Hello
I have a diy speaker using a Seas P17REX paired with a Morel MDT29 tweeter. I like the overall sound of both of these units. They are in a vented enclosure, modelled using lspcad i believe many years ago.
Recently the amplifier used to drive these has changed, to something with a lot less power output. My idea now is to buy another pair of P17REX and build a new enclosure. I thought of making a transmission-line type this time, and realise i know nothing of how to model these speakers. I think i remember a software being mentioned somewhere, to atleast give oneself a basis for experimentation.
Are the woofers particularly unsuited for a transmission line?
Their stats are, i believe, Fr=34Hz, Vas=30.5, Qts=0.24, Z=8ohms with a sensitivity of 89dB/SPL.
I was thinking about a design similar to the "Thor" kit,
http://www.madisound.com/pdf/Kits/audioXpress Thor Review.pdf
Ive only read that review in passing, however, and it seems that some(all?) of the math required can be derived from that document. My main question is, is there a piece of free software which simplifies experimentation with TL speakers?
I have access to a non-professional measurement setup, so i might be able to utilize this in the crossover tweaking and enclosure adjustments.
rgds Kristian
I have a diy speaker using a Seas P17REX paired with a Morel MDT29 tweeter. I like the overall sound of both of these units. They are in a vented enclosure, modelled using lspcad i believe many years ago.
Recently the amplifier used to drive these has changed, to something with a lot less power output. My idea now is to buy another pair of P17REX and build a new enclosure. I thought of making a transmission-line type this time, and realise i know nothing of how to model these speakers. I think i remember a software being mentioned somewhere, to atleast give oneself a basis for experimentation.
Are the woofers particularly unsuited for a transmission line?
Their stats are, i believe, Fr=34Hz, Vas=30.5, Qts=0.24, Z=8ohms with a sensitivity of 89dB/SPL.
I was thinking about a design similar to the "Thor" kit,
http://www.madisound.com/pdf/Kits/audioXpress Thor Review.pdf
Ive only read that review in passing, however, and it seems that some(all?) of the math required can be derived from that document. My main question is, is there a piece of free software which simplifies experimentation with TL speakers?
I have access to a non-professional measurement setup, so i might be able to utilize this in the crossover tweaking and enclosure adjustments.
rgds Kristian
Quick reply, just about to eat my dinner, good place to start is Martin J King's website: http://www.quarter-wave.com/
yacuza said:
Qts is rather low for a TL... other important TL resources are my site (URL below) and Bob Brines
http://geocities.com/rbrines1/
dave
Low Qts
As Qts is an equation using qes and qms, how does one know when qts is smallish for a given TL enclosure? A smaller driver is going to have a smaller Qts. Wouldn't that make it unsuitable only for a large TL enclosure?
Or does the science of TL teach us that "Transmission" only works in certain physical circumstances?
Dave
As Qts is an equation using qes and qms, how does one know when qts is smallish for a given TL enclosure? A smaller driver is going to have a smaller Qts. Wouldn't that make it unsuitable only for a large TL enclosure?
Or does the science of TL teach us that "Transmission" only works in certain physical circumstances?
Dave
The Qts from a smaller speaker isn't lower per definition, compared to the qts of a large driver. The Qts is the total quality factor for a speaker and is a good way to see the strenght (and efficiency) of any speaker.
Basically a low Qts means a big magnet and vice versa.
I use a driver with an Qts of 0.29 in a TL and it works very nice, while giving good low end extension. It has a smaller mouth as throat and was easily modelled with AJ-horn. Modelled with the mouth the same size or larger than the throat it gave poor results.
Mvg Johan
Hello
The first TL I built using a 6.5in driver was with the P17REX, using an " IPL Acoustics " design, and the results were excellent. It was this that proved to me that low Qts drivers were the way to go for lines. So I would go ahead and try them. I still have the original plans, but you might achieve even better results by using Martin Kings worksheets
Roy
The first TL I built using a 6.5in driver was with the P17REX, using an " IPL Acoustics " design, and the results were excellent. It was this that proved to me that low Qts drivers were the way to go for lines. So I would go ahead and try them. I still have the original plans, but you might achieve even better results by using Martin Kings worksheets
Roy
Roy Lewis said:
If you happen to have these in some sort of electronic format, and would not mind, i would very much like to receive those plans, if nothing else to use as a sanity-check for my own design. Email is kristian (at) renegade (dot) se
After reading some more on various places it seems to me one of the more important factors to decide upon is the length(and thus resonant frequency) of the line. Using two drivers, should i start with a point between the two drivers, or at the topmost driver? Perhaps it is best to use a middle point, so that both drivers experience an equal amount of "detune" from their optimum tl length.
Or should i just shut up and build a line, it seems to me it is a good idea to build it from cheap board on the first go, and get a sense of it, then build it again with higher-quality woodwork.
I want to thank everyone for replying, and hope to receive more of them, in spite of my tendency to lengthy posts.
rgds,
Kristian
Truly, truly, check Martin King's site for his Mathcad worksheets. He has one that is specifically designed to help you determine line length, driver placement, and line cross-sections.
I suggest that it would be profitable to learn how to run the mathcad worksheets in order to build with confidence.
I suggest that it would be profitable to learn how to run the mathcad worksheets in order to build with confidence.
kneadle,
If you look at attachment 1 of my alignment tables, you will see a study of TL response and enclosure size as a function of driver Qts. The results show that a low Qts driver will work in a TL and as always there are compromises to be made with both low and high Qts values.
My personal preference would be for a Qts between 0.3 and 0.4 with the extended range 0.2 to 0.5 also acceptable. I have not had much luck with Qts values above 0.5, this usually means a small magnet which does not seem to have a strong enough grip on the driver's moving mass to move the driver quickly or work against the line enough at the fundamental quarter wavelength resonance.
Just my preferences,
If you look at attachment 1 of my alignment tables, you will see a study of TL response and enclosure size as a function of driver Qts. The results show that a low Qts driver will work in a TL and as always there are compromises to be made with both low and high Qts values.
My personal preference would be for a Qts between 0.3 and 0.4 with the extended range 0.2 to 0.5 also acceptable. I have not had much luck with Qts values above 0.5, this usually means a small magnet which does not seem to have a strong enough grip on the driver's moving mass to move the driver quickly or work against the line enough at the fundamental quarter wavelength resonance.
Just my preferences,
That's very interesting. It happens that I have been trying to design an enclosure based on an ideal geometry for the Tang Band 881s (I have several Mark McKenzie modified 881s drivers). It has proved to be very difficult, and I was trying to track this down. The Qts for this driver comes to .661.
By your own personal preferences, and some other designs I've seen with other drivers, this Qts is too high. By others' preferences, this Qts is too low. In either case, this Qts is not ideal.
Now I'm wondering if anyone has found success in modeling the 881s. And I'm also wondering if I should start another thread.
Thanks,
Dave
By your own personal preferences, and some other designs I've seen with other drivers, this Qts is too high. By others' preferences, this Qts is too low. In either case, this Qts is not ideal.
Now I'm wondering if anyone has found success in modeling the 881s. And I'm also wondering if I should start another thread.
Thanks,
Dave
Here are the specs I used, downloaded from Partsexpress when I ordered the things.
I measured them using SW, and if they're different, I didn't note it, so they must not have been too different.
Dave
Re 6.87 [ê ]
Fs 111.15 [Hz]
F1 81.75 [Hz]
F2 153.99 [Hz]
Zm 37.25 [ê]
D 62.00 [mm]
Qms 3.60
Qes 0.81
Qts 0.66
Bl 3.31 [N/A]
L1K 0.60 [mH]
L10K 0.24 [mH]
Ms 1.86 [g]
Vas 1.40 [l]
dBSpl 86.36 [dB]
Cms 1.10 [mm/N]
Ma 16.00 [g]
FsMa 35.88 [Hz]
I measured them using SW, and if they're different, I didn't note it, so they must not have been too different.
Dave
Re 6.87 [ê ]
Fs 111.15 [Hz]
F1 81.75 [Hz]
F2 153.99 [Hz]
Zm 37.25 [ê]
D 62.00 [mm]
Qms 3.60
Qes 0.81
Qts 0.66
Bl 3.31 [N/A]
L1K 0.60 [mH]
L10K 0.24 [mH]
Ms 1.86 [g]
Vas 1.40 [l]
dBSpl 86.36 [dB]
Cms 1.10 [mm/N]
Ma 16.00 [g]
FsMa 35.88 [Hz]
GM said:
What GM says...
The back driver can be run FR or can be rolled off at or just above the baffle step (ie used as a 0.5 driver). The 2 drivers can be placed at the same place along the length of the line so no fudging or averaging of offset required.
The tighter you cn mechanically couple the drivers the greater the "active force cancelation"
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.