"Unitized" Image Control Waveguide

[pelanj]

I did some very crude first measurements (without box or baffle), the distance is less than 1/2 m, I think that is why there is the bump in the HF response (=measurement error). The off axis is nice and flat.

The latest crossover is in post #203, is that correct?

What is the lowest possible crossover point on the low end (below WG response)? I can't wait to listen to these, so I will just use something I already have around below the WG.

Edit: Something does not seem right - the MF is something like 9 dB below the HF. I think I need to check the MF wiring (connected series/parallel - or at least it should be).

[Patrick Bateman]

Originally Posted by pelanj

I did some very crude first measurements (without box or baffle), the distance is less than 1/2 m, I think that is why there is the bump in the HF response (=measurement error). The off axis is nice and flat.

The latest crossover is in post #203, is that correct?

What is the lowest possible crossover point on the low end (below WG response)? I can't wait to listen to these, so I will just use something I already have around below the WG.

Edit: Something does not seem right - the MF is something like 9 dB below the HF. I think I need to check the MF wiring (connected series/parallel - or at least it should be).

Yes, current crossover is on post 203.



Here is the response of your midranges and tweeter



Here is the response of my mid and tweeter

I think the issue may be that you're not using a baffle whatsoever?

If you look at my speaker and your speaker, the response shapes are fairly similar down to about 1500Hz, and then they diverge.

If you just want to listen immediately, you could cut a hole in a piece of plywood, and then mount the waveguide into the baffle using something that's temporary, such as hot melt glue from a glue gun.

If you want to go with something permanent, it's time to build a box!

Either option is going to improve the response quite a bit; if you measure the waveguide with no baffle, the response will be rough because it will be behaving as a dipole.

Off the top of my head, I believe I used a high pass around 350hz, give or take 100hz. I would need to re-read my own thread, but it's 3:07am and I should probably get some sleep

[pelanj]

Thanks! Time to get some XPS for quick prototyping

[1hiep0]

Patrick,
What is the max SPL between UICW vs Metlako ?

[Patrick Bateman]

I'm on a work trip, so I don't have time to look up the figures.

If I recall correctly, Metlako V1 and V2 are capable of about 5dB more output, something like 115dB maximum.

But the two speakers distort in a very different fashion:

The main thing holding back the UICW speaker is the enclosure itself. Basically the foam CLD construction is probably more hassle than it's worth. I found that the enclosure was starting to rattle long before the drivers hit their maximum.

There's a simple fix for this : brace the enclosure and add more fiberglass.

But by the time I got to that point, of reinforcing the enclosure, my attention had shifted to the various versions of Metlako. (As insane as this sounds, there's something like five Metlakos; I only published two, and there's more on the way.)

'Metlako' has kinda become my 'white whale', because it's really interesting to see if it's possible to get full-range performance out of a two-way Unity horn. It's really pushing the limits of how much bandwidth and efficiency you can get out of two drivers.

All of the most recent iterations of 'Metlako' have used two midbasses instead of four, simply because the tweeter will run out of output long before the midbasses will. I was kind of shocked by how loud Metlako V1 and V2 can get.

The only obvious downside to the Metlako V1 and V2 designs is that they're just a nightmare to print. No exaggeration - I threw away 75% of the prints, and some of the prints took more than a DAY to finish. Imagine spending EIGHT DAYS making TWO waveguides. And I know some of you are probably thinking "eight days, that's no big deal, some people wait a month to get their waveguides in the mail." But the thing is, when you're doing 3D printing, you have to babysit the printer. I work 50-60 hours a week, I don't have time to invest eight days of my life babysitting a printer.

I suppose there's probably an argument to be made, that the most efficient use of my time would be to get a much larger printer, instead of using my time to figure out how to make Metlako V1 and V2 smaller.

[sugercube]

Originally Posted by Patrick Bateman

Sorry about the slow reply on this one.

I measured the T/S of my TC6s, and my Gentos, and took some pics. And was going to make a big post about it, but I've been buried at work. (Literally at work right now.)

In a nutshell, what I found is that the TC6 will fit on the waveguide, just barely. You might have to use a file to take about a millimeter off the basket, or file down the mounting places on the waveguide itself.

The trickier part is that it looks like the Tymphany TC6 may need a back chamber.

Whether you opt to build one really depends on whether you're trying to get the efficiency up.

Basically the Gento has a very high FS and a very high QTS; it behaves almost as if it has a sealed back but it doesn't. The TC6 has a lower FS and a lower QTS.

If you put the TC6 in a small enclosure the FS will go up and the QTS will go up. You would have to measure the TC6 with a DATS to determine the optimum chamber size. Or you could simulate it for free in Hornresp.

Then again, you could use the TC6 and skip the back chamber entirely, but this would mean that the efficiency would go down around 1-3dB.

Hi Patrick, I have been following this thread closely for a while. I am interested in building your waveguide speaker. I am interested in using the TC6s so hacked your waveguide in Tinkercad to accommodate the TC6s and re-oriented it to print on it face. My plan is to develop a loudspeaker using your waveguide (and tc6 + sb19) with midbass firing through slots on either sides of it.

I'm posting the STLs of my hackjob on your file. I can modify the sealed chamber cap optimise the internal volume

I'm wondering if you will be kind enough to post the FRD & ZMA files of the Gentos, TC6s & SB19 so I can play around with the crossover, when you have some free time.

[pelanj]

Just a note - did you check if the SB19 fits the flange with the midrange enclosures?

[sugercube]

Originally Posted by pelanj

Just a note - did you check if the SB19 fits the flange with the midrange enclosures?

Yes I did. I will need to shave some of the SB19's fange. I needed a circular shape because I wanted a screw top for sealing the enclosure.

[Patrick Bateman]

Originally Posted by sugercube

Hi Patrick, I have been following this thread closely for a while. I am interested in building your waveguide speaker. I am interested in using the TC6s so hacked your waveguide in Tinkercad to accommodate the TC6s and re-oriented it to print on it face. My plan is to develop a loudspeaker using your waveguide (and tc6 + sb19) with midbass firing through slots on either sides of it.

I'm posting the STLs of my hackjob on your file. I can modify the sealed chamber cap optimise the internal volume

I'm wondering if you will be kind enough to post the FRD & ZMA files of the Gentos, TC6s & SB19 so I can play around with the crossover, when you have some free time.

Tymphany TC6 is only three milimeters larger than the Gento woofers

So you could use the "stock" print, with one of three modifications:

1) file down the TC6

2) buy a Dremel and grind away some of the plastic of the waveguide

3) just create a gasket out of mortite

Any one of those three should work.

The QTS of these small woofers tends to be high, so I generally don't bother using a back chamber. My thought process is that I can control their excursion using the high pass filter on my DSP. That's why I generally don't use back chambers in my Unity horns. I figure if the QTS is between 0.7 and 1.5, there's no real point in using a back chamber.

The SPL Unity horns require a back chamber because the driver has a very low QTS, so they need a back chamber to raise the QTC and the FS.

The Gento and TC6 midranges already have a FS and a QTS that's where we want it, so I don't see any real benefit to a back chamber.

One exception to this rule is if the midbasses modulate the response of the midranges, but I haven't determined if this is a real concern. It might be, I simply don't know, one way or the other.

I'll dig around on my hard drive for FRD and ZMA files.

[sugercube]

Thanks Patrick. I "Digitally" filed the the mount on that model. Hopefully you can find those FRD & ZMA files.

I'm also following your Metlako thread very closely as well (for the dual midbass version). I am fascinated by your dedication/obsession with horns and synergizing/unitizing them. Your experiments and findings have been very educational.