robi Posted April 8, 2012 Share Posted April 8, 2012 Hi guys, some years ago I've built a pair of Cornscalas, here the tread: http://community.klipsch.com/forums/t/125302.aspx Now I'm looking for a simple and big basshorn. The following design it's OK ??? (Conical horn with triangular section) http://imageshack.us/photo/my-images/4/tromba1rev0.jpg/ To calculate the dimensions the following theory is OK??? http://www.bd-design.nl/contents/en-us/d26_Bass_Horn_Design.html Thank you in advance for your answers and sorry for my written english!!!!! Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 9, 2012 Share Posted April 9, 2012 Sorry that you did not get a quicker response. This sort of speaker was described in a letter to SpeakerBuilder in about 1990 if I recall correctly. OTOH, I've never seen any report of one being built or tested. The equations look okay without me checking them. Most people use an exponential expansion. The big equation with CosH and SinH reduces to the typical exponential formula with some substitutions. I don't know how the high frequency response of the bass horn will perform. Typically bass horns have the last section of the horn facing the listner and if I nnderstand correctly, this causes some beaming to make up for the fact that all but the most strong driver will roll off well below 400 Hz. You should be able to calculate the roll off of the driver you chose with the equation in the article. This 400 Hz upper end is needed because that is about as low as a midrange horn and driver will go. The design shown is pretty much aiming at the floor except for the reflector at the floor level. I can't say whether or not it will work. It is really inaccurate to predict results unless a working model is tested. OTOH, "scoop" type designs including the Klipsch LB somewhat make a turn near the end like your target. So perhaps it will work. The other issue is that the back chamber will probably be difficult to build. But that is up to your skills. It looks like the horn is easy (good) but the back chamber is difficult (bad)! One other issue is that this looks like it is possible to build an 8-foot long horn reaching from ceiling to floor. But this is going to cause a sizable time delay. Maybe it is easier to build a Belle or LaScala or even a Jubilee. = = = = = Addition by Edit. I'd like to point out that the drawing is more of a concept than something which well designed. Things which have to be large look small in the drawing or not matching. For example, it looks like the driver in the back chamber fires down into a round hole in the "motor board" The motor board is going to be quiite a large trriangle to take a 15-inch driver. The hole is going to have a cross sectional area of about 80 square inches for a 15-inch driver. It could work out but the dimensions are something to take a closs look at. The next problem is that the "throat" (small) end of the horn in the drawing does not mate to the hole as far as size. The throat should be about the same area as the hole but in the drawing it could be up to ten times as large in area. That discontinuity is a problem. A final issue is that the mouth (big) end of the horn is going to be about four square feet (but look at your equations). Perspective and dimensions are tough to gauge. But the drawing certainly doesn't look that large. The conclusion is that the drawing makes things look easy. A working design is going to be different. I'd suggest you sketch out your design and perhaps make a model out of foam core. Best, WMcD Quote Link to comment Share on other sites More sharing options...
DrWho Posted April 9, 2012 Share Posted April 9, 2012 Conical horns tend to have poor low frequency extension, or a lot of passband ripple. However, exponential or tractrix horns tend to be more complicated to build. I would recommend learning how to use hornresp and model the expected performance (that way you can make the tradeoffs between ripple and low frequency extension):http://www.hornresp.net.ms/ The tool is completely free too. If you need some more help on the design front, then I would recommend visiting this forum: http://www.avsforum.com/avs-vb/forumdisplay.php?f=155 Quote Link to comment Share on other sites More sharing options...
robi Posted April 10, 2012 Author Share Posted April 10, 2012 Thank you guys!!!! I'm a novice so I'm trying to divide the problem in several little problems...to be solved separately. Quote Link to comment Share on other sites More sharing options...
robi Posted April 10, 2012 Author Share Posted April 10, 2012 Let's speak just about CONICAL horn. What is the formula to calculate the lenght of the horn??? In the link above I see just the sections (throat and mouth) but not the overall lenght of a conical horn (in the case of T=00). I've already calculate: 1)Sections 2)Back chamber 3)Compression chamber I need to calculate the lenght. Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 10, 2012 Share Posted April 10, 2012 Nevermind what I wrote earlier. I think it is best that you put the area equation in a spreadsheet. WMcD Quote Link to comment Share on other sites More sharing options...
robi Posted April 11, 2012 Author Share Posted April 11, 2012 Thank you WMcD !!!! Here my spreadsheet. I can't understand the formula for the lenght of a conical horn on the link above. (T=infinity , ecc....) Could anyone post this formula??? I'll put it on my spreadsheet. Thank you again!!!! Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 11, 2012 Share Posted April 11, 2012 Well, I've been working on it -- but for the next ten days I have to do other things. One problem is that the equation for area as we go down the horn (x) makes sense as long as we use T=1 for an exponential. But if T= infinity (or a very big finite number), it looks to me that it blows up. I checked Barnek's book on acoustics and it says the T must be limited to a value of 1. I can post that tomorrow. The equation shows up in Salmon's patent on hyperbolic horns. I'll post that, too. There is no mention of a limit. There is someplace an article by Salmon but it may take a week to find in my collection. WMcD Quote Link to comment Share on other sites More sharing options...
djk Posted April 12, 2012 Share Posted April 12, 2012 Download Hornresp from McBean (free). http://www.dmcbean.bigblog.com.au/index.do Here is a nice little dual 15 cornerhorn (of course it can only play 146dB at 35hz): Quote Link to comment Share on other sites More sharing options...
robi Posted April 12, 2012 Author Share Posted April 12, 2012 @WMcD: Ok WMcD, I'll wait you infos!!! @djk: before start with software I'd like to understand the theory.... ;-) Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 12, 2012 Share Posted April 12, 2012 Dear Robi, I too will have to go step by step. It is easy to be a know-it-all and I've found that there are gaps in my understanding. I do believe that the author of the article you are relying upon is stating equations and what is stated is at least over generalized. Attached is the U.S. Patent by Salmon. To my knowledge it is the source of the equation for the shape of several horns. It is true that conical horns are mentioned. However, his major work was in midrange horns where T is 1 or lower. In the patent there is a drawing of the sidewalls of several T value horns,including the conical. However, it seems to me that T= infinity gives the wrong impression. Of course anyone can search for patents on the www.uspto.gov website. But it is sometimes difficult to set up a TIFF reader plug in. Quicktime kidnaps the plug in setting and while you can see pages,you can't save them. Alternitiff is needed. WMcD USP 2338262 to Salmon.pdf Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 12, 2012 Share Posted April 12, 2012 My next step was to put the Salmon equations on a spreadsheet. Actually I used metric at first (see next pane). But I wanted to check that it made sense. So I shifted to Imperial inches and T=1 for exponential,and used the LaScala as a model. The LaScala is easy to understand. -- at least in my way of thinking. First, though, we must understand that in an exponential horn, like the LaScala, the area going from the the throat to the mouth doubles every increment of X distance. In the LaScala, that increment is 12 inches or a foot. (Aside: the exponential equation descibes this increase -- like in science fiction movies where the toxic bacteria blob doubles itself by reproduction every hour until it consumes the entire world -- Scream!. It is the same math. Smile.) Returning to the LaScala: Generally, the throat area is 1/2 square foot. After 1 foot it has doubled to 1 square foot. After 2 feet along the path it has doubled to 2 square feet.. After three feet, it has become 4 square feet. I must apologize to metric people because im mixing feet and inches. But Imperial people will see that the progression of area in square inches equates to a progression in square feet of 0.5 to 1, to 2, to 4. The mouth of an LS is 2 feet by 2 feet or 4 square feet. People who know Excel (looking into cell equations) can see that the complicated Salmon equation with hyperbolic sine and hyperbolic cos is accomplishing this result. WMcD Spreadsheet for Exponential Horn (inches).zip Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 12, 2012 Share Posted April 12, 2012 Attached is my first spreadsheet using metric meters. It is a bit different than the previous one posted in that instead of the side of a square it uses the radius of a circle to show a linear dimention from the center axis of the horn to the side wall. People out there may be wondering about the units overall. Essentially, in the calculations, the C (speed of sound) used dictates the results. WMcD Spreadsheet for Exponential Horn.zip Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 12, 2012 Share Posted April 12, 2012 So, where are we and what is this all about? One issue is whether we can use T greater than 1. Or, more specifically, can we use T = infinity, or a big number? Try putting in 100 for T, which is not even close to infinity. The spread sheets allow for that and I've put in the nifty graphs. Nonetheless, you have to look at the resulting numbers. The areas are impossibly large. More, in a few days. WMcD Spreadsheet for Exponential Horn.zip Quote Link to comment Share on other sites More sharing options...
robi Posted April 13, 2012 Author Share Posted April 13, 2012 Thanks for sharing those very very interesting documents !!!!! I'm going to read all !!!! I'm a novice and I'm surprised that La Scala bass horn is "exponential". I supposed to be conical because the walls of the horn are planar (!!!!?????!!!!) See you after my vacations (without internet and computers...) in 10 days... ;-) I've to study a lot :-( Quote Link to comment Share on other sites More sharing options...
WMcD Posted April 13, 2012 Share Posted April 13, 2012 Here is one last thing to take on vacation, if you have not already left. Note that on page 277 it mentions that M (same as our T) never exceeds unity. It seems to me that Beranek recognized the problem that we have been discussing. WMcD Beranek pgs 276-280.pdf Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.