Paul, there is lots of knowledge about using waveguides... 148mm diameter wg is way too small to be paired well with a 8" midrange! It will not have loading effect/directivity at appropriate xo. And actually it should work also a bit before xo.
Dave Pellegrene used to make super fine wgs, but looks like he is out of business. My suggestion is CD+horn or Beyma TPL-150H (with horn), and horn should be 200mm wide!
For a 1" CD I would choose a rectangular horn that mimics JML M2's horn with those bulbs near throat JBL M2 for The Poors
- or an elliptical like XT1086 from Eighteensound.
Dave Pellegrene used to make super fine wgs, but looks like he is out of business. My suggestion is CD+horn or Beyma TPL-150H (with horn), and horn should be 200mm wide!
For a 1" CD I would choose a rectangular horn that mimics JML M2's horn with those bulbs near throat JBL M2 for The Poors
- or an elliptical like XT1086 from Eighteensound.
Juhazi, yes you are completely right, you can see it in the plots of my post 476, the DI of 148 mm waveguides is lower than the TD8M from high frequencies down to 1.5 kHz.
A compromise could be an X-over crossing around 1.5 kHz with high order filtering, if possible for the tweeter excursion. Not perfect, but maybe an acceptable compromise.
But indeed a larger waveguide will be a better solution. We have to find and check directivity.
Beyma TPL 150-H is a nice driver, but we also have to look to the single hemisphere directivity with that driver, if it matches well with the AE TD8M. Vertical directivity is larger with it.
A compromise could be an X-over crossing around 1.5 kHz with high order filtering, if possible for the tweeter excursion. Not perfect, but maybe an acceptable compromise.
But indeed a larger waveguide will be a better solution. We have to find and check directivity.
Beyma TPL 150-H is a nice driver, but we also have to look to the single hemisphere directivity with that driver, if it matches well with the AE TD8M. Vertical directivity is larger with it.
^Vertical directivity will be dictated by vertical separation/distance of mid and tweeter at xo anyway, if we use modest xo slopes like LR4.
Most large waveguides are rectangular or ellipsoid (smaller diameter vertically) giving higher limit of controlled directivity and boost. That is to keep interdriver distance minimal at the expense of directivity.
Most large waveguides are rectangular or ellipsoid (smaller diameter vertically) giving higher limit of controlled directivity and boost. That is to keep interdriver distance minimal at the expense of directivity.
You want it cheap but good? Take a look at the Omnes T25 H dome. Waveguide, 92 dB@2.83V@1m, 5.9Ω impedance minimum (so just a hair below the "nominal 8Ω" rating), and measures well. 15 Euro a piece!
Attachments
Juhazi,
Of course the distance between the drivers will affect the DI in the vertical single plane and in that way also the DI in the single hemisphere of the total speaker. There will become DI peaks in the vertical plane on the crossover point and they will be wider with lower slope filters.
But my statement is, if you choose the midrange and the tweeter in a way they have equal DI in the crossover overlap zone, horizontal and vertical, you get also the best total DI result in a single hemisphere of the speaker with X-over. I think, if you choose a tweeter which is more vertical directive, you create a room power dip, resulting in even a higher DI peak of the total speaker with X-over.
Suppose you like to make a compromise for example for a AMT tweeter which is more vertical directive. I think you have to choose the tweeter that the DI in the hemisphere (horizontal + vertical) is equal to the DI of the midrange also in the hemisphere around the crossover point. And that you make the most flat DI and room power for the speaker with X-over in that way.
Of course the distance between the drivers will affect the DI in the vertical single plane and in that way also the DI in the single hemisphere of the total speaker. There will become DI peaks in the vertical plane on the crossover point and they will be wider with lower slope filters.
But my statement is, if you choose the midrange and the tweeter in a way they have equal DI in the crossover overlap zone, horizontal and vertical, you get also the best total DI result in a single hemisphere of the speaker with X-over. I think, if you choose a tweeter which is more vertical directive, you create a room power dip, resulting in even a higher DI peak of the total speaker with X-over.
Suppose you like to make a compromise for example for a AMT tweeter which is more vertical directive. I think you have to choose the tweeter that the DI in the hemisphere (horizontal + vertical) is equal to the DI of the midrange also in the hemisphere around the crossover point. And that you make the most flat DI and room power for the speaker with X-over in that way.
I have tried to make a first preliminary conclusion which type of tweeter would be the best choice for the Tower XL with AE TD8M midrange, IMO at this moment.
I have considered as tweeter, a dome without waveguide, a dome with waveguide and a CD horn tweeter.
Used drivers:
- Midrange AE TD 8M Leap model
- Dome tweeter Satori TW29BN
- Dome tweeter with waveguide R2904 – Visaton WG148 – measurement Pida in enclosure.
- CD tweeter with horn 18Sound XT1086 + ND1075
I have done a DI simulation of 2 drivers in the cabinet in the single horizontal plane +/- 90 degree. For the AE TD8M midrange and the Satori TW29BN.
In the first plot you see the DI responses in the cabinet for the AE TD8M (in green) and the Satori TW29BN (in red). I have added also the DI of the drivers mounted on infinite baffle (in grey) to show the cabinet impact on the directivity.
You can see that the bafflestep gives a DI boost below 1 kHz , almost the same for both drivers. The DI dip at 1 kHz of the midrange is caused by the bafflestep boost below. Not so visible in the tweeter DI, I will look at that later.
For the tweeter there is diffraction ripple above 1.5 kHz, because a dome without waveguide has a wide dispersion, which causes diffractions. Not for the AE TD8M above 1.5 kHz because that driver is more directional there.
Suppose a speaker with X-over at 2 kHz is designed with these two drivers, green to red curve at 2 kHz, a DI dip between 2 and 4 kHz can be expected.
For the dome with waveguide and the CD horn, I have no Leap model and cannot simulate them in the cabinet. So I have used the previous DI responses, that are calculated with 0 – 90 degree off axis measurements. In a cabinet there will probably appear some extra diffraction ripple but less, because these drivers are more directional above 2 kHz.
In the second plot I have added dome tweeter with waveguide (in yellow) and CD tweeter with horn (in blue)
Now, making a design AE TD8M – waveguide tweeter, green to yellow curve at 2 kHz, the DI wil have a flat behavior.
Making a design AE TD8M – CD horn tweeter, green to blue curve at 2 kHz, the DI wil have a boost between 2 and 4 kHz.
Remark that this study is done in the horizontal single plane. In a hemisphere a DI peak can be expected at the crossover frequency. Also another bafflestep behavior to be expected but no spectacular difference. Later we will simulate with final X-over.
Using a ATM tweeter with the AE TD8M, such tweeter is more vertical directive than a dome tweeter without waveguide.
It can be expected that the total DI has also a DI dip between 2 and 4 kHz but not so much as the dome tweeter without waveguide.
My personal preference is the Scanspeak R2904 with Visaton WG148. I also expect acceptable results of the AMT tweeter.
The AMT with horn like the Beyma TPL150-H for example, we have to calculate also the DI of that one. This driver gets very good reviews.
That's all for now
.
I have considered as tweeter, a dome without waveguide, a dome with waveguide and a CD horn tweeter.
Used drivers:
- Midrange AE TD 8M Leap model
- Dome tweeter Satori TW29BN
- Dome tweeter with waveguide R2904 – Visaton WG148 – measurement Pida in enclosure.
- CD tweeter with horn 18Sound XT1086 + ND1075
I have done a DI simulation of 2 drivers in the cabinet in the single horizontal plane +/- 90 degree. For the AE TD8M midrange and the Satori TW29BN.
In the first plot you see the DI responses in the cabinet for the AE TD8M (in green) and the Satori TW29BN (in red). I have added also the DI of the drivers mounted on infinite baffle (in grey) to show the cabinet impact on the directivity.
You can see that the bafflestep gives a DI boost below 1 kHz , almost the same for both drivers. The DI dip at 1 kHz of the midrange is caused by the bafflestep boost below. Not so visible in the tweeter DI, I will look at that later.
For the tweeter there is diffraction ripple above 1.5 kHz, because a dome without waveguide has a wide dispersion, which causes diffractions. Not for the AE TD8M above 1.5 kHz because that driver is more directional there.
Suppose a speaker with X-over at 2 kHz is designed with these two drivers, green to red curve at 2 kHz, a DI dip between 2 and 4 kHz can be expected.
For the dome with waveguide and the CD horn, I have no Leap model and cannot simulate them in the cabinet. So I have used the previous DI responses, that are calculated with 0 – 90 degree off axis measurements. In a cabinet there will probably appear some extra diffraction ripple but less, because these drivers are more directional above 2 kHz.
In the second plot I have added dome tweeter with waveguide (in yellow) and CD tweeter with horn (in blue)
Now, making a design AE TD8M – waveguide tweeter, green to yellow curve at 2 kHz, the DI wil have a flat behavior.
Making a design AE TD8M – CD horn tweeter, green to blue curve at 2 kHz, the DI wil have a boost between 2 and 4 kHz.
Remark that this study is done in the horizontal single plane. In a hemisphere a DI peak can be expected at the crossover frequency. Also another bafflestep behavior to be expected but no spectacular difference. Later we will simulate with final X-over.
Using a ATM tweeter with the AE TD8M, such tweeter is more vertical directive than a dome tweeter without waveguide.
It can be expected that the total DI has also a DI dip between 2 and 4 kHz but not so much as the dome tweeter without waveguide.
My personal preference is the Scanspeak R2904 with Visaton WG148. I also expect acceptable results of the AMT tweeter.
The AMT with horn like the Beyma TPL150-H for example, we have to calculate also the DI of that one. This driver gets very good reviews.
That's all for now
.Attachments
Last edited:
Here is a nice exaple of the difference of 4" vs. 8" waveguide. SEAS H1212 is nice, but too low sensitivity for this project. One might be happy for the 7dB increase of sensitivity with wg, but it must be eq'd away. Actually it means 7db more directivity at 2kHz!
Seas 27TBFC-G in a waveguide -
Techtalk Speaker Building, Audio, Video Discussion Forum
Seas 27TBFC-G in a waveguide -
Techtalk Speaker Building, Audio, Video Discussion Forum
Juhazi, I agree with you that a larger waveguide is better for a higher directivity below 2 kHz.
The curve of the Scanspeak R2904 and Visaton waveguide in my post 487, is measured in a smaller enclosure than the Tower XL one. So the DI gain because of the bafflestep will be less in the more wide Tower XL.
From the other side, using a larger waveguide will increase the vertical distance between the drivers, which is worse for the vertical DI.
Edit:
If you know such 8 inch waveguides examples that can be mounted on actual drivers, we have to look at also.
The curve of the Scanspeak R2904 and Visaton waveguide in my post 487, is measured in a smaller enclosure than the Tower XL one. So the DI gain because of the bafflestep will be less in the more wide Tower XL.
From the other side, using a larger waveguide will increase the vertical distance between the drivers, which is worse for the vertical DI.
Edit:
If you know such 8 inch waveguides examples that can be mounted on actual drivers, we have to look at also.
Last edited:
Sorry I don't know of any 8"wgs since Dave quit. That is the real problem I didn't want to say...
DIY it? Practical DIY Waveguides - Part 3
Lots of 8-10" horns for compression drivers though... and TPL-150H
DIY it? Practical DIY Waveguides - Part 3
Lots of 8-10" horns for compression drivers though... and TPL-150H
I have tried the WG148 with my Scan 6600, which showed good dispersion. However, it also showed quite some wiggles in the on-axis response, and some more and stronger dips/peaks above 10 kHz. These issues are most likely due to the WG148, as they are also observed with other tweeters in the WG148. While the WG148 may be a useful test dummy to study the suitability of a waveguide for the Tower XL, I don't think it's the best waveguide out there.
I have some R2904 on the way, which I will forward to the guys in the Open Source Waveguide for 3D Printing thread in order to make custom designed waveguide for the R2904. I have high hopes for this!
I also measured SS6600 and 7100 in WG148R. The same conditions and baffle that I used for R2904/7000 so the results are directly comparable, except SPL. It has been quite some time ago, but I still see it usable, when CNC is not available.
SS6600 shows some air leak => dip around 500Hz.
SS6600 shows some air leak => dip around 500Hz.
Attachments
Pida,
I have checked DI of Scanspeak 6600 also, same DI as with the 7100 as expected. In the plot the 6600 in blue and the 7100 in red.
I also found this tweeter, the Seas noferro-900 – H1025-06 and it can be used with the Visaton W148R. There is a 4 and 6 Ohm version. I cannot find it on the Seas website. It looks an OEM version.
And ... it is cheap!
I can find it here:
NOFERRO900 | Lautsprecher Intertechnik - Shop
Strassacker, Komponenten: Lautsprecher, Frequenzweichen, Bauelemente
Here are some tests, they look good and this driver has just enough sensitivity.
Test Seas Noferro-900 am Visaton Waveguide WG-148R – Heissmann Acoustics
Test Seas Noferro 900 (H 1025-06) – Heissmann Acoustics
The SPL above 10 kHz looks better than with the Scanspeak types.
Anybody knows more about these drivers?
Matthias, what do you mean by the wiggles in the on-axis response with the Visaton waveguide? Do you also see that in the measurement of Pida?
I have checked DI of Scanspeak 6600 also, same DI as with the 7100 as expected. In the plot the 6600 in blue and the 7100 in red.
I also found this tweeter, the Seas noferro-900 – H1025-06 and it can be used with the Visaton W148R. There is a 4 and 6 Ohm version. I cannot find it on the Seas website. It looks an OEM version.
And ... it is cheap!
I can find it here:
NOFERRO900 | Lautsprecher Intertechnik - Shop
Strassacker, Komponenten: Lautsprecher, Frequenzweichen, Bauelemente
Here are some tests, they look good and this driver has just enough sensitivity.
Test Seas Noferro-900 am Visaton Waveguide WG-148R – Heissmann Acoustics
Test Seas Noferro 900 (H 1025-06) – Heissmann Acoustics
The SPL above 10 kHz looks better than with the Scanspeak types.
Anybody knows more about these drivers?
Matthias, what do you mean by the wiggles in the on-axis response with the Visaton waveguide? Do you also see that in the measurement of Pida?
Attachments
Matthias,
You were first, I also found the Seas
92 dB is enough, with the AE TD15X the sensitivity will be 90.1 dB
And with the Faital 15FH520 90.5 dB.
...but impedance is low.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.