Budget Duelund Three-way

[gregrueff]

Thanks for the kind words.
The midrange enclosure (PVC pipe) is acting as a brace of sorts, extending all the way back to the rear wall.

I'm declaring success today on the TG9 rebate! Not as good as a CNC, but good enough!

Cheers,

[gregrueff]

Some progress.
The black strip is neoprene sheet. It is sandwiched between the sub-baffle and the baffle.
Getting the chamfer on the backside of the TG9 was tricky, and I had to create a whole separate jig for the router bit to guide off of. But it is totally necessary.
Wiring has been started and in-box measurements will commence shortly.

Cheers,
Greg

[gregrueff]

Wiring is complete!
All three drivers' wiring terminate at the cobbled-together rear binding terminal, allowing me to prototype the crossover out-board of the box.

[gregrueff]

Initial impedance sweeps were completed.

As expected with an unstuffed box, the woofer exhibits three blips on the impedance curve corresponding to the 1/4 wave length of each box dimension. The midrange exhibits one of the same blips corresponding to the depth of the midrange tunnel.

Once the boxes were fully stuffed as planned, the blips are pretty smoothed out and the impedance drops substantially.

Stuffing is Ultratouch Denim Insulation and Acousta-Stuf.

[gregrueff]

I managed to take some in-box FR and HD measurements today of the first speaker.

Mainly I wanted to confirm my earlier position that my test baffle had a resonance around 100 Hz that was only really excited once the relatively beefy DA215 really moved it.
In box measurements confirm the DA215 will work very nicely for the bottom end!

Also, the predicted diffraction dip at 2 kHz showed up as expected in the TG9's response. HD measurements in-box to make sure I didn't screw anything up.

Finally, an in-box HD measurement of the DX20BF00-04. Not that I distrusted HiFiCompass' measurements, but now they can be directly compared to the TG9. I think these two drivers' HD will blend very well at the crossover point.

Interestingly enough, diffraction resulting from the recessed baffle (no grill cloth frame installed or designed yet) mostly affected the DX20 more than anything. The TG9 dip at 2 kHz was there, but apart from that it was ruler straight!

Note that all the edges are perfectly square, have no round-overs, and the entire front baffle is currently recessed a good 5/8". This should be a diffraction nightmare in theory.
I have some ideas how to deal with this nonetheless...

[gregrueff]

I did some experiments with grilles today (or at least the grille frames).

The inspiration comes from the always controversial "xmax" (norman druce) who used a similar technique on his monitors. When I saw the pictures he posted, I always thought it was an interesting idea with merit and wanted to try myself. The first attachment to this post are his monitors.

Mocking-up in cardboard is a quick and easy way to test this without laboriously cutting wood. I know xmax subb-ed his work out to someone with a burn-table and I can totally see why now. This sucks! I think the precision of a laser is also justified for these types of cuts too.

The cool thing is this seems to work relatively well.
This dreamt-up geometry does a pretty good job smoothing out the dip at 2 kHz on the TG9. I'm sure experimenting with iterations upon this could yield some minor improvements as well. The tweeter's response also looks better in my eyes- not quite as large a dip and peak as before.

I'm still trying to figure out how I want to apply this technique to a wooden frame.

[Steve Lee]

Wow! - Those response curves are FLAT!

[buggers]

If I recall correctly Jon made a star cutout made of felt to surround the tweeter on one of his projects.

[dkalsi]

Hmm..........it just so happens that I have a pair of the same tweeters, mid-range, and woofers on hand. I also have plenty of baltic birch plywood on hand :-).

Thanks for sharing this design. Do you plan to make the crossover public? Also - what is causing the dip at 2Khz? Is that just the natural response of the TG9?

[gregrueff]

Of course, I plan to share everything!

The only reason I haven’t published the crossover yet is because it was all simulated and based on no real measurements. I expect it to change, but the first iteration provided a proof of concept.

The dip at 2kHz is diffraction related to the driver being smack in the middle of the baffle. I did this for aesthetic reasons, understanding the diffraction trade-offs.
The TG9 is ruler flat otherwise.

Cheers!

[Billet]

About the 2k dip. First of all, 2k may not be a bad place at all for a dip. I defer to the endless smiley face curves on old school equalizers in the 70's and 80's. Second, and probably far more important. If it is caused by diffraction, then I suspect the dip will average out in the room response.

In other words, it all looks good to me!

[Steve Lee]

YES! around 2KHz is an overly sensitive listening frequency which frequently needs attenuation so I say it is fine as is.

[gregrueff]

Totally agreed guys.

Also to elaborate further, I did favor off-axis diffraction over the on-axis dip. I rarely get the opportunity to sit down in the sweet spot. I'm usually listening doing something else.
You can see from the diffraction curves that the dip smooths out pretty darn well off-axis.

I've also seen Curt Campbell say he designs with the 15 deg. off-axis tweeter FR because he doesn't toe his speakers in. Similarly, I do not intend to toe these in, so on-axis may well be considered off-axis in this case.

[gregrueff]

The TG9 is ruler flat otherwise.

I have to correct myself here. I had forgotten a lot, as I've been thinking about this design for over a year and I need to refresh my memory on some things.

The TG9 actually has some built in baffle step so on this baffle it appears ruler flat apart from the 2 kHz dip. Refer to the quazi-infinite baffle FR curve from my earlier testing. Once on the baffle it becomes very nice.

[gregrueff]

Progress today on the grilles.
I've never made so much sawdust.

[gregrueff]

So much time and effort into these things.
But the look of the cabinet is as big a deal to me as the performance.
I'm super curious to see what grille cloth does to the response.

[Steve Lee]

Fabulous!

Most impressive wave guide.

[dkalsi]

So much time and effort into these things. But the look of the cabinet is as big a deal to me as the performance. I'm super curious to see what grille cloth does to the response.

That’s some serious work on the grills!! Sent from my iPhone using Tapatalk