Can't decide: 5" in optimal box size OR 6.5-7" in smaller than optimal box size?

[SQconstable]

I'm desigining a pair of 2-way speakers for my wife. For the tweeter, that's still up in the air but I do love these Usher 9950's and feel they're worth the money. I just haven't thought out the crossover design for them with whatever woofer I end up choosing. I'm pretty sure I'll keep the 9950's in this design.

For the woofers.... My problem is that I normally like larger than optimal (or just at "optimal") enclosure volumes for a given woofer. I guess I could say that I like a Qt lower than .707 because it sounds more natural, and, well.. I like more bass in the mains so I can have more flexibility in subwoofer xover points (and localization). My wife, although she understands the purpose of larger airspace to get the most bottom end out of any woofer, wants a compromise: Something that isn't visually obtrusive. I am already thinking "scratch the 3/4 MDF" standard haha. I'll use 5/8" and use dowel rods for bracing I guess. As for cabinet dimensions, I'm thinking it should be no taller than 11", no deeper than 6 or 7", and the width needs to stay under 7". I know that I can use a roundover bit to make the box appear smaller and might be able to get away with more size hehe.

So.. it seems that with 5/8" MDF and some losses due to bracing, I'm lookin' at around .21 cu. ft. max. I read in another thread here about some guy wanting to make Dayton RS-150 drivers work in .25 cu. ft., and some people flamed him for wanting to do it. Well, I took a look at both the 8 and 4ohm versions on the RS-125's and I get different results for recommended airspace. The 4ohm's would work in a smaller box (due to lower vas and qts) compared to the 8ohm versions. However, the 4 ohm version seemed to have a higher Fb. Lower inductance on the 4's too.. less power handling? Although the 4's seemed to be more sensitive based on looking at the Sensitivity spec, I was told by a tech at PE that you really have to take off 6dB when using an 8ohm amp.. then it comes to about the same sensitivity as 8ohm. I assume they made the 4's just for the car audio ppl. I'm understanding that 4ohm drivers may cause more cost on the xover components too... depending. So I guess I'll want 8ohm.

I looked at the RS-125's and RS-150's. I can't decide whether I should try stuffing the RS-150's into .21-.25 cu. ft. and use EQ to compensate for bass rolloff, or if it would be better to use the RS-125 in .21-.25 and get better natural enclosure-produced bass (I'm assuming ported is the way to go since EBP is higher than 100 for either of these drivers.. but there goes more extra volume needed for the port!). In either case, I'll be using a Tymphany LAT250 for the sub, powered by a Dayton 100w plate amp (which I have yet to purchase both, nor any of this stuff), so it's not ABSOLUTELY critical to get the MOST from a woofer, however, the more, the better. I'm open to suggestions on the sub amp. I do want to use the LAT250 since it is something I've never heard and I'd love to see how such a slim box would sound in an unobtrusive office space.

I guess I want to hear opinions on how you like the sound of a larger woofer in a smaller box vs. a smaller woofer in an optimal box and if the pros/cons outweigh whatever... especially if not running the speakers as "large".

Has anyone designed something like this which they took more time trying to make compromises between woofer diameter and small-space?

Oh! I will be using some sort of bass management with either a receiver's high-pass options or an active high-pass before some sort of amplification. I was really disappointed that PE didn't have those older, 2-way plate amps that I used to see in their flyers. I was hoping to be able to use something like that for these 2-way mains instead of some huge bulky receiver.

If you have any other recommendations for some good woofers at least 5" diameter that don't need much box volume, I'm open to anything.

[rc white]

A method of making satellites smaller withought compromising too much on output can be found in this article..

http://sound.westhost.com/qb5align.htm

I originally developed this technique for the same reason, to maintain domestic harmony.
Methods of adapting given drivers to specific needs are found here..

http://sound.westhost.com/articles/cscaling.htm

Using these techniques you can fit a given driver into a given box, and adapt it for instance to a given high pass output found on 5.1 and 7.1 receivers.
rcw

[SQconstable]

Thanks for the info! You have just shown proof that EQ (high-pass) can correct sub-optimal enclosure design.

I always thought that a sealed enclosure would be better than ported in a given high-pass'd satellite speaker. I'm thinking maybe that ported somehow keeps the curve flat on the bottom end better than sealed in this situation since Fb is raised. Ported design does require more volume inside though. But the articles above are showing how ported works better in this scenario. Am I reading that right? I do remember people saying that sealed doens't "move as much". But If ported is louder in general, it doesn't move as much from what my eyes see.

It looks like the high-pass xover method (bass management) was going to be in the plans the whole time. I wonder if I could build a sealed box first, then if it doesn't have the "low" end that I'm wanting, use the ears while tuning with a port later (at whatever port length). I find myself using the ears more and more as I build loudspeakers and it is harder to tell if ported was better than sealed when dealing with increased highpass filters, but I do end up doing ported. I guess because even though your xover point may be set to 100hz, you can definitely feel more air down there when adding the port. I've been accustomed to trying several different port lengths and choosing the one that gives that low end the right balance of "oomph" without making one note louder than others (hard to exactly say it right).

Hmm I still dont know whether to go for the 5" or 6". I gues should I go for whichever has the better midrange?

Hmmmmm... would the larger driver sound better playing 100Hz in the small box compared to a smaller driver playing 100Hz in the optimal box? That seems to be a better question to ask. Of course, there's so many theoretical (and real) factors and variables like differences in T/S parameters, but surely there's some sort of standard when you bring in this type of alignment (and bass management high pass filtering) into play, eh?

[rc white]

Here is a description of a satellite I made using this scheme..

http://sound.westhost.com/articles/ssl1.htm

The 830860 is a nominally 5.5 inch driver and when used in this box a pair can produce peaks of 110db. down to 100Hz. just at their rated cone excursion.

To get down to 80Hz. you usually need a 6.5 inch driver.
rcw

[owdi]

Since you are crossing over to a sub, a larger box and lower q will gain you almost nothing. Get WinISD Pro and model a few woofers, you'll see the greatest difference between different box sizes is below your xover frequency.

I have the Seas L18 already open in WinISD.... let's see. In .2 cu ft you get q=.742 and f3=90hz. That should integrate well with a sub and 80hz xover point.

Dan

[SQconstable]

Cool that's what I was thinking. I see that the Seas L18 is 7"... is there any reason to want to go for a larger woofer in this application? One advantage I'm seeing is that power handling increases with larger woofers. I'm gonna check out some more 7" drivers!
My only concern with larger drivers is that the baffle is still constrained to about 7" width. I understand that bass response is affected by baffle width, but does it come into play above 75Hz? The last speakers I built have a baffle that's actually skinnier than the diameter of the woofers used (if you can picture that), but people were saying you need it to be a certain width for the best sound, which is vague to me. Playing with WinISD Pro right now...

update: I see a truncated frame woofer here that is cheaper than the others: http://www.partsexpress.com/pe/showd...number=295-342

I know these are designed for arrays of these woofers touching eachother, but if someone was trying for a skinny enclosure (even skinnier than the 6" diameter of woofer), this would be of interest. It would make the smallest width in a baffle ever though... would that be a problem? This is another aspect I'm trying to outweigh.. larger speakers will take up the baffle width whereas smaller drivers will allow more baffle width. Or is this really miniscule?

[SQconstable]

After reading https://www.htguide.com/forum/showth...ighlight=rs150 I have a feeling that rc_white is trying to say there would be more distortion with a smaller box (excursion limits, power handling limits). Of course, the source material would be crossed over but I suppose it still comes into play...

Hmm.. what about a passive radiator on a 6" RS150-8? My Mackie HR824's has 8.75" woofer on a 9" x 12" radiator and many people like the bottom end. Are you guys here fans of PR designs? I dont know enough about PR's to do it, but just thinking outside the box for now... I'm curious to know how the sealed .25 cu. ft. RS150-8's sound. Jeff, wazzup!

A part of me feels like an L18 would be a waste for utilizing it only for midbass above 80hz. The Dayton drivers are a great price...

Ok I'm just gonna do this one ThomasW has posted here (image) and be done with it: https://www.htguide.com/forum/showth...6&page=4&pp=35 I'll be trying it in .25 sealed, on a desk against a wall (actually a large window with blinds that will eventually have a curtain). I'll let ya know how it works out.

While I'm waiting for the parts, I'll be building the SR-71 kit I've had under my futon here for monthssssss!

[rc white]

The basic principle I used in my article, is based upon Smalls parameter, "kp", the power output constant..

W = f3^4 kP Vd^2

In this kP is a constant dependent upon box type and alignment, it varies between .8 and .85 for sealed boxes in the q=.5-1 range, for reflex alignments it is usually 3, but for filter assisted ones it can be as high as 30 for the type of use we are discussing.

Vd = the volume displacement and this is the product cone area x peak cone displacement.

As you can see the greatest effect upon the acoustic power output of a given system is the cut off frequency f3, the same system can produce 2.44 times the acoustic power just by raising the f3 from 80Hz. to 100Hz.

If we want a satellite that is not being overdriven on program peaks it is desirable that the driver not be driven above its linear excursion region, so the Vd^2 product is constant for any driver, since for any cut off frequency f3^4 is also constant, for a given acoustic power output we need a given kp.

As to exactly how much spl. is needed for the average has been a matter of debate in this forum, everyone has different opinions. Mine is that a stereo pair with any pretensions to "hi fi", should be able to reproduce 110db. peaks withought exceeding linear cone excursion, this is about .4 acoustic Watts each.
rcw

[Curt C]

I guess I want to hear opinions on how you like the sound of a larger woofer in a smaller box vs. a smaller woofer in an optimal box and if the pros/cons outweigh whatever... especially if not running the speakers as "large".

So your real question is the audible effect of box Q.

For a stand alone speaker, I prefer a lower Q response. When you add the sub and HP filter, the issue is much less important in my opinion. This is only slightly different than a 3 way, and the same issues of proper summation apply. As long as they are properly integrated, the difference in Q will not be an issue, and I doubt you will be able to hear any audible differences.

There will be a marked difference in max SPL and power handling however.

For example: Compare the RS150 in in 2 sealed alignments: One with a Qtc of 0.707, and one with a Q of 0.9. Yes, the higher Q box will have a slightly higher f3, and probably around 1 dB of peaking. However this 'extra' dB may assist in baffle step compensation and possibly raise the system sensitivity by an equivalent amount.

Adding a high pass filter at 80 Hz, the f3's of the two boxes are now within a couple of Hz of each other. The low Q box will exceed Xmax over 35 watts, while the hi Q box takes 47 watts for the same Xmax. This will result in a max SPL at that level 2 dB higher for the high Q design. Without the HP filter, the Low Q box exceeds Xmax with only 12 watts applied, while the high Q will play 5 dB louder for the same Xmax, with 30 watts applied.

Impulse response is slightly worse for the Higher Q box. However, its effect is swamped by the much greater effect on impulse response of the 80 Hz active HP filter.

C

[SQconstable]

Ok I think I understand what's going on in the two scenarios now.

It makes sense to say that a smaller box will generally accept more wattage to reach a particular Xmax value compared to a larger (lower Q) enclosure. However, there will be more distortion below the high-pass crossover point allows too much energy to play below f3. I've experienced this when I built my very first 12" sealed subwoofer enclosure 15 years ago. I built a box that was way too small and while the upper bass and midbass freq's played beautifully when run full range (as a test), the lower bass freq's surely didn't "breathe" and were distorted at high volume. No amount of boosting subbass freq's to compensate would truly let the system sound as natural as it could have if it were larger. Now, it makes sense how if that too-small-12"-sub-box were used as a midbass woofer to compliment, say, a 15" or 18", then I see how it would become useful ONLY if the highpass x-over on that 12" was set at or above f3.

I think my ears will find the more natural bass from a 6" than a 7" in .25 cu. ft. I appreciate all the explanation.

[rc white]

A point that should be made is that a sealed box for minimum distortion should be no larger than Vas/3, for this to be true then the Qt must be .35 for an overall q= .7.
Just putting a given driver in a box that gives it a Qt of .7 does not mean the box is optimal if the Vas/v < 3.
Putting a given driver into a sealed box that is too large can result in increased distortion, but apart from this, and my previous comments about spl. and kp, there are no hard and fast rules I can think of regarding driver/box size and sound.
Another thing to consider is that at an 80Hz. crossover frequency the average room is well into its standing wave/sidewall reflection region and any small differences through the crossover region are swamped by these effects.
Non linear distortion products however are present right up to midrange frequencies and if the driver is being driven above its linear region these increase at a very rapid rate.
rcw