Can you EQ this with a compression driver?

[DanielT]

I will start my DIY projects soon. I have two different, and by different I really mean different, tweeters. Part of my project is to figure out what kind of sound, dispersion among other things, I like best. Another part to learn plus of course to have fun.
In any case, to my questions. Regarding this driver, which I have:
Eighteen Sound XD125:

Eighteen Sound - Professional loudspeakers

www.eighteensound.it

This is what the FR looks like, I suspect Eighteen Sound measured on a large baffle, in an anechoic chamber:

Question 1:
Is it possible to smooth the FR over the entire register from around 3kHz and up to 16kHz? So it will be straight throughout that area?

Question 2:
IF it is possible to equalize the FR between 3kHz to 16kHz, does it make sense also off axes? Does the off axes response also follow the on axes when EQing this driver? Can it be interpreted based on this data:

Question 3:
FR as measured by Eighteen Sound under those conditions, but if I don't do anything to the FR how do you think it will be in a normally furnished listening room, living room? Is it possible to say something general about it?
I have acquired a measuring microphone that I will plug in, once I have updated the programs in my computer so I will be able to measure, but right now I was just generally curious.

The other tweeter is this:

SB Acoustics SB26ADC-C000-4 | HiFiCompass

The THD measurements at distance of 315 mm and voltage levels 2.83 - 8 Volt were carried out with High Pass Filter "on" (2nd order Buttreworth type, 500 Hz cutoff, HPF2-500). The THD measurements at distance of 315 mm and voltage level 11.2 Volt were carried out with High Pass Filter "on" (2nd...

hificompass.com

Edit:
That the XD125 rolls off at around 16-17kHz doesn't bother me. I still don't hear anything above those frequencies.

 

[ppataki]

I can assure you that you will be able to make that FR fully flat
I would just measure the real-life response and apply correction to become flat and then apply an additional high shelf to taste (to decrease the high end a bit)

 

[Flaesh]

Question 1:
Is it possible to smooth the FR over the entire register from around 3kHz and up to 16kHz? So it will be straight throughout that area?

Question 2:
Does the off axes response also follow the on axes when EQing this driver?

1. Yes. It's always done in one way or another.
2. Yes. Directivity remains the same after equalization.
What crossover will you use?

to have fun.

Comparing comression tweeter and dome tweeter can be entertaining.

 

[DVDdoug]

I'd probably EQ it unless my ears didn't like the result. I'd probably do just a gentle boost. Don't try to fix the wiggles. It's probably OK if that little 16kHz bump ends-up a little high as long as everything averages-out about flat. I would NOT try boosting 20kHz by 30dB! Even if I could hear that high it would likely cause problems "somewhere else".

That tweeter is super-sensitive (most horns are) so you'll probably have to attenuate the tweeter signal. That extra sensitivity would allow you to build EQ into the crossover (attenuating the highest frequencies a little less, or maybe not at all). Or if you are bi-amping (or tri-amping) with an active crossover it's "easy" to build-in some EQ (if you're building the crossover yourself) and also easy to separately adjust level of all the drivers.

 

[fpitas]

EQing the HF stuff is easy. I think your biggest challenge will be getting the acoustic slope right in the crossover region.

 

[RammisFrammis]

That is a typical raw HF compression driver curve and it can be corrected with EQ. With my horns which are Klipsch I attenuate the broad hump below around 5k rather than boosting above that frequency. That takes care of the excess efficiency of the driver at the same time.

 

[DanielT]

Thank you all for the response, as well as tips and advice. I suspected it could be done, EQ that is but was a bit unsure about smoothing such a large range from 3kHz to 16kHz.

 

[DanielT]

I can assure you that you will be able to make that FR fully flat
I would just measure the real-life response and apply correction to become flat and then apply an additional high shelf to taste (to decrease the high end a bit)

Yep, I will measure the real-life response.

1. Yes. It's always done in one way or another.
2. Yes. Directivity remains the same after equalization.
What crossover will you use?

Comparing comression tweeter and dome tweeter can be entertaining.

Crossover, at the moment I have a miniDSP (not HD unfortunately). Plus an active crossover, LD X-223, which I managed to buy at a good price.
This one:

LD Systems X 223 LDX223

The LD Systems X223 is a stereo 2 way / mono 3 way crossover in standard 19"/1U rack format. Its rugged construction, high quality and robust knobs as well as the integrated power unit are rarities in this price range. It provides a...

www.adamhall.com

But at a lower price.
However, I am open to suggestions. That's why I started this thread to get tips and advice:

Suggestions for three-way crossover filters?

For a pair of DIY three-wave speakers I was thinking of building this autumn.Crossover points, 300 Hz and 2500 Hz. At least 24 dB HP-LP filter.. I have several amplifiers so it is no problem to let different amps take care of the bass, midrange and tweeter respectively. Preferably in the...

audiosciencereview.com

I'd probably EQ it unless my ears didn't like the result. I'd probably do just a gentle boost. Don't try to fix the wiggles. It's probably OK if that little 16kHz bump ends-up a little high as long as everything averages-out about flat. I would NOT try boosting 20kHz by 30dB! Even if I could hear that high it would likely cause problems "somewhere else".

That tweeter is super-sensitive (most horns are) so you'll probably have to attenuate the tweeter signal. That extra sensitivity would allow you to build EQ into the crossover (attenuating the highest frequencies a little less, or maybe not at all). Or if you are bi-amping (or tri-amping) with an active crossover it's "easy" to build-in some EQ (if you're building the crossover yourself) and also easy to separately adjust level of all the drivers.

Nop, anything above 16kHz I will not EQ. I thought bi, or rather tri ampa. Three-way DIY speaker.

That compression driver has high sensitivity, are you kidding me. I've never had one like this before. A really fun experience. See the attached picture of a 50 watt NAD receiver where I test play with the XD125 drivers at normal listening volume. You can see how little I turned the volume.

Image from this thread:

Pre out + RCA splitter

I will test this today: Pre out from my NAD receiver, split the signal and feed it to two different amplifiers. One of these amplifiers (a low powerd tube amp) will take care of my XD 125 compression drivers. The other amplifier (HK 670) will take care of the bass element in a couple of other...

audiosciencereview.com

EQing the HF stuff is easy. I think your biggest challenge will be getting the acoustic slope right in the crossover region.

I think so too. I will come back to that slope question.

That is a typical raw HF compression driver curve and it can be corrected with EQ. With my horns which are Klipsch I attenuate the broad hump below around 5k rather than boosting above that frequency. That takes care of the excess efficiency of the driver at the same time.

Aha! Smart.

 

[fpitas]

For the crossover, one way is to flatten the response down to an octave below the desired crossover point, then use a standard crossover HPF like LR4.

 

[Flaesh]

Crossover

I'm readig https://www.audiosciencereview.com/...-for-three-way-crossover-filters.36416/page-2

 

[DanielT]

I'm readig https://www.audiosciencereview.com/...-for-three-way-crossover-filters.36416/page-2

Many different interesting solutions are suggested in that thread, I think so anyway.

 

[DanielT]

I found this about compression drivers interesting:

#134 in this thread:

As for distortion, as the fundamental gets louder, low order distortion products become less and less audible. Compression drivers have second order dominated HD profile. According to Geddes Lee's research, participants of the research couldn't hear even %20 distortion of compression drivers while they are playing loud

Genelec S360 Review (Studio Monitor)

The review compares to the 8361a, but isn’t the 8351b the better comparison at the same price point? Which is the better speaker then?

audiosciencereview.com

 

[Flaesh]

Compression drivers have second order dominated HD profile.

Many dome tweeters also. SB26ADC-C000-4 e. g.

 

[Jon AA]

Yup. That's one of the great things about the headroom a good compression driver gives you. Example--full spin of the speaker in my avatar to the best of my ability:

As you can see, that speaker needs a bit of EQ help. And after EQ:

Not a bad speaker--certainly considering the price paid and output capability. That's not an REW simulation either, I actually measured that spin with the EQ applied via a MiniDSP HD.

 

[Flaesh]

Not a bad speaker

I predominantly support medium format woofer+waveguide active 2way as simplest and funniest [first] DIY loudspeaker project.
Massmarket example:

Sorry for offtopic.

 

[DanielT]

I predominantly support medium format woofer+waveguide active 2way as simplest and funniest [first] DIY loudspeaker project.
Massmarket example:
View attachment 225313
Sorry for offtopic.

That's my idea of how a pair of compression drivers should be used. Big bass drives together with a big horn driver. My XD125s have a bit too small a horn for it to work, I think.

Yup. That's one of the great things about the headroom a good compression driver gives you. Example--full spin of the speaker in my avatar to the best of my ability:

View attachment 225307


As you can see, that speaker needs a bit of EQ help. And after EQ:

View attachment 225308

Not a bad speaker--certainly considering the price paid and output capability. That's not an REW simulation either, I actually measured that spin with the EQ applied via a MiniDSP HD.

Super duper.

... as Taco sang

 

[DanielT]

In any case.

The fact that I have what I have now is due to a combination of chance, impulsiveness and luck.

My thoughts, plans so far:
It started with me listening to these DIY speakers at a fair last year. Those speakers sounded really good, I think. With these elements:

With these elements:
SB SB26ADC-C000-4
SB SB15NBAC30
Satori WO24-4

From the thread:

Hifi Vintage DIY fair with pictures

I visited this fair yesterday! DIY & Vintage meeting in Mölnlycke Here is a short report. When: 6 November 2021 Where: Country: Sweden City: Gothenburg I took pictures myself BUT most were so damn blurred so I borrow pictures from this thread, taken by other participants. Pictures taken from...

audiosciencereview.com

Then I read this thread and became even more curious, interested:

Revel M105 Copy DIY build

After seeing the success of Directiva, But knowing the price, I wanted a nice small pair of passive bookshelf speakers for use in the second living room. I didn't need massive SPLs (if you need more SPLs the Directiva is much better - however that build is DSP XO + Multi Channel amplified). The...

audiosciencereview.com

Same drivers minus Satori WO24P-4. I have now bought a pair of SB26ADC-C000-4 and a pair SB SB15NBAC30 (4 Ohm variant).
Bass like in a three-way like Samuels in the Vintage DIY thread I'm interested in but the Satori WO24P-4 is a good, but expensive, driver.That's why I bought a couple of sensible bog standards instead:

Experiences of and / or measurements Peerless SLS series?

Seems to be speaker element with good performance - price ratio.:) You who have used these speaker elements, what are your experiences? Are there any distortion measurements on any of the elements in the SLS series? I tried to find some but I did not find any.

audiosciencereview.com

SLS 10, 8 Ohm 89 dB sensitivity. Then, luck again, this weekend by chance. Saw ad for a pair of nice, good condition SLS 10's in two pairs of boxes for $60. Bought them:

Plug them in here and test run them:

So now I have four SLS 10. My thought. Two in each box. Connect two SLS 10 in four Ohm, then I gain 6 dB. BUT lack of baffle support, I reduce 6 dB. It should be good, with a sufficiently wide baffle so this happens at the crossover point SLS to the midrange around 300-350 Hz. Then the midrange SB15NBAC30 with 88 dB is supported by the baffle. Plus really round corners on the speaker, to avoid edge diffractions, something like this, from this thread :

faktiskt.io • Visa tråd - 1: Frekvensgång, 2: Spridning, .. Spinorama?

My friend has a 3 d printer, so I want to fix a waveguide for my SB SB26ADC-C000-4, if want to do that.

The XD125 compression drivers was just luck. I bid on an auction site and won them. Actually had no major thought behind it other than I became interested in compression drivers after this thread:

Compression drivers?

How common are Compression drivers for Hifi speakers, for home use, home hifi that is? I do not see them that often. How is it that? Too expensive? Difficult to design good speakers with? What about Compression drivers regarding FR, directivity and distortion? They have a high sensitivity, I...

audiosciencereview.com

So that's the situation right now. Ends with a picture of Samuels from his home. When he measures:

Edit:
Those are my thoughts, but I haven't calculated it all in any program yet. Or test measured outdoors, for example the elements in some test baffle. But it will happen in the future.

 

[Plcamp]

I think it’s true that with a horn loaded compression driver you need big eq anyway just to match bass driver output…normally best case at about 97 db/ watt and sometimes much lower.

You may find out dsp can’t eq certain things, if for example the response has a resonance, but generally dsp should control very well.

I was most interested in Joseph Crowe’s latest incarnation with SB audience drivers that one conclusion, or at least insinuation, was that large (15”) for midbass granted very significantly better distortion performance (when mating to a circa 600 hz horn). I can see why that might be true, but can’t yet fully understand it.

I now think of my new speaker being a floor level 15” woofer in an 18” deep damped uframe for cardioid-like bass to 30 hz, another of the same driver above that in an 18” deep sealed enclosure, and a 600 hz horn sitting atop that covering up to 16 kHz, all controlled by a Minidsp flex eight.

 

[tmuikku]

^ bigger transducer has much less excursion for same volume displacement. A 15" cone area is usually ~850cm^2, while ~6" drivers have around 130cm^2, about 6x less. Non-linear distortion in driver is related to excursion, current, due to non-linear properties of the motor and suspension. Hense, keeping excursion low makes less variance in the attributes that are involved and less distortion happens.

You could think it like this as well. If one 6" driver was enough for some SPL level you want to listen to, it would require x Watts from amplifier.
If you swap 15" in for the same duty making the same SPL, it would now move 6x less and also have 6x less power from the amplifier, simplified.
Instead of using one 15" you could use 6x 6" drivers for similar net effect, its just more expensive that way.

 

[beagleman]

The usual way is to NOT EQ, but to use a capacitor to roll off the response starting at around 8-10 kHz and that will reduce the whole region below it enough to make it mostly flat.

One cap, at a 6db/octave roll off.