open backed Tapered TL? reverse-taper-into-BIB? zero-net-taper-1/2 wave resonator?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
imagine a tapered TL, folded once, but designed as a non folded line.

basically like a one fold design built within a rectangular box, with a centre partition that forms the fold, EXCEPT that back is left off the enclosure.

the line area at the bend in the line is equal to the desired vent area, and the first half of the line is calculated, with taper, to be the desired total line length.

the 2nd half of the line would expand with the inverse ratio to the first.

Along the lines of this:

An externally hosted image should be here but it was not working when we last tested it.



or maybe even something a little more like this look wise:

An externally hosted image should be here but it was not working when we last tested it.


if anyone out there understands me, would anyone like to hazard a guess as to the effect of doing this? or could hornresp model it ?
 
Last edited:

GM

Member
Joined 2003
Wasn't making light of your idea, just there's not really anything new in driver alignments, just variations on a theme with 'breakthroughs' due to advancing materials, manufacturing technology and increasingly powerful digital signal processing.

Anyway, these are a lot of fun, especially the corner loaded variants. I built a single freestanding one 6-7 yrs ago for a mono setup loaded with a 6.5"/~65 Hz Fs 'FR' RadioShack driver and the few folks that auditioned it were stunned at its 'slam' and extended LF. A properly done Karlson would probably have a little more 'slam', but wouldn't have as much extended LF as the trade-off. Not many 'free lunches' in audio.

GM
 
Ex-Moderator R.I.P.
Joined 2005
It's just a freestanding version of the simple little Klipsch/Jensen/EV BLHs of the '50s.

GM

I dont know that Klipsch

But fore corner load, or using backwall, maybe turn it the other way round
 

Attachments

  • TL corner.JPG
    TL corner.JPG
    12.4 KB · Views: 270
@ GM : NO offence taken. i didnt think i had invented anything, i was just asking about what the effects on response would likely be, given that the first part of the line is optimised for TQW alignment, as is the vent area.

lol it looks like a nice experiment, i should try and download hornresp and give it a try, assuming i can get my head around the program.......:D

@ tinitus: excuse me if im clutching at straws here, but isnt what you propose in your drawing a voigt pipe?
 
Last edited:
really???

:eek:

wow well thats about as big a compliment as im ever going to get on here!!!

:eek:

um...

:eek:

if only i could find an idiots guide to using hornresp.......as ive had a play with it and the SPL response doesnt look too clever.... big 10dB peaks etc....

area 1 = 290 area 2= 20-30 and area 3 = about 1000-1500 L1 = about 100 thus L2(if applicable...im a bit lost here) = about 100 also.

looks a rather rough though.....
 
Last edited:
@ GM : NO offence taken. i didnt think i had invented anything, i was just asking about what the effects on response would likely be, given that the first part of the line is optimised for TQW alignment, as is the vent area.

Ah, OK. Well, because the line is much more efficiently loaded it will have a wider gain BW and due to having a too 'fast' a flare (too large a mouth for its length) it will have a much higher Q (stronger) HF 'ripple' BW that causes much of the 'slam' part of its gain BW.

For it to 'be all it can be' the line side must be optimized for efficiently loading the secondary horn, so the TQWT part is optimized for the LF and the 'fast' horn for its HF BW. Fortunately, one no longer needs to be conversant in TL/horn theory to do it now that DB has expanded his 'slider' horn Wizard to include such esoterica, though of course knowing it will shorten time spent simming.

Don't forget to combine the driver/port responses, HR doesn't automatically do it as other programs do.

Note too that HR apparently assumes ~infinitely rigid boundaries/whatever as its peaks/nulls are much higher in amplitude than measured plus once you damp the system to 'taste' and combined with the room's response they don't sound anything like the sim except in the most general of terms, so when looking at it, draw an averaged mental line through all the peaky 'hash' to get an idea of its outdoor half space response, i.e. keep the hash' high Q with no deep, wide BW nulls since we tend to key on just the peaks otherwise.

GM
 
In the SPL RESPONSE window click on TOOLS, then COMBINED RESPONSE or use CTL + O. This will open a pop-up window that let's you dial in the path-length difference between the driver's front radiation and the vent's on the rear to the listening position. Fill it in, click OK and it will calculate/display it.

See also Pg. 15 of the HELP file.........

GM
 
In the SPL RESPONSE window click on TOOLS, then COMBINED RESPONSE or use CTL + O. This will open a pop-up window that let's you dial in the path-length difference between the driver's front radiation and the vent's on the rear to the listening position. Fill it in, click OK and it will calculate/display it.

See also Pg. 15 of the HELP file.........

GM

righty ho!

well after a bit of tinkering, i have come up with this:

An externally hosted image should be here but it was not working when we last tested it.


to my eyes it doesnt look great but considering its using a driver i have planned for TL at the mo, and that its only 87dB/W, the flatness and SPL gain was rather compromised. mind you i am primarily looking at this design from the stand point of a small as possible sub app, and i havent tried other drivers which may be much much better.whats the opinion of others?
 
Assuming this is a 2pi space sim, then it probably has enough BSC to be damped without losing too much and the high Q peaks will be ~gone, so looks good enough to try. If 1pi or 0.5pi it will probably require a lot of stuffing to keep from sounding 'boomy'.

FWIW, here's a corner loaded example I just calc'd for an AE AV12-x. Damped and 1/3 octave smoothed, IME its response above ~100 Hz will be a fairly smooth +/- 2-3dB 'ripple' till it begins to decay away:

GM
 

Attachments

  • AE AV12-X small Klipsch style corner horn.gif
    AE AV12-X small Klipsch style corner horn.gif
    21.2 KB · Views: 107
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.