modeling H-frame and U-frames, excel acting up, any advice?

[thadman]

I'm really not sure whats wrong with this excel spreadsheet, it mentioned something about macros? If it means anything "spl_max1" works perfectly fine along with edge.

Here is a screenshot, anybody have any ideas?


I'm trying to model the baffles to discover which route is best for dipole bass. I can take the RSS390HF from 35-40hz up to 300hz, but a flat baffle (or very shallow H-frame, although I wont know without modeling) is required and it can only be so wide before it becomes visually distracting. I could use a pair of RS270s on a flat baffe (100-300hz), and then a huge displacement woofer (Mach5 18s anyone?) in a deep h/w frame to cover the very lowest octaves. The problem with this setup is the RS270s would be covering only about 1.5 octaves and the subwoofers would be covering even less...also I'm introducing even more crossovers and the amplification costs associated with powering multiple woofers (ie im going active) starts to become a factor.

[MJKing]

The file is a MathCad worksheet not an Excel spreadsheet. You need at least MathCad Explorer 8 to run the worksheet.

[thadman]

Is there anyway to get mathcad for free? I couldnt find it on download.com

[MJKing]

You can download a free version from my website. It does have a few limitations, you cannot save your work or print directly (print screen and paste into Word works) but it is free. Most people use this version to run my MathCad worksheets. There are new U and H frame worksheets, as well as OB worksheets, which extend what was done in the version you have downloaded from the FRD site. They are not quite free and require a modest fee to access.

[joecarrow]

I'll just add in that I got Martin's worksheets back in 2003/2004, back when they were free and slightly less capable. If I was designing cabinets from scratch right now, I'd say that the updated worksheets would be worth paying for. They'd go into my toolbox along with the measurement mic and SPL meter, and at that point I'd probably also buy some crossover software to more conveniently measure and model drivers and crossovers.

I really think that this forum has outstanding crossover design resources, and the people here are overall some of the most practical and knowledgeable on the net- but I don't see as much cabinet innovation here as I've seen globally. I still say that a properly designed transmission line could improve a lot of the boxes you see in the Mission Accomplished section. Heck, Zaph's ZD5 ended up in a transmission line- I don't think that this forum or this school of thought is married to the idea of traditional ported boxes.

Having said that, No, Martin, I still haven't given a single driver an honest chance. I have a surplus 0.38 cu ft box I've had no luck selling; perhaps once I find employment I should go ahead and load it with an FE167e, eh?

Back on topic-

Yes Thadman, there's a limit before it becomes visually distracting. There's also a point of baffle width where there is some reinforcement at a particular frequency, and excursion requirements are reduced. As I understand it, it's the point where the output of the dipole is equal to the output of the driver in a closed box. If your baffle is wide enough, this frequency can reinforce the bottom end of the midrange and allow it to more gracefully handle a lower crossover. Here's a start for reading about this effect.

http://linkwitzlab.com/faq.htm

There is a LOT to sift through, so please be patient- Linkwitz has already answered a huge number of your possible questions. You'd be surprised what's on his site if you read long enough.

[Dennis H]

I really think that this forum has outstanding crossover design resources, and the people here are overall some of the most practical and knowledgeable on the net- but I don't see as much cabinet innovation here as I've seen globally. I still say that a properly designed transmission line could improve a lot of the boxes you see in the Mission Accomplished section.

Hey Joe,

I think you might be surprised at the amount of enclosure 'innovation' here. Jon and Thomas built a lot of TLs back in the day but, after a lot of experimentation, didn't find any significant advantage over a good ported design. With proper tuning and internal damping a ported box can give very similar response to a good TL. And Jon's M8TA enclosure is actually very close to one of Martin's quarter wave pipe designs. I forget the thread but the TL guy from DIYAudio (Scottmoose?) was here for a while cranking out box designs for one of the projects (Modula MTM?) similar to the Fat Thor concept. His best one ended up having one big chamber with a port although he insisted it wasn't really a ported box.

[ThomasW]

I really think that this forum has outstanding crossover design resources, and the people here are overall some of the most practical and knowledgeable on the net- but I don't see as much cabinet innovation here as I've seen globally.

Jon and I went through our TL phase back in the 1970's, when we designed and built a boatload of them (I had a discrete 4 channel 4-way system where the woofers and the midwoofers were mounted in TL's). In one extreme design we put a single Eminence 10" DVC woofer in a really big TL. Powering it up brought a Phase Linear 700 to it's knees, must say though it was really LOUD.... :B

[thadman]

Got it up and running, Thanks MJK! :T :B

I am playing around with it right now, and have entered in the thiele small parameters of the driver, but am not sure what units Im supposed to enter for the "enclosure geometric definition". Specifically how do I calculate length, area of the driver end/open end (cubic inches?).

[MJKing]

One of the great features of MathCad is the way it handles units. Length can be entered in inches, feet, yards, miles, mm, cm, m, or kilometers since it automatically converts to a consistent set of units to do the calculations. Same for areas. What you cannot do is enter an area in cubic inches. You decide what enclosure dimensions to analyze and iterate the design by hand until you get a response you are happy with and meets your goals for the speaker.

[MJKing]

Having said that, No, Martin, I still haven't given a single driver an honest chance.

I am sure one day you will, and then ............................

[thadman]

One of the great features of MathCad is the way it handles units. Length can be entered in inches, feet, yards, miles, mm, cm, m, or kilometers since it automatically converts to a consistent set of units to do the calculations. Same for areas. What you cannot do is enter an area in cubic inches. You decide what enclosure dimensions to analyze and iterate the design by hand until you get a response you are happy with and meets your goals for the speaker.

How do I calculate length though? Does it include the width of the baffle...AND the depth and the length from wing tip to wing tip?

What do you mean I cant enter area in cubic inches? Would I just put 100 instead of 100^2?

[MJKing]

How do I calculate length though? Does it include the width of the baffle...AND the depth and the length from wing tip to wing tip?

What do you mean I cant enter area in cubic inches? Would I just put 100 instead of 100^2?

The dimensions you enter are for the air volume, so they are internal dimensions. The thickness of the baffle in the center of an H frame or the front of a U frame is ignored.

Area is not expressed in cubic inches, that is a unit of volume. Area is expressed in square inches.

[thadman]

The dimensions you enter are for the air volume, so they are internal dimensions. The thickness of the baffle in the center of an H frame or the front of a U frame is ignored.

Area is not expressed in cubic inches, that is a unit of volume. Area is expressed in square inches.

Sorry, I didnt mean cubic inches originally. I meant square (as you can see in my post).

Let me get this correct, L is the distance from the center of the baffle to free space?

[MJKing]

Yes, I checked the worksheet and your figure is correct.

[Mark K]

For what it's worth, you can get the full version of MathCad 14 at an academic discount for ~125. Of course, you or someone in your family has to be in school. Believe or not, I think K-12 qualifies. I guess they're pushing higher math pretty early these days. I won't mention the website specifically, but one site starts with academic and ends in superstore.com...

My wife is teaching anatomy this semester, so I've been thinking of picking up some discounted software.

[cjd]

I am sure one day you will, and then ............................

You'll go back to multi-driver, because you just can't replace missing frequency range any other way?

[joecarrow]

You'll go back to multi-driver, because you just can't replace missing frequency range any other way?

I'm trying to keep an open mind on this matter; design is full of trade-offs, and enough people are into the single-driver thing that it seems to me that there's got to be some kind of a benefit present or else they wouldn't be doing it. It may be purely psychoacoustic, in which case it would be an interesting thing to observe.

I've had several very successful job interviews this past week, so hopefully I'll have an income again and will be able to feel free to experiment once again.