JX92 + Ribbon + VTL enclosure: measurements

[neil_kaye]

Several years ago i built a pair of JX92's in a VTL enclosure. I was not completely satisfied with the results so the enclosures and drivers sat unloved for sometime.
Fast forward a few years and i decide to re-look at the project using Jim Griffins famous ribbon + Jordan combo and crossover, but employ it in a VTL enclosure.
I embarked on this project knowing full well that it would in all likelihood end in total failure. As a full disclosure i have almost NO experience building speakers. Sure i have nailed together a few pairs using tried and tested designs (always the smart way to go) but nothing as foolhardy as this.
I completed the first channel a few weeks ago and fired it up. They sound OK, certainly better than the Jordan alone in the VTL enclosure. I decided to measure the built speaker using True RTA. I would really appreciate it if some of you might give me a pointer using this software.
I took several measurements with no signal generator and averaged them out. This became the room base line, ambient noise etc.
I then took measurements with the signal generator (pink noise) on. I subtracted the first curve form the second and the attached file was the resultant in closure speaker measurements.
What do these measurements look like?
Thanks
Neil

[Scottmoose]

Originally posted by neil_kaye
What do these measurements look like?

Roughly what I'd expect & suggests they could do with either a shelving filter (or some other form of Eq), to compensate for step-loss, or pushing against a rear wall like my own MLTLs are.

[Bob Brines]

Now you have to understand that this is my opinion and reflects how I do business. It my well not jibe with the majority of posters here, but....

1. The speaker rolls of too soon at the bottom. The bottom half-octave of normal music is totally missing. It will sound very thin on orchestral music and anything with stand-up bass or bass guitar. As long as you intend to roll in a sub at 60Hz, no issue, but at least for me, it won't cut it as a stand alone single driver speaker. This is why I have no designs for anything smaller than a 6" driver.

2. You have ~4dB of baffle step and ~3dB of baffle hump. A BSC filter will fix this, or as Scott suggests, push them up against the wall.

3. You really have to do something about that big bad hump starting at 3500Hz. It looks like a Lowther! This gives a harsh, sibilant sound to the speaker. Wait. You have a tweeter. You probably have the tweeter crossed in too low. Rerun you measurements on the Jordon only and the tweeter only.

4. Once you get the above fixed, you are going to wind up with ~3dB BBC dip. This is not necessarily bad. It is just that the overall sound will be somewhat laid back. To me it is sure a better sound than the forward sound a lot of single driver speakers have because of a reverse BBC dip.

Bob

[neil_kaye]

This is what i imagine the next few steps would look like.
1. Keep this channel as is
2. Load the drivers into the second channel enclosure and take measurements of each without a cross over involved.
3. Import these measurements into something like speaker workshop.
4. Work on a cross over that can maybe help with a baffle step compensation for the lows and cross over the tweeter higher up.
5. Ask tons of questions on how to use speaker workshop... anyone know of a good tutorial?

[Bob Brines]

Speaker Workshop was abandoned by its author some six years ago. There are some diehards that continue to press on knowing that there will never be an update. Try Claudio Negro . I think that he is the current keeper of the documentation.

I was never personally able to get consistent results with SW and eventually bought Sound Easy. SE is no joke to learn, but at least I get consistent measurements without fooling around with the jig and settings. I'm trying to make a buck at this and just can't afford to get bad data. Interestingly, though, I do use SW to publish charts. SE has no good way to present data for publication.

Bob

[graaf]

let me link to some independent measurements of the JX92s for a comparison: http://www.hifisound.de/oxid/out/oxb...JJ-1110858.pdf

no humps nor bumps
"ganz sicher eine Ausnahmelautsprecher"

best,
graaf

[neil_kaye]

I will run the same tests (over the next week or so) with no cross over and publish the graphs.

As an aside is it correct to take room / ambient sound measurements and then subtract them from the speaker / driver / enclosure measurements. It is just that without an sound isolation chamber don't you need to compensate for microphone and room distortion?