Three-way tower with Dayton RS225's, Help needed.

**cjd** · 23 March 2005, 11:53 Wednesday

I have a design sketched out that uses a pair of the RS225's, a RS150 (could probably go with a pair of these too), and the Seas 27TDFC. I don't think it made your budget though. Close-ish (IIRC it was ~$750?) I haven't looked at it in a while and never really pushed it all the way, so that estimated cost may be way off base.

The other thing is I would suggest you give serious thought to a two-way with a pair of the RS225's and either the forthcoming RS28A tweeter or the Seas 27TDFC. At this point I would wait.

There are two trouble spots I see with the drivers you've picked. The big one - unless someone has them, or you have the ability to build a box and get accurate measurements, it's less likely you're going to get a crossover that's anywhere near what it should be. The second is that they seem to be a mismatch to me. The tweeter isn't what I would pick at all - I'd go for a Fountek or similar if you want a ribbon. The mid is on the small side and I can't find any specs to see if it will even like crossing as high as required to cross to a ribbon.

As to bettering what you have available (the JBL), I think the RS225 is far and above better. If your fixed goal is sticking with your budget and the RS225, go two-way. A 3-way is trickier and at least double (if not triple) the cost of a good two-way crossover. To limit it to a budget while picking drivers that require care in crossing is an injustice and may end up compromising the end result. And I have absolutely no doubt that one of Jon's 8" two-ways would blow the doors off that JBL.

Oh yeah. On your SPL numbers, you are neglecting to take baffle step into account. If that's needed. No 94dB sensitivity is going to happen with a pair of 225's though.

C

**m_bathke** · 23 March 2005, 18:49 Wednesday

Hey CJD thanks for the Feed Back. One thing that I forgot to mention is that the Reciever that I am going to use for my home theater will make the out put of the speaker flat by the way of Yamaha's Parametric Audio Optimization, which is a parametric equalizer that it will set. A two way would be fine too, I just couldn't find a tweeter to cross that low, and I wasn't going to try to tune a 4th order or higher by ear.

So where would one find one of John Marsh's designs on a two way, I have seen them mentioned reguarly but, I haven't been able to find the a build sheet for any of them.

Again, thanks for the feedback.

**cjd** · 23 March 2005, 19:00 Wednesday

1. You want the Yamaha magic to have nothing to do except deal with room issues, not speaker issues.

2. By ear? Your first ever? With this much invested? You're insane. :P

3. Couple threads to peek at... though I'm not finding BOM etc. in these on a quick skim. This basic design probably could be tweaked to use the RS225 and the RS28 without too much work, if you ask Jon really nicely.

I know he has the RS225 in use in his TM variant and is already working with the RS28, so it may be there just waiting to be published.

Ported M8n MTM designs..... - HTGuide Forum

https://www.htguide.com/forum/showthread.php?t=3606

People have been interested in building a ported (fullrange) version of the MTM that Tibor built a couple of years ago. For those that don't remember, Tibor's sealed design speaker, here's a pic. To date the 'best' ported version using the Hi-Vi drivers is Jon's M8a-T&A Just so people don't get confused, either

MTMs for my brother this time! - HTGuide Forum

https://www.htguide.com/forum/showthread.php?t=8780

DIY (Do it yourself): Cabinetry, speakers, subwoofers, crossovers, measurements. Jon and Thomas have probably designed and built as many speakers as any non-professionals. Who are we kidding? They are pros, they just don't do it for a living. This has got to be one of the most advanced places on the net to talk speaker building, period.

Also, check out the stickied "Modula MTM" which uses a pair of the RS180.

If you want to peek in on a 3-way in progress with the Daytons... The process is kind-of being laid out a little here, which you may find interesting.

C

**m_bathke** · 23 March 2005, 19:12 Wednesday

Well you are not the first person to tell me I am a few bricks short of a load. But who isn't at some point in there life. Again thanks, I am beginning to get the feeling that if I just jumped into this project I might of ended up with a pretty big surprise.

**m_bathke** · 23 March 2005, 19:22 Wednesday

So I have one more question for you, if you don't mind. If you take two drivers and wire them is series the dB is a wash, two drivers using half of the power but with twice the resistance, +3db for two drivers together, but -3dB for half the power. But if you take those same two drivers and wire them in parrallel the resistance gets cut in half, the power doubles, and you get a 6db gain in theory. How true is that in actual implementation, is it anywhere close to this, or did I miss something?

**JonMarsh** · 23 March 2005, 19:23 Wednesday

Yeah, I think it's inevitable that there will be a two way (similar in driver crossover frequency range to the M8ta using the RS225) but with the dual driver MTM configuration), but I didn't want to "broadcast" that idea until I'd had a chance to test the production RS28a tweeters in that configuration- another one of those "lucky" things is that the impedance curve of the Hales Transcendence I'm using in the M8ta and the RS28A are very similar, as well as the frequency response. Both use Faraday sheilds to lower the VC inductance as well as the upper harmonic distortion. The Hales Transcendence, of course, is not available (I have a few pairs through someone who knew Paul Hales, who went to QSC), but the RS28a should be available some time in April.

So, what you may want is the evil offspring of the following speakers:

MTM using HiVi M8a's and Vifa XT

MT using Dayton RS225 and Hales Transcendence

RS28a Prototype

Mate the woofer of the M8ta with the RS28a into the MTM configuration we used for Paul H's speakers and several others... that will be a new crossover, but a bloodline relative to the M8ta and MTM, which are closer than first cousins, more like half siblings.

The one other 8" midwoofer I plan to investigate for this type of speaker, a new Peerless Nomex Exclusive series (due out by May, probably) would be too expensive compared with the Dayton RS225 for your price target.

**m_bathke** · 23 March 2005, 22:59 Wednesday

So is the Dayton RS28a going to have a much better sound then the Seas 27TDFC tweeter at about the same price? From what I have read on this forum it seems like the 27TDFC is a budget favorite, I guess I would just like to hear what you think of the RS28a in comparison.

**JonMarsh** · 24 March 2005, 00:24 Thursday

The 28a seems to be cleaner both in power handling down low, and in having low levels of higher harmonic distortion; i.e., less grit or sszz overtones that don't belong their. But I won't argue with anyone about the 27TDFC being one of the $28 champs of the world. But I think the RS28a is better, and worth the cost difference in the right applications. It's definitely an "A" class tweeter, whereas the 27TDFC is a B+, with extra points for value.

**m_bathke** · 24 March 2005, 01:18 Thursday

Thanks a bunch Jon, I think I will wait to see the RS28a, at about $50 each that is something I can live with. I guess what is $40 and a few weeks anyway, when I am probably going to have these speakers for the next few decades.

One last question, and I think I will have everything covered.
If you wanted to do a TMWW setup with the RS225's, and not use one of the RS drivers for the mid, would something like one of Sea's midranges work? Specifically the MP14RCY/P which has a response rated from 100hz-4000hz and is really smooth the whole way until about 3.5kHz and has a 90dB sensitivity. From what I gathered from its use in other projects that I have seen on the net, its a pretty sweet mid if you cross it over below 3kHz.

I guess I am just stuck wanting to go with a three way design because I have never heard a 2-way that I liked for music, this is probably because I have never heard one of your designs though. And also because I was under the impression that with a three way system you can use XO's with lower orders if the drivers you used were able to cover wider frequency ranges. Thus keeping problem area's for drivers an octave or two from the XO frequency and not taking someone with a BSEE degree to figure out.

P.S. I see that Pat,one of the moderators for this Forum, is from Iowa. Wonder if he is from the Cedar Falls/Waterloo area, as I am a student at UNI. I just saw that he bought some of his stuff from Hawkeye Audio which is in Waterloo so I figured he has to live somewhere around here.

**JonMarsh** · 24 March 2005, 13:12 Thursday

I've never worked with the MP14RCY/P, though the factory data looks like it would be fairly easy to work with. It would be interesting to see some ETC plots on it higher up, though... note that the dipsersion starts to narrow at 2 kHz. How high would you want to take this, realistically? And how low? (considering that you probably want to stay away from the baffle step region unless you want a fairly complex midrange crossover- that means running it in the range of 800 Hz to perhaps 2.5 kHz?

From a review of MarkK's midrange test data, I have to conclude that many 5" midranges aren't any better behaved above 2 kHz than most of the better 7's; in fact, the W18 seems to have most of them beat for energy storage up to 2.5 to 3 kHz.

Other midrange drivers I'd consider looking at are some of the smaller Peerless models with nomex diaphrams -

HDS NOMEX 106 WR 26 72 NWP AL TV 8 OHM - 830872

830872.gif Image not available

HDS 106 WR 26 81 PPB AL 8 OHM - 830870

830870.gif Image not available

HDS EXCLUSIVE 106 WR 26 81 NWP AL CU PH 8 OHM - 830881

830881.jpg Image not available

830881.gif Image not available

Because of the Farady rings, phase plug construction, and cone design, I expect the Nomex Exclusive series to be very low in non-linear distortion.

I'm going to be looking in detail at some of the larger diameter Nomex series components later this year...

~Jon

**cjd** · 24 March 2005, 14:02 Thursday

Originally posted by m_bathke

If you wanted to do a TMWW setup with the RS225's, and not use one of the RS drivers for the mid

However you look at things, you've blown your budget out of the water, and this is where I suggested you start looking at a two-way concept. It's not entirely an issue of trying to use RS drivers for the mids or not since you still have center to center considerations, etc. That's not to say that other drivers might not be better suited, or that you may find you can use lower order crossovers to simplify things and potentially lower cost even with a more expensive driver (though, I'm not convinced it would happen).

I suspect a 3-way using the RS150 at 1800Hz on the upper end will still be fairly reasonably priced (OK, maybe 25% less? maybe better since the RS225 goes higher) compared to the RS270/RS180/RS28 3-way, which is looking at roughly a grand in components.

C

**m_bathke** · 25 March 2005, 00:54 Friday

You two sure gave me alot to think about, and thanks for all the info. I realize that you have jobs and lives and really appreciate you taking the time to evaluate my delema. I think I will wait a few weeks and see what people come up with for a RS two way or three way. It won't be too bad if I have to put a little bit more money into it, as long as I get something that sounds better then the JBL's it will be worth it.

**Kramer** · 25 March 2005, 09:54 Friday

Not an RS design, but ...

Did you look at building Dennis Murphy's 3-way based on the M-130 mid? He has 3 versions, the least of which will run around $400.

Peerless 850146 + GR M-130 + GR T2
Peerless 850146 + GR M-130 + Hiq OW1
Peerless 850146 + Seas CA15 + Hiq OW1

**cjd** · 25 March 2005, 11:01 Friday

Originally posted by m_bathke

I realize that you have jobs and lives and really appreciate you taking the time to evaluate my delema.

Yes, well... the job often likes to leave me sitting in the dark corner all alone with nothing to do. For days, weeks. . .

I've been getting lots of circuit design done! What should be some pretty slick amps at an astoundingly good price for what they deliver. I hope!

I don't do much speaker modeling here at the moment, though that's not to say I couldn't. . .

As for a life, I'm alive.

Crossing that line into another year today too. :P But yeah. This is the easy stuff.

If you ask for such a design, THEN it takes a bit more time. So, let me know and I'll dive in and push the TMWW with the RS150/RS225's further. Heh.

C

**JonMarsh** · 25 March 2005, 11:28 Friday

Happy Bday, Chris... hope you have a good one, whatever you do!

**m_bathke** · 27 March 2005, 17:53 Sunday

Does anyone have the math for doing Linkwitz-Riley 2nd order crossovers? I have been looking around the net to try to find them and have been pretty unsucessful for finding a version that is larger than a two way. From what I can gather this is the best lower order cross over to use, or maybe I am wrong, but it is -6db down at the cross over and it doesn't give you a ripple at the cross over point if you put the midrange out of phase in a three way set up. Anyway if anyone knows where to find it, could you please point me in the right direction, Thanks.

**Dennis H** · 27 March 2005, 22:13 Sunday

Does anyone have the math for doing Linkwitz-Riley 2nd order crossovers?

There are "textbook" formulas on the net for LR crossovers but they are worthless.

What you want is for the acoustical output of the crossover and the driver to equal a Linkwitz-Riley response. The textbook formulas assume that the driver has perfectly flat response but it won't. They also assume that the driver has a constant impedance (8 ohms or whatever), as the frequency changes, but it won't.

It's better to go buy a pair of Polks or Athenas than to use a textbook crossover in a DIY speaker.

**m_bathke** · 27 March 2005, 23:07 Sunday

I wasn't going to use the actual textbook crossover in the speaker. I just wanted the math so I could understand how they worked. I have found many sites on the net with the 2nd order two way, but the third driver from what I can figure will throw a kink in the whole setup, so I would like to know how a three way influences things. I figure no matter how well you can do something, its not going to be its best until you understand why and how it works.

**JonMarsh** · 28 March 2005, 12:04 Monday

For a good basic treatment of crossover filters and response profiles, as well as some reasonable advice about the pitfalls and perils of three way crossover networks, pick up Vance Dickason's "Loudspeaker Design Cookbook". Another recommended book for someone getting started who already has a little design background is "Speaker Building 201". Both are available from AudioXPress/Old Colony Sound.

If you're not ready to buy desing software, you might try downloading SpeakerWorkshop (it's free), becuase some practical experimentation with measured speaker data and the crossover filters you want to implement, in combination with the above books, should get you moving in the right direction.

Just doing the math for a filter network into a fixed impedance is not very illustrative or productive (IMO); I can do that all day long in MathCAD (I'm an EE for SMPS technology, quite familiar with control systems and filter networks), and the reason I use a program like LspCAD is that it allows me to stay focussed on the big picture and results without spending a lot of time crunching numbers the hard way... Been there, Done that, have better things to do with my time.

If you want to play around with filter stuff outside of a purely audio context, try Filter Workshop

If you start with some good references, it will give you a solid footing from which to proceed.

Last, hint: L-R networks have filter coefficients which result in a response profie the same as two cascaded Butter worth networks of half the order. Go back to Linkwitz's orignial AES paper (should be available as a pre-print, and is included in the AES Louspeaker Anthology (another good source book - had that one for decades). For reasons of copy right, I can't share, but AES has on line ordering.

Audio Engineering Society Home Page

~Jon

**cjd** · 28 March 2005, 13:39 Monday

I would suggest Jeff B's PCD spreadsheet (Check the FRD Consortium) before Speaker Workshop. It's a lot more obvious what the heck to do with it, and you can tweak values and see the change immediately.

It does not support the more complex filters (you can sort-of get it to handle some of them).

Of course, you can probably crack open the sheet and see some of the crazy stuff it's doing too.

C

**Dennis H** · 28 March 2005, 13:48 Monday

Okay, now that Jon has given you all the caveats, here are the textbook formulas. Note that the author says to use the DC resistance of the driver. That is incorrect. You should actually use the AC impedance at the crossover frequency and that impedance should be the same at least an octave above and below Fc (it seldom is.)

Yahoo | Mail, Weather, Search, Politics, News, Finance, Sports & Videos

http://www.geocities.com/loudspeakerguru/TwowayNetworks.htm

Latest news coverage, email, free stock quotes, live scores and video are just the beginning. Discover more every day at Yahoo!

**NMyTree** · 28 March 2005, 21:42 Monday

Is that curved cabinet a custom made cabinet, or is that something that one could buy?

**m_bathke** · 29 March 2005, 03:56 Tuesday

Thanks Guys, I actually used the Crossover sim program and the PCDC worksheet that are availiable from FRD to get a ruff numbers for a crossover. Then I would put them into Flexsystems to see how they changed the response, crossover point, and phase. It really gave me an idea of the cross over points that are really needed if one were to use a low order cross over. I would have used the PCD worksheet that CJD suggested but it was pretty confusing, way more advanced then the PCDC, but way more confusing.

What I really wanted to see was how one component related to its counterpart, and with the numbers for the Linkwitz Riley I think I get it. Impedence is really the variable that is important, the rest is just a constant and the designated frequency which doesn't change. Now I get why it is important to measure in the box before you build the XO, and and to figure in the series resitance of components. I wanted to do a sealed enclosure and from what I can gather impedance would go up and the crossover point would move higer for the woofers, and the mid's XO points would widen, so I would basically have two humps in the response. Also I now see the importance of the Zobels and R-L-C, they basically let the crossovers function like it is a flat impedance.

If I am off on any of this please correct me.

CJD, I would really like to see the ruff design that you did on the 27TDFC, RS150, and RS225s. Hopefully it would give me an idea of what the XO should look like for these guys. I think these will be the drivers that I will use unless the new Dayton RS28a comes out in the near future and isn't over $50.

Anyway thanks again for your time, you guys would make great teachers, maybe you should put together your own book on DIY audio, that would be awsome

I say that because of everything I have learned just from reading other threads in this Forum, the knowledge base here is pretty darn impressive. I guess its nice for me to see that because I am used to having to explain to people why a decent bookshelf speaker is better then the bose cube's and acoustimass systems. Its funny people are so hooked on Bose stuff that they think that is how a good system is supposed to sound, its sad I know.

**NMyTree** · 30 March 2005, 08:59 Wednesday

JonMarsh,

Could you please tell me, if that curved cabinet is a custom made cabinet, or is that something that one could buy? The one with this written underneath. " MTM using HiVi M8a's and Vifa XT "

I'm refering to just the cabinet, not the drivers.

I'd like to try my hand at building a pair of speakers.

**JonMarsh** · 30 March 2005, 11:28 Wednesday

Those were MTM's built using the HiVi and Vifa XT25 custom built by Paul H, using the acoustical design I developed.

There's more info about that project in this thread, including a lot of construction pics.

~Jon

**cjd** · 30 March 2005, 11:55 Wednesday

Originally posted by m_bathke

CJD, I would really like to see the ruff design that you did on the 27TDFC, RS150, and RS225s. Hopefully it would give me an idea of what the XO should look like for these guys. I think these will be the drivers that I will use unless the new Dayton RS28a comes out in the near future and isn't over $50.

I'll see what I can dig up this evening. I was hoping to have a computer upgrade to do, but the motherboard box didn't happen to contain what it *said* it contained which is entirely perplexing and a little disconcerting (same brand, same line, but different feature-set - and a less expensive model).

And, there are many ways to put a crossover together. There is nothing that says only one way is correct - though I know full well the value in having something to compare to, I also know that it doesn't really teach you many things. So, at some point you'll have to share for tips.

C

**m_bathke** · 31 March 2005, 05:52 Thursday

So here dementions that I came up with for the box, all sealed and stuffed, divided internally into 4 different enclosers, and a 1.5" thick X 13" wide baffle.
Internal volumes are: RS225= .998264 cu.ft. each(F3=50,F6=40),
RS150= .239583 cu.Ft(could be made smaller).

Also there will be a 4.5"w x 6"t x11.5"L spot behind the tweeter, good place for the crossover I thought, probably will make a door there for easy access.

These will be placed 24" off of the floor so the tweeter center will be at 41 1/2" off the floor, which is exaclty ear level for me when I sit on my couch(I measured it twice just to make sure :wink

Other then that these will basically be within 1"(yes INCH, okay may two tops) from the wall, I know its not optimul but for now space is pretty limited. So I am not sure how much BSC they will need, and it might make a difference that each one will be paired with a 12" sub. I figure bottom end won't be that big of a deal, so less might be better as not to piss off the neighbors.

any and all comments welcome.

**m_bathke** · 31 March 2005, 05:55 Thursday

By the way I did the box drawing in MS paint, if there is a freeware cad program out there I would really like to know.

**cjd** · 31 March 2005, 11:59 Thursday

I'm not gonna let you do that to yourself.

That's a great CC topology (if you turn it 90 degrees). For a vertical speaker, you will be wanting to put the tweeter/mid relationship vertically, or you'll have really nasty off-axis troubles.

Box volume is a little less than optimal but not out of the question at all. If you've been box modeling, you need to take into consideration the series resistance of crossover elements.

My suggestion: Keep the baffle the same size. Center the woofers, 7" from the top/bottom to center. Mid, 14 3/8" from the bottom, offset from center 3/8". Tweeter, 13.5" from the top, offset from center 3/4".

Will you be able to chamfer or round-over the edges? Without, there are some pretty nasty diffraction ripples that show up in the tweeter response. The best response I got in that respect was actually a 1.5" chamfer on all four baffle edges, but it still puts a troubesome diffraction peak right in the range where the mid and tweeter may cross.

The other thing is we can move away from the WTMW setup and possibly go to a tower - is there any reason NOT to go with a tower if you're already planning on using stands? If we target similar box volume and go TMWW there are some pretty big advantages, among them the fact that the speakers will not have to be as deep for the same volume and can have a little more breathing space (or as like as not, be moved closer to the wall). Probably can be slimmed down a tad as well. A TMWW with external dimensions of 12Wx45Hx15D gives you more volume in-box, and with a 3/4" roundover or chamfer on the side edges of the baffle gives you much better diffraction results on the tweeter in the area where the mid and tweeter are crossing. Tweeter at 41" (plus spikes and it'll be right where you want it). Add a roundover/chamfer on the top edge and it gets even better - this would be my ideal (if you chamfer, it's also easy to veneer - if you roundover, not so much). This box will be a bit more elegant looking IMHO and will have less in-room presence (though that depends somewhat on your choice of finish).

C

**m_bathke** · 31 March 2005, 22:09 Thursday

CJD
Thanks for pointing out the problems with the box design, better to know now then after I have the boxes done. After looking into your suggestions on the box design, I think that a tower would definately suit my situation the best. I was initially looking at pairing these with 12" subs, but after modeling the boxes, there is really not an advantage using the subs. With the boxes ported and tuned to about 22Hz or 23Hz they will have about the same response down low as the subs, they just won't be able to hit 122 dB, but I think that would make the neighbors a little mad, as I know it did in the dorms. :wink: It was kinda cool to watch the door jump about a quarter inch off of the door jam when the bass hit though.

The room size is only about 15' by 20' so that should add to the room gain a little bit.

As for construction of the boxes, I can do pretty much anything you suggest, chamfer, roundover, or anything else. As for their finish, it will probably be veneer.

**cjd** · 01 April 2005, 10:52 Friday

So you're wanting to go ported now? Let me get back to you on some box options.

I investigated a ported alignment fairly thoroughly at one point, but I have that stuff at home. I'm kind-of not allowed in the office at home these days since my keyboard clacking distracts my wife working on her homework (also on conference call with classmates, since it's a group final). :P I will say that it reverberates pretty badly on the phone, so hey. Lots of reading and listening to music happening right now.

A real sub *would* contribute. In fact, I would cross a set of ported RS225's at 40-50Hz to a sub which should cover down to 10 or 15Hz. 10-40Hz is two octaves - not insignificant depending on the music you listen to.

On your cad question - I do most of my work in Fireworks, since I sketch in that for work stuff all the time it's just easy for me. But it's a graphics program. Modeling I do in Maya but that always plasters big watermarks since I'm not about to shell out the bucks for the non learning version for the rare times I use it.

How big a roundover can you do? 3/4 is the largest reasonable roundover for most folks (between bit cost and equipment's ability to handle it) but sometimes a larger roundover is particularly worth doing. If you prefer the look of a chamfer, let me know.

C

**m_bathke** · 01 April 2005, 11:22 Friday

I perfer the Chamfer, it will just be alot easier to veneer. The sub would be something I would be able to live without for one year until I move into a different larger place. I guess when I build these if I don't feel the bottom end is there maybe I will be able to find a way to hide one of the subs in the room. I decided I wanted to go ported in an effort to not have to have the subs, I saw that if I went big and ported I would get alot better bass response. Again thanks, sorry about changing things up on you. I am beginning to realize that with DIY loudspeakers the more you learn, the more you realize you don't know.

**cjd** · 03 April 2005, 19:08 Sunday

Alright! I should note, this is with full baffle step. Pair of RS225's, RS150, and Seas 27TDFC. Tossing the RS28A tweeter into this should not be difficult, but I want to wait until final specs are released.

Box: 12"Wx48"Hx?? Deep, 1.5" chamfer on the sides and top of the baffle. This is going to be a rather ideal size for a sealed box (at 16" deep) but could be ported if volume is pushed by making it deeper. Altering the general box dimensions is also an option but I would need to incorporate the new diffraction/baffle step results (it'll be quite close, possibly only requiring the second cap on the tweeter to be tweaked.) I'll work up a diagram of the baffle layout I used and post it in a bit, but I've gotta go make pizza dough right now!

Response looks pretty good to me.

Impedance: This qualifies as a 4ohm speaker. Minimum gets down to ~3.2ohm and that's quite reasonable.

Network: This uses 15ga Jantzen inductors anywhere they are in series with the driver, and 18ga Jantzen any time they're in parallel. Target was to keep the cost more reasonable, so no impedance comp or cauer filter on the tweeter, and the minimum on the rest.

C

**m_bathke** · 03 April 2005, 20:06 Sunday

Wow

, that looks nice.

**cjd** · 03 April 2005, 20:48 Sunday

Baffle layout, ported. Tuned to ~22Hz. 12"Wx48"Hx22"D In-room response should put it solidly flat nice and low. Watch the power - excursion at ~16Hz goes over quite a bit with only 25W (you'll also be at ~106dB/1M though...) This would go *really* well with a nice LM3886 chip-amp IMHO. Hmmm. heh.

Shorten to 16+" deep for a sealed alignment.

It is expected this will go on spikes for another inch or two of height.

3/4" material, so it'll have to be braced well. Sub-enclosure for the mid, sub-enclosure for the tweeter (both sealed). The top/bottom of the tweet box will need to be full of holes where they're not sealing off the tweeter, as will the other full-length/width braces (you could just do them around the perimeter even). As I'm pondering, a brace across the top panel may be beneficial somewhere between the back of the mid/tweet box and the back of the main box.

I recommend a round-over on the port (1/2" works nicely) both sides.

C

**cjd** · 03 April 2005, 21:02 Sunday

List of parts from Parts Express...

027-418 DAYTON 3.0uF-250V POLYPROPYLE.. $1.67 $6.68
027-428 DAYTON 10uF-250V POLYPROPYLEN.. $3.15 $6.30
027-430 DAYTON 12uF-250V POLYPROPYLEN.. $3.55 $7.10
027-436 DAYTON 20uF-250V POLYPROPYLEN.. $4.95 $9.90
027-440 DAYTON 30uF-250V POLYPROPYLEN.. $7.88 $15.76
027-442 DAYTON 40uF-250V POLYPROPYLEN.. $10.45 $62.70
027-458 DAYTON .47uF-400V BY-PASS CAP.. $1.38 $2.76
005-1.5 MILLS 1.5 OHM 12W NON-INDUCTI.. $3.50 $7.00
005-6 MILLS 6 OHM 12W NON-INDUCTIVE.. $3.50 $7.00
005-3 MILLS 3 OHM 12W NON-INDUCTIVE.. $3.50 $7.00
255-220 JANTZEN .33mH 18 GA AIR CORE .. $2.41 $4.82
255-280 JANTZEN 4.00mH 18 GA AIR CORE.. $6.74 $13.48
255-198 JANTZEN .025mH 18 GA AIR CORE.. $1.47 $2.94
255-406 JANTZEN .39mH 15 GA AIR CORE .. $4.10 $8.20
255-414 JANTZEN .68mH 15 GA AIR CORE .. $5.19 $10.38
255-422 JANTZEN 1.00mH 15 GA AIR CORE.. $6.42 $12.84
255-434 JANTZEN 2.70mH 15 GA AIR CORE.. $12.30 $24.60
027-443 DAYTON 50uF-250V POLYPROPYLEN.. $11.29 $22.58
027-447 DAYTON 100uF-250V POLYPROPYLE.. $19.51 $39.02
295-362 DAYTON RS150S-8 6" REFERENCE .. $28.55 $57.10
295-366 DAYTON RS225S-8 8" REFERENCE .. $40.80 $163.20
Subtotal: $491.36

Not listed, you'll need a pair of the Seas 27TDFC (~$60)

Plus good binding posts, wire, box materials, wall covering (you can stuff the mid enclosure but for the big box just cover all the walls well) and, of course, your choice of finishing materials.

The 10.47µF cap is a 10 and a .47 parallel (I went for the film/foil bypass here on the .47 value)
The 23 is 20 and 3.
The 140 is 100 and 40

You can add a small bypass to the 30µF on the tweeter without significant change and it's probably worth doing.

C

**m_bathke** · 03 April 2005, 21:51 Sunday

Again, Thanks for all the effort, not to bad of a price either all things considered. Do you think that these would pair well with the CC you built using the RS150's and the 27TDFC?

One last question, and I know it might seem a little silly. How much do you think something like this would cost if you where to buy it from a comercial vendor like say Klispch or Infinity, etc?

**cjd** · 03 April 2005, 22:20 Sunday

1. CC - yes, once I update the crossover. It needs a little work to really get where I want it though it does work quite well as-is - that is to say, my sister quite likes it and reports no problems, but I know that it has some off-axis troubles. I am also pondering tossing a CC design together with a pair of the RS225's, a 150, and the 27TDFC but we'll see. That would cost a bit more.

I want to lower the crossover point to where I went for this WWMT - that should blend beautifully.

2. Cost - very difficult to say. Given the value of the RS drivers, consider that it's already playing in a field where DIY would have previously cost perhaps $850-900 (or more, since you could easily sink a LOT more money into caps alone - something you may wish to consider spending a bit more on regardless). You then have to realize that Klipsch and Infinity would not be offering these speakers - the components are higher end than that. I might arrive at a $3000-4000 price tag, perhaps that much each. Or more. Maybe less. See, I haven't been in a retail store in ages, and I really have no idea how things are priced these days.

Or how this might compare. So, maybe you're getting a pair of $1200 speakers (retail) for $700 all said and done (binding posts, box material, stuffing, etc.) Maybe you're getting a pair of $20,000 speakers though.

I just realized I need to re-work this for you! I went full baffle step and you need next to none. Let me give that a quick whirl and see how bad it makes things.

C

**cjd** · 03 April 2005, 22:39 Sunday

Ahhh... This looks promising indeed.

To compensate for the need for less baffle step, there are three things that need to be done - increase the series resistance on the woofers, lower the value of the parallel cap on the woofers, and drop the value of the second inductor on the woofers.

For ~2dB baffle step (probably the minimum)

1. Use a smaller gauge 2.7mH inductor on the woofer (instead of 15ga) (something with ~1ohm resistance)
2. Drop the 140µF cap to 115µF
3. Drop the second inductor from .68mH to .60mH

That's it.

I think.

That means that, in theory, for ~4dB BSC you'd need a 2.7mH inductor with ~.8ohm resistance, ~127µF cap, and a ~.64mH inductor on the second spot.

C

**m_bathke** · 04 April 2005, 08:59 Monday

Well I was pondering your suggestion on using higher quality caps, and maybe spending more money on them. I thought I would keep with the theme of getting more then what you paid for and suggest using GE caps on everything except the bypasses. From what I can gather they are as good/better than the Solen Caps (at least thats whats been said a few times on this forum). I figure I can measure and then parrallel 10uf's to get close to the values and then finish things off with a smaller Solen or GE cap. It ends up being about the same price as using Dayton caps too($170.04 for dayton caps, $161.20 for GE and Solen), so its not going to kill my pocket book.

If its a bad idea please let me know, as I really like to know about my mistakes before I actually make them.

Also I was thinking of using AudioCap Theta Film and Foil for the bypasses on the 10uf and 30uf on the tweeter, but I am really unsure of how much this will really benifit me besides just having a lighter wallet. How much difference can a .1uf or .47uf cap make to the sound quaility as a bypass? I will definately at least us the Dayton Film/Foil caps, there $4.50 for all of them, but the Audio caps would total to $33.00 for what I need.

**cjd** · 04 April 2005, 09:14 Monday

The GE caps are a fine idea. The Dayton poly caps are not bad, just easy to work with when putting together a price list. I think they rank in there with Bennic caps? The GE's just get big piles when you're talking the larger values. On the other hand, this means you may have more ability to tune baffle step a bit more easily.

The 10.47 value wasn't arrived at simply to allow for the bypass - dropping to straight 10µF alters the contour with a shallow broad dip (~1dB). The 30µF is far less sensitive to small changes like that.

Around here, I believe the norm is to split the values with something like the Audio Theta's, so 5µF/5µF + whatever, 15µF/15µF, etc. If you're going with the GE's, 10+.47 and 10+10+10 with one being film/foil is probably fine. You'll have to get the tips on that process from Jon.

If you're considering the Theta's, I would consider holding off for the RS28A tweeter I think - it's supposed to be priced at $50 (or less, but I'm not holding my breath for less) and available soon. Very soon. I believe it will be a step up from the 27TDFC (which is still a fine tweeter choice).

The box won't change, so you could get rolling on that and see what happens in the next week or two.

C

**m_bathke** · 04 April 2005, 09:37 Monday

Originally posted by cjd

The GE's just get big piles when you're talking the larger values.

Yeah big piles, I started adding them up, I will need 30 10uf per side. But for the price I think I can live with having a little bigger XO board.

I was looking at wire to do the internal wirind and probably for hooking them up. I was thinking of using Monster XPHP-CI100, its has time correct windings(whatever that means?), and magnetic flux tube construction(again no idea), and it is double jacketed with the pair twisted (it looks kinda cool too

). Now if it wouldn't be for the fact that I can get it for $52.00 (if you would buy it it would be $130) for a 100ft. spool I would probably get something else. My question is would you suggest something else for about the same money?

**Dennis H** · 04 April 2005, 10:28 Monday

I've heard the Dayton caps are rebadged Bennics. Some people on the Mad board prefer them over Solens. Personally, if the price were about the same, I'd go for one big cap over a bunch of small ones just for the hassle factor.

**cjd** · 04 April 2005, 10:36 Monday

Go to apexjr.com and pick up some of the teflon insulated silver-plated wire. You could go simple 12 or 14ga, or you could grab 16 or 18ga and braid it (three 16ga is equivalent to ~11-12ga). 100 feet of 16ga is $15 - that's $30 for a pair of 30 foot speaker wires. For runs from your amp to the speakers, the braided 16ga is a great option - put quality spades on the end (Cardas has some great stuff) and wrap the whole mess in some tech-flex. The simplest way I've seen to braid is to run ~1 foot of each color braided, then pair them up (one + one -) and braid the pairs.

C

Three-way tower with Dayton RS225's, Help needed.

Three-way tower with Dayton RS225's, Help needed.

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment