“Loading” of tweeters in large drivers Coaxes

[otto88]

nb: the no longer available Unity horn was a clever multiple driver single point source arrangement . .

 

[otto88]

Apparently baffle step is much less smooth with a single point source . .

http://www.diyaudio.com/forums/showthread.php?s=&postid=438905&highlight=#post438905

 

[John Busch]

Hello all...

Some more mussings.....

OK... We may be long here again.. imagine that!! Again plz forgive any spelling or syntax errors...

Several things... as for "Single Point Source" across the entire audio spectrum, this of coarse is not possible in actuality IMHO... with dynamic coax speakers... many believe you can get closer with a single magnet assembly powering both the HF and LF voice coils... and or tack a small dome tweeter on the top of the LF center pole... they are mechanically aligned or very closed to being mechanically aligned in both the horizontal and vertical planes.... but it is funny how any speaker's point of acoustic origin will vary with frequency... especially at the lower end of ANY drivers response. This is the Zero Delay Plane (ZDP) aspect of things. Vance Dickason has some discussion in his Cook Book Series about this. What does help a typical pro coax though is the ZDP is usually BEHIND the LF on a typical coax... so with the HF also mounted behind you can come that much closer to a Single Point Source around an octave or two above and below the crossover frequency. This assumes the crossover frequency is fairly low .... in the 700-1100 Hz area.

So mechanical alignment does not necesarily translate into a true point source across the audio spectrum. It makes things easier (closer to point source) to be sure. This is where ... again... the crossover can be SO important!!

Just as a reminder.... The BM series of P.Audio coaxes are not mechanically aligned in the horizontal plane. The HF is bolted on to the rear of the woofer. IIRC.. the horzontal voice coil separation is around 1-3/4".. give or take. Excellent engineering IMHO to get it that close.

Now with an HF driver that can really play and load to below 1000 Hz.... the phase shift at the crossover point is not all that much. This is also a great help crossover wise. Assuming a 800-900- Hz acoustic crossover point with the BM18CX38... wave lenghts are around 16"... by the time you get to a frequency that approaches the voice coil separtion distance (1-3/4" = 7,500 Hz) you are long past the crossover region.

The main thing to keep in mind is that most Coax's are so much closer to the ideal "Single Point Source" than most other multi-way speakers. In my opinion... they are maybe closer than many single voice coil full range speakers because of the ZDP issue. I am sure that I will get a lot of flack for saying this... but I am stating my opinion... not claiming it to be fact. More on this some other time...if I find the courage to further elaborate.

One more thing before I get back to my day job.... and that is power response. IMHO... after tonal balance... power response has much to do with a speakers ability to image and approach a live music goal. Here again ...on the proper baffle... with a proper crossover a good coax has a big advantage. The big wave guide cone provides a controlled radiation pattern. And the natural cancellation that occurs in the lower frequencies with an OB LF set up.....you end up with most of the energy foward of the speaker and much less reflected of the side walls in the room.... you get a lot more direct ... reflection free sound at the listening position. There is less smear... more time coherency. Controlled Directivity if you will. Again I may get some flack over this view... but in my many years of experimentation... this seems to be true.

Best... JB

 

[Pano]

..Jeez. You almost make it sound like I'm being ganged up on for a smack-down or something.

Naaawww.... I just mentioned it because I've had exactly the same doubts!

Thanks JB, for the further elaboration. The point of the acoustic center, or ZDP is an interesting one.

Big coax drivers certainly have a fan club. I feel I may soon be joining the club.

 

[BHTX]

Anyone know if Hawthorne Audio's Silver Iris drivers are true single point sources?

Might do yourself a favor and skip on the SI, unless you're just really tight on money (like I am)..

The 15" woofer is very similar to the Beta-15CX (which no longer exists), except with a much higher Qts. The cheapo compression driver is crossed way too high (no other choice, as it can't go anywhere near low enough).. like 2.5 KHz, if I remember correctly (yuck!). I'll see if I can find anything on my hard drive that might be of interest..

..Yep, looks to me like the kapton former is actually sticking up over the metal flare. Just imagine what it looks like playing a low frequency tone, lol. Aside from that, look at all the hard transitions and mess the HF has to go through. Not trying to bad mouth, just saying. Actually, I think these kinds of things are pretty common, although hopefully less so with more expensive models. I have my doubts though.

 

[Pano]

Intersting graphs. The top one seems to be close to what I heard.

Can you tell us a little more about the measurements?

1. At what distance was the outdoor measurement done?
2. What's the big dip at 650Hz on the top graph? Tweeter out of phase?
3. The in room looks like it was done without crossover - right?
4. What is the 3rd graph?

Thanks!

 

[otto88]

> The top one seems to be close to what I heard

. . not good. How much smoother is the P.Audio BM-..CX38?

Thanks

 

[BHTX]

Can you tell us a little more about the measurements?

1. At what distance was the outdoor measurement done?
2. What's the big dip at 650Hz on the top graph? Tweeter out of phase?
3. The in room looks like it was done without crossover - right?
4. What is the 3rd graph?

Nah, unfortunately, this is just stuff I've saved off the net over time, so my guesses are as good as yours. Mic was probably around 1 meter looks like. No idea about that dip @ 650hz, but I'm pretty sure it has nothing to do with the tweeter. As for the crossover, at first glance I thought he might have used the xo that comes with them, but yeah it looks like they might have just been raw responses. In that case, I seem to remember reading somewhere way back about the crossover point actually being as high as 3.5 KHz or something (rather than the 2.5 I mentioned earlier). Anyway, it's way up there somewhere. I remember saving the 3rd graph from Darrel Hawthorne's "member gallery" on his website.. freq response and impedance of the SI woofer, likely could have been measured by Eminence.

If you're looking for a true point source with constant directivity, this probably doesn't come anywhere close. The 15" woofer is obviously WAY past the point of beaming and breakup at the crossover point, at which point it goes to the cheap tweeter with much wider dispersion. Granted, that's just what it is though.. cheap. A fairly cheap way for newbies (or anyone) to easily get into DIY with open baffles and end up with something that's hopefully somewhat bearable after some fun on a Sunday afternoon. Other than that, I wouldn't expect much from it.

 

[BHTX]

Just for the heck of it, here's some B&C stuff I've had saved. All the links are still on the web though, so rather than hosting all these, I'll just post the links.

Here's some B&C coaxial measurements from prodance.cz, along with their spec sheets from B&C if available on their website. Prices are each, USD, and obtained from either usspeaker, partsexpress, or a quick Google search.. unless they've been discontinued, in which case I didn't bother searching for a price cuz I'm too lazy right now. And the prices listed usually aren't exact, since they often seem to differ among different places. These numbers are just to give an idea. None of them are exactly cheap though, that's for sure..

15CX40 ($440)
http://prodance.cz/protokoly/15cx40.fin..pdf
http://bcspeakers.com/download/prodotti/PDF/15CX40.pdf

15CX (discontinued?)
http://prodance.cz/protokoly/bc_15cx.pdf

12CXB (discontinued)
http://prodance.cz/protokoly/bc_12CXB.pdf
http://bcspeakers.com/download/prodotti/PDF/more/153.pdf

12CXL (discontinued)
http://prodance.cz/protokoly/bc_12cxl.pdf
http://bcspeakers.com/download/prodotti/PDF/more/164.pdf

10NCX ($350)
http://prodance.cz/protokoly/hrub..pdf
http://bcspeakers.com/download/prodotti/PDF/10NCX.pdf

8CXT ($255)
http://prodance.cz/protokoly/B&C_8CXT.pdf
http://bcspeakers.com/download/prodotti/PDF/8CXT.pdf


And here's all the spec sheets from the rest of B&C's current line of coaxials..

15CXN76 ($600)
http://bcspeakers.com/download/prodotti/PDF/15CXN76.pdf

12CXN76 ($550)
http://bcspeakers.com/download/prodotti/PDF/12CXN76.pdf

12CX32 ($425)
http://bcspeakers.com/download/prodotti/PDF/12CX32.pdf

12NCX ($365)
http://bcspeakers.com/download/prodotti/PDF/12NCX.pdf

12CXT ($260)
http://bcspeakers.com/download/prodotti/PDF/12CXT.pdf

8CX21 ($240)
http://bcspeakers.com/download/prodotti/PDF/8CX21.pdf

 

[Pallas]

Agreed the midband is much more important than the top octave.
The benefits in midband power response you referred to – do you mean single point source Tannoy vs the rest, off axis?

Not "just" off-axis. I mean that a Dual Concentric or similar driver with a properly designed crossover won't have the wretched-sounding-in-room midrange that a conventional speaker with a flush-mounted tweeter will have, because of the latter's mid-tweeter directivity mismatch in the crossover region.

Concentrics are not unique in this regard. Separate waveguides (see, e.g. the GedLee Summa series) offer the same benefits.

Which make me think, has anyone quantified or verified the advantages of a single point source?

There are lots of polar plots for different concentric driver based systems out there.

- A true Point Source gives the same sound from seat to seat.
- Dual Concentrics can use simpler, better sounding and more efficient crossover
- "Harmonic relationships" are better preserved using a single point source (I think is the same as saying phase response is better)

with your wide experience, do you agree?

Not directed at me, but yes within reason (way off axis all you lose is highs, and that in a consistent, controlled manner), no, and I think the main advantage is in the consistent directivity through the midband rather than any phase response issue.

 

[otto88]

I'm flat out to respond much, but thanks to all for the responses so far.

I learnt today that the local Aussie P.Audio dealer Cannon Sound has some BM15CX38 left at half price (probably 4 left now). When they sell out, the normal prices will rise – in the last few months the $A is down about 25% vs the Thai baht, and 35% against the $US. Ie in a few weeks the price compared to the current sale price will be about 2.5! times higher (No 18 CX38’s in stock).

So based on the enthusiasm for the general approach and in particular its bigger brother (pano and JB), I plumped for a pair. I'll put them in an open baffle . . I’ll be looking for how to optimise the crossover . .

Cheers

 

[BHTX]

Good to hear, and congrats on the half price deal.
One bright side of not being able to get your hands on the 18" version is that you shouldn't have to worry the least bit about having to push that 2" exit compression driver to match the 15" woofer and still have everything line up perfectly, although I guess it hasn't been a problem with the 18".

Keep us updated on your progress, as I'm really interested to see how it all turns out!

Do you already have a good way to measure freq responses and whatnot?
Just like you probably are, I'd definitely be most interested in seeing that as well.

 

[John Busch]

Hello Otto 88....

Congrats on your purchase!!

Spent some time refamiliarizing myself with the BM15CX38 specs.

The main difference is the smaller cone area. The rest is pretty much the same. In terms of performance... you will have a little less dynamics...less bass output capabilties (displacement) but with a smaller... lighter cone your mid range resolution maybe better... don't know for sure of coarse... and just to be clear I make the above statements relative to the larger BM18.

Once you break these guys in (the woofer section) ... need to see where you end up QTS and FS wise. My 18 have a slightly higher than spec FS (34Hz spec - 35.8 Hz actual) and a higher QE and resulting QTS ( .32 spec - .455 actual). Long and short the higher fs and QTS works to my OB advantage while retaining close to spec'd efficiency.

So.... please let us know what you actually have. And what size baffle you intend to use.... wings if any... so on.

This kind of info... reallity will have much to do with a workabe crossover.... one last.... do you intend to augment with a sub woofer or run alone?

JB

 

[Pano]

Congates Otto! Way to go!

Do please keep us updated on the new drivers. Being able to measure the T/S will help a lot in designing the baffle and especialy the x-over.

You're going to have fun, 4 sure.

 

[otto88]

Thanks guys

I need to set up a measurement system. After much planning, borrowing, building and fitting out, I'm only just getting a proper workspace now . .

> a higher QE and resulting QTS ( .32 spec - .455 actual).

That's quite a difference, but yes higher is better for OB. I'll borrow Woofer Tester 2 to measure the T/S.

I calced the current baffle plan, it's probably likely to be 1.0 m2 = 1520 sq inches. There will be modest wings, maybe just 6 inches deep. The front profile will step in towards the top, like the Empire State Building.

I haven't run this by my wife yet, but it’ll be on castors so I can push it back against the wall if that helps. It will help with positioning.

“And honey, you like Art Deco - itll be more attractive than other speakers I’ve had”

Cheers

 

[Russell Dawkins]

I think if you angle the top of the baffle one way or the other you can call it art deco...

 

[theAnonymous1]

I've had both the 15" and 10" SilverIris coaxials. I found the midrange to be extremely harsh and lacking. The compression driver sounded awful under 3k and the larger drivers could not play high enough to make a good match. I think I found the 15" only played well to ~900hz and the 10" to ~1200hz (it's been awhile so I might be off some).

That experience ended my adventure with coaxials.

I should probably say the above is just my subjective opinion and I'm not trying badmouth anyones product. I don't want anyone jumping down my throat for an opinion I paid good money for.

 

[otto88]

Off topic now, but attached is a draft profile. It'd be on castors to move easily, and the baffle size could accommodate several other drivers I have, acting as a comparator/ test bed.

For OB rather then maximising baffle size, to get extension lower I believe the critical factor is width. So I think I’ll do folding wings, to roughly double width. When the wife is using the room she can fold the wings back

Cheers

 

[BHTX]

Off topic now, but attached is a draft profile.

Not exactly sure what that xls was supposed to look like, but I'm assuming I didn't see what you intended people to see. I wonder if anyone else will have the same problem. Just thought I'd let you know. I opened it in Office 2007 after extracting with 7zip btw. Kinda strange.

 

[otto88]

> Not exactly sure what that xls was supposed to look like, but I'm assuming I didn't see what you intended people to see. I wonder if anyone else will have the same problem. Just thought I'd let you know. I opened it in Office 2007 after extracting with 7zip btw.

An Excel file was created with a grid to a scale, that was saved with (old!) Excel 2000, then zipped.

If saved then opened with Win Zip, I was able to re-open it Excel 2000 . .

Any other reports?

Cheers