Proposed TL enclosure design and driver parameter comments & questions

[Texan]

A while back I started a series of threads inquiring about what known designs people recommended and liked, so that I might gain some insight on how to choose one to build. Then I asked if anyone had a design for an rs125-4 mtm with the woofers in series.
http://techtalk.parts-express.com/sh...d.php?t=209027http://techtalk.parts-express.com/sh...on+line&page=2 Last night. I have found some mistakes already, but nothing past appearance. Oh ya my old excel did not have many color choices that will be obvious.

If any of you have more accurate data for the parameters of this driver, and the new Bl for a series connection I would appreciate it.

[Paul K.]

Re: Proposed TL enclosure design and driver parameter comments & questions

Using PE's published T/S specs for the RS125-4, I modeled an ML-TL for a pair wired in series using Martin's ML-TQWT worksheet (I have the last version of this worksheet he released). F3 came out to ~56 Hz, which I believe would satisfy you. The cabinet I modeled has internaldimensions of 38"H x 5"W x 5"D which I somewhat arbitrarily chose. Since this would be an MTM, the tweeter's center is the TL design center and I chose that to be 7" below the top (internally). I also arbitrarily modeled with 0.5 ohms assumed in series with the RS125s. Lastly, I used a stuffing density of 1 lb/cu.ft., which requires ~5.5 ounces of stuffing in the top 24" of the cabinet. The mass-loading port's center is located 2" from the bottom of the cabinet and has a diameter of 2" and a length of 1-3/4". In the attachment, if it shows up, is the system response I modeled. I hope this helps, and don't hestitate to contact me directly for more info, etc.

[Pete Schumacher]

Re: Proposed TL enclosure design and driver parameter comments & questions

Using PE's published T/S specs for the RS125-4, I modeled an ML-TL for a pair wired in series using Martin's ML-TQWT worksheet (I have the last version of this worksheet he released). F3 came out to ~56 Hz, which I believe would satisfy you. The cabinet I modeled has internaldimensions of 38"H x 5"W x 5"D which I somewhat arbitrarily chose. Since this would be an MTM, the tweeter's center is the TL design center and I chose that to be 7" below the top (internally). I also arbitrarily modeled with 0.5 ohms assumed in series with the RS125s. Lastly, I used a stuffing density of 1 lb/cu.ft., which requires ~5.5 ounces of stuffing in the top 24" of the cabinet. The mass-loading port's center is located 2" from the bottom of the cabinet and has a diameter of 2" and a length of 1-3/4". In the attachment, if it shows up, is the system response I modeled. I hope this helps, and don't hestitate to contact me directly for more info, etc.

Just curious Paul, how are you combining the Ts params for the worksheets? You're doubling Vas, Sd, Re and Lvc and leaving Qes, Qms and
BL alone?

[Paul K.]

Re: Proposed TL enclosure design and driver parameter comments & questions

Just curious Paul, how are you combining the Ts params for the worksheets? You're doubling Vas, Sd, Re and Lvc and leaving Qes, Qms and
BL alone?

I also doubled Bl, which is the correct thing to do, and Qes and Qms are the same as for a single driver, also the correct thing to do AFAIK. I arbitrarily added 0.5 ohms in series with the pair of series-connected drivers to represent the d.c. resistance of crossover inductor(s), which the worksheet automatically incorporated to give the corrected value for Qts. I know in past modeling with series-connected drivers, if I don't also double Bl, you get some really weird and ugly response graphs.

[Pete Schumacher]

Re: Proposed TL enclosure design and driver parameter comments & questions

I also doubled Bl, which is the correct thing to do, and Qes and Qms are the same as for a single driver, also the correct thing to do AFAIK. I arbitrarily added 0.5 ohms in series with the pair of series-connected drivers to represent the d.c. resistance of crossover inductor(s), which the worksheet automatically incorporated to give the corrected value for Qts. I know in past modeling with series-connected drivers, if I don't also double Bl, you get some really weird and ugly response graphs.

So, for parallel connected drivers, you'd halve BL, Re and Lvc, double Vas and Sd?

[Paul K.]

Re: Proposed TL enclosure design and driver parameter comments & questions

So, for parallel connected drivers, you'd halve BL, Re and Lvc, double Vas and Sd?

Nope, Bl for two paralleled drivers is the same as the Bl for a single driver. Yep, Re and Lvc would be halved, and Vas and Sd would be doubled.

Are you trying to set me up for something here?
Paul

[Pete Schumacher]

Re: Proposed TL enclosure design and driver parameter comments & questions

Nope, Bl for two paralleled drivers is the same as the Bl for a single driver. Yep, Re and Lvc would be halved, and Vas and Sd would be doubled.

Are you trying to set me up for something here?
Paul

No, I've got the "original" worksheets and was going to attempt to do the same thing you did.

So, for drivers in series, BL doubles, but doesn't for being in parallel. Weird.

[Paul K.]

Re: Proposed TL enclosure design and driver parameter comments & questions

No, I've got the "original" worksheets and was going to attempt to do the same thing you did.

So, for drivers in series, BL doubles, but doesn't for being in parallel. Weird.

Yeah, weird, and that's what I thought the first time I had to address it, but it definitely is the way Bl must be handled (for any type box).

If you have the older worksheets that do not automatically take into account the effect of added series resistance on Qes and, therefore, Qts, you'll have to manually calculate the new value of Qes and enter it instead of the specified Qes in order for Qts to be correctly calculated by the workhsheet.

[Texan]

Re: Proposed TL enclosure design and driver parameter comments & questions

It seems like this Bl adjustment or lack of adjustment works correctly with the table method as well. I discovered this by using the Tri Trix and the Triune as examples. I guess force = Bl*i so for series if i is held constant force from 2 drivers =2*Bi*i even though voltage has doubled since resistance has doubled. From the tables it looks like if Bl is increased the area of the line can be decreased and if resistance is decreased then the area of the line can be decreased. I am glad I was not the only one who was unsure about this.

Tonight I will mess around with it a little more. Thanks for modeling this stuff for me. If you guys are willing I will get you to model my proposed alignment when I have it.

[Texan]

Re: Proposed TL enclosure design and driver parameter comments & questions

Would you guys be kind enough to stick these dimensions in to the Mathcad model.

Length of line=33.5
SL/SO=0.333
Inside width of baffle=4.5”
Inside depth = 9”
Opening =4.5”x3”
SL =40.5 in^2
SO=13.5 in^2

You can adjust things as you see fit. I would like to preserve the narrow 6” outside baffle and keep the depth to a minimum, but length can be extended if necessary. I think I would like to keep a relatively large amount of taper if possible. I want to stay away from the mass loaded designs, because I am hoping my final design can be duplicated with the Alignment Table Spreadsheet. Thanks a lot. I guess I have said adjust as you see fit, and then in the next breath said don’t adjust.

I have noticed you guys “Paul K, and Pete Schumacher” have showed continued interest in transmission lines, so I am glad ya’ll chimed in.

I have been working on a few things and trying to learn more and more as I go. After discovering that Curt’s Tritrix Triune enclosure was significantly reduced in volume from what the tables suggested I have been trying to learn what parameters could help me find a range in amount of volume reduction that could be utilized for a specific design. I chose this enclosure to look at since he said it was started using the Tables

[Paul K.]

Re: Here's what I got...

The attachment shows what I got by modeling your proposed 3:1 tapered TL. As you can see, the response is much more lumpier than the ML-TL I originally modeled. I used the same stuffing density but a bit longer into the line to minimize the lumps as much as I could. From my viewpoint the taper needs to be increased or even more stuffing added in order to smooth out the response more. Adding more stuffing, however, will raise F3 and increasing the taper will require a shorter line. In general, I don't even think about tapers less than 10:1 because anything less usually is way too bumpy for my tastes. I do prefer tapered lines when they are appropriate, however.
Paul

[Texan]

[Paul K.]

Re: Proposed TL enclosure design and driver parameter comments & questions

Okay, the attachment shows the modeling results for the 10:1 tapered TL you dimensioned above. Again I used a stuffing density of 1 lb/cu.ft. and the stuffing is uniformly distributed by density in the first 20" of the line's length starting from the closed end. The response is better than the 3:1 tapered line but still not as good as for the ML-TL. The terminus could be on the baffle as you want. I did also try earlier today an ML-TQWT but couldn't get very good results for that, either.
Paul

[Texan]

[Paul K.]

Re: Proposed TL enclosure design and driver parameter comments & questions

Yep, we can reduce the internal width to 4.5" and increase the depth appropriately to end up with the exact same system response curve.

There is no tapering involved with an ML-TL. An ML-TL has a constant cross-sectional area and can be described as a tall, vented cabinet, but it functions differently because the height of the cabinet is utilized as a 1/4-wave resonator along with the mass-loading port's size to obtain the desired system tuning frequency.