a pair of TLAHs (rectangular)

[fitzwaddle]

Some measurements with REW - take them with a grain of salt, as they were measured standing in my garage next to the open garage door, with only ballpark mic placement, speakers were a couple feet away from shelving on one side and a table on the other, yada yada. I just wanted to see approximately where I stand as far as effectiveness of the crossover and how the speaker components and build were behaving.

Test setup: Behringer ECM8000, AudioBuddy preamp/phantom power, UCA202 USB sound "card", laptop, Lepai TA2020 amp.

mic'd from 3' away, with mic vertically aligned with middle midwoofer:



mic'd from 6' away, with mic vertically aligned with middle midwoofer:



mic'd from 6' away, with mic vertically aligned with midwoofer 2nd from top (53"):



comparison of L speaker on axis, and ~30 degrees off axis on either side:



So dips around 1800 and 5500, but nothing horrendous I think - will remeasure when they're actually in place in the room, near wall boundaries, and then deal with EQ. I was actually pleasantly surprised that the midwoofers measure as well as they do on the low end, was expecting to have to cross them over to sub fairly high, but maybe I don't need to.

Crossover is 3rd order Butterworth @ 4000, fwiw.

Any thoughts?

[Tom Smit]

Can you throw up one more chart, with either both left or both right, at the 3' and 6'? The one you have shown is one Left at 3' and one Right at 6'. I'm curious about the drop off. Thanks.

[fitzwaddle]

Sure, here's the L from 3' and 6':



And the R from 3' and 6':



So does that look more like -3 dB or -6 dB per doubling of distance? Your guess is as good as mine. I guess more like a line source higher up, and less down low, as would be expected. I guess. Or not.

[Bill Fitzmaurice]

Reverse the polarity of the tweeter bank, that may fill in the dip in the transition area. And being a line source you can't get an accurate measurement unless the mic is at least four times the array height distant. preferably more.

[slavedave]

Love the look of your labour.
Can I ask a stupid question? If you put veneer on the sides (and back), did you have to slightly oversize the width of the baffle beyond the veneer and then flush trim it with a router, and then roundover? Or do you do it a different way? I only ask because I am thinking of building a tlah (or even an SLApro) and I would love to prodce something of such visual beauty as opposed to black duratex Of course, I would make a lined case for storing the speaker in between its use in a community centre for a couple of hours each week.
And I am very interested to see any graphs recorded following Bill's recommendation about distance from array.

PS. You do know that they say imitation is the greatest from of flattery!

[fitzwaddle]

If you put veneer on the sides (and back), did you have to slightly oversize the width of the baffle beyond the veneer and then flush trim it with a router, and then roundover?

Yes, that's exactly what I did. It was a bit of a pain to get the veneer edge to line up with that baffle edge, as the veneer wanted to stretch as I ironed it on. If I was to do it again, I might opt for veneering the side panels before assembly, and then use pocket screws from the inside to assemble.

The backs aren't veneered, just painted the same color as the face, in satin, over the plywood. And I painted the exposed edges of the plywood sides on the back black. Its the back, doesn't show, so didn't put a lot of effort into that part.

Look forward to seeing yours!

[fitzwaddle]

Reverse the polarity of the tweeter bank, that may fill in the dip in the transition area. And being a line source you can't get an accurate measurement unless the mic is at least four times the array height distant. preferably more.

Thanks Bill. So I'll try another measurement about 16' away then.

Not too excited about taking the backs off atm, so will hold off on trying the polarity flip until after I've tried and measured them in the house.

[fitzwaddle]

Got REW working with the ability to measure results of software EQ.

Here's the before:



Here's after EQ (w/ house curve):



I didn't bother trying to boost the L below 80, will likely eventually add a sub and crossover there - difference probably caused by some room and LP/mic assymetry.

[Bill Fitzmaurice]

Boosting below the natural LF corner at 80Hz would likely stress the drivers, so do a sub or two for sure.

[subharmonic]

And being a line source you can't get an accurate measurement unless the mic is at least four times the array height distant. preferably more.

How does that affect listening distance?

[Bill Fitzmaurice]

And being a line source you can't get an accurate measurement unless the mic is at least four times the array height distant. preferably more.

How does that affect listening distance?

Your ears aren't as fussy as a microphone, so even twice the array height away is fine.

[GirlyHandedDog]

And being a line source you can't get an accurate measurement unless the mic is at least four times the array height distant. preferably more.

How does that affect listening distance?

I would guess this has something to do with it:

[subharmonic]

Thanks Bill.

GHD you have some mad drawing skills. Maybe a bigger smile and the hair being blown back more would have been more descriptive.

[GirlyHandedDog]

Thanks Bill.

GHD you have some mad drawing skills. Maybe a bigger smile and the hair being blown back more would have been more descriptive.

Yes, I know. Please do not print this out and hang it on your wall without my consent. Thanks.

[Jon Barnhardt]

fixed