What's the difference between a TL, a ML-TL and a TQWT?

Transmission Line speakers are generally assumed to be using quarter-wavelength standing waves inside the enclosure to produce the bass emanating from the port so they are also referred to as Quarter Wave tubes. ML-TL resembles a bass reflex but is utilizing these standing waves to load the port. A classic TL may appear to be a long pipe with the driver at one end and the other end open. Usually some poly-fil to help damp the driver. A ML-TL may be a box with the driver mounted on the front baffle near the top of the box and the port at the bottom of the box very much like a BR but calculated to use a quarter-wave. This enclosure may use damping material. The TQWT acts as a cross between a horn and a TL. By tapering the pipe the resonances will spread over a wide range of frequencies theortically producing smoother response. The top of the pipe is usually filled with damping material. The results you get will vary depending on the driver used and how well the pipe has been designed. The rest is personal preference. Some think TQWT are honky like some horns but I've heard some very good ones.
Nice response, I think that you have summed it up very well. But I think that you missed one point.

The TQWT acts as a cross between a horn and a TL. By tapering the pipe the resonances will spread over a wide range of frequencies theortically producing smoother response.

Actually, when the line expands like a TQWT, I agree a cross between a horn and a TL, the lowest standing waves increase in frequency and start to bunch up and overlap giving the TQWT more SPL efficiency at the bottom end but TQWT bass will not be as low in frequency as the same length TL. This can also lead to a rippled response.

If you think about it, a 30 Hz TQWT will be much longer then a 30 Hz straight TL. The TQWT's lowest standing quarter wave is at a frequency that is higher then the length would indicate if you used the equation for a straight TL.

f = c/4L

By about the 5th or 6th quarter wavelength standing wave, the equation above becomes more accurate for the TQWT. So the first quarter wavelength mode is higher in frequency then would be expected for a straight TL and each of the following modes are still higher in frequency but not by as much until they converge with the TL frequencies at about the 6th mode.

Boy that was a real babble. Maybe an example will help. I just made this up for demonstration purposes. Suppose we had a TL and TQWT both with a length of 112 inches. The frequencies might look like the following table.

Mode__Straight TL __TQWT (SL>S0)
1/4_____30 Hz______80 Hz
3/4_____90 Hz_____130 Hz
5/4____150 Hz_____175 Hz
7/4____210 Hz_____220 Hz
9/4____270 Hz_____275 Hz
11/4___330 Hz_____330 Hz

The closer spacing of the first three modes in the TQWT causes the peaks to overlap a little and provide a summed bass SPL response that is sligthly higher then would have been achieved with the straight TL. Again the bass will not go as low for this sample TQWT.

The opposite trend will happen if you use a tapered TL (SL<S0).

Adding a port, or mass loading, to a quarter wavelength enclosure allows the length to be shortened for the same fundamental standing wave. This is like adding a lump of clay to the end of a ruler clamped to the counter in your kitchen. When the lump of clay is added, and the length is unchanged, the ruler will vibrate at a lower frequency. The type of vibration (displaced shape) remains the same but the frequency is reduced.

I think I just confused myself. I hope that helps,
a little OT

Would adding a little bracing in an ML TL disrupt the standing wave used to load the port? I find my ML TLs are rock solid (I used 3/4"mdf and 5/8" Baltic Birch laminate) and are fine without bracing but a friend is interested in a pair, he does not want them built as thick as mine. I was thinking to use a few dowels to reduce the resonances large panels may create.

P.S. Thanks for all your help so far mr.King. The congestion I was trying to solve earlier(if you remember) has been diminishing with further driver break in and the use of felt on the inside of the basket legs. I am really impressed with the ML TL design.
Here is how I label quarter wave enclosures.

Tapered TL : SL/S0 < 1
TL : SL/S0 = 1
Expanding TL : SL/S0 > 1 (also known as a TQWT if SL/S0 >> 1)

Typically in the first two the driver is placed between the closed end and 1/3 of the length. In the last one the driver is placed closer to the 1/2 length point.

You can add a port to any of these geometries which I call mass loading. Others might call this something else which is OK. So if you use my definitions you can have a ML TL or a ML TQWT both of which I have used in several of my projects.

But please, don't get all hung up on these labels. The box does not know what it is, it just obeys the laws of physics. It does not matter what the geometry is called, it is just an attempt to conveniently describe the configuration that has been designed and built. If you tell somebody the following variables S0, SL, Length, driver position ratio, stuffing density and location, R_port, and L_port then you have described the system. Don't struggle with names.


2002-11-11 6:52 am

But the ML-TL is something I don't really understand. I know that the driver/box/port form a "resonator" tuned to a given frequency.
But that's the same for a BR box. And the 2 designs look very similar

When does a ported box start to act like a TL? Is it defined by the geometry? (the length of the line)

Bob Brines

2003-01-31 10:11 pm
May I chime in.

You are getting hung up on labels rather than physics. The difference between the bass reflex box and a quarter-wave pipe is the dominant resonant mode. If the dominant mode is the helmholtz resonance, then the box is a bass reflex. If the dominant mode is the quarter-wave standing wave, then the box is a quarter-wave pipe. You can easily see the difference using Martin's worksheets, or by measuring the nearfield frequency response of your cabinet.

As a practical matter, if your box is close to golden ratio dimensions, you are assured of helmholtz resonance. Once the height to depth and width exceeds 3:1, you start to get quarter-wave modes. Simply stated, tall, thin floor-standing vented boxes are going to be quarter-wave resonators. As an example, my FT1600 cabinet for the FE167E is 45"x9"x9".

One other interesting point, if you use either the ML TQWT (setting SL/S0 = 1) or the Ported Box worksheet to model a ML TL then you will get an accurate prediction that includes the standing waves in the long direction of the enclosure. If the height is not tall enough to support a quarter wave resonance for the particulaer tuning frequency, and the box acts as a Helmholtz resonator then the worksheet will also accurately calculate this type of resonant response.

If you use one of the standard free ware packages, and model a ported box using lumped parameter theory, you will miss the quarter wave resonances in the long dimension. So if your box long dimension generates a quarter wave mode, you will not see it or any of the higher harmonics. One other nice feature of the Ported Box worksheet is the ability to specify the location of the port and the driver along the length. I think that this is a real advantage when designing a ported enclosure.

Bob Brines

2003-01-31 10:11 pm
So, many of the commercial floorstanders are ML TLs, without knowing it?

Indeed! Also, there are any number of DIY floor-standers with inexplicably low tuning or short ports. These are quarter-wave pipes.

To amplify what Martin said about using the "Ported" worksheet with straight MLTL's: If you model a straight MLTL with the "TQWT" or "Sections" worksheets, the port is by definition sticking out of the bottom of the pipe. You will notice port harmonics above ~500Hz that are equal in magnitude the fundamental resonance. Now, if you model the same dimensions in "Ported" and move the port up the pipe a few inches, so that the center harmonic of the group of three is suppressed, you will have a speaker that is +/- 1-2dB all the what to 1000Hz with very little stuffing. (Of course, you have to have the driver position correct also.)



2002-11-11 6:52 am
If I understood correctly, all mathcads models are based on the same equations (that seems logical), but allow you to change other parameters, in function of what kind of box you want to build?

So, if you model the same box in different worksheets, you should have the same results

Bob, you're talking about values for driver-port distance.
But I have no idea about the value to start with, neither for the driver-port distance, nor for any other parameter.

The alignment tables are such an usefull tool, can somethink like this be calculated for ML-TLs?
So, if you model the same box in different worksheets, you should have the same results

That is true, almost all of the worksheets are just TL Sections with a simplified set of input.

TL Offset Driver = TL Open End = ML TQWT = TL Sections

The one exception is Ported Box, this worksheet uses the same method but has different equations and a different solution sequence. Ported Box is the next step up from the TL worksheets listed above.