Selah Audio RC3R 3-way Speaker Review

[napilopez]

The smoothness variable (SM) in Olive's formula includes the 'slope' variable (SL) in its calculation, which he defines as the difference between the 'target slope' (-1.75 for the predicted in-room response i.e. a gentle downward slope) and the actual measured PIR slope. I suspect it's this larger difference from the target slope for the Selah (basically too flat PIR, and so perceptually too bright) compared to the other speakers, which appreciably lowers its smoothness score, as opposed to any significant deviation from the regression line (what we would intuitively call smoothness). So 'smoothness' is a bit of a misnomer here, as it implicitly includes spectral tilt in the form of the slope variable in Olive's definition.

Thanks for reminding me!

Interestingly this is also one part of the Olive score I tend to have more leeway with, because Olive himself says in the paper that the PIR tilt might need to vary with directivity.

Specifically, speakers with wider directivity may require less of a tilt, although it's unclear to what degree. But for example, the target slope for Test one (the one with 13 speakers but more controlled tests) was -2.1 vs -1.75 on the newer one. Of this he says:

The degree of tilt varies among curves for Test One and the larger sample. Test One includes mostly 2-way designs whereas the larger sample includes several 3-way and 4-way designs that tend to have wider dispersion (hence smaller negative target slopes) at mid and high frequencies. This suggests that the ideal target slope may depend on the loudspeaker’s directivity.

Furthermore, Dr Toole's has also said:

Cone/dome loudspeakers tend to show a gently rising directivity index (DI) with frequency, and well designed horn loudspeakers (like the M2) exhibit quite constant DI over their operating frequency range. There is no evidence that either is advantageous - both are highly rated by listeners.

Translating that to the predicted in-room curve, constant directivity speakers tend to be flat rather than sloped downwards in the upper half of their Early Reflections/PIR curve, as can be seen in the M2's spins. They don't have a smooth curve, but this doesn't seem to be a bad thing.

So I think the Olive formula is probably most accurate for speakers with fairly typical radiation patterns. If I were a betting man, I'd be willing to bet that the choice of ideal target curve and reference axis accounts for a decent chunk of the remaining 14 percent of the preference score that isn't accurate. It'd be interesting to see which speakers were the outliers.

 

[Juhazi]

Selah measurements look ok to me - valid. They show very well how this kind of construction works. Sealed bass has smooth roll-off, dome mid is crossed too high propably to avoid distortion/damage. Tweeter is "hot", serial resistor should be bigger. Woofer's impedance drops too low , which might be problematic for some amps. This kind of construction aims for low and smooth horizontal directivity, but this speaker is not perfect.

Directivity mismatch around 1kHz is a big problem, sound will change a lot if listener moves sideways and if sidewalls have high reflectivity.

Discussion of high/low overall directivity is a basic difference and a matter of preference to listeners. People who are picky about imaging and "clarity" prefer high directivity. Toole has this nice thread of listening test between JBL M2 vs. Revel Salon 2. Low directivity was voted beter.
https://www.avsforum.com/forum/89-s...ccurate-well-reviewed-speakers-ever-made.html

 

[thewas]

Furthermore, Dr Toole's has also said:

Cone/dome loudspeakers tend to show a gently rising directivity index (DI) with frequency, and well designed horn loudspeakers (like the M2) exhibit quite constant DI over their operating frequency range. There is no evidence that either is advantageous - both are highly rated by listeners.

Although it must be also said that in a big blind comparison of another forum there was a significant preference for a top Revel model which measured less smooth but had wider and constantly increasing directivity to the JBL M2 which had higher but constant directivity.

Similarly I personally also liked the classical Buchardt S300 tonally more to the constant directivity S400, it can be a coincidence and maybe also depending on listening distance, but I think its a factor that could be investigated more in the future.

 

[MZKM]

the SM_PIR value aka 'smoothness' of the predicted in-room response is partially dependent on spectral tilt

That’s what I initially thought, as the ideal slope is part of the calculation, but when I change the target slope the final result is the same (multiplying and diving by the same thing will cancel out, and slope is part of the numerator and denominator). However, we should wait until we get a speaker with an in-room response that is very smooth but not tilting down to be extra sure.

 

[q3cpma]

Looking at Revel and JBL, what are the advantages of building 3+ way speakers these days? If they can build the M2 in a 2-way, why is top of the Revel line a 4-way? While they share a philosophy for output, construction is wildly different.

The M2 is kind of a 3-way with its "double diaphragm" compression driver. Anyway, 3-way allows for higher SPL with less THD/IMD; and personally, I think it's best since the drivers can really be designed to do different things (low, medium and high) while in a real 2-way, the woofer and tweeter share the medium load which brings compromises in their physical design.

 

[maty]

To listen to music, 2-ways design is less problematic or better option. Maybe 3-ways as coaxial + woofer, but 2-ways with two subs or more I think is more interesting, if there are no complaints about so much box, you know.

 

[Thomas_A]

FIY. The in-room response is shown with a tight scale on the graph. 400 Hz-20 kHz is +/- 1 dB or so.

Edit: saw that this was mentioned already...

 

[Hiten]

Thanks for the review.

Regarding 2 way and 3 way speakers.
Since this is the only 3 way speaker reviewed and number of other of 2 way designs, and assuming low frequency extension is moderately close of each design. Wouldn't it be interesting to see low frequency distortion level at appropriate sound level of 3 way and any other suitable 2 way already reviewed ?
Thanks and regards.

 

[thewas]

The measured Kali and Revel are also 3 way speakers.

 

[Rick Sykora]

Looking at Revel and JBL, what are the advantages of building 3+ way speakers these days? If they can build the M2 in a 2-way, why is top of the Revel line a 4-way? While they share a philosophy for output, construction is wildly different.

A 3-way speaker can avoid having phase and driver integration issues in the midrange that are difficult to avoid in a 2-way speaker. That said, trying to find midranges to cover the range where are hearing is most sensitive is not all that easy either. Thereby rationalizing the way towards a 4-way design. All along the way, almost all the drivers have directivity, cone breakup and distortion issues to avoid or your crossover becomes more complex and expensive.

Even though I like the greater transparency that DIY designs offer, a well designed COTS speaker has the advantage that the cost of design iterations (or expensive software/equipment to reduce iterations) can be (hopefully) spread out over its sales volume. You have to love DIY a lot to throw away your cabinet or driver that does not meet your goals! Selah Audio fits somewhere between open DIY and COTS speakers. So, while the RC3R design is a bit older, thanks for sharing as it adds an interesting value point to the ASR database.

 

[ctrl]

You have to love DIY a lot to throw away your cabinet or driver that does not meet your goals! Selah Audio fits somewhere between open DIY and COTS speakers.

Thank God these times are also over for DIY. Before the first cabinet is built, simulations of different loudspeakers are created and compared.

It is possible to simulate different types of crossovers, crossover frequencies and even the smallest changes to the cabinet in advance.
All that is needed is a fast CPU, many cores and a lot of RAM

 

[Rick Sykora]

Thank God these times are also over for DIY. Before the first cabinet is built, simulations of different loudspeakers are created and compared.

It is possible to simulate different types of crossovers, crossover frequencies and even the smallest changes to the cabinet in advance.
All that is needed is a fast CPU, many cores and a lot of RAM

Agreed, DIY has certainly been able to leverage better tools for designing and measuring too. If the designer cares enough, the results may be better than a COTS one that does not care so much. My point however was that some of the tools are expensive (Klippel, laser interferometry, etc.) and only a large corporation can readily absorb that cost typically. Also, you still have to build it in the end, and a smaller DIY shop takes a harder hit when the simulation comes up shy. Would love to hear what tools you are using, but suggest we take over to the DIY forum.

 

[napilopez]

Although it must be also said that in a big blind comparison of another forum there was a significant preference for a top Revel model which measured less smooth but had wider and constantly increasing directivity to the JBL M2 which had higher but constant directivity.

Similarly I personally also liked the classical Buchardt S300 tonally more to the constant directivity S400, it can be a coincidence and maybe also depending on listening distance, but I think its a factor that could be investigated more in the future.

Indeed, I've brought up that comparison many times myself to advocate for wider directivity. However, overall, I consider directivity width to be a matter of preference rather than an almost-absolute as I do with linear frequency response.,since a few studies have shown some people have a clear preference for more or fewer sidewall reflections. As I'm sure you know, usually more recreational listening, fewer for mixing. In a stereo setup, wider directivity also has an advantage in that it helps fill in the Interaural crosstalk dip in a typically reflective room. It'd be interesting to repeat a comparison like the M2 vs Salon2 and then see if the people who chose the M2 also preferred other constant directivity/narrow directivity speakers. I'd be willing to bet they would.

 

[jhaider]

Selah Audio RC3R PIR curve overlaid Spinorama on 50dB scale plus polars normalized at right side:

View attachment 48311

I wonder what happens to the curves if you set the reference axis at 30 degrees, and bias the "listening window" to be say 0-60º instead of ±30º. A couple years ago I heard Rick set up a demo of speaker with Satori midwoofer and his OEM ribbon tweeter. He did not toe them in. Overall in an unfamiliar room with "audiophile" type music I thought they sounded good as long as I was seated. Standing up they dulled. I don't usually like tall strip tweeters for that reason.

The measurements remind me of the first speaker I reviewed, Bryston's Mini A. I think that review was the first polar map to appear in SECRETS. Bryston Mini A was also a 3-way bookshelf, but of slightly different design. The tweeter was a dome in a short round waveguide, and the cone midrange was more of a "filler driver" than a full output midrange. (See measurements.) However, it had the same general on axis response profile. They were darned near unlistenable toed in, not great toed in to cross ahead of the listening position, but very good pointed straight ahead.

PS: General question: what's COTS?

 

[ctrl]

point however was that some of the tools are expensive (Klippel, laser inferometry, etc.) and only a large corporation can readily absorb that cost typically

Agreed, DIY cannot afford the complete development including own loudspeaker chassis. Klippel units are particularly helpful in chassis development.
If you want to develop loudspeakers just for yourself, you can get along with the mostly free tools.

Would love to hear what tools you are using, but suggest we take over to the DIY forum.

All right! Will write something about this in your thread "Tools for building better DIY speakers".

 

[thewas]

Indeed, I've brought up that comparison many times myself to advocate for wider directivity. However, overall, I consider directivity width to be a matter of preference rather than an almost-absolute as I do with linear frequency response.,since a few studies have shown some people have a clear preference for more or fewer sidewall reflections. As I'm sure you know, usually more recreational listening, fewer for mixing. In a stereo setup, wider directivity also has an advantage in that it helps fill in the Interaural crosstalk dip in a typically reflective room. It'd be interesting to repeat a comparison like the M2 vs Salon2 and then see if the people who chose the M2 also preferred other constant directivity/narrow directivity speakers. I'd be willing to bet they would.

Wide vs. narrow directivity is one very interesting topic and comparison, but I would also like to see similar wide constant vs. increasing directivity comparison, personally I have the feeling that I have not so much problem with a narrow directivity as I have with a directivity which is increasing to a frequency and then remains constant and thus has a "kink".

 

[napilopez]

Wide vs. narrow directivity is one very interesting topic and comparison, but I would also like to see similar wide constant vs. increasing directivity comparison, personally I have the feeling that I have not so much problem with a narrow directivity as I have with a directivity which is increasing to a frequency and then remains constant and thus has a "kink".

As in the M2's response?

 

[thewas]

As in the M2's response?

Yes, or even worse when the kink is higher in frequency like on the Buchardt S400.

 

[Icboschert]

PS: General question: what's COTS?

Commercial off the shelf

 

[dshreter]

In concept, this speaker is very similar to the

The core part of the M2 is the waveguide. Design of these is not trivial, nor is construction. There are not many manufacturers with the expertise and resources to do this. But a critical problem is that the performance of the waveguide with frequency is tied to its size. You simply cannot make a waveguide go low if it is not wide enough. The width of the M2 cabinets is defined by the low frequency cutoff of the high frequency driver. So this places constraints on the low frequency driver. As well as being able to perform all the way up to the crossover, its directivity must match at the crossover, and this places limits on its diameter. One will note that pretty much all waveguide two ways have the same morphology. Without a scale they all look the same. This is partly determined by matching of waveguide cuttoff and bass driver high frequency directivity.

Using a stack of drivers of different diameters allows control of directivity simply by choosing the crossover points so that you cross over before a driver becomes too directional. The intrinsic directivity determined by the driver diameter allows design of the overall horizontal directivity.

One feels sure market expectations, WAF, and economics have a great deal to do with what is marketed. I'm sure we will be seeing more and more large waveguide speakers as the technology becomes better appreciated. But there is no free lunch. Waveguides have engineering tradeoffs of their own, so don't expect more conventional designs to vanish.

Thanks, I just learned a ton! For a speaker being crossed over to a sub at 80 Hz anyway, what are good examples of speakers that do this well? Is that 305 territory?