Three way horn system

[pelanj]

Now when I look closer, you are right. I thought the line around it is just a frame from a magazine picture. Facepalm Thank you both

 

[pelanj]

GM, I see your posts about the pioneers' approach quite often. And over the time, I get a feeling that they were mostly right and it is actually amazing in how many things they were right with the technical limits of their time.

 

[peterbrorsson]

Hello Pelanj,
How does the throat look like on your ARB2, are they oval as on my so-called-quality horn JBL/Selenium HM4750SLF?

 

[pelanj]

Hi Peter, as far as I can tell, it looks perfectly round. On the ones I sanded and painted, they have a small flat part on the top and bottom, but that was done by me and could be made round again if I wanted to

 

[peterbrorsson]

Here's how mine looks, it's the good one The other which I have in use is worse...

 

[pelanj]

If I can judge on the picture, the yellow ones looked better. For 12 Eur, I cannot complain

 

[peterbrorsson]

12 Euro is ok
Have you tried to run them all the way up in treble? I use mine from 400Hz and have "some" dips starting at 7kHz due to mismatch in throat and of course, I use a Visaton FRX5s.

I tried to upload measurements but it's not possible for the moment...

 

[pelanj]

Here are two measurements of the same driver (Beyma CP385/Nd) on the ARB2 horn (with JBL2445 below with passive XO around 4.5 kHz) and on some 1" horn (I do not remember exactly which one that was). To me it seems the ARB2 has no problem reaching high. That is with my custom 3D printed 1" to 2" adapter.

 

[peterbrorsson]

Sorry, but I don't understand what is what! 2445 with passive x-over and printed adaptor?

 

[pelanj]

Ok, I was not too clear. I could not find separate measurements of the 1" CP385/Nd on the ARB2 horn. Therefore I posted the brown curve, which is two ARB2 horns on top of each other. The lower horn is JBL2445, the upper is the 385 with a conical adaptor from 2" to 1". All above ca 4 kHz is the 385. For comparison, I posted measurement of the 385 in the red chart - on another smaller horn - to prove that there is not much difference in the HF. Please ignore the absolute dB level, they were measured at different distances and volumes. And there is also different scaling. The upper cut off looks the same on both horns and is caused by the driver, not thr horn.

 

[peterbrorsson]

Now it's crystal clear Thanks!

 

[pelanj]

Some in-room polar plots of the ARB-2 with the 1" to 2" 3D printed adapter and Beyma CP385Nd. All ca 10 degrees spaced. I wonder if the ca 2.6 kHz dip is caused by the room or something else. I could see it also on the K-Tube measurements, but there I thought it is related to the tube length. But as a CD tweeter above 3 kHz it looks OK. I also wonder if the dips above 3 kHz are caused by the imperfection of the adapter - it needs to be matched a bit better to the throat diameter, maybe even to the real shape since I sanded it a bit into an oval, not a circle

 

[DonVK]

Did you gate the measurements to minimize room reflections? You also need about 1m free space on the sides and 2m freespace in front of the speaker.

 

[pelanj]

I did some gating, but the room I measured in is definitely not good for measurements. This was just to try out and see the general trends. Unfortunately, it was too damp/cold to measure outside. I would not trust anything below 3 kHz, but above it shows the trends somehow.

Edit: There is definitely room influence. In the attachment, there is another measurement of the same horn in another place of the room, slightly off axis.

 

[DonVK]

If you gated [-1,5ms] and used 1/12 smoothing you should get a reasonable curve. I'm assuming the horn was lifted off the floor and measured at 1m, rotated about the horn mouth center line.

 

[pelanj]

Yes, 1 meter off the floor, rotated (almost) at the mouth center line. When I look your suggested gate settings and smoothing, the graph looks definitely better, thanks!

 

[DonVK]

Looks the same to me

Usually plastic parts are not perfectly flat. Is there a gasket to ensure a really good fit between the adapter and horn.

 

[pelanj]

The things are still there, but it is easier to read I will have a better look at the mounting. There is no gasket at the moment between the adapter and horn, there is gasket on the CD. Both surfaces should be very flat, I sanded them.

 

[DonVK]

They might be smooth from sanding, but may not be flat, and may not seal properly at the throat interface. A gasket may fix that type of problem.

 

[pelanj]

Today I tried ca 8 cm deep conical 60x40 horns that I made from some half destroyed horns I got for free. Yesterday, I made some quick and dirty measurements and they are surprisingly flat and have nice directivity. Enjoyable sound, maybe even more than with the large ARB-2 horns.

Of course, there are problems due to the placement and reflections - but I think I could easily live with conical horns - these may well one day serve as the entry part of a Synergy/Unity horn.