Open Source "Tower XL"

[mbrennwa]

I heard such kind of tweeter in the Magico mini about 5 years ago.
Magico Loudspeakers | Mini
That was really not good. Ok on axis, but if you listened a little off axis, a real disaster, very directive for high frequencies.
At that time they said, it was a Scanspeak type, made for Magico.
Just an an information, maybe the R2904 is completely different.

The tweeter on that Magico Mini looks like a normal ScanSpeak R2904 with a standard faceplate. I don't see any waveguide.

Here's some data of a R2904 in a Visaton WG148 waveguide.

 

[Paul Vancluysen]

Gable, that would be super, if you will make an IEC baffle and measure the drivers.

I made the IEC 225Hz version. I have chosen that large one to measure infinite baffle responses as low frequent as possible without diffractions of the baffle.
In my setup the measurement is diffraction free above 650Hz. Good enough to make a good infinite baffle model of the driver together with TSP data for low frequencies.

I give you more information later.

In attach the IEC baffle dimension standard.

Important to respect the driver position to be conform standard.
I made a square hole of 40 x 40 cm in the baffle, to mount drivers in an easy way. And make a separate little baffle for every driver that is mounted in the IEC baffle, very practical. Maybe you can choose a little more than 40 x 40 cm.

I made the baffle in two parts and screwed them together at the back side.
In one piece it is very large to handle it.

 

[Paul Vancluysen]

The tweeter on that Magico Mini looks like a normal ScanSpeak R2904 with a standard faceplate. I don't see any waveguide.

Here's some data of a R2904 in a Visaton WG148 waveguide.

That already looks better on paper.

But that tweeter in the Magico mini was real bad, I can assure you. And I was not the only one with that opinion. Indeed without waveguide. Let's forget that and look to actual measurements.They look very promising.
If the Volt DI is also around 2 dB in the horizontal plane, then you have a very good match for the Monkey coffin.

 

[pcgab]

Gable, that would be super, if you will make an IEC baffle and measure the drivers.

I made the IEC 225Hz version. I have chosen that large one to measure infinite baffle responses as low frequent as possible without diffractions of the baffle.
In my setup the measurement is diffraction free above 650Hz. Good enough to make a good infinite baffle model of the driver together with TSP data for low frequencies.

Hello Paul,
I'm looking at the document you attached, and working out which size I can build reasonably. I would love to build the 150hz one, but it's enormous, even by Texas standards.

I _think_ I can pull off the 175hz, if not, then the 200.

I give you more information later.

In attach the IEC baffle dimension standard.

Important to respect the driver position to be conform standard.
I made a square hole of 40 x 40 cm in the baffle, to mount drivers in an easy way. And make a separate little baffle for every driver that is mounted in the IEC baffle, very practical. Maybe you can choose a little more than 40 x 40 cm.

I made the baffle in two parts and screwed them together at the back side.
In one piece it is very large to handle it.

I am definitely going to have to make it in several pieces. I'll make a frame for the back to hold the pieces together and for attachment points for some sort of stand and base, etc.

I think 40x40cm may be a bit small to test the AE 15TX, so I'll likely make the 'driver plate' 45x45cm.

Do you have a plate, or some sort of jig that you use to make your on-axis and 15,30,45deg, etc. measurements?

Cheers,
Gable

 

[Paul Vancluysen]

Gable, I have chosen for the IEC 225Hz, because I had no more place on my attic. If you can make a larger one, is always better.
My ceiling is well damped by the isolation of the roof.
I have chosen the on axis for the drivers at the heigth for a maximum distance from wall, floor and ceiling to become the longest time before early reflections arrive.
Therefore my baffle is placed about 30 cm above the floor.
In that way the position of my microphone is at about 150 cm height from the floor.

Edit: I will make a plate on wheels turning on a 1m circle to measure off axis, later.

 

[pcgab]

Gable, I have chosen for the IEC 225Hz, because I had no more place on my attic. If you can make a larger one, is always better.
My ceiling is well damped by the isolation of the roof.
I have chosen the on axis for the drivers at the heigth for a maximum distance from wall, floor and ceiling to become the longest time before early reflections arrive.
Therefore my baffle is placed about 30 cm above the floor.
In that way the position of my microphone is at about 150 cm height from the floor.

Edit: I will make a plate on wheels turning on a 1m circle to measure off axis, later.

That makes good sense, thank you for the explanation.

I'll do some similar measurements of our living room. Thankfully it is a very large space, >600sq/ft, with coffered ceilings over 12', IIRC, I'll need to measure to confirm.

I was thinking something similar to move the microphone for off axis measurements.

Cheers,
Gable

 

[ILikeFoodz]

What are the units for distances in the iec standard pdf? Cm?

 

[Juhazi]

A thread here IEC 268-5 Baffle Dimensions

Diffrent baffle sizes are for different frequency range or driver diameter.

PDF from IEC https://webstore.iec.ch/preview/info_iec60268-5{ed3.1}en.pdf

 

[KaffiMann]

...

Also the dip at 6.5 kHz, even without the waveguide looks not so good.

...

The measurements on Mr. Gravesens site that I linked to previously suggest that the dip is slightly reduced when removing the magnetic oil, the measurements before/after modifications are the first measurements on the page. It should also help the low end performance, making a 2khz filter a quite realistic option, the Monacor 300+300 combo is cheap and fulfills most of the criteria, unless someone has something equally good or better to bring to the table, I suggest the Monacor 300+300 is chosen.

 

[mbrennwa]

unless someone has something equally good or better to bring to the table, I suggest the Monacor 300+300 is chosen

I *think* there was a test of the DT300+WG300 in the Hobby Hifi magazine (along with some other waveguide tweeters?), possibly here. I'll check if I still have this issue...

Here's another test of the DT300+WG300. Looks good, but consistency could be an issue.

 

[PKAudio]

DT300 used to be quite popular here in Czech. But quality went down, some samples needed tuning (excessive glue, tight wire leads, rubbing and off-centered voice coil, no damping material) and some were still useless even after tuning. In measurement it has shown as double or triple impedance resonance peaks, distortion profiles of defective units, FR irregularities. Out of 10pcs batch almost each needed tuning, and even after tuning, with resulting clean impedance profile, it was veeery difficult to obtain matched pair.
I have seen something similar on Focal TC90TD5 (used and even brand new pair) and BW805S tweeters. Though situation with DT300 was much worse.

 

[Paul Vancluysen]

What are the units for distances in the iec standard pdf? Cm?

The upper left table in inch and the upper right table in cm.
The unit is mentioned at the top of each dimension table

 

[Paul Vancluysen]

I suggest the Monacor 300+300 is chosen.

Kaffimann,

With the Faital 15FH520 the speaker sensitivity will become 90 dB minimum.
If you look to the SPL plot of Troels you can see also that this Monacor isn't 90 dB up to 20 kHz.
Also if you look to the impedance curve and other things Pida is mentioning, this is not a good choice. Anyway the sensitivity is too low above 10 kHz.

 

[Aatto]

that is a lot of simulations Paul I appreciate your dedication and patient.
DT300 could be a good candidate.

 

[Aatto]

DT300 used to be quite popular here in Czech. But quality went down, some samples needed tuning (excessive glue, tight wire leads, rubbing and off-centered voice coil, no damping material) and some were still useless even after tuning. In measurement it has shown as double or triple impedance resonance peaks, distortion profiles of defective units, FR irregularities. Out of 10pcs batch almost each needed tuning, and even after tuning, with resulting clean impedance profile, it was veeery difficult to obtain matched pair.
I have seen something similar on Focal TC90TD5 (used and even brand new pair) and BW805S tweeters. Though situation with DT300 was much worse.

Thanks man, good to have this info, can save a lot of headache.

 

[PKAudio]

I would not mention it if this was just one speaker pair project.

I considered some project with DT300/WG300 too, in the past.

Thanks man, good to have this info, can save a lot of headache.

I think one of Augerpro WGs will be the best option.

 

[Aatto]

Aatto, well it seems the Faital/regular PA brand build is cheaper by quite a bit. So it would probably be okay if one build just went "screw the budget!" And the other kept a more sane and sober approach. These speakers will be quite the performers at any rate, and it's nice to see a little bit of diversity. It is mostly the same project, just one version goes all-in and the other aims to achieve much the same performance on a smaller budget

that is a good point KaffiMann I do not think i can build it myself tho (at list for the time being) but it would be nice to match the AE version with that Beyma 150.

 

[Aatto]

this would be interesting if it was sensitive enough: Wavecor TW030WA12

 

[Paul Vancluysen]

Hi Pida,

Thanks for all your information, very useful.

I have a question about the off axis measurements you placed in post #186 in the “Open source Waveguides for CNC & 3D printing” thread. I got this link here of Mbrennwa.

- I suppose that it is an off axis measurement of a waveguide tweeter mounted in an enclosure. Because the DI stays about 2dB below 1 kHz with some increase around 1 kHz, probably due to the bafflestep. On a large baffle the DI would tend to zero below 1 kHz.

- How is the DI calculated in that plot? I suppose that it is the DI in a single horizontal half plane and not in a full plane or hemisphere. Do you know the formula behind your DI calculation or where to find it? It should be interesting to know to compare my DI calculations with yours. I use the calculation as explained in the article “On the calculation of Full and Partial Directivity Indices” by Joseph G. Tylka. At this moment I calculate the DI for a horizontal half plane.

Maybe this information can be found in the other thread, but it is so much to read it all now.

Thanks.

 

[Paul Vancluysen]

this would be interesting if it was sensitive enough: Wavecor TW030WA12

It is already better than the Monacor TD300-WG300 sensitivity. Maybe it is just possible. But we have also to look at the DI of that tweeter. If possible we have to choose a tweeter with the best DI match with the midrange.

Edit: no forget it, for the second time I looked to the 4 ohm version instead of the 8 ohm one, and that isn't sensitive enough.