Salvage DIY speaker you don't like.

[mightycicadalord]

Well a bit of a bummer but I have to say I don't really care for the current pair of DIY speakers I have. I bought and built the Amiga kit from PE and some days they're ok but most days I don't care for them much. I find them quite overly bright and EQ doesn't really fix it. I have two subs and treatments and I've gotten things quite flat (with a nice slope) and still find them to have this sort of glare to them. They have a graininess to them at times. I'm a little bummed because I helped my friend pick up some monitors and chose the kali lp6 v2 and they frankly sound better.

What would you do? Try to make something else with the drivers? Sell the kit in parts? No one is interested locally for the whole package it seems.

Definitely one of the downsides of DIY that I didn't really take into consideration.

 

[hex168]

Look at it as an opportunity! Any other tweeters out there you've been hankering to try? The designer posted enough information for you to re-engineer the Amigas:

Paul Carmody's DIY Speaker Pages - Amiga

Home > Floorstanding Speakers > Amiga

sites.google.com

First thing I'd do is check to see if the resistor in the crossover matches he one specified for the tweeter model you have (see above link). If that is not the problem, you have a nice test-bed to try out tweeters.

Me, I'd try a semi-Econowave with a JBL waveguide (a real one) and a PRV 290s compression driver.

JBL 338800-001 Replacement Horn Lens

Brand New Genuine JBL Replacement Screw-on Horn Lens #338800-001. Used in JRX100 Series, JRX200 Series, PRX600 Series, Sound Factor Series. Used in the following models: JRX112M, JRX112MI, JRX115, JRX115I, JRX125, JRX212, JRX215, JRX225, SF12M, SF12MI, SF15, SF15I, SF25, PRX612M, PRX615M...

reconingspeakers.com

Home

PRV Audio D290Py-S 1" Polyimide Horn Compression Driver 8 Ohm 1-3/8"-18 TPIPRV Audio's top-of-the-line D290Py-S polyimide 1" screw-on compression horn driver is designed to offer unsurpassed natural upper midrange and high frequency response, with accuracy not found in most manufacturer's stock...

www.parts-express.com

You can look through the enormous Econowave threads for passive crossover hints, but I'd start with a 2K crossover and leave the woofer crossover alone, and modify from there if it seems to have potential:

Zilch's AK Design Collaborative - Econowave Speaker

******** Project Summary: http://audiokarma.org/forums/showthread.php?p=1856214#post1856214 Member Projects and Acknowledgments: http://audiokarma.org/forums/showthread.php?p=2260584#post2260584 Member Listening Impressions...

audiokarma.org

Flex Your PCD Mettle: - Techtalk Speaker Building, Audio, Video Discussion Forum

Want a second or third opinion about your speaker cabinet design or other audio related problem? Post your question or comment on the Technical Discussion Board. Hundreds of technicians, engineers, and hobbyists, nationwide read and discuss electronics related questions each week. We welcome...

techtalk.parts-express.com

Or a GRS planar:

Home

GRS PT6816-8 8" Planar Slim Tweeter 8 OhmGRS PT6816-8 8" Planar Slim Tweeter 8 OhmDesigned to deliver high efficiency with extremely smooth response.With wide dispersion capabilities and minimal mounting flange, the PT6816-8 is an excellent choice for use in high output line arrays. The...

www.parts-express.com

or (back-ordered):

Home

GRS PT2522-4 3-1/2" Planar Tweeter 4 OhmGRS PT2522-4 3-1/2" Planar Tweeter 4 OhmDesigned to deliver high efficiency with extremely smooth response.The PT2522-4 is a perfect choice for use as a drop-in replacement in existing systems or for use in new designs where low distortion and accurate...

www.parts-express.com

Another thought - maybe, due to manufacturing variation, the breakup on the woofer is not aligned with the notch in the crossover and that peak is coming through. Can you close-mic it and see if that's the problem?

Good luck!

 

[mightycicadalord]

Oceanway looks to be using the RS180 with a seas TDFC, I though that might be something to try, though they don't have a great track record as far as making good speakers.

I have some close mic measurements but the db scale cuts off right as the break up is visible, can do another.

There's also a design by toids diy where he uses an RST28f. The sim is at least a flatter than the amiga sim but not sure how that would translate to real world.

 

[hex168]

Hmm. If you are only 20 dB down at the woofer breakup, that might be bothering you. Try playing with the notch filter (I think it's the cap across the 2.0mH inductor), or adding a tank circuit.

 

[HammerSandwich]

Carmody notes that the RS180 is difficult in a 2-way, because breakup. In addition, the Amiga's 2kHz XO implies directivity mismatch with these drivers, which would give too much off-axis output in the sensitive 2-3kHz range.

I'd test if uncontrolled polars cause the brightness by using absorbers to constrain the beamwidth. It's only a temporary setup, so who cares if it's ugly? Just set up panels (or thick layers of blankets/towels) flanking the speakers to narrow the HF dispersion.

Regardless of test results, both problems could be addressed by adding a midrange. Cross the RS180 around 500Hz, and use a ~4" mid to match the tweeter's pattern at 2kHz. Instead of building new boxes, put the mid in a small enclosure which rests on your current speakers. A couple TC9s & foam/cardboard boxes would allow a cheap & easy test, especially if you can run it temporarily via DSP filters.

 

[mightycicadalord]

I have this set I did outside not too long, it's gated so nothing below like 300hz is accurate. Sorry they aren't labeled, it's 0 degrees to like 90 I believe. Orange would be directly on axis. I wanted to try and generate my own spin but still learning on how to do that. A 3 way does sound awesome but I am not well versed in xover design at all. I would definitely need someone else to help with that if that ends up being the route taken.

I put some panels to sides of the speakers and it's hard to say but they seem a little darker? The imaging improved a ton though, like I mean a ton.

 

[hex168]

Carmody notes that the RS180 is difficult in a 2-way, because breakup. In addition, the Amiga's 2kHz XO implies directivity mismatch with these drivers, which would give too much off-axis output in the sensitive 2-3kHz range.

I'd test if uncontrolled polars cause the brightness by using absorbers to constrain the beamwidth. It's only a temporary setup, so who cares if it's ugly? Just set up panels (or thick layers of blankets/towels) flanking the speakers to narrow the HF dispersion.

Regardless of test results, both problems could be addressed by adding a midrange. Cross the RS180 around 500Hz, and use a ~4" mid to match the tweeter's pattern at 2kHz. Instead of building new boxes, put the mid in a small enclosure which rests on your current speakers. A couple TC9s & foam/cardboard boxes would allow a cheap & easy test, especially if you can run it temporarily via DSP filters.

Good thought! The TC9 is perfect for this test, you don't even have to reconnect the tweeter.

If directivity and excess off-axis output from the tweeter is part of the problem, the JBL waveguide in post #2 should give you a good directivity match somewhere between 1.3KHz and 2Khz. (Easier to try than to model.)
Also:

Test Peerless DX25TG09-04 (XD-270 F / 4) on the waveguide

The Vifa / Peerless XD-270 F / 4 was combined with Monaco's WG-300 waveguide using a CNC milled adapter plate.

heissmann-acoustics.de

He used to sell a CNC adapter for this waveguide that should work with your tweeter, but I do not see it on his website anymore.

 

[HammerSandwich]

<off-axis chart>

That clearly shows the hotter off-axis from 2-4kHz or so.

I put some panels to sides of the speakers and it's hard to say but they seem a little darker?

That makes sense. Does it tame the "overly bright" issue better or differently than your EQ attempts?

The imaging improved a ton though, like I mean a ton.

Amazing, isn't it? You're hearing more recording & less of your room. Now try sitting 2' closer.

 

[mightycicadalord]

Yes I would say the brightness improved in a way that it didn't with eq. Sitting close improves imaging, I have always felt these speakers are great nearfield but suffer in midfield.

 

[HammerSandwich]

So, what's your plan? Waveguide, 3-way, other?

 

[mightycicadalord]

Not sure, I am going to PE tomorrow though to pick up some things for someone. What would you guys do?
The tc9 is out of stock buy they do have the JBL (dayton clone) horns. I am maybe partial to the 3" because I have an ikea eneby I've been listening to in the garage while I work on stuff, and that little thing reveals a lot of things in the mid range I wish I could hear on the Amiga's. I know it's a little bluetooth speaker, but there are parts of the mid range on the Amigas that feel "hidden" if that makes sense. My only experience with CD's is, not great so I'm a little hesitant on that.

 

[hex168]

For a 3", PE has this in stock (Dennis Murphy likes it):

Home

Tectonic TEBM46C20N-4B BMR 3" Full-Range Speaker 4 OhmThe Tectonic TEBM46C20N-4B 3" full-range speaker achieves the maximum amount of quality sound from small enclosure sizes. Aside from the benefits to acoustic engineers, the BMR's compact, square form factor also appeals to designers. BMR...

www.parts-express.com

Since PE raised the price of the JBL clone waveguide, I'd prefer the real one I linked to.

Going back to this thread:

Compression drivers?

translation for 'warmduscher' would be "pussy", I guess on a sidenote: I live in the tropics and dusche warm lol

www.audiosciencereview.com

You could also try the JBL/PRV combination with your Dayton PA380s, it's a two-fer. Daniel T posted the 380's off axis response and it might work up to 1300 Hz, the lower limit of the JBL waveguide. The uncertainty is because Dayton, I believe, does their off axis measurement on a large (IEC?) baffle and the off axis behavior on your baffle size will be different. Of course, you can go 3-way there, too, but a 3" would not be a good match with a PA380.

 

[mightycicadalord]

I would love to try a build with these drivers and the BMR driver. I'm not too keen on using my PA380's, as I currently have the wood I'd need for a three way and not looking to go grab another sheet considering the cost of wood at the moment. I will be building new cabinets and tuning the woofer higher, no sense in trying to pull 34hz roll off out of them like the current build does.

I would like to build some sub cabinets for the PA380 to replace the dayton budget powered subs I have, although I find them to fine.

 

[HammerSandwich]

The tc9 is out of stock...

Maybe PC83 for cheap experiments? For $25-30 each, you could move up to the RS75 or LY302 as tested by Feleppa? Faital 3FE or 4FE is always an option.

@Wolf & @mtg90, any ideas?

 

[HammerSandwich]

Also:

Test Peerless DX25TG09-04 (XD-270 F / 4) on the waveguide

The Vifa / Peerless XD-270 F / 4 was combined with Monaco's WG-300 waveguide using a CNC milled adapter plate.

heissmann-acoustics.de

Wow, I'd forgotten how good that looks. OP, you need a pair of WG-300s.

 

[mightycicadalord]

Wow, I'd forgotten how good that looks. OP, you need a pair of WG-300s.

That is seriously impressive, might have to go with those. I can make my own adapter.

 

[Wolf]

Maybe PC83 for cheap experiments? For $25-30 each, you could move up to the RS75 or LY302 as tested by Feleppa? Faital 3FE or 4FE is always an option.

@Wolf & @mtg90, any ideas?

The RS75 is not that sensitive, so it becomes harder to use in singles.

Heard the LY302 a couple weeks back, and it takes some notches to smooth it out. Otherwise, it was easily x'd at 100Hz or so.

PC83 is a gem at its price point.

 

[Rick Sykora]

Some hack modeling (using manufacturers measurements) in VituixCAD suggests that increasing the tweeter cap to 10-15 uF would help with the 2.5 kHz dip (and improve directivity).

Might be worth a try before an entire rebuild.

 

[mightycicadalord]

I'll try the cap mod, but I did leave PE with some BMR drivers today.

 

[hex168]

Gathering some hints from BMR speaker threads, they seem to need a VERY well sealed box, stuffed.
I look forward to reading about your results!