=== "$100 Reference System" ===

[enigmaticEntity.]

I wish to build a cheapish "reference" system using Dayton drivers.

rs100-4
nd20fb-4

I designed the enclosure around Zaph's specs, fr graph found here:

Click on rs100-4.
Click on nd20fb-4.

The formulas for the ported enclosure:

DIYaudiocorner.

The box will be made out of 3/4'' MDF and will have a rather large separate chamber for the crossover, which will also make the box taller.
The crossover will be set at ~3,000Hz to ~3,500Hz.
If the response graphs and formulae are accurate, I should end up with an almost ruler flat speaker that should look rather nice.

Open to suggestions.

P.S. (The bad thing is that the small peak is going to be right in the middle of the sensitive hearing range...)

[woody]

When I clicked on your links they came up as a pair of Scanspeak drivers .

I think the rule of thumb is to cross a tweeter over no lower than 4x it's resonance
frequency so for this tweeter with a resonance of 2072hz that comes out to ~8k.
I would think you would be better off just using the woffer and using a notch filter
for that 10 to 12 db peak at 16k.

[enigmaticEntity.]

Zaph's comparision application is a javascript application, so I can't link directly to the Daytons. You have to click on the driver name in that list (woofer is near bottom left, tweeter bottom right) or else you will only see the first driver, i.e. the Scanspeak.

Damn about the tweeter - is there any reason for not going lower than 4*Fs?

[alspe]

I think tweeter XO point is related to slope. And 4 x Fs is far too high, at least with 12-24 dB slopes.

[danny1974]

I would cross that tweeter LR4 at 3500hz. This combo should make a nice mini monitor with low listening level.

[woody]

Usualy you don't cross over any where near the resonance frequency because the
impedience peaks at that frequency. This makes designing a crossover harder.

[dantheman]

I'm listening to the RS 100-8 right now. What an awesome little speaker! Forget that tweeter unless you have the hearing of a bat. The 16k peak isn't even troubling for my dog. I know I sure don't hear well up that high and nether do most people. Heck, most people can't hear anything that high and there's a good chance you are one of them if you are over 30, male and you've behaved like a typical guy at any point in your life. The power response of that peak is even going to make it less audible and the fact that it doesn't ring makes it a non issue. It may actually help make the speaker sound flat if anything.

The more important end to put Augmentation on is the lower end. These speakers run out of gas quickly on the low end. This driver makes for excellent augmentation on the cheap: Parts-Express.com:*Peerless 830634 10" SLS Woofer | Peerless P830634 10" driver SLS Woofer subwoofer bass mid midbass mtm 2-way Tymphany genelec 342-6001 2-way 3-way midrange mid vifa studio monitor active
You'll quickly get over your $100 limit, but only by a bit and a year from now you'll be so happy you did.
I'm running the RS100-8 and the Peerless 10" SLS on a 40cm wide by 75cm tall open baffle with absolutely incredible results. I almost hate telling you about it b/c I kind of want to buy more of those Peerless buyouts--even though I have no purpose for them.
More on the design: I just got this up and running last night and it's not complete, but this is a good starting point. I'm actively crossed over 24dB/octave 4th order LR and just over 200Hz. There's a 6dB baffle peak right above the x-over, but the little driver has started it's roll-off at this point anyway. The little speaker is around 67cm off the floor and 3.8cm off center and the woofer is 40 or 45cm off the floor centered on the baffle. The basics of the design come from here: http://www.quarter-wave.com/OBs/OB_Design.pdf

My wife said that this combination sounds as good or possibly slightly better than any of the Tannoy Heritage speakers we heard while on holiday in Akiba-Tokyo, Japan with much more expensive equipment behind them. They are many thousand USD. I'd actually say the same when comparing this design to the Feastrex field coil on the Nelson Pass's BOB speaker I just heard at Burning Amp. In some ways I'm sure I prefer these--the price! Actually I think I may prefer the sound as well, but there are too many variables to say resolutely. After hearing these things it's really apparent how much speakers have limited my system up until this point. Actually I bet they're still the most limiting part even if less limited than ever before. I mean they are very inefficient and are probably compressing peaks, the crossover hasn't been optimized, the FRers haven't been flush mounted, the openings and baffle edges are not rounded over yet, etc... So there's still a long way to go before they are finished, but they are a class act for $60 in drivers per side. Clever implementation of the tweeter you mentioned might just be the icing on the cake, but definitely not the most relevant first augmentation to start with if you plan on listening at normal, moderate levels. I really think if you do what you're planning on doing you'll regret it in a week. I can listen very loud with this pair if I want to. No way, no how with the FRer going into the low end. Another thing, don't forget that power response may be (I now think it is) the most important factor in speaker design. Narrow OB does power response better than most designs.

I hope this helps,

Dan

[danny1974]

Originally Posted by woody

Usualy you don't cross over any where near the resonance frequency because the
impedience peaks at that frequency. This makes designing a crossover harder.

yes you are not suppose to cross near the Fs but 3500hz is almost an octave over the Fs (the Fs of the Dayton Neo is not consistent, I measured from 1600hz to 2000hz). LR4 (24db/octave slope) will work fine. I would use the RS100-4 because it has more sensitivity.

[dantheman]

If I had it to do over again I would use the 4 ohm version as well. It may not work out as well in every situation, but most of the time it will.

[chris661]

Originally Posted by danny1974

yes you are not suppose to cross near the Fs but 3500hz is almost an octave over the Fs (the Fs of the Dayton Neo is not consistent, I measured from 1600hz to 2000hz). LR4 (24db/octave slope) will work fine. I would use the RS100-4 because it has more sensitivity.

Changing impedance may mean it's been getting warm.

Chris