try.
Size is the problem.

Better directivity control than round would be good.
I've learned that from the SH-50.

 

[AllenB]

It took a compression driver I really like to get it there sound wise.

Are you EQing the SH-50?