Dave Fabrikant measures the RAAL 70-20XR

[kmibb]

measurements were a bit limited and it measured as good or better as a $60 dome tweeter.

[r-carpenter]

nice one.

[Pete Schumacher ®]

measurements were a bit limited and it measured as good or better as a $60 dome tweeter.

You don't listen to plots. But if you insist, the $60 dome (ANY dome) does not have the impulse response of the RAAL, nor the extended high end, or the low level detail. None of them sound like one either.

exactly. All these drivers including Raal are tools and if used properly are benefit.

True. But not all tools deliver the same results. If you really think the $60 dome is the equal of the RAAL, you've never sat down and actually listened.

[r-carpenter]

You don't listen to plots. But if you insist, the $60 dome (ANY dome) does not have the impulse response of the RAAL, nor the extended high end, or the low level detail. None of them sound like one either.

True. But not all tools deliver the same results. If you really think the $60 dome is the equal of the RAAL, you've never sat down and actually listened.

I think you'd be really hard pressed to hear the difference between Raal and a nice dome if both had same liner on and off axis response (which apparently is a strong point of Raal)and worked within HD below 40db. Blind of cause and level matched. I am not interested enough in such a test to actually buy Raal but if someone has it already I could set up something. Using the same mid woofer, same cabinet and of cause controlled conditions.
This way we can definitely say: it measures so and so and it sounds so and so compare to such that measures as this. Know what I mean?

[Pete Schumacher ®]

I think you'd be really hard pressed to hear the difference between Raal and a nice dome if both had same liner on and off axis response (which apparently is a strong point of Raal)and worked within HD below 40db. Blind of cause and level matched. I am not interested enough in such a test to actually buy Raal but if someone has it already I could set up something. Using the same mid woofer, same cabinet and of cause controlled conditions.
This way we can definitely say: it measures so and so and it sounds so and so compare to such that measures as this. Know what I mean?

You'd think so, but it's not the case Roman. We have the Scan Speak Illuminator Air Circ 6600 here to compare with the RAAL, arguably a true world class dome tweeter. The Air Circ has a very extended top end, and smooth response, but even so, doesn't SOUND like a RAAL. And I really have no doubt I could pick out the RAAL from a dome when listening to material I select.

You don't actually LISTEN with measurements. They give you an idea on performance, but not on how they sound. It's the amount of low level detail that you get from the RAAL that sets it apart from a dome tweeter. I don't know how you measure that, other than with your ears, but it's obvious to them, even if not to a mic and measurement software.

[NyxOne]

I don't know how you measure that, other than with your ears, but it's obvious to them, even if not to a mic and measurement software.

+1

Ear and measurement gear are not on the same page. Eg: Voicing is done with my ears, not my gear... and i don't listen to distortion plot.

Note: i don't own any Raal product, i just believe measurement don't tell everything.

Chuck

[GranteedEV]

I think you'd be really hard pressed to hear the difference between Raal and a nice dome if both had same liner on and off axis response (which apparently is a strong point of Raal)and worked within HD below 40db

A dome is an inherently flawed HF transducer. When placed on a baffle, the re-radiation of its bottom end off the baffle is inconsistently delayed compared to the radiation of the driver itself. How do you presume to diffuse the delayed reradiation off the baffle of the dome? THe result is classic "dome hiss".

The soft dome material measures great steady state, but when it's playing back real material it doesn't maintain rigidness at higher excursions. Sure you can cross it over high, but then your speaker has a power response issue. Cross it over low (like the 1.6khz of the RAAL) and you run into other issues due to the increased acceleration. You can measure these things, just not with the typical steady state stuff. You need laser holography.

Switch to a metal dome and you get intermodulation distortion from the ultrasonic breakups, so you really need a metal that breaks up so far above the audible passband that its IMD is above the audible passband - beryllium - and it ain't cheap. Of course DIYers are perfectly happy with silk and that's fine, but it's not without its issues.

Also, in a flat baffle you've also got heavy vertical reflections. A ribbon can work in a waveguide thanks to its flat diaphram. A ribbon will also have narrowing vertical dispersion which may help reduce floor/ceiling reflections in the treble ~4-6khz. So it won't have the same off axis response anyways. A dome just won't work optimally in a waveguide again because of its shape and the wavefront that loads the waveguide.

Sorry but HD/On-Axis/Off-axis response are only three factors in what we hear. It's not untrue that everything can be measured if it's there. It's untrue that everything that's measured by DIYers with DIY equipment is what's there.

[Pete Schumacher ®]

A dome is an inherently flawed HF transducer. When placed on a baffle, the re-radiation of its bottom end off the baffle is inconsistently delayed compared to the radiation of the driver itself. How do you presume to diffuse the delayed reradiation off the baffle of the dome? THe result is classic "dome hiss".

The soft dome material measures great steady state, but when it's playing back real material it doesn't maintain rigidness at higher excursions. Sure you can cross it over high, but then your speaker has a power response issue. Cross it over low (like the 1.6khz of the RAAL) and you run into other issues due to the increased acceleration. You can measure these things, just not with the typical steady state stuff. You need laser holography.

Switch to a metal dome and you get intermodulation distortion from the ultrasonic breakups, so you really need a metal that breaks up so far above the audible passband that its IMD is above the audible passband - beryllium - and it ain't cheap. Of course DIYers are perfectly happy with silk and that's fine, but it's not without its issues.

Also, in a flat baffle you've also got heavy vertical reflections. A ribbon can work in a waveguide thanks to its flat diaphram. A ribbon will also have narrowing vertical dispersion which may help reduce floor/ceiling reflections in the treble ~4-6khz. So it won't have the same off axis response anyways. A dome just won't work optimally in a waveguide again because of its shape and the wavefront that loads the waveguide.

Sorry but HD/On-Axis/Off-axis response are only three factors in what we hear. It's not untrue that everything can be measured if it's there. It's untrue that everything that's measured is what's there.

You might be interested to know that Alex at RAAL recommends never using his tweeters in a waveguide, because it messes up the impulse response. That's not to say it couldn't work of course, just perhaps there's no real advantage by doing so.

Something very interesting to me, was when measuring the 70-10D on an 8" wide baffle with no roundover, there was very little of the typical diffraction ripple you get with a dome tweeter on the same baffle.

[Paul Carmody]

This has gone on too long. Is Roman not allowed to be skeptical? Or must we just bow down and worship the Raal, no questions asked?

And even if it is the greatest tweeter ever made. SFW? It's just a speaker, guys. Lighten up.

[GranteedEV]

You might be interested to know that Alex at RAAL recommends never using his tweeters in a waveguide, because it messes up the impulse response. That's not to say it couldn't work of course, just perhaps there's no real advantage by doing so.

Designing the appropriate waveguide for it would require some serious R&D but I think it could work extremely well, + with geddes style foam. You have to remember that even a flat baffle itself is a 180 deg waveguide` so that too messes up the impulse response. But since the driver is flat not dome, it creates the ideal wavefront for loading an optimal waveguide. After that you need to make sure that the waveguide itself is optimal and it would be a big change from the waveguides designed for round drivers of course. You're trying to keep diffraction low so I don't know if it would give much if any boost in sensitivity but the main thing is that it would give wide-but-controlled dispersion in a live room

I know for yourself that's not a big advantage but i like the idea of mating a 140-15d to a TD10M for 90 degree directivity.

Additionally, you can see that alexander`s own raal 2010 omni appears to be in a waveguide of its own sort:

[Face]

I know for yourself that's not a big advantage but i like the idea of mating a 140-15d to a TD10M for 90 degree directivity.

I have 2 pairs of TD10M's and a pair of 140-15D's here. Crossing the large RAAL at 3K sounds good to me, but I haven't tried lower yet. IMO, crossing them where their directivity would match(1.5-1.8K?) may put too much of a strain on the ribbon and the 10M's would be at the point of beaming, no?

[GranteedEV]

I have 2 pairs of TD10M's and a pair of 140-15D's here. Crossing the large RAAL at 3K sounds good to me, but I haven't tried lower yet. IMO, crossing them where their directivity would match(1.5-1.8K?) may put too much of a strain on the ribbon and the 10M's would be at the point of beaming, no?

Well, I don't know just how loud you expect to push it, but according to the RAAL website,

*Recommended crossover: 4th order L-R @ 1600 Hz

and I know Dennis Murphy chose to cross the 70-20XR (which I believe has less radiating surface area) at 1.8khz 4th order which matches up to the RAAL website's recommended frequency.

But I suppose an 8" woofer might be an option... you'd have to measure and experiment. At the end of the day though you'd have to start with a waveguide optimized for the RAAL.

[johnnyrichards]

This has gone on too long. Is Roman not allowed to be skeptical? Or must we just bow down and worship the Raal, no questions asked?

And even if it is the greatest tweeter ever made. SFW? It's just a speaker, guys. Lighten up.

Nope, because it is RAAAAAAAAAL.

[gregbegland]

Don't forget that those numbers are the recommended MINIMUM crossover frequency. Usually best to stay above that point for optimum performance.

Greg

[GranteedEV]

This has gone on too long. Is Roman not allowed to be skeptical? Or must we just bow down and worship the Raal, no questions asked?

... we must bow down and worship the RS28F since tweeters above $60 all sound the same?

I get that not everyone wants to shell out the money for these things - I don't either. But that doesn't mean we just go around attacking them because of that fact.

And even if it is the greatest tweeter ever made. SFW? It's just a speaker, guys. Lighten up.

Like any driver it has its limitations. This one does too but it has lots of positives.... I'm not allowed to make a thread about it here because it's not a $60 dome tweeter?

Don't forget that those numbers are the recommended MINIMUM crossover frequency. Usually best to stay above that point for optimum performance.

Not really - you'd have to define optimum performance. It's an arbritrary number elsewise. Like I said, the Salk Soundscapes use exactly the recommended crossover frequency, and DM is a better crossover designer than I could ever hope to be. Are you saying he didn't get optimum performance?

[r-carpenter]

You don't listen to plots. But if you insist, the $60 dome (ANY dome) does not have the impulse response of the RAAL, nor the extended high end, or the low level detail. None of them sound like one either.

hence the name of the thread.
http://www.youtube.com/watch?v=FIAiUvAv46c&feature

and nobody is interested in doing DBT against a dome? Reminds about DBT of amplifiers then people can't recognize which one is playing but ohhhh one is so much better then the other one.
http://www.youtube.com/watch?v=zFl6p4D59AA&feature

BTW, Thanks Paul.

[gregbegland]

..
Not really - you'd have to define optimum performance. It's an arbritrary number elsewise. Like I said, the Salk Soundscapes use exactly the recommended crossover frequency, and DM is a better crossover designer than I could ever hope to be. Are you saying he didn't get optimum performance?

Just saying that a point that low is really cutting it close to where the distortion profile begins to rise on these things. I have no direct experience with the 70-20, but here are sweeps of my 140-15. If its distortion is rising just below 1.5KHz, then what of the smaller 70-20? In my own experience, the 140-15 just sounds more composed and less strained crossed a little higher, like say about 3KHz. Not allowed to talk about sound on this thread though, so here is something a little more concrete.

Greg

[gregbegland]

Here we are with a 2KHz/24dB slope and then a 3KHz/24dB slope in place. Both crossovers performed in the Digmoda DSP, not passive.

Sorry about the switch from % to dB scale. I messed up when I saved these plots.

[Pete Schumacher ®]

Roman, I've no doubt that I could pick the RAAL from among the domes blindfolded.

You could too.

They sound that different.

What it boils down to here, is that those of us who've actually listened to both will stand by the status of the RAAL as the tweeter to beat.

The rest need to go listen to a Salk Soundscape or Vapor Cirrus before telling those of us who have listened that we don't know what we're talking about.

And yes, about the title of the thread, yes, it's got measurement in the title.

That measurement of the RAAL in the time domain is the one that won't be replicated by any dome you name. It's already been pointed out that steady state measurements like THD and response, only tell part of the story. Music isn't steady state, but a series of transient events. And it's during MUSIC that the RAAL comes into its own, and where the details in the difference in sound come in.

I've got the RS28F on hand, the RS28A, the 27TBFC/G, AirCirc 6600, XT25 . . . and I can hear stuff on the RAAL system that none of these other tweeters can deliver. All the speakers are EQ'd to flat in the XO, and the same songs on the different speakers have different presentations, especially in the tweeter range. And it's not subtle, it's obvious.

Bring on the double blind Roman. You build and voice a couple of monitors, one using the dome of your choice, and another using the RAAL. When you get them done and have convinced yourself that you can't tell a difference, let us know and we'll stop by for the challenge. However, my guess would be that you'd call off the test once you built the two sets and actually listened after getting the response plots within .5dB of each other.

[billschulte]

Don't forget that those numbers are the recommended MINIMUM crossover frequency. Usually best to stay above that point for optimum performance.

Greg

The information on the 20xr is limited on the web and correct me if I am wrong but it seems that the crossover frequency is sort of built into the design to be at 1800hz(?)