As the origin of this speaker is a measurement reference speaker that was one of the design goals. And there are not many components on the market which can meet these goals, so I'm sure it's one of the lowest distorting speaker available.
Compression drivers often have high H2 values, even at lower volumes. And even well known chassis have somtimes high THD at low volumes - people don't test for that or have noisy measurement mics. Like the 18Sound 8NMB420, it has 0,5-1% THD at 90dBSpl. And never get's <0,1%.
But of course it should be possible with the right components and design! You postet some measurements of your system when I remember right - can you put a link here then we can discuss there to not clog up this thread.
High sensitivity systems have a lot of potential for home use! They just need some care to behave well at low levels and take care of the resonances.
I know these features of REW - sadly Audio Precision is very inconvenient in doing similar analysis (as good as they are in all other regards, waterfall and similar analysis is not their strength).
Hope I have time to play around a little with REW and let's see what I can find.
Not at all! A normal measurement microphone takes 140dBSpl. Even cheap ones like MicW M215 are specified to 135dBSpl. My 1/4" mics go up to 160dBSpl.
Only very cheap electret mics are not able to take level ... they have a pretty narrow range of use cause noise is also bad.
Noise is also a problem with professional mics, esp the 1/4" which I sometimes need for high frequency measurements. MicW M215 is pretty good in that regard and is my recommendation for a good mic up to 20kHz. Take 2 and you can also do some excellent sounding AB recordings with them.
Only very cheap electret mics are not able to take level ... they have a pretty narrow range of use cause noise is also bad.
Noise is also a problem with professional mics, esp the 1/4" which I sometimes need for high frequency measurements. MicW M215 is pretty good in that regard and is my recommendation for a good mic up to 20kHz. Take 2 and you can also do some excellent sounding AB recordings with them.
It’s a balance……if the ground plane response has a slight null in the 8k region, the object rises above the plane. As you add 8k to the overhead, the object rises further. Like I said, it’s a mixing tip folks have been observing and sharing. For an engineer to cross fade automate an EQ for thousands of objects?………..that’s not a reality that’s going to happen until AI mixing algorithms arrive (warning….this won’t be long before). For now, an Atmos or Immersive engineer is more likely to add EQ pronunciations to taste for object impact provided of course that the object is on its own track. In many so these instances, these are synthesized renders and already very linear…..a little phase smear is good thing.
You folks out in the listening and design world have some odd notions on how our senses work, assuming there’s a standard of reference when in fact, the difference a person hears listening to a song with their glasses on and then off would have a greater variance than a narrow Q 2b notch anywhere from 800-8k…….hats, scarfs, heavy sweaters, collar vs non collar shirts?…….heavy earrings that pull the lobe down? These physical variances pale in comparison to the biological….blood pressure, age , fatigue, hydration, blood sugar……and yes….gender.
With advances in software and processing power, we’re shifting towards sighted analysis of nearly everything….and our dominant suggestive bias is becoming all powerful in our reasoning. This isn’t a positive trend because it ignores the fundamental fact that we are biased and do not behave anywhere near linear on the individual level or less so as a group. In other words, don’t sweat the small stuff……sensory engineering is a rabbit hole. Problem is, engineers read more math papers than they do biology because there’s a long standing wall that’s been growing for decades between the two. Sadly that wall is crumbling under its own weight instead of the two communities opening windows in it. The wall is called Scientism and now all scientists from all fields are getting behind it……thats what happens when you build a wall.
Environmental limitations always tip that scale…….the fact that high output horn systems are simply better all around isn’t a fact anymore when you‘s space is a 10ft by 12ft room crammed with other stuff. Same could be said for 20” wide open baffle systems……there’s a different performance factor here……these systems are typically loaded with HD but the presentation and soundstage outweigh it…..and yet you don’t see a lot of 20” wide open baffle systems……sorta the elephant in the room kinda thing. Same could be said for compact horns, the elusive and rare Wolverine of the audio kingdom.
Let's consider a pair of SB34RNXL. Max SPL is 120dB. Suppose you want to measure distortion.
How do you do it? If you measure at 1m you'll get all dips due to room modes. If you take it outdoors to measure, you need to get off the ground as high as you can do get a long gate eg. 10ms. Can be challenging to hoist such a large and heavy speaker a few meters above the ground. So then there's the ground plane - but of course if needs a large open space. If you do a nearfield measurement eg. 1cm you'll likely overload your mic- 120dB @1m is 160dB at 1cm.
If you measure at 31.6cm the it's not clear to me which is responsible for the distortion - is it the mic that "sees" 130dB, or the twin 12" woofers that are reaching their limits. ]
If you
How do you do it? If you measure at 1m you'll get all dips due to room modes. If you take it outdoors to measure, you need to get off the ground as high as you can do get a long gate eg. 10ms. Can be challenging to hoist such a large and heavy speaker a few meters above the ground. So then there's the ground plane - but of course if needs a large open space. If you do a nearfield measurement eg. 1cm you'll likely overload your mic- 120dB @1m is 160dB at 1cm.
If you measure at 31.6cm the it's not clear to me which is responsible for the distortion - is it the mic that "sees" 130dB, or the twin 12" woofers that are reaching their limits. ]
If you
What's with the rest of the speaker? It looks like your THD is burried in noise - so it's likely your real THD is lower.
What dirver and horn do you show here?
How much THD is ok for the ear is a different debate (and a good one!) - but we are here to see what a modern 3" can do and in this case how a big horn compares.
SPL goal is ATMOS - 105dBSPL at listening position! That's pretty hard for a direct radiating speaker ...
It's really not that hard.
*) First - you can measure only one driver - then add 6dB for 2.
*) Use ground plane measurement, get a lonly parking lot. Or your garden if you have nice neighbours. A gym at late hours, a company building after working hours ... ask around, you will find something. But also a good position in the room (middle/middle or 1/4-middle) can sometimes do it. Get the mic in 1m distance and correct for the 6dB gain. Or 50cm distance and correct 6dB more (but check that, make 1m and 50cm measurements and check if the level behaves as it should). Tilt the speaker and put the microphone capsule directly on the ground, not 5mm in the air. You can do fullrange measurements that way.
*) Nearfield - as you see the 2 12" don't reach 120dBSpl at low frequencies. But be careful your mic doesn't get kicked
And 120dBSpl aren't 160dB at 1cm ... don't you do nearfield measurements? Nearfield SPL depends on the membrane area ... I'm sure you will find the formular to calculate the corrected level at 1m. Usually it's ABOUT 20dB higher. A M215 could handle the task.*) Instead of putting your speaker to big heights you can pile absorption to the ground at the reflection point. I usually do that 50-60cm high with 50x50cm big squares. Works well.
I measured this speaker at 50cm distance to reduce room influence and recalculated SPL levels to 1m (just put in double the sensitivity of the mic - done).
Get an MicW M215 and start measureing instead of guessing ;-)
(and don't get me started about the SM58 crowd cause they are to stingy to buy a €500,- microphone ...)
I have done indoor ground plane measurements of woofers. If the woofer is near the floor and the mic is on the floor, the effect of room modes is reduced. reduced by a lot actually.
When using STEPS, I find it helpful to smooth the fundamental before calculating the percent distortion. 1/3 or 1/2 octave smooths the response out quite a bit. Many people do not like STEPS because it is slow, but I am not bothered by this.
When using STEPS, I find it helpful to smooth the fundamental before calculating the percent distortion. 1/3 or 1/2 octave smooths the response out quite a bit. Many people do not like STEPS because it is slow, but I am not bothered by this.
I am wondering what your THD looks like at 105db/1m in this case. I know THD is complicated topic but I am curious is all. In my 1m indoor measurement above is a large exponential horn with the Axi2050.
at that point why not just measure at 1cm and do the same?
When you have a look at the graphs on the first page ...
You need a bigger tweeter to keep up at these levels but the rest will do fine. Only in a pretty large room it's still not enough ...
Good to know that the Axi behaves at low levels! I heared the Klipsch Jubilee in a not good room - the midrange is really good! But of course top end can't compete with teh resolution of a 1" Beryllium dome, but is good for a membrane of this size. I really hope they will bring a smaller Axi driver with perfect behaviour up to 20kHz and still low crossover (800-1kHz) - that would be an instant buy.
Cause you can't measure a 3way at 1cm ... and expect meaningfull results (of course you CAN ...
)
It's the cheapest reference mic with "real" capsule, good electronics and calibration sheet I know. You can use a standard 1/2" calibrater (there are cheap china ones, mine works perfectly) and have a class 1 setup for "no" money.
I often need to measure >20kHz so I mainly use Earthworks M50 - which has a little more noise but behaves perfectly at high frequencies.
No difficulties at all. I also use it with a passive crossover, no sights of overpowering it. 2kHz is not low for a modern tweeter at all.
Found this speaker on FB just yesterday!
There is a 2nd woofer on the backside. These ScanSpeak units are as good as the SB Acoustics I use - as long as Purify doesn't make a 12" the best units available ;-). It's actually a good idea to put it on the back but my speaker is designed to be installed at the wall - backside woofer is not a good option there. Actually I planed 4 12" for a free standing speaker ... but this get's massive.
It's probably a 3.5.5 design - the backside woofer can't go to high and the 2nd 3" also acts as a .5 way. With a wider baffle it wouldn't be needed but is still nice to keep THD even lower.
I used this ScanSpeak tweeter in a custom faceplate in my old reference speaker design - also a very good driver! It can go lower as the Bliesma, takes more power but has less sensitivity, so about the same output. For my use as measurement speaker the Bliesma is superior >30kHz. And I dislike the big front baffle, I used the smaller neodym Version before the T25B was available - but the faceplate is not removeable on the small one.
As I know these drivers - this is for sure a top notch speaker. I really dislike the hard edge close to the lower midrange and design is a matter of taste it seems ;-) Price of this speaker is interesting ... I really need to charge more
(for a custom project of 2 speakers price would be absolutely fair. Not sure how many they build and sell)