Bricolo said:
I was actually thinking about this is i drove to send off the mail today.
This could be done at a number of levels.
Level 1: Fix the box, leave the MTM + existing XO.
Reading between the lines in Joe's article, this would on 1st pass entail restoring the volume he threw out. So extend the external depth dimension to 18.25". Now we note that he original choose the lower bound of Augspurger's taper rate, 3:1. King's work shows that the larger bound, 4:1 is sometimes to small, so perhaps even larger is a good idea -- without actually plugging the numbers into King's (or Augspurger's) model we are still running on feel. 4:1 would push the depth out to about 23".
Higher Levels would involve changing the XO, the ultimate probably being a tri-amped 2.5 way with the 0.5 way driver on the anterior of the box.
dave
My agree'ing thoughts
Hi Dave,
Sorry for the long post, it kept growing.
I agree that from people's collective reviews, the Thor could use some work. One thought I had was that in the article, the line was overstuffed in order to reduce the ripple. Tapering it more will help to solve this problem, but one significant thing that will happen is that the fundamental of the pipe will shift to a lower frequency, so the line length would likely need to be shortened. But I think that this could be the root of the problem since the current tuning is too high!
The line length is't sufficient to have the fundamental close to fs. If the line were straight, the line should be about 13500 in/s / 4 / 31 Hz = 108 in. The line's pretty close to straight right now.
The Sd of the driver is 126 cm^2 * 2 units = 252 cm^2 (39 in^2).
The starting, closed end cross section is currently 8.25" * (9" - two wall thicknesses@1.5") = 61.875 in^2. This is only 1.58 Sd.
For a single driver, my other simulations have shown good results with the starting area equal to 4 Sd and the finishing area equal to 0.5 Sd. Now that there's two drivers in there, it could make this a bit rediculous for some. 4 Sd would result in a starting depth of the pipe of 20.8 in (by the 7.5 in internal baffle width). This doesn't include the other end of the pipe which is about four more inches. But that four end depth could be reduced to the same length as the height of the terminus (2.75") to save some overall depth. This is pretty much the same as you mentioned - 23 in.
The article's terminus area is 2.75 * 7.5=20.6 in^2 which is 0.5 Sd, so this is looking good.
Perhaps a better solution in the end is to simply lengthen the line from what it currently appears to be about 76 in (although mentioned to be 81 in). This would also cause to offset the drivers a bit more in the line which has been shown to help with the ripple problem too.
Brendon 🙂
Hi Dave,
Sorry for the long post, it kept growing.
I agree that from people's collective reviews, the Thor could use some work. One thought I had was that in the article, the line was overstuffed in order to reduce the ripple. Tapering it more will help to solve this problem, but one significant thing that will happen is that the fundamental of the pipe will shift to a lower frequency, so the line length would likely need to be shortened. But I think that this could be the root of the problem since the current tuning is too high!
The line length is't sufficient to have the fundamental close to fs. If the line were straight, the line should be about 13500 in/s / 4 / 31 Hz = 108 in. The line's pretty close to straight right now.
The Sd of the driver is 126 cm^2 * 2 units = 252 cm^2 (39 in^2).
The starting, closed end cross section is currently 8.25" * (9" - two wall thicknesses@1.5") = 61.875 in^2. This is only 1.58 Sd.
For a single driver, my other simulations have shown good results with the starting area equal to 4 Sd and the finishing area equal to 0.5 Sd. Now that there's two drivers in there, it could make this a bit rediculous for some. 4 Sd would result in a starting depth of the pipe of 20.8 in (by the 7.5 in internal baffle width). This doesn't include the other end of the pipe which is about four more inches. But that four end depth could be reduced to the same length as the height of the terminus (2.75") to save some overall depth. This is pretty much the same as you mentioned - 23 in.
The article's terminus area is 2.75 * 7.5=20.6 in^2 which is 0.5 Sd, so this is looking good.
Perhaps a better solution in the end is to simply lengthen the line from what it currently appears to be about 76 in (although mentioned to be 81 in). This would also cause to offset the drivers a bit more in the line which has been shown to help with the ripple problem too.
Brendon 🙂
Hi Bricolo,
If you go to my website you can find my models and latest documentation. You will need to download the free MathCad explorer, make sure your PC is configured like an American PC, and then open any of the worksheets that are of interest. As far as I know Augspurger's models are documented in AES papers but are not available for general use. Both models provide the same answers.
Hope that helps,
If you go to my website you can find my models and latest documentation. You will need to download the free MathCad explorer, make sure your PC is configured like an American PC, and then open any of the worksheets that are of interest. As far as I know Augspurger's models are documented in AES papers but are not available for general use. Both models provide the same answers.
Hope that helps,
Hi Bricolo,
Under my computer, go to regional settings and change them to English (US). This seems to solve a lot of the problems that some people in Europe are having running the worksheets. You can try running the worksheets without doing this and see if they function correctly. If things don't function right you can always make the change.
Hope that helps,
Under my computer, go to regional settings and change them to English (US). This seems to solve a lot of the problems that some people in Europe are having running the worksheets. You can try running the worksheets without doing this and see if they function correctly. If things don't function right you can always make the change.
Hope that helps,
Has there been any progress in your redesign plans for the Thor ?
Dying to know if it is possible to improve the bass performance of this speaker, if so then I'm more than willing to start building one myself 🙂
Dying to know if it is possible to improve the bass performance of this speaker, if so then I'm more than willing to start building one myself 🙂
I'll have to understand some basic things about TLs before 😀
maybe planet10 has some infos
I'd also like to see some activity in this topic 🙂
maybe planet10 has some infos
I'd also like to see some activity in this topic 🙂
I'd also be interested in seeing Thor *updated*.
Since there are a few people around this forum who have a good
grasp of TL principles , maybe someone could come up with something ?
If I was to build it with the "original" dimensions before He reduced them ie: depth = 16.25" etc , and change nothing else , what would be the results ??
Anyone care to comment ?
Cheers
Andrew
Since there are a few people around this forum who have a good
grasp of TL principles , maybe someone could come up with something ?
If I was to build it with the "original" dimensions before He reduced them ie: depth = 16.25" etc , and change nothing else , what would be the results ??
Anyone care to comment ?
Cheers
Andrew
Ignoring the original TL design and using the driver's specs posted in the article, the Seas driver isn't all that suited for TL loading due to the lowish Q/Vas IMO, but it's what we've got to work with to keep the rest of the design ~intact.
Assuming the sims are reasonably accurate, a 10:1 taper in a 54" pipe with the drivers down so the tweeter is at ~22.58", S0 = 4.546*Sd (assumes 39.06in^2/252cm^2), SL = 0.4546*Sd, and 0.25lb/ft^3 stuffing density, this damps Fs with a ~aneochoically flat response down to ~50Hz, with a fairly room curve friendly roll off to a ~40Hz F3. A significant increase.
A simple straight ML-TL yields virtually the identical roll off except it has a much smoother upper BW response due to the smaller/shorter vent and F3 = Fs, Fb = ~27Hz. A really significant increase.
L = 36.62"
drivers centered at 16.63"
S0 = 2.1157*Sd
SL = S0
Density = 0.2lb/ft^3
rport = 1.5"
Lport = 8" (FWIW, 6" gives a more T/S max flat roll off, with a negligible efficiency gain).
I leave it to whoever wants to try either of these to adapt it to the original baffle size, etc., to keep the overall design as intact as possible, and of course, caveat emptor/YMMV/etc.. 😉
GM
Assuming the sims are reasonably accurate, a 10:1 taper in a 54" pipe with the drivers down so the tweeter is at ~22.58", S0 = 4.546*Sd (assumes 39.06in^2/252cm^2), SL = 0.4546*Sd, and 0.25lb/ft^3 stuffing density, this damps Fs with a ~aneochoically flat response down to ~50Hz, with a fairly room curve friendly roll off to a ~40Hz F3. A significant increase.
A simple straight ML-TL yields virtually the identical roll off except it has a much smoother upper BW response due to the smaller/shorter vent and F3 = Fs, Fb = ~27Hz. A really significant increase.
L = 36.62"
drivers centered at 16.63"
S0 = 2.1157*Sd
SL = S0
Density = 0.2lb/ft^3
rport = 1.5"
Lport = 8" (FWIW, 6" gives a more T/S max flat roll off, with a negligible efficiency gain).
I leave it to whoever wants to try either of these to adapt it to the original baffle size, etc., to keep the overall design as intact as possible, and of course, caveat emptor/YMMV/etc.. 😉
GM
I am beginning to design an MTM TL very similar to the Thor this week using Vifa drivers. I plan on using the P17's and the new XT ring radiator. The dimensions will be slightly different, but all of the theory and application notes should transfer just fine to the Seas drivers.
Also, I could probably use some help from all the TL guys out there as I go. The only TL I've ever built was for a 6 1/2" sub driver that had previously been in a tuned ported box. I was blow away by how good this little 6 1/2" speaker sounded when I was done. I used a small box made of 3/4" MDF, and mounted the driver to that. Then, I used a 55" long 3" diameter PVC pipe stuffed with about 2/3 lbs/^3 ft of poly fill. It sound overstuffed right now, and I plan on removing some of the stuffing to get the response I had before putting this much poly in it.
Anyway, I'm now REALLY excited about TL's! I plan on trying out a few different variations, but the MTM one will be first.
Cheers,
Zach
Also, I could probably use some help from all the TL guys out there as I go. The only TL I've ever built was for a 6 1/2" sub driver that had previously been in a tuned ported box. I was blow away by how good this little 6 1/2" speaker sounded when I was done. I used a small box made of 3/4" MDF, and mounted the driver to that. Then, I used a 55" long 3" diameter PVC pipe stuffed with about 2/3 lbs/^3 ft of poly fill. It sound overstuffed right now, and I plan on removing some of the stuffing to get the response I had before putting this much poly in it.
Anyway, I'm now REALLY excited about TL's! I plan on trying out a few different variations, but the MTM one will be first.
Cheers,
Zach
I'm not too good with the TL maths , so can someone put me straight about this.
If I leave the width & height at 9" & 41.25" respectively , but increase the depth to 16.25" (from 12.5")......what do I do with the internal baffle ?
Does the angle of 86 degrees remain the same and just move the baffle back by 3.75"..???
Do the proportions of the two volumes of v1 & v2 stay the same?
Help !!!
Andrew
If I leave the width & height at 9" & 41.25" respectively , but increase the depth to 16.25" (from 12.5")......what do I do with the internal baffle ?
Does the angle of 86 degrees remain the same and just move the baffle back by 3.75"..???
Do the proportions of the two volumes of v1 & v2 stay the same?
Help !!!
Andrew
the deepth must change to 16.25" (internal!, external is 18.75")
As for the internal bafle, just put it in order to split the top to 1/5 and 4/5 sections (taper of 4/1, the initial taper is 3/1)
As for the internal bafle, just put it in order to split the top to 1/5 and 4/5 sections (taper of 4/1, the initial taper is 3/1)
Taper rate
IS one supposed to taper the line all the way from driver to terminus? I read somewere that there had to be a clear demacation between the line and coupling chamber.
IS one supposed to taper the line all the way from driver to terminus? I read somewere that there had to be a clear demacation between the line and coupling chamber.
- Status
- Not open for further replies.