[sfdoddsy]

Yes, I'd be interested in seeing if the measurements match the hype, so to speak.

[bill13]

Would also appreciate seeing Dan's Neo10 measurements - would like to see off axis response for the quad arrangement in his new speaker.

New thread is a possibility but not yet sure.

Looks like it’s going to be lot easier to get the RAAL 140-15D to work with four vertically stacked Neo 10s (a taller version of Greg’s bamboo OB), and that's of course what I'm going to evaluate first. 
   
It would be great if Dan Richie’s custom Neo 3 tweeter were available for DIY – just in case the horizontal RAAL doesn't work out with an experimental 'square' quad Neo 10 D’appolito array arrangement.   Perhaps acoustic compromises, like some beaming, can be tolerated.
Waveguide mounting of a tweeter is DIY-doable, I believe.

Anyway, it's still not obvious to me that a horizontal RAAL with the foam pads, and some EQ (per Greg's comment) would not ‘work’ – it seems that the vertical spacing of the top and bottom Neo10s acoustic centers would not be all that different from using a custom Neo3.   If I'm wrong, well, the four vertically stacked Neo10s with vertical RAAL should be great, so I believe.

Bill



   

[Danny Richie]

Okay fellows here are some goodies for you.

The funny thing is after trying several different networks and really giving each type a good listen, the one that had the worst response had the best sound. The one that stood out to me as the best all around sound was a fairly low order series network. The higher order networks with the lower crossover points had slightly smoother on axis response and more even off axis response due to the lower crossover point, but just didn't sound as good. Less phase shift might have had something to do with it. So instead of a steeper slope and a 2kHz crossover point the slopes are more gradual and the crossover point fell at 2.5kHz. To me that crossover point is a bit high, but the Neo 10's handle it MUCH better than typical woofers. And the coherency was great.

These measurements are without the lower woofers playing. With them on the lower range is lifted up a little making the overall balance very good. There is a .045uF Sonicap Platinum in line with the RCA input to the amp. This rolls off the lows to the Neo 10's without having to use a passive high pass filter. This makes a HUGE difference in performance. So there is you guys huge tip of the day.

Anyway here is the horizontal off axis.



On axis is the red line. Orange is off 10 degrees. Yellow is off 20 degrees. We are still super smooth here and still about +/-2db. Green is off 30 degrees. And here we start to see a little drop just above the crossover range. And blue is 40 degrees off axis. So at 40 degrees off axis we have a 6db swing from 3kHz to 9kHz. That is still +/-3db. Not too shabby.

Here it is on a 10db scale like everyone else likes to use.



And the vertical off axis. This is taken at 1 meter. Red is on axis. Orange is moving up the mic 4". Since the speaker tilts 3 degrees this still does not get too much above the acoustic center of the speaker. This is about the same as standing from the back of the room verse sitting. Yellow is up 8" at 1 meter, and still looking very good. Green is up 12" at 1 meter and now we see a little bit of a drop out, but it is a smooth wide band drop showing that the crossover has a lot of driver overlap. It is not a sharp and thin suck out like I see on most speakers. And this is real good considering how far the acoustic centers are apart. If the Neo 10's were stacked up vertically then this would be much worse.



I wish I would have saved the in room response measurements that I took when dialing in the servo subs. It was one of the flattest in room responses that I have ever seen from any speaker. I may have to go measure all of those all over again just to save the measurement data.

So see, great off axis response.

[studiotech]

Danny, thanks for providing the info. 

Are these meant to be listened to on or slightly off axis, cause on axis looks like the treble is tilted up a little hot for my tastes.  So the horizontal are at 1 meter also?  And all smoothed to 1/3 octave?  I find 1/6 and 1/10 a little more closely related to how specific I can hear.  The dips around Xover would be be more pronounced are 1/6 or 1/10.  Those broad dips off axis, basically pushing down the upper mid-range would seem to be more noticeable than more narrow dips, no?

[Danny Richie]

Are these meant to be listened to on or slightly off axis, cause on axis looks like the treble is tilted up a little hot for my tastes.

I am very sensitive to a speaker that is even a little too hot in the top end. I tried listening to these speakers on axis and with a slight notch filter pulling down the top end. But that configuration lost out to no notch filter and tilting the speakers out about 10 degrees or so and letting the speakers face just a few degrees off from being aimed straight into the room. This produced the best sound stage and imaging. And the top was not hot at all.

Also, adding the lower drivers lifts the bottom end rang up a little bit and then the total response looks real balanced. We are not seeing that in this graph.

Those broad dips off axis, basically pushing down the upper mid-range would seem to be more noticeable than more narrow dips, no?

Narrow dips are usually much deeper and a greater disruption. Also wall and ceiling reflections tend to produce a lifted area in the 2kHz to 5kHz range compared to the range right above it. So usually a little broad softening in that range in the off axis helps even out the room reflections that are typical. Ranges above that tend to beam more and not reflect off the walls, and those shorter wavelengths are easily soaked up by carpet and other furnishings.

So the horizontal are at 1 meter also?

Yes it is.

[bill13]

Dan's off-axis frequency response plots look very encouraging - looks like no problem running his quad D'appolito Neo10 'square' array up to 2.5 kHz using a low order crossover.  Again, using the Raal foam pads hopefully might make listening in 'the sweet spot' OK.

Just wondering:  Does this thread removal mean I would have to post my previous OB stuff again if I create a new thread  ?   Or, can Greg just re-post his Bamboo OB thread again ?

No workshop - I only cut and drill wood outdoors (we don't like the sawdust indoors & garage).  With the coming of winter, it looks like it would take another few months to construct and evaluate an experimental quad D'apploito Neo10 arrangement with a horizontal RAAL.   
This would occur after adding fourth Neo10 on top of my three-Neo10 OB (evaluate a taller experimental version of Greg's bamboo OB speaker).

Bill

[bill13]

forgot to say:  try crossover shallow slope at say, 2.5+ kHz crossover freq and roll the Raal off faster at lower frequencies.

Bill

[Rclark]

But then again, I've never really been a fan of single way, full-range driver systems.  I've never heard the so-called holy grail of "coherence".  I just hear piecing upper midrange, dynamically lacking bass response and in general, more problems than any lack of a crossover could ever fix.  I bought one of the Tang Band Lowther looking clones to try out as a midrange for this project.  Talk about an ear burner.  Ouch, even after steep filtering and EQing to be flat in it's intended passband, there was some nasty resonance coming through that just killed it.  I think there's a reason that fullrange fans seem to usually be part of the AARP crowd......just sayin'

Greg

I lol'd hard at this. Noted and you're not the first to say it. I will spare my wallet any single driver DIY.

[Tyson]

My favorite single driver speakers are the ones with bass modules and a "high frequency extender".  In other words, a 3 way speaker!

[scorpion]

Danny,

In Europe we have (at least some of us reading the German Hobby HiFi magazine) had access to some Neo10 measurements since 2008. This first one is an IB measurement made by Proraum AG in their Anechoic Chamber, link is here: http://www.audax-speaker.de/index.php?module=explorer&displayAction=download&downloadFile=imported/tests-reviews/pro20d-hobby-hifi.pdf for a kit presented in the magazine.

Comparing this measurement to Zaph's or your own they all make sense. In fact I suspected your baffle to be less supportive for the low mid/high bass than it probably is. I also think that the different baffle
measurements do make sense when compared to BG's own without any baffle at all. Given your comments I suspect that the small 10 K hump is emanating from the Neo10 rather than the Neo3.

I do not doubt that within your chosen crossover frequencies both horizontal and vertical dispersion can grant a wide listeners stereo window.

That said I certainly also agree that listening is the real proof and to me the Neo10 and Neo3 driver combination has the proven performance to combine for an outstanding design.

/Erling

[HT cOz]

Danny,

In Europe we have (at least some of us reading the German Hobby HiFi magazine) had access to some Neo10 measurements since 2008. This first one is an IB measurement made by Proraum AG in their Anechoic Chamber, link is here: http://www.audax-speaker.de/index.php?module=explorer&displayAction=download&downloadFile=imported/tests-reviews/pro20d-hobby-hifi.pdf for a kit presented in the magazine.

Comparing this measurement to Zaph's or your own they all make sense. In fact I suspected your baffle to be less supportive for the low mid/high bass than it probably is. I also think that the different baffle
measurements do make sense when compared to BG's own without any baffle at all. Given your comments I suspect that the small 10 K hump is emanating from the Neo10 rather than the Neo3.

I do not doubt that within your chosen crossover frequencies both horizontal and vertical dispersion can grant a wide listeners stereo window.

That said I certainly also agree that listening is the real proof and to me the Neo10 and Neo3 driver combination has the proven performance to combine for an outstanding design.

/Erling

Sure would be nice to know German.  A TM kit using these drivers would be cool!  Something that is PE box friendly.  I'm sure the new 8" servo subs would fill out the bottom end very nicely. 

[Danny Richie]

Given your comments I suspect that the small 10 K hump is emanating from the Neo10 rather than the Neo3

The Neo 10's aren't allowed to play up near that high. That is just a natural response of the tweeter. It is a customer version with slightly higher sensitivity. I notched it out once with a shunted LRC trap. But it really wasn't worth it. It sounded cleaner without it. And with it slightly off axis the response was real smooth.

[lowtech]

IMO if you want planar midrange you should just buy a Magnepan.  They're dipole to boot and have resale value should one ever decide to sell them.

[studiotech]

IMO if you want planar midrange you should just buy a Magnepan.  They're dipole to boot and have resale value should one ever decide to sell them.

Been there, done that...for several years in fact.  They're sensitivity is very low and they've got their own set of problems with distortion and durability.  They tend to suffer from a delamination of the wire from the diaphragm material.  Happened to mine after about 4 years.  BG can play louder, cleaner and has a conductive trace rather than a glued on wire.  Besides, we are talking DIY here, not store bought systems.

[face]

forgot to say:  try crossover shallow slope at say, 2.5+ kHz crossover freq and roll the Raal off faster at lower frequencies.

Bill

2.5K is too low for the large RAAL.  If you need to cross that low, the OEM model can be crossed around 2K. 

[bill13]

I have the RAAL 140-15D (the big one) --  but with the amorphous core which makes it sound even better.
 
I read that the RAAL 140-15D  goes lower than the smaller OEM ribbon.  This seems believable because the big ribbon has more surface area than the OEM (the one that requires a non-disclosure agreement).   Wish that the OEM model was not ‘unobtainium’ for DIY like me...

Sure, if you listen at very loud levels (I don’t, and my wife doesn’t permit) then perhaps a 3 kHz crossover is warranted.

My opinion: Even 1.6 kHz @ 24 dB per octave sounds great IMO on my quasi “16+16” A3 array provided most listening is at low/moderate  volume level (70 dB typical).  See my previous posts in this thread with comments & picture regarding my modified “16+16” A3 speaker array.

BTW, The RAAL PDF spec shows a recommended 1600 Hz crossover design.   
OK, you can take that ‘with a grain of salt’ but 1600 Hz is OK under the right conditions.   IMO, As long as distortion is not evident, in its range, the RAAL 140-15D with amorphous core sounds better than anything else I’ve heard so far. 

My spectral analysis of sine waves indicated distortion harmonics that reach a low plateau above 2.5 kHz.  I wonder if somehow the amorphous core transformer makes the ribbon sound better at the low frequencies?

My two cents worth.

Bill

[studiotech]

Thanks for the info Bill.  You're making me wish I had sprung for the Amorphous core for myself.  Mine is not the amorphous core, but measured distortion taken at 95dB/SPL is rising by 2.5KHz, so the 1.6KHz recommendation(remember this not where he actually recommends to cross, just the lowest possible point) would only be suitable for lower level listening.  And yes, there is quite a bit more surface area than the OEM, so the OEM is in no way any MORE capable lower than the 140-15, despite what the spec sheet may allude to.  This is directly fro personal experience with both AND from Aleksanders own words to me. 

I'm crossing the OEM at around 3.5KHZ, 24dB/oct, but this is for a studio monitor that will probably take some abuse.  Better safe than sorry and the Neo8-S is just fine that high.  See here:

http://www.audiocircle.com/index.php?topic=110721.0

Greg

[face]

Thanks for the info Bill.  You're making me wish I had sprung for the Amorphous core for myself.  Mine is not the amorphous core, but measured distortion taken at 95dB/SPL is rising by 2.5KHz, so the 1.6KHz recommendation(remember this not where he actually recommends to cross, just the lowest possible point) would only be suitable for lower level listening.  And yes, there is quite a bit more surface area than the OEM, so the OEM is in no way any MORE capable lower than the 140-15, despite what the spec sheet may allude to.  This is directly fro personal experience with both AND from Aleksanders own words to me. 

I'm crossing the OEM at around 3.5KHZ, 24dB/oct, but this is for a studio monitor that will probably take some abuse.  Better safe than sorry and the Neo8-S is just fine that high.  See here:

http://www.audiocircle.com/index.php?topic=110721.0

Greg

That's odd that Aleksander would say that.  I've taken measurements of both the large RAAL and the OEM.  I wouldn't use the large RAAL under 3K either, but the OEM model is usable down to 2.2K, maybe even a little lower with some crafty crossover work.

[bill13]

Greg,

I wonder what you think of the following DIY project posted in the Parts Express forum.

In the Parts Express forum there's an interesting open baffle design with discussion by 'Professor Niko' at

http://techtalk.parts-express.com/showthread.php?232905-New-Dipole-Experiment

This OB speaker uses a pair of B&G Neo 10s,  a Neo 3 PDR in the middle, with a side-vertical array of $8 Focal 6 1/2" woofers (total of eight woofers per speaker - see the pictures at PE).  Interesting that the baffles were made from 4 each 12"x48" stair treads bought at Lowe's for only 20$.

Speaker builder 'Professor Niko' said that he could hear the speed of the Neo 10 as slower than that of the Neo 3 tweeters above 3300 Hz.  There were other interesting observations, as well.

The Aura 6" Neo woofer is/was now selling at very low price (around 6 or 7 bucks?) - maybe a bunch of these woofers might work (perhaps use sixteen woofers for a pair of OB speakers) - anyway, this OB project seems like a relatively inexpensive way to get a big 'bang for the buck'.

Bill

[studiotech]

That's funny.  Those responses on that thread are from me...

I think it looks like it could have good potential.  I just wonder how well they are adjusted due to being done by ear.

Greg