Fast, fun, Inexpensive OB project

[mige0]

Hi

Originally posted by BWRX
Thank you guys for sharing the design and the subsequent analysis with us. Cool stuff

An active crossover version of this would be interesting to compare to the passive version.

BWRX, here it is....

-----------

Panomanic, thanks for documenting this nice project of John Bush here.
As I had a Peerless at hand I thought I give it a try. Good bang for the buck indeed.


As I didnt have the " right " tweeter for the Manzanita 12, I built it with the Seas Millenium ( and an old ScanSpeak to the rear )





Built in half a day.





Total frequency response. Just white noise averaged 50 times at a distance of roughly 1m





The XO is basically the same as in passive.





FR of the Peerless SLS12 alone




EQ1 was done for flat FR – and FUN
+6dB / 28 Hz Q=2 will be plenty for normal listening though




EQ2 shows the need of high shelving. The notch at ~ 50 Hz is to equalise my room resonance.



Its fun to build and fun to listen at !

Greetings
Michael

[Pano]

Hey Michael - that's great!

I'd love to hear your version, but the hills of Austria are just a little too far from the hills of Hawaii. =)


Nice work. And thanks so much for the graphs and measurements - they go a long way to explaining what is going on. What is the crossover software you use?

I'm suprised at how flat you got it. Just seems to be a peak at 3Khz. Cone break-up? Baffle nasties?

My main question would be this:
Given that your measurements show flat (more or less) at 1M in front of the baffle, do they not sound too bright in room? Is there a lot of mid energy coming off the back of the Peerless and filling the room?

The original pair have a big midrange dip in the crossover to counter the midrange energy coming from the back. I see that your soft crossover does the same thing, but the resulting FR looks a lot flatter than I would have expected!

Thanks for pursuing this fun little speaker. Maybe it will encourage others to give it a try.

[ttan98]

Originally posted by panomaniac
Hey Michael - that's great!

I'd love to hear your version, but the hills of Austria are just a little too far from the hills of Hawaii. =)

you have all the fun in Maui, how is the weather there, recently have you been experiencing global warming there or just normal? Just curious...Cheers..

[BWRX]

Nice job Michael

[Pano]

Originally posted by ttan98
you have all the fun in Maui, how is the weather there

We don't have weather here. We leave that for you folks who live on continents. (I'm serious, actually)

Oh, just looked at the graphs again, it's a 300Hz peak, not 3Khz. My mistake. That points to the baffle edges, I suppose.

[terry j]

Originally posted by mige0
Hi

Total frequency response. Just white noise averaged 50 times at a distance of roughly 1m

Its fun to build and fun to listen at !

Greetings
Michael

Hi michael, could you (and others) comment on this please. I'm curious about using white/pink noise when setting up speakers to be flat.

Basically, when do we use which?? They are very different sounds, and so when we make the speaker flat on each, well they will sound very different won't they??

Additionally, what is the difference again if we use a swept signal?

Originally posted by panomaniac



Thanks for pursuing this fun little speaker. Maybe it will encourage others to give it a try. [/B]

I have this thread bookmarked for a later date Pano!! I doubt (at the moment) I would go passive with it, but hey the network is so simple that why not?? ha ha.

[Bratislav]

Originally posted by panomaniac

Oh, just looked at the graphs again, it's a 300Hz peak, not 3Khz. My mistake. That points to the baffle edges, I suppose.

I was curious about this too, and your guess is spot on. Would experimenting with baffle edge (making it progressively narrower, to a point) help, or should we attack it differently (moderate Q anti resonant LCR targeted at 300Hz) ? I'm extremely intrigued by this development as I still like the idea of multiple SLS's on a large-ish baffle as bottom end of a Arvo/R909 like 'clone'.

Bratislav

[MJK]

Originally posted by panomaniac


Oh, just looked at the graphs again, it's a 300Hz peak, not 3Khz. My mistake. That points to the baffle edges, I suppose.

Originally posted by Bratislav


I was curious about this too, and your guess is spot on. Would experimenting with baffle edge (making it progressively narrower, to a point) help, or should we attack it differently (moderate Q anti resonant LCR targeted at 300Hz) ? I'm extremely intrigued by this development as I still like the idea of multiple SLS's on a large-ish baffle as bottom end of a Arvo/R909 like 'clone'.

Bratislav

I do not believe the peak at 300 Hz is generated by the baffle or the sharp baffle edge, my guess is it comes from the driver or is an artifact in the measurement equipment/set-up or input signal.

[Pano]

Originally posted by MJK
I do not believe the peak at 300 Hz is generated by the baffle or the sharp baffle edge, my guess is it comes from the driver or is an artifact in the measurement equipment/set-up or input signal.

Is 300Hz too low to be caused by this size baffle, or is it the shape of the peak that points elsewhere?

Really, the FR looks very good. Just wondering about the little bump.

[MJK]

Is 300Hz too low to be caused by this size baffle, or is it the shape of the peak that points elsewhere?

I believe it is both. If I model the biggest possible baffle size, include the rear wings as part of the front baffle, the contribution to the response from the baffle peaks at about 400 Hz and produces a broad hump that spans 150 Hz to 700 Hz. A smaller baffle representation will shift this hump higher in frequency. The relatively narrow and sharp peak at about 300 Hz just does not look like anything generated by a baffle, it looks more like a driver or room resonance of some kind.