Opinion - would this driver combo work with 2nd order xo slopes?

[brianpowers27]

Opinion - would this driver combo work with 2nd order xo slopes?


woof Rs225-8 (x2) parallel
mid rs100-4 (x1) 500hz
tweeter (dayton silkie dc 28 2500hz

[cjd]

Forget the silkie, it's junk by comparison. You'd be better off with one of the $5 dome tweets PE is selling and crossing ~3kHz.

You should be able to go 2nd order though it'll be tough on the tweeter - so better choices than what I've just mentioned would likely be in order. Find something with a small face so you can comfortably push the XO higher.

[JonMarsh]

You will need some trap filtering for the RS225's. I'm with Chris, get a real tweeter! D26NC55 are in stock at Madisound for under $30. Just my 0.02. And if you can afford a little more refinement for the crossover, go with a Duelund alignment instead of just LR2.

[brianpowers27]

I'm with Chris, get a real tweeter! D26NC55 are in stock at Madisound for under $30. Just my 0.02. And if you can afford a little more refinement for the crossover, go with a Duelund alignment instead of just LR2.

I am trying to run a tight budget. I may have to pony up on the tweeter though. I wouldn't do a trunc flange since I am trying to keep the woodworking simple. I am debating whether or not the 2nd order slopes are worth the effort when the .75" dayton nd tweeter sounds so nice and costs so little.

Perhaps I could go 3rd order between a nd20 and the rs100 @ 3500hz...

[JonMarsh]

I'd suggest developing and modeling an overall system target function before even selecting the drivers, just keeping in mind what you know about the driver limitations. That's what I've been doing the last several projects, and it helps channel the design process and choices.

But, I'm an engineer, and maybe I'm thinking too much like an engineer in my approach- afterall, we do this for fun, right? :W

[brianpowers27]

I usually do the slow and steady modeling approach before driver selection.... Usually.

This time around, I already own the rs225s-8(x4) and rs100-4(x2). I am also thinking about purchasing at the tent sale tomorrow.

Side note: How do you feel about the Dayton nd28 tweeter? It is very small and appears to be well suited for crossing low ~2k.

[JonMarsh]

Well, as Zaph notes, the sensitivity is a bit low, and the Qts bump is something of an issue. It's a decent performer distortion wise in it's price class (~$20), but I'd rather have a stellar performer distortion wise for a few dollars more (that D26NC55), with much cleaner CSD to boot. Considering the time and effort put into designing and building, why be penny wise and pound foolish? It's a much better match to the other drivers than the ND28AF.

Again, just my opinion- it's a free country. Unless you ask Evil Twin.

[brianpowers27]

Considering the time and effort put into designing and building, why be penny wise and pound foolish? It's a much better match to the other drivers than the ND28AF.

You make a good point

What else is in that price class that doesn't have a trunc flange?

[JonMarsh]

Why is the truncated flange a problem? Just make a felt diffraction smoother; it doesn't need to be rebated.

I've looked at a LOT of tweeters in that price class and above; below $50, the Vifa is IT. Again, YMMV, but if it's performance that counts...

[PoorboyMike]

I can't remember where I saw it, but I think there is an adapter for the D26 to make it a round flange. Now if I could just think of where it was at????

On another note, I made a template for the D26 that works fantastic. You'll need a Whiteside 1/8" inlay kit to make it work though. Wouldn't be hard at all for me to copy it.

[fjhuerta]

Get a DQ25 by Vifa.. an amazing tweeter for $15. I'm crossing it 2nd order at 3.8 KHz without issues on my ZDT3.5 design.

[Jed]

And if you can afford a little more refinement for the crossover, go with a Duelund alignment instead of just LR2.

Some values of Aleph in the Duelund 3way result in almost perfect Linkwitz LR2 slopes. Still refined? :W

[Curt C]

The North Creek D2506 is a nice sounding little tweeter, able to cross 2K 2nd order. I'm pretty sure its still available from George at North Creek For $36/pr.

C

[JonMarsh]

Some values of Aleph in the Duelund 3way result in almost perfect Linkwitz LR2 slopes. Still refined? :W

All of the Duelund variants regardless of initial slope or final slope in stop band will have good phase tracking across all drivers, if properly implemented. See, I'm the local Duelund nut so I get to adovocate and wave my arms while standing on my soapbox, as long as I don't overbalance and fall off...'

When you're that close, why not go all the way? :W

[cjd]

Get the D26 and check out Zaph's notes on rear-mounting it to a "built-in" waveguide of sorts. No funky flanges and all the good stuff.

[augerpro]

When you're that close, why not go all the way? :W

I used that line a lot in my youth.

Whether you go Duelund or LR2 you should do some modeling using DennisH's Duelund excel spreadsheet so you can see what lower crossover point you can get away with and not run into trouble with the mid sensitivity. My gut inclination is to cross fairly low (and fairly high to the tweet for that matter), but this sensitivity issue will work against that. Some modeling with the Asymetric Radiation worksheet at FRD tools could be very helpful on selecting crossover points.

Also consider BW3 in the lower crossover to help this transition.

The DQ25 is an excellent suggestion depending on what crossover and slope you use, and if price is a big factor. The D26 is bit more money, but more or less a step up too. Again modeling it might help you decide. If you need more low end grunt the SEAS 27TDFC, 27TBFC, or 29TAFW are reasonably priced possibilities.

BTW cool idea. This sort of alignment with small mids and low order slopes has captured my interest since using the Morel MD55 in a a Duelund alignment. Now if I can just get the response I want out that dang TB 3" dome midrange on a waveguide...

[Dennis H]

Some values of Aleph in the Duelund 3way result in almost perfect Linkwitz LR2 slopes. Still refined? :W

Hi guys, it's me, the annoying old broken record. :B

All Duelund crossovers use LR2 filters at the same frequencies for all the drivers. Forget aleph and all that. Choose your XO frequencies, say 500 and 2500 if that's what works 2nd order with the drivers.

Woofer: LR2-500 lowpass + LR2-2500 lowpass
Mid: LR2-500 highpass + LR2-2500 lowpass
Tweeter: LR2-500 highpass + LR2-2500 highpass

Simple, huh?

[Jed]

I was thinking of you, Dennis, when I made the comment.

[Dennis H]

Thanks Jed, you're a good straight man. We should take our act on the road.

[Dennis H]

When you're that close, why not go all the way? :W

Ahem. I think we need ET to start supervising the compliance division in the Duelund sector of the galaxy.

[Evil Twin]

The old man is often not aware of the "deeper" meanings in what he says... the memory banks may be purged of unused connotations and meanings in order to free up working buffer space for current projects.

I will have a discussion with him and if necessary we will install the equivalent of a droid restrainer chip, to assure he does not offend the delicate sensitivities of the spiritual souls participating on this forum board. If that doesn't work, we'll suspend his forum privileges temporarily- it may well be that pictures as posted on fjhuerta's recent thread have triggered latent suppressed memories.

[brianpowers27]

I ended up finding deals.

New drivers:

4x rs225s8
2x rs28a
2x rs150s8

I may try for some level of dueland 2nd order. I really need to find a good xo for all of this. I went with the rs150 for sensitivity. I realize it may not reach as high.

I may even consider turning this into a 4.5 way with a 5/8 neo dome on the top.

[JonMarsh]

The RS28a is a nice driver to work with; goes low and is very neutral sounding as metal domes go; very similar sonically to the SS 98000 which is my favorite metal dome (if you don't push it too hard).

[brianpowers27]

I saw the calculator for choosing the dueland frequencies once. I can't remember where I found it.

I am thinking that something along the lines of 300-2200 hz xo points would work well. I can't push the rs150 too high though.

[JonMarsh]

Yeah, that sounds like some good choices for the crossover points. The RS150 shows energy storage issues (linear distortion on shaped sine tests) from 1400 Hz on up, though there are hints of it even at 1000 Hz (the W15 beats it, but many other drivers in that class are worse). I presume you've chosen the RS150 because of SPL levels you want? Otherwise, I'd suggest the RS100 if you want more midrange resolution- two of the RS100-4 in series would be nice, though with dual woofers, maybe you'd want two of the 8 ohm in parallel. It's kind of a wash as far as sensitivity, and the RS100-4 is flatter. The sensitivity issue may be manageable, if you use an alignment where the tweeter and woofer overlap fill in the last two dB of output needed in the midrange.

Just some thoughts... have a good Sunday!

[brianpowers27]

Yeah, that sounds like some good choices for the crossover points. The RS150 shows energy storage issues (linear distortion on shaped sine tests) from 1400 Hz on up, though there are hints of it even at 1000 Hz (the W15 beats it, but many other drivers in that class are worse). I presume you've chosen the RS150 because of SPL levels you want? Otherwise, I'd suggest the RS100 if you want more midrange resolution- two of the RS100-4 in series would be nice, though with dual woofers, maybe you'd want two of the 8 ohm in parallel. It's kind of a wash as far as sensitivity, and the RS100-4 is flatter. The sensitivity issue may be manageable, if you use an alignment where the tweeter and woofer overlap fill in the last two dB of output needed in the midrange. But mostly they will just be high fi home speakers.

--I think the rs150 will be more of a filler.
--I considered multiple rs100s I think I would prefer to keep the polar pattern uniform.
--I think the rs150 will be better up to the abuse.
--I chose the RS150 because of the kiler deals I got at the tent sale.
--I still need to find the exactxo points for 2nd order. Dennis mentions that the true alignment is all 2nd order. Can I use John K's quasi TP 2nd order spreadsheet or would another alignment achieve better results?
-- One day I may take these out somewhere and try doing some smaller PA addresses.

I walked out of the sale with the following for $75

--2 x rs225-8
--2 x high low adapters
--50ft 16 gauge spk wire
--2 x rs28 tweeters
--2 x rs150-8 woofers
--1 x rs28 with "perfect" dimple (Prototype or dent? )
--One free accelerometer (courtesy of FBOV.)

[JonMarsh]

--I think the rs150 will be more of a filler.
--I considered multiple rs100s I think I would prefer to keep the polar pattern uniform.
--I think the rs150 will be better up to the abuse.
--I chose the RS150 because of the kiler deals I got at the tent sale.
--I still need to find the exactxo points for 2nd order. Dennis mentions that the true alignment is all 2nd order. Can I use John K's quasi TP 2nd order spreadsheet or would another alignment achieve better results?
-- One day I may take these out somewhere and try doing some smaller PA addresses.

Using an LR2 is an approximation for one area of Duelund alignment; the band pass is 2nd order high and low, but the high pass and low pass are 4th order with substantial pole spread. You can get close with just LR2- close.


Witness, this is the target transfer function for my current project, (this is a very exact fit to a Duelund Target I calculated in MathCAD, aleph = 3.2, center frequency = 900 Hz )




And this is the network required to realize the transfer function correctly for nominal resistive driver loads (first phase of modeling).



Some typical variations in a Duelund alignment, depending on the value of aleph; all are technically 4th order high pass and lowpass, though for some portions of the transition region could be approximated with a second order.



Here's a link a Danish site for calculating Duelund transfer functions online- use this as a target for your speaker software, or use DennisH's Excel sheet to generate PRN files for the targets.

http://pro.speakerbuilder.dk/synkron/