SB is developing a coax Satori driver, based on the MW19.
what do you guys think of it?
SB Acoustics :: 71/2” SATORI COAXIAL
what do you guys think of it?
SB Acoustics :: 71/2” SATORI COAXIAL
Attachments
Last edited:
I'd seen it.. but that 3+ db depression even ON-axis between 7-14kHz, isn't good. (..and it starts even lower in freq., practically speaking this driver combo looks good to about 4.5 kHz, before higher freq. loss, with modest correction and a good filter.)
The nice thing is that its moderatly efficient (though it is 4 ohms) and the lower freq. response for the tweeter actually extends pretty low, presumably low enough to faciliate a good crossover design.
The problem then is you'll be p!ss!ng away all that efficiency to correct the response.. OR you'll be adding a higher freq. tweeter to it (perahaps around 7 kHz). (..and the former idea sounds ridiculous to me, while the later idea can be difficult to acomplish with a good objective result because of the driver distance between drivers.)
The nice thing is that its moderatly efficient (though it is 4 ohms) and the lower freq. response for the tweeter actually extends pretty low, presumably low enough to faciliate a good crossover design.
The problem then is you'll be p!ss!ng away all that efficiency to correct the response.. OR you'll be adding a higher freq. tweeter to it (perahaps around 7 kHz). (..and the former idea sounds ridiculous to me, while the later idea can be difficult to acomplish with a good objective result because of the driver distance between drivers.)
Last edited:
With standard domes and coaxials you always get on axis issues near the top octave and nothing can be dome about that save for fancy phase shields or dome geometries. It's never normally much of an issue.
The trouble here is that the tweeter itself has a rising response towards low frequencies (as expected because of the wave-guide loading), but it also has a rising high frequency response. You'll see that the dip gets less of a problem as you go off-axis (typical), but the real problem here is that the tweeter has low overall sensitivity, most likely to give it its low frequency extension whilst remaining very compact.
You aren't suppose to listen to a coaxial on axis anyway and if designing 30 degrees off axis you could end up with an 88dB sensitive 2 way with full BSC and a ~2dB dip centred around the general tweeter dip. Not too shabby, but it would be better if the tweeter was 94dB too.
The SEAS C18EN002/A also has terrible tweeter efficiency if you EQ it completely flat to 20KHz. As I've said elsewhere it would make make for a better driver if they increased the voice coil diameter even more and included a dimple dome in the middle, or at least did something to increase the sensitivity of the tweeter. Personally the crap tweeter sensitivity isn't a huge deal as I would always go active with this driver. Or replace the tweeter...but I'm not sure what with.
The trouble here is that the tweeter itself has a rising response towards low frequencies (as expected because of the wave-guide loading), but it also has a rising high frequency response. You'll see that the dip gets less of a problem as you go off-axis (typical), but the real problem here is that the tweeter has low overall sensitivity, most likely to give it its low frequency extension whilst remaining very compact.
You aren't suppose to listen to a coaxial on axis anyway and if designing 30 degrees off axis you could end up with an 88dB sensitive 2 way with full BSC and a ~2dB dip centred around the general tweeter dip. Not too shabby, but it would be better if the tweeter was 94dB too.
The SEAS C18EN002/A also has terrible tweeter efficiency if you EQ it completely flat to 20KHz. As I've said elsewhere it would make make for a better driver if they increased the voice coil diameter even more and included a dimple dome in the middle, or at least did something to increase the sensitivity of the tweeter. Personally the crap tweeter sensitivity isn't a huge deal as I would always go active with this driver. Or replace the tweeter...but I'm not sure what with.
OTOH, a typical tweeter on a typical baffle will have well more than 3dB dip somewhere between 2-4khz. And if you eq it flat you have a big hump off axis. What it really comes down to is a wide spot in the dispersion shows up somewhere, I'd rather have it up near 10khz than down at the upper midrange/lower treble.
-that's sounds (looks) about right. But that's 4 ohms.
Depressing (litterally) IMO.
I'd tend to think of this in the "inverse": that you end-up with some lower freq. gain (..though at the top of that you do typically have some negative pressure narrow-band "dip" as it starts into a ripple alteration.)
Totally dependent on baffle shape/size + driver position though.
(For those who are less familar with this: )
dlr has a good example of what you are describing with that classic narrow baffle hard-edge design (vs. felt modifiers):
David Ralph's Speaker Pages - Felt Effects on Baffle Diffraction
Understanding Cabinet Edge Diffraction
Last edited:
This reminds me I forgot to type my main point: that the baffle won't matter anymore with a "waveguide" tweeter like in this coax. That is the big advantage IMHO.
Would be nice to see something in SB's 16cm size though.
An MR13P coax would only require a 3.5kHz xover for a decent directivity match with its coaxial tweeter. This is something even small 19mm domes can do, so hopefully it could have decent sensitivity.
An MR16P coax could work with a 2.5-2.8kHz xover frequency, still within the realm of a good 19mm dome.
If SB could fit their 22mm dimple dome onto a small neo motor and whack it into the voice coil of those two they'd be onto a winner. Hopefully with better sensitivity too.
-that's sounds (looks) about right. But that's 4 ohms.
Depressing (litterally) IMO.
Indeed Scott, it is depressing
Although to be fair it's in line with most other speakers of its size, you cannot get something for nothing. I mean that would be a full bafflestep 88dB 4 ohm speaker, or 85dB 8 ohm. Built as a 2 way that's in line with, and actually better than, a lot of other 6-7" + 1" two ways.
A full BSC design with SBs Satori MW19P would = 83dB, with a SEAS Excel magnesium cone 80-82dB, Scanspeaks top drivers 79-83dB.
Doesn't look so bad when you compare it to those. Of course they'll probably give you more bass but stick the coax in a small sealed box and use it along side a decent small sub...I'd take that vs any standard 6-7" + standard 1" dome. If SB have done their job right the woofer on the coax will have stellar HD performance and by the looks of things the flat, corrugated surround, actually helps with the usual surround resonance issue present in all the other Satori drivers.
Exactly, this is often overlooked when considering waveguides, the help they provide with giving freedom from cabinet edge diffraction.
I believed that too and a wg does reduce diffraction but it doesn`t eliminate it, well, at least the shallow types. I recently build a medium sized 2-way using Monacor drivers and with a xover point of 2Khz LR2 and an insignificant roundover of 1/2", there was a good 3db bump at 1.2Khz caused by the tweeter. I bought some more of these waveguides (it sounds really nice) and will carry on experiments very soon as to what extent it does affect edge diffraction. Perhaps if its more narrow, it will eliminate it but then it will offset the tweeter too much in the z-axis. I`ll try to carry on some experiemtns on edge rounding too as there`s still a debate if smaller radiuses do provide any audible benefits (yesterday got a massive 7/8" Freud roundover bit which costed an arm and a leg...).
A bit of variation around 2-4kHz is often easy to fix though as you can just alter the Q of the highpass filter to suit. A dip at 7kHz is somewhat more difficult to fix. The most logical thing to do is bring down everything below 7kHz with a low Q filter but that really trashes the sensitivity.
FWIW: On a 9" wide baffle, with careful placement of a 1" dome tweeter you can achieve about +2-3dB at around 1kHz and virtually flat (+/-1dB max) above that. A good rule of thumb is golden ratios between the distance from the centre of the tweeter to the left, right and top edges of the baffle i.e. 1:1.618:1.618^2. Zaph's ZD5 shows this with it's offset tweeter placement:
http://www.zaphaudio.com/ZD5-measured-FR-IMP-rawinbox.gif
Last edited:
If you imagine that whatever the diffraction signature of a baffle is on-axis, it will become the inverse (nearly) as you move off axis. So you can see that if you use the filter Q to make that on-axis dip flat, you will have produced a very large bump off axis, because the diffraction character is already causing a bump rrelative to the on-axis response. This can happen pretty rapidly, by 20 degrees, so that the entire "first reflections" have a pronounced woofer dip/ tweeter hump.
So from my POV the question then becomes, if you are going to have a power response anomaly either way, where is more acceptable? My personal preference would be to have a smooth 2-4khz area over a smooth 7-10khz. YMMV.
And in this case adding to that a concentric, time aligned, point source, this will definitely do a lot of important things very right.
Offsetting can work well as you mention, I just don't care for asymmetric radiation in principle, so I try to avoid. Besides it is critical to measure and model off axis when you do this, and almost no one does.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.