joeybutts said:
If you look at the diagram in the first post, you'll see the thicknessses for each part. "1,2" means 1.2cm = 12mm = 1/2". "1,6" = 16mm = 5/8", although you can use any thickness you want for any piece as long as the internal dimensions stay the same.
For mine I used 16mm for A,B,E,F and 12mm for the rest.
dzzmiller said:
We all share a common interest of making genuine quality stuff from scratch. Once you remove brand/locality prejudice, people can make a lot of really good progress together
(Also everyone here is doing it for the pure sake of enjoyment, not for profit, which distorts things a lot.)
kristleifur said:
you're welcome.... I'm happy to make another diy-er happy... just like was
chylld said:
We all share a common interest of making genuine quality stuff from scratch. Once you remove brand/locality prejudice, people can make a lot of really good progress together
Hey All!
I've been lurking this thread for some time to find tips building a pair of needles. Well, it helped me a great deal and it's just fair that I post a picture of the completed couple.
They still are without shoes, havent really decided on the size, they sound great for being so small, that frs8 isnt so bad as it looks.
Here goes:
Thanks guys!
I've been lurking this thread for some time to find tips building a pair of needles. Well, it helped me a great deal and it's just fair that I post a picture of the completed couple.
They still are without shoes, havent really decided on the size, they sound great for being so small, that frs8 isnt so bad as it looks.
Here goes:
An externally hosted image should be here but it was not working when we last tested it.
Thanks guys!
HornResp model of Needles
A friend of mine was looking for some reasonably cheap sound in her small dorm room and since I have a pair of Monacor SPH-60X fullrangers gathering dust for over 4 years, I offered to "see what I can do". (A simple and not too powerfull second-hand amplifier can be found on the internets for under 50€, bringing the whole project at around 100€.)
Since I have some experience with HornResp and it can model unstuffed/undampend Mass-Loaded Transmission Lines, I first simulated Cyburgs Needle with the Tangband W3-871S parameters I found on the PDF-specsheet. After choosing the Combined response option in the FR plot, I got a huge dip between 50-100Hz, not to mention the spike around 45Hz ... I am pretty sure I modeled it right, but somehow i think this is not the way the needles are supposed to sound.
I then adjusted the width of the cabineth from 10cm --> 13cm (i.e. increase the cross-sectional areas by 30%) to fit the larger Monacor SPH-60x and planned to adjust the model untill I got a FR plot that looked acceptable. Strangely enough, I got a perfect plot from the first time! Compare this to the plot of the TB W3 needles ...
I never expected the SPH-60x to be a drop-in replacement (with width adjustment), so what is going on here?
To simulate stuffing somewhat, I increased the length of the 2 large conical sections/folds and then it goes deeper with a gradual slope that can fit room gain to some extent, as is expected, so i think the problem is not in HornResp but in the parameters.
What is wrong with my TB W3-871S simulation? And if that would be ok, what does this say about the validity of the SPH-60x simulation?
I feel like I'm missing something here ... any second opinions?
Aside from AkAbak - which fits HornResp for equivalent models anyway - I don't have other modeling programs that can do a ML-TL, but maybe someone can do a quick comparison? Thanks in advance!
http://users.telenet.be/Hellhouse/public/HR-models/
A friend of mine was looking for some reasonably cheap sound in her small dorm room and since I have a pair of Monacor SPH-60X fullrangers gathering dust for over 4 years, I offered to "see what I can do". (A simple and not too powerfull second-hand amplifier can be found on the internets for under 50€, bringing the whole project at around 100€.)
Since I have some experience with HornResp and it can model unstuffed/undampend Mass-Loaded Transmission Lines, I first simulated Cyburgs Needle with the Tangband W3-871S parameters I found on the PDF-specsheet. After choosing the Combined response option in the FR plot, I got a huge dip between 50-100Hz, not to mention the spike around 45Hz ... I am pretty sure I modeled it right, but somehow i think this is not the way the needles are supposed to sound.
I then adjusted the width of the cabineth from 10cm --> 13cm (i.e. increase the cross-sectional areas by 30%) to fit the larger Monacor SPH-60x and planned to adjust the model untill I got a FR plot that looked acceptable. Strangely enough, I got a perfect plot from the first time! Compare this to the plot of the TB W3 needles ...
I never expected the SPH-60x to be a drop-in replacement (with width adjustment), so what is going on here?
To simulate stuffing somewhat, I increased the length of the 2 large conical sections/folds and then it goes deeper with a gradual slope that can fit room gain to some extent, as is expected, so i think the problem is not in HornResp but in the parameters.
What is wrong with my TB W3-871S simulation? And if that would be ok, what does this say about the validity of the SPH-60x simulation?
I feel like I'm missing something here ... any second opinions?
Aside from AkAbak - which fits HornResp for equivalent models anyway - I don't have other modeling programs that can do a ML-TL, but maybe someone can do a quick comparison? Thanks in advance!
http://users.telenet.be/Hellhouse/public/HR-models/
those are the spikiest graphs i've ever seen
from what i've read on the internet, the needles are definitely far from a flat/neutral response speaker. it's just that there's something charming about their simplicity and popularity that makes them such a joy to build, and hey, for real music they sound quite nice too!
that said, it seems you've found one relatively easy way to extract more out of the basic design, and i definitely recommend you go ahead with it. just make sure to post back with results!!
from what i've read on the internet, the needles are definitely far from a flat/neutral response speaker. it's just that there's something charming about their simplicity and popularity that makes them such a joy to build, and hey, for real music they sound quite nice too!
that said, it seems you've found one relatively easy way to extract more out of the basic design, and i definitely recommend you go ahead with it. just make sure to post back with results!!
Asides from the measured FR graph from Cyburgs himself somewhere on the first page in this thread, i found this simulation with a few different drivers.
http://www.coolcat.dk/bjoern/needle_analysis.pdf
Both lack the spike-and-dip on the bottom of the graph and I haven't found much comments about uneven freq response. Off course about the lack of bass, but with these drivers this should be much less of a problem than with the smaller ones they use.
Hornresp and most other modeling software usually exaggerate high Q peaks, but even with some dampening I think this is too much to smooth out.
If I would go ahead with it, I will definately post results here, but for now I am a bit wary and i can wait a few more days to see if some interesting remarks pop up around here.
http://www.coolcat.dk/bjoern/needle_analysis.pdf
Both lack the spike-and-dip on the bottom of the graph and I haven't found much comments about uneven freq response. Off course about the lack of bass, but with these drivers this should be much less of a problem than with the smaller ones they use.
Hornresp and most other modeling software usually exaggerate high Q peaks, but even with some dampening I think this is too much to smooth out.
If I would go ahead with it, I will definately post results here, but for now I am a bit wary and i can wait a few more days to see if some interesting remarks pop up around here.
Thanks Henkjan, I have seen more flat curves now than the one I have of the Needles, so the problem definately lies with my simulation.
I see you have made different simulations with various drivers. As I don't have the opportunity to install or use this spreadsheets, would you mind plugging in the parameters for the SPH-60x (with the increased cabinet width) to see how it models? I suspect the FR curve will be mostly the same, but the low end is even more raised than with the other smaller drivers.
thanks in advance/dank bij voorbaat!
Joris
I see you have made different simulations with various drivers. As I don't have the opportunity to install or use this spreadsheets, would you mind plugging in the parameters for the SPH-60x (with the increased cabinet width) to see how it models? I suspect the FR curve will be mostly the same, but the low end is even more raised than with the other smaller drivers.
thanks in advance/dank bij voorbaat!
Joris
The T/S parameters were in the HornResp settings I posted above , but I got the parameters for you, 'fresh' from Monacor's site:
http://www.monacor.de/typo3/index.php?id=84&L=1&artid=2152&spr=EN&typ=full
P.S. (I did the same for the Tangband:
http://www.tb-speaker.com/detail/1208_03/w3-871sc.htm and there was a large difference in Le and Rms, but even with the updated parameters, the 45-hz spike remains ... curious...)
On the other hand, I have stumbled upon these posts in some fullrange-forum, and if the GM and the Scottmoose from there, are the same guys from here, I should definitely ask Scottmoose how his MLTL turned out.
http://fullrangedriver.com/forum/viewtopic.php?id=556
/me is off to send a pm ...
Edit: doesn't work- doesn't accept emails. Maybe he will notice it anyway?
, but I got the parameters for you, 'fresh' from Monacor's site:http://www.monacor.de/typo3/index.php?id=84&L=1&artid=2152&spr=EN&typ=full
P.S. (I did the same for the Tangband:
http://www.tb-speaker.com/detail/1208_03/w3-871sc.htm and there was a large difference in Le and Rms, but even with the updated parameters, the 45-hz spike remains ... curious...)
On the other hand, I have stumbled upon these posts in some fullrange-forum, and if the GM and the Scottmoose from there, are the same guys from here, I should definitely ask Scottmoose how his MLTL turned out.
http://fullrangedriver.com/forum/viewtopic.php?id=556
/me is off to send a pm ...
Edit: doesn't work- doesn't accept emails. Maybe he will notice it anyway?