making cabinets match hornresp

[jbell]

One of my least favorite topics, but it's the gift that keeps on giving -- Translating hornresp into buildable dimensions.

In this case, what I built EXCEEDED my hornresp prediction... Arrgh. It seems like the bigger the cabinet, the less my translations errors affect the outcome, but the smaller the cabinet, and/or smaller the dimensions... the bigger the differences become.

Can anyone help me out on this one? I keep making the same mistake over and over again.

Attached is a drawing of what I built, and what I simmed.
Here is what I measured -- the sim and the measurement really don't have much in common.

40 95
43 96.5
46 97
50 97.5
53 98.5
57 99.5
62 102.5
66 105
71 106.5
76 105.5
82 104
88 102.5
95 101.5
102 101
109 107
117 109
126 109
135 105
145 101
156 97

[tb46]

Hi jbell,

I gave it a quick try to redraw your enclosure, but cannot find the dimensions necessary to do so. I'll need at least the internal height and depth to take a more informed look at this.

Regards,

[jbell]

oops, missed the height, 21.5 above the baffle, 8.5 below. All other dimensions are there.

[tb46]

Hi jbell,

Thanks for posting the additional information. I had to make some minor changes to my ACAD sketch, but nothing major.

Basically, I come up with similar numbers for the areas and length, but find some differences with the chamber dimensions, but, there are some differences.

The throat chamber volume is relatively well defined @ 8.5"x23"x4" (minus the portion taken up by the driver, plus the driver cone volume), so that gives you about 12 liter. What to pick as throat chamber area is more difficult to me, do we look at it as the area the drivers are directly working into, e.g.: 8.5"x23", or, as the smaller cross-section of 8.5"x4"?

The rear chamber volume is 8.5"x23"x3.6" (minus driver portion, plus cut-out volume) for about 11.5 liter. I figure the volume of the 55-2421 to be about 1.6 liter (each) that leaves me with about 8.5 liter Vrc. I set the Lrc @ 3.6"/9cm (does driver fit?). I would delete the rear chamber vent (Ap=0 - Lpt=0 - it is in your Hornresp Input, but not your sketch).


Finally, (it may not have anything to do with the question of matching Hornresp with reality, but it would make it much easier to pick specific cross-sectional areas and length for Hornresp) I would make some changes to the layout; it would work better with the horn starting in the middle of the throat chamber.

Anyway, I like your general layout, it reminds me of the old Lowther Side Vivace and the Schmacks. Nice work. I hope this helps.

Regards,

[jbell]

Thanks oliver, good to know I didn't have things WAY off. Yes I know the drawing shows a front chamber of 12liters, I filled it with open cell high density foam, to bring it down to 5 liters. A 12 liter front chamber doesn't work at all in this design according to hornresp.

Your drawing shows I was off a few cm here and there, but nothing major. The one major place where your drawing and my thinking are different, is the location of L23-L34. I would have said that they would have both been about 52cm, instead of 45cm and 59cm. When I put your dim's into hornresp... I still get within a db of my earlier sim... which is nothing close to 104db@60hz, or 109@120hz.

This time I'm on the good side of the measurements and sim not lining up, but other times it's been the other way around.

Still don't know what I'm doing wrong... I swear it has to be the fact that I'm not building a real conical expansion... and that's what I'm simming in hornresp.

[littlemike]

Oliver,

I came up with a similar idea with respect to your modification to Jim's fold. If you elect to split the back chamber, one could also orient the drivers face to face, making the throat chamber very small.

Jim,

This is an impressive little box. Unfortunately, I'm too tied up with the rest of my projects to play with speakers right now, or I'd have built one already.

[tb46]

Hi jbell and littlemike,

jebell, your results are really not that far off your simulation with the exception of the peak around 120Hz, and that may be a measurement problem. I agree, that it is important where in the horn flare we pick our S3, etc.. It looks to me that some points are clearly defineable, whereas something located in the middle of a turn is not as accurate, and the length has given me fits even in the 70's when I was messin' with TLs. Oh, well.

littlemike, I was thinking about a TH around a similar arrangement, but, have not had time to put real numbers to it, I'll attach a really bad sketch. :-) And, talking about projects and time........

Regards,

[jbell]

resurrecting this thread, because I need to build some 55-2421 'flat to fourty' horn subs.

Size no issue (up to about 500 liters or so with up to 4 drivers per box) What's the loudest design anyone has done? 2pi only please... (this particular cabinet didn't go low enough, but it is loud)

I've done a lot of sims, and several cabinets -- but before I make a more sawdust, I want to see if someone has a 'home run' design.


(and if you look, I just ran MCM out of 55-2421's.. so anyone without, will have to wait awhile...)

[tb46]

Hi jbell,

I just posted a 1cft TH using the MCM 55-2421 (single, obviously) on the "Smallest TH" thread. It just happens to approximately meet you lower corner spec. I just tweaked it up to 40 Liter for a little higher output. It would be easy to boost it up into a Quad. Heck, with a 500 Liter upper box limit, how about an Octo? :-)

I got 4 of the 55-2421s before you struck, they are resting just fine.

Regards,

[jbell]

Thanks oliver, However I was thinking more in terms of large PA cabinet, not small wife approved type of cabinet for house duties, although 8 of those would be pretty fun.

The attached isn't flat to 40 as a single cabinet, but does get there in blocks of 4, or in a corner. 2 Of these sim about as good as my big cabinet with a 3015lf, and at about the same size. (which makes sense, as a quad of 55-2421's move the same amount of air as a 3015lf.)

Yes, I know CR is way high, but I've built insanely high CR cabinets with this little 8" driver, and just can't seem to destroy the cone... Even with 40 volt, 40hz sine waves...