[panomaniac]

Just a fun thought experiment here - I have not yet  built this.  But I think it would work well. Let's see what you think. 

For those of you not familiar with the OB concept of Lukasz "Lampizator" Fikus , please see his website.
http://www.lampizator.eu/NIRVANA/nirvana.html
I've used Lukasz' approach and have to say that it works well and is simple to do.
Basically he likes to use a wide range alnico driver on open baffle supported by box bass down low and paper tweeters up high. Crossover is very simple.
Lukas likes vintage radio drivers, especially SABA greencones.  But what if you wanted to use new drivers, what could you pick?
Here is my idea:

Eminence 10" alnico guitar speaker. http://www.eminence.com/pdf/legend-1028k.pdf
Dayton 12" DVC sub.   http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=295-490
Pioneer paper cone tweeter: http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=270-032

That brings your total driver cost to a whopping $280 for the whole set.
I figure about $100 for crossovers. Most of that is inductors, which are expensive these days.
So it should be under $400 a pair minus lumber costs.

[panomaniac]

  Part 2
Here is the layout:

Eminence Legend guitar speaker on a 400mm wide baffle (15.75").   This is the heart of the system.
If you look at the specs for the driver, it's an odd one.  Very high Qts and a rising response.  But we can use this to our advantage.  Simply put a ~1.2mH coil in series and it should be flat on baffle from 95Hz to 4.5Khz.  Just bring the woofer up to match and the tweeters to fill in the top end.  Easy!

Dayton DVC woofer.  This is a great little woofer and will work well in a sealed or aperiodic box.  I would go with aperiodic at about 1.6 ft^3.  I have run a pair of these for years in a 2.5 ft^3 sealed box and they worked beautifully.  Aperiodic will allow a somewhat smaller box and more benign impedance curve.  With the voice coils in parallel it's a 4 ohm driver - but will give you 96dB @ 2.83V in a modest sized box.  (A bigger box will give you a lower F3)

Pioneer tweeters.  This is the tricky part.  Lukas likes to use vintage paper tweeters with open backs, so do I.  Nice if you can find them, but there are not any made today, AFAIK.  So we cheat.  We use 2 Pioneer sealed back tweeters -- 1 front and 1 back.  I've done this with Foster and Rat Shack paper tweeters and it works well.

[panomaniac]

   Part 3
The crossover:

Simple.  1.2mH on the 10".  8uF on the tweeter pair.  12 or 13mH coil + 100uF cap on the woofer. Basta!
Of course those values would need to be adjusted a little for the actual drivers, the room and your taste.  But that should be pretty close.

If I get the chance, I may build this crazy thing.  I have all the parts in stock - minus the 10" guitar drivers.  It could be quite nice sounding a rig and would not need much voltage to get loud.  Good for tube and small T-Amps.
And if it isn't good  - for example, the Eminence guitar driver doesn't sound good, there are lots of other 10s that could drop in.

Anyway, there is the hare brained idea for your amusement. 

[Graham Maynard]

Hi Michael.

My thoughts would be to get the 10" sounding 'best' on its own first and then add the augmentation drivers.
Further, that 1.2mH in series with the 10" is likely to take too much 'life' out of its reproduction.

Also the 10" frequency range is where the BSC will need to be effective, so possibly a better arrangement would be with a 2x 8 ohm potential divider between live and ground at the LS terminals, with your choke across the top (live) resistor and the 10" driver across the lower (earthy) resistor.  If there is still too much of a rising characteristic, then this could be further tamed by a series R+C Zobel connected across the LS terminals, say 8 ohm and 1uF.

Of course the exact component values would need to be chosen *by ear*.


Cheers ............ Graham.

[ttan98]

Just a fun thought experiment here - I have not yet  built this.  But I think it would work well. Let's see what you think. 


Eminence 10" alnico guitar speaker. http://www.eminence.com/pdf/legend-1028k.pdf

Hi,

Have anyone you know tried this driver before? ie you or any of your friends, the other augmented drivers will follow suit or replace with possible substitutes.

I am keen on these drivers they relatively cheap as well.

[Dmason]

Its frequency response definitely indicates "guitar."

I would be concerned about a ~15db variance between one octave above Middle C, and C three octaves above.

Having said that, i have noticed that light cone drivers with ALnico magnets and good sensitivity ALL sound really pretty special.

[panomaniac]

Hi guys, thanks for the replies and comments.  Spot on, I think.

@Graham.   Yes, 1.2mH might smother the driver too much, but it's were I'd start.  The sims say that's about the value that will flatten the FR.   Have a look at the graph below  which shows the Legend 10" filtered and unfiltered.
This is with 1.3mH.  Don't worry too much about the bump on the bottom end, there will be a little more roll off than that on a 400mm wide baffle, so that bump should diminish.


Here we see that the simple series coil has brought the driver to a usable range of about 70Hz to 4KHz.
We don't worry about a perfectly flat FR, that's the point of the Lampizator approach.
Open baffle simulations show that the bottom end of this curve looks a lot like what is shown in the factory graph.
In this graph it looks like a 4th order acoustic roll-off on the top, 2nd order on the bottom.

One caveat.  This is the on axis response.  Off axis is not going to look like this.  A 10" driver will start beaming above about 2KHz, and will be very narrow by 5K.  So this rising response will be seen only on axis.
As Graham notes, the big 1.2mH inductor may suck a lot of the life out of the driver.  It's going to depend on your listening position and your room acoustics.  That's where the tuning by ear comes in.  Tweeters with a 1st order HP (1 cap) will help to fill in the polar response from 2Khz up thru the x-over point.

@ Dmason.  Yeah, the rising response at first lead me to reject the driver.  But then it dawned on me that the fairly straight rising response would be easy to take care of.  See above.  This driver is certainly going to have "character", that's for sure.  We can hope that with the alnico magnet and paper cone will give it a good and musical character.

If the 10" Legend driver doesn't work out, there are lots of other nice 10s that could drop in.  The 10" Alnico Tone Tubby, for example.  Or build an adapter board and pop in an 8" driver.  Open baffle is forgiving that way.

[scorpion]

I say, this is a very interesting idea. The 1028 Legend would simulate much like the TT I published in the 'B200 - Eminence Alpha' thread. An about 40 cm wide baffle will in fact produce a hump to help the bump between 200 - 400 Hz. A 1.3 mH Jantzen foil inductor will set you back 23$ at Parts Express and this will certainly not harm the driver. You put the lowpass at about 800 Hz and work from there. OK Pano, go on with the design, build it and report. 

I own two pairs of Vintage Saba Greencones à la Lampizator 8" and 9 x 6", they are fabulous drivers soundwise.

/Erling

[panomaniac]

OK Pano, go on with the design, build it and report.

I may just do that.  Tho it has occurred to me that integrating the sub may be harder than it looks.
Getting anything below an F3 of about 45Hz is going to be hard with an efficient 12" sub.  So that leaves me with only an octave between the sub and the 10".  Hard to do that and not make it peaky.

Will think about it.......

[scorpion]

Look at Lampizator project 19: http://www.lampizator.eu/SPEAKERS/PROJECTS/P19/P19%20alnico%20open%20baffle.html and do it like that.
If I may suggest something for the design, I would go for Eminence Alpha15A in a 'topless-U' á la my 'Volks-OB', 60 x 40 cm baffle and 60 x 30 cm 90 degree wings. This would be down 5 dB at 35 Hz and match the Legend 1028 perfect in efficiency. Crossover 13 mH - 80 uF Linkwitz-Riley. The Ledgend I would cross at 300 Hz (giving an efficient crossover frequency at about 250 Hz, likewise L-R 40 uF - 6 mH, then comes Pano's brilliant low-pass idea, but I think 1 mH will do. So only to put in the two tweeters, cross them over and you are off. 

/Erling

[panomaniac]

Hey Erling - sounds like a pretty good idea.

I have never used a high Q driver like the Alpha 15, so I just don't know how much bass it can make.  I chose the Selenium 15WP3 because if its "ideal" specs  - Qts 0.61  But in fact it measures at 0.54, so a tad lower.  It produces very little bass on a 41cm baffle, so it's hard to go passive.  In active mode, tho, it really thumps!

The 10" Legend may need a high pass - a big cap.  Hate to do it, pity not to use the low end extension it will have.  But Lukasz always uses big film caps on his Saba drivers, soooo.....  (and I have used motor run oil caps, too)

Will think about the deep wing Alpha version.  Sure would be nice to keep it all open - right?

[scorpion]

In fact, I have tested the Alphas in my Volks-OB, I have bought 4 of them to build a pair of  H-baffles after MJK's latest paper.
They produce as good as the A&D1524s going about 5 Hz lower according to measurements. R1524s have Qts of 0.72, ideal, and the Alphas has 1.26 but bassquality and transients are very good with any of these speakers. Alphas is a very safe bet. And you can go with the big Obbligato caps for the Legend like I did in the 'Volks-OB'. I do like those caps.

Yes, OB allover !

/Erling