ShinOBIWAN said:
A single Shiva that's virtually flat from 125hz to 16hz at 108dB, with no EQ and the response is that flat?
I'll take that with a pinch of salt even with room gain
You'd probably get a a flatish response in a larger room but then it certainly wouldn't be flat to 16hz at 108dB and no EQ or driver destruction for that matter.
I could work at the p-p required for a 12" at 16hz 108dB but I already know the answer and the Shiva isn't it.
Also why do you have the sub running all the way upto at least 125hz? Bet you can hear that sucker booming away in the corner
It was a subwoofer test with the x-over knob at the highest frequency. When in the system, the subwoofer is crossed over at 70 Hz.
It's a small room, the box is large and well built, following exactly Adire Audio EBS plan. If you model it in WinISD, you'll see you can achieve a solid 103 dB anechoic at 16 Hz and you're below xmax rating. 5 dB of room gain at 16 Hz is very reasonable.
There's absolutely no EQ even if you don't believe me. This wasn't even a maximum power test, wait for the next one...
PAging Mr. Johnstone
Yep that is where I saw the original image to. Looked through a whole bunch of back issues when I was sinking my teeth in speaker design in the late 80's. Seems like yesterday.
I would post pics if I had the slightest idea how. THere was some thread or something on how to do this but I have never found it. Nor did I figure it out on on my own.
Help !
MArk
Yep that is where I saw the original image to. Looked through a whole bunch of back issues when I was sinking my teeth in speaker design in the late 80's. Seems like yesterday.
I would post pics if I had the slightest idea how. THere was some thread or something on how to do this but I have never found it. Nor did I figure it out on on my own.
Help !
MArk
For images, go below the typing input window for a reply and click on browse. Just find the file in your computer and click save, which attaches it. Use JPEG files so they are small.
At least now I feel alot more confident in my idea, since some horn expert came up with the same thing decades ago.
At least now I feel alot more confident in my idea, since some horn expert came up with the same thing decades ago.
Hunting and Fishing
Ok all you savy horn designers. I have done my home work. Modeled the hell out of the horn in Horn Response. I understand what to do. But due to system crach a couple of months ago I lost my program that generates the horn area per lenghth increments. I found a great paper by MArshal Leach and a great program that will work along side. But it won't spit out the design on paper. Just in a box that I have to copy by hand. It's the MLutil from the single driver website. Does anybody know of a spreadsheet file that will spit out the numbers?
Mark
Ok all you savy horn designers. I have done my home work. Modeled the hell out of the horn in Horn Response. I understand what to do. But due to system crach a couple of months ago I lost my program that generates the horn area per lenghth increments. I found a great paper by MArshal Leach and a great program that will work along side. But it won't spit out the design on paper. Just in a box that I have to copy by hand. It's the MLutil from the single driver website. Does anybody know of a spreadsheet file that will spit out the numbers?
Mark
Just highlight and copy from the tool on the single driver website and paste directly into excel if you want to manipulate the numbers or simply notepad to just print them.
Also note that HornResp gives you the corresponding flare rate in Hz for each exponential segment. Then convert to conical at the end. Varying the flare rates of the different segments is a common way to use space more efficiently in a bass horn. For example the Jensen Imperial hypex horn flare has a 20hz 1st segment, a 50hz second segment, and about 40hz final 2 segments with a 2m overall length and it's a real champ in the low bass department. Of course it cheats to get the really low bass with the extra large compression chamber.
Also note that HornResp gives you the corresponding flare rate in Hz for each exponential segment. Then convert to conical at the end. Varying the flare rates of the different segments is a common way to use space more efficiently in a bass horn. For example the Jensen Imperial hypex horn flare has a 20hz 1st segment, a 50hz second segment, and about 40hz final 2 segments with a 2m overall length and it's a real champ in the low bass department. Of course it cheats to get the really low bass with the extra large compression chamber.
Thanks for the reply John
I guess that I should have been more specific.
I'm really looking for something that will output the specs from the Leach model.
I have found that there is quite a difference between the Leach model and the others available. I have read that Horn Response uses Leach as a source for the calculation to. Everything that I have found so far ges me a hyp/exp flare rate based on different formulas and gives me a very different horn. The ML util is quite interesting to use. Really quite powerfull. But from it's data I cannot print the horn section table.
If I am going to build such a beast I would like to do it right. I have built around a dozen horns using tractrix calculator on the single driver web site and the hyper/exp calc on the same site. You don't usually get the response that you are shooting for when you design with these. I never knew about the Leach horn paper untill quite recently. Many people who's websites I take as being well researched say that Leach is one of the most accurate methods. See this site:
http://ldsg.snippets.org/HORNS/index.php
http://users.ece.gatech.edu/~mleach/
Here is the paper.
“On the Specification of Moving-Coil Drivers for Low-Frequency Horn-Loaded Loudspeakers,”
Thanks for the tip about copying and pasting I appreciate it. I'm just a little more computer savy than that ( though at times it doesn't show) I spent two years in graphic design. Just getting rusty.
Mark
I guess that I should have been more specific.
I'm really looking for something that will output the specs from the Leach model.
I have found that there is quite a difference between the Leach model and the others available. I have read that Horn Response uses Leach as a source for the calculation to. Everything that I have found so far ges me a hyp/exp flare rate based on different formulas and gives me a very different horn. The ML util is quite interesting to use. Really quite powerfull. But from it's data I cannot print the horn section table.
If I am going to build such a beast I would like to do it right. I have built around a dozen horns using tractrix calculator on the single driver web site and the hyper/exp calc on the same site. You don't usually get the response that you are shooting for when you design with these. I never knew about the Leach horn paper untill quite recently. Many people who's websites I take as being well researched say that Leach is one of the most accurate methods. See this site:
http://ldsg.snippets.org/HORNS/index.php
http://users.ece.gatech.edu/~mleach/
Here is the paper.
“On the Specification of Moving-Coil Drivers for Low-Frequency Horn-Loaded Loudspeakers,”
Thanks for the tip about copying and pasting I appreciate it. I'm just a little more computer savy than that ( though at times it doesn't show) I spent two years in graphic design. Just getting rusty.
Mark
This thread may very well have been dead for a while, but I feel that rather than starting a new thread, posting here would be more relevent (and to show I did at least do some reading)
It seems the original intent of the poster was to produce a cheap subwoofer by using smaller drivers 8" but in higher quantity. I agree with the argument to go with a single or double 12" instead of several 8". With my project (or more like pipe dream at the moment... until someone responds to my emails...), I feel that perhaps it might work in my case where I plan to use 4, 12" quality drivers instead of a large 18" (or 24" which is required to equal surface area) which is usually very expensive.
A possible enclosure would be similar to
http://www.ultimateavmag.com/features/604way/index4.html
But I also thought about 2 drivers in front, and 2 drivers in the rear also facing the front while out of phase (push-pull out of phase) to cancel non-linearities in a sealed enclosure.
I looked into TL such as the sonotube for lower cutoff-frequency, but the size of the enclosure, construction complexity, output from the port creating even more complicated room interactions, and reasoning supposedly sealed has the ideal transient response has persuaded me otherwise (unless someone else has something to add in favor of a TL).
I figure that the multiple driver setupwill help get the low frequency of 20hz easily, and 15hz @ >-3db
I am not concerned with max spl, just 15hz-60hz (actually even as low as 40hz) type of output at low distortion levels and moderate listening levels 80-90db
I'm not sure how much wattage I will be needing either (regarding adding more drivers). My configuration for wiring would be 2 parallel drivers in a series, to get back to 8 ohms.
Comments and subwoofer advice please!
It seems the original intent of the poster was to produce a cheap subwoofer by using smaller drivers 8" but in higher quantity. I agree with the argument to go with a single or double 12" instead of several 8". With my project (or more like pipe dream at the moment... until someone responds to my emails...), I feel that perhaps it might work in my case where I plan to use 4, 12" quality drivers instead of a large 18" (or 24" which is required to equal surface area) which is usually very expensive.
A possible enclosure would be similar to
http://www.ultimateavmag.com/features/604way/index4.html
But I also thought about 2 drivers in front, and 2 drivers in the rear also facing the front while out of phase (push-pull out of phase) to cancel non-linearities in a sealed enclosure.
I looked into TL such as the sonotube for lower cutoff-frequency, but the size of the enclosure, construction complexity, output from the port creating even more complicated room interactions, and reasoning supposedly sealed has the ideal transient response has persuaded me otherwise (unless someone else has something to add in favor of a TL).
I figure that the multiple driver setupwill help get the low frequency of 20hz easily, and 15hz @ >-3db
I am not concerned with max spl, just 15hz-60hz (actually even as low as 40hz) type of output at low distortion levels and moderate listening levels 80-90db
I'm not sure how much wattage I will be needing either (regarding adding more drivers). My configuration for wiring would be 2 parallel drivers in a series, to get back to 8 ohms.
Comments and subwoofer advice please!
Hara said:
In my exprimenting with TL's I found long straight TL's can have issues with either end exciting different room modes. I've cut said lines in two and folded them with a smooth 180 degree curve so the driver and opening are together and eliminated the weird mode issues. (normal room mode behaviour still applies though)
Though folding a sonotube based TL is easier said then done
Thats what I thought, I just couldn't simulate anything... especially since I have no driver parameters at the moment, I can only get those once I get the drivers and find them out by hand. I was hoping to draw from the experience of others.
For sealed designs, I am getting in the works to learning Speaker Workshop. Any other recommended resources specifically for sealed designs?
Also any critique on the strategy? Perhaps I'll have to wait until I get the drivers...
For sealed designs, I am getting in the works to learning Speaker Workshop. Any other recommended resources specifically for sealed designs?
Also any critique on the strategy? Perhaps I'll have to wait until I get the drivers...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.