A few thoughts:
Question for all, but particularly @waxx - Is the argument for a tweeter without a waveguide (which was discussed a long time ago) more about sticking with the original design concept and respective sound characteristics, or (because we were talking about 3D printed waveguides at the time) was the argument more that many people would not want to deal with hassle of printing a 3D printed waveguide?
If the latter, then the Peerless H26TG/SS H2606 may be the best choice - it will make directivity matching easier AND help with the total cost. If the former, then the SB26ADC is probably the best choice, because it gives us the most flexibility among the non-waveguided tweeters AND gives us the option to put it into a 3D printed waveguide.
- SB Acoustics SB26STWGC-4: This is a good tweeter but ruins the value proposition of the target build. It is a little better priced in Europe at €84 but it is $94 in the US. We went a little over our target price for the midrange, but we should be able to get exceptional sound quality across the most sensitive regions with the Scanspeak 15W. This SB26STWGC would be the most expensive driver in the build and would add 21% to the total driver cost relative to the SB26STAC (it is double the cost of the SB26STAC). I do have this tweeter, so if we decide to use this one I don't even have to go out and buy it, but I would argue we have drifted from the "Bargain" build to just a good performing classic 3-way that is not expensive.
- Peerless H26TG45-06/Scanspeak H2606/9200: If we want to go with a small waveguide to attain better directivity, then this one makes more sense to me. The Peerless is actually less expensive than any of the SBA's in the US, so we are decreasing the total cost rather than increasing (in the US). I don't know if the Peerless is available in Europe or if it is a perfect drop in replacement for the Scanspeak Discovery, but I'm sure we can make the design interchangeable between the two tweeters with very minor modifications (if any).
- SB Acoustics SB26ADC: It was discussed early in the thread that there is likely some builders with a preference to build a "nostalgic" speaker to keep closer to the original design(s), regardless of the potential flaws. (See question below, regarding this aspect.) The SB26ADC is a very good tweeter and gives us some flexibility in that (1) it can be crossed quite low IF NEEDED, and (2) can be put into a 3D printed waveguide easily as an option.
- SB Acoustics SB26STAC: Similar to the ADC above, but a little less expensive and maybe there is a preference for fabric vs aluminum dome. (My thinking is if we use either of the SB26xxx tweeters we can tweak the crossover to accept either since I have both on hand.)
Question for all, but particularly @waxx - Is the argument for a tweeter without a waveguide (which was discussed a long time ago) more about sticking with the original design concept and respective sound characteristics, or (because we were talking about 3D printed waveguides at the time) was the argument more that many people would not want to deal with hassle of printing a 3D printed waveguide?
If the latter, then the Peerless H26TG/SS H2606 may be the best choice - it will make directivity matching easier AND help with the total cost. If the former, then the SB26ADC is probably the best choice, because it gives us the most flexibility among the non-waveguided tweeters AND gives us the option to put it into a 3D printed waveguide.
By the way...we have measured both the SB26ADC and SS 15W in the test cabinet.
Here is a simulation with actual tweeter and midrange measurements but traced woofer. This is about as simple a xo as possible, in about 30 minutes. LR4 at 3 kHz. I would think when we get to rigorous xo design, matching the midrange and tweeter (without a waveguide) is not going to be a significant constraint on getting a good sounding and good measuring speaker.
Here is a simulation with actual tweeter and midrange measurements but traced woofer. This is about as simple a xo as possible, in about 30 minutes. LR4 at 3 kHz. I would think when we get to rigorous xo design, matching the midrange and tweeter (without a waveguide) is not going to be a significant constraint on getting a good sounding and good measuring speaker.
To put that into comparison to the Wharfedale Linton 85 (8 inch woofer, no waveguide on tweeter) and JBL L100 (12 inch woofer, waveguided tweeter), both of which are reported to sound good...
I don't know at this point, I am not familiar with the details of the JBL.
I just wanted to make the point that getting "good enough" directivity doesn't seem like it will be particularly difficult, even without a waveguided tweeter. Note for the Project DI, (1) everything below 400Hz or so is inaccurate as it is just simulated in VituixCAD, and (2) the dip at 1.4 kHz is from the baffle step diffraction (more clearly seen in the SPL graph of post #1,142) and we can alleviate that by offsetting the midrange (if we want and further xo tweaks. I also used LR4 crossovers for simplicity but I am sure we will explore different topologies ) and different xo details.
I actually find the comparison to the Wharfedale Linton more interesting. There is a clear directivity mismatch between 1 kHz and 2 kHz that people who pay a lot of attention to DI like to point out, and would claim that this speaker must sound bad (even though they have never heard it). Yet this speaker is widely accepted to sound very good.
I just wanted to make the point that getting "good enough" directivity doesn't seem like it will be particularly difficult, even without a waveguided tweeter. Note for the Project DI, (1) everything below 400Hz or so is inaccurate as it is just simulated in VituixCAD, and (2) the dip at 1.4 kHz is from the baffle step diffraction (more clearly seen in the SPL graph of post #1,142) and we can alleviate that by offsetting the midrange (if we want and further xo tweaks. I also used LR4 crossovers for simplicity but I am sure we will explore different topologies ) and different xo details.
I actually find the comparison to the Wharfedale Linton more interesting. There is a clear directivity mismatch between 1 kHz and 2 kHz that people who pay a lot of attention to DI like to point out, and would claim that this speaker must sound bad (even though they have never heard it). Yet this speaker is widely accepted to sound very good.
That waveguide is a shallow one, and one integrated in the tweeter itself (aka not difficult to get). Other waveguides often need 3D printers and knowledge of/acces to CAD software, even if it's only to make an adaptor. This is as easy as mounting a flatplate tweeter.
The sound is also not very limited in directivity like many waveguides, it's fairly wide, but a bit more controlled than on a flat plate. So this kind of waveguide is not a problem for me and many (and was also in some of the originals like the L100).
I'd be quite interested to hear opinions on whether the significant additional price for the SB26STWG would be worth it. Keep in mind, if you don't actually plan to build these, look at it from the perspective of a likely builder. If you are a likely builders, please share your thoughts.
As I have said, these are not a build for myself, but because I have the tools and most of the drivers to help move the process along. I have the SB26STWG on hand, so it is no extra cost for me. But I know when I first got into speaker building the differences between $300, $400, and $500 was not insignificant.
As I have said, these are not a build for myself, but because I have the tools and most of the drivers to help move the process along. I have the SB26STWG on hand, so it is no extra cost for me. But I know when I first got into speaker building the differences between $300, $400, and $500 was not insignificant.
I did a Google search and looks like the Peerless H26TG45-06 is available with Digikey - that should help with availability; however the shipping and taxes will vary of course.
Digikey is to hard to order from if you only want one or two pieces, they add significant shipping costs. It's not an option for most diy builders.
Nice, thread, guess this is where the party is at, have browsed through, quite some invested time. The target house curve looks great and is quite tempting, we can assume primary sources will be vinyl. Is there danger of boominess near boundaries and certain rooms? Modern music mastering and rew will already have engineered that target curve into the mix, share your vertical dispersion? heres the frequency , spl and power of a smaller 3 way
About those DIs, some of them could be from only horizontals, or 2pi vs 4pi.
Monkey box 3-way with 12" woofer and 5" mid is what it is, tweeter and it's xo plays the most important role in DI
Monkey box 3-way with 12" woofer and 5" mid is what it is, tweeter and it's xo plays the most important role in DI
It’s practically the same as the Scanspeak. About half the price of the SB with waveguide. And I really doubt if -properly implemented- anyone would hear the difference with the SB in a double blind test.
So let's do an old school monkey coffin with a Monacor ht-22 dome tweeter at 15eu ^^.