how do i find thiel small parameters for the radio shack 40-1035 15" woofer

[zuki]

the papers in the box explain only
watts db @ 1 w 1 m
and fs @25 hz

the needed specs are not there.

anyone used this wooer before and know what box works ?

any info will help.

thank you

[dooper]

From a google usenet search:

The JAES has some theoretical stuff on this, but if you want the
practical scoop, issues of Speaker Builder from the early 1980s had
some great articles by Robert Bullock and others on this. I recommend
getting as many back issues as possible from the 1980s. Accurately
determining speaker parameters is more difficult than it seems, and
Never accept the manufacturer's published specs. Bullock's articles
are compiled in "Bullock on Boxes", which I think Old Colony Sound Lab
in Peterborough can sell. They can also give you the Speaker Builder
phone number, since they are under the same management.

In a nutshell, you need a signal generator, an AC voltmeter with good
accuracy and a frequency response down to 10 Hz or so (almost impossible
to find) or an oscilloscope, along with a Wheatstone bridge for DC
resistance. Alternatively, a loudspeaker measurement system for PC
like CLIO is lots easier, if a bit more expensive, but will also give
you SPL and phase measurements. I am discovering you can't take
manufacturer's data for SPL either, nor can you assume the drivers are
flat within their usable range.

I use the constant current method to check impedances, and a dual trace
scope to find the phase null at resonance. A 600-700 ohm resistor in
series with the driver works pretty well, and will give you about a 1%
accuracy plus whatever your DVM error is. The trick is to measure the
voltage across a known resistor in series with the 600 ohm resistor, then
replace the known resistor with the actual driver and measure the voltage
without changing the drive level. If the know resistor is fairly close
to the driver impedance, the impedance should be proporional to the
voltage. For example, if a resistor of exactly 5 ohms reads 25 mv (I've
vound 5 mv/ohm is a good choice), and then the driver shows 30 mv, at
that frequency the impedance is 6 ohms.

Find the resonance in free air, and the impedance at resonance, Rmax.
Divide Rmax by the DC resistance of the voice coil to get r0. Calculate
R1 to be the DC resistance times the square root of r0. Then find
the frequencies on each side of resonance where the impedance drops to
R1.

The Q formulas are:

Qms=Fs x SQRT(r0)/(F2-F1) where F1 and F2 are the two frequencies found
above.
Qes=Qms/(r0-1)
Qts=Qms x Qes / (Qms + Qes)

Vas is more difficult to find and requires either adding mass to the
woofer or putting it in a sealed or vented box.

Note that reading frequency off the generator scale is not accurate
enough, in my experience. Ither count ticks on the scope scale or
get a frequency counter. To check your accuracy, SQRT(F1xF2) should be
within 2% of the resonance.

Vented box method for Vas:

Put driver in vented box. Find two impedance peaks, and the minimum
in betweehem (call these FL, FM and FH, going from lowest to
higheshest). Vas is:

Box volume x (FHxFH-FmxFm)x(FmxFm-FLxFL)/(FHxFHxFLxFL)

(I'm not sure how to get across "squared" so I factored it out).

I hope this helps.

- Don

Dickason has more detailed instructions in the LDC. There is also some good info here:

http://sound.westhost.com/tsp.htm

Like the man said, hope this helps!

dooper

[dooper]

In place of the signal generator and the frequency counter, it would be quite possible to use an inexpensive synthesizer, even something as basic as an old Yamaha DX100 4 operator FM synth. The 4 op synths can output pure sine waves, and you can transpose across the entire audible range. Here's a chart that I did in Excel - middle C = 261.63 Hz.

dooper

[kelticwizard]

The above posts are excellent about measuring your own Thiele-Small specs.

However, here are the published specs from the Radio Shack 2002 catalog. The 40-1035 is the first woofer listed. Fo equals Fs, the speaker's free air resonance frequency.

[zuki]

thank you

although i have little knowledge of using programs and software to reach an enclosure size.

with the info given on the catalog page.


any guess what a best/decent sized box would be for this driver ?
i can make any box once i know size and sealed/ported rec. and port dia and length.

i could even post pictures once i finished the box

thank you very much for your time.

[kelticwizard]

Zuki:

The high Qts of 0.81 pretty much rules out any boxes, except under special circumstances.

Many people here are building dipoles, for which this would be an excellent candidate. A dipole is a large board, not a box, with a speaker simply mounted on it. However, there are special dipoles which sort of snake the board around angles. These are called folded dipoles. Many people here are building these. If nobody posts here about them, I suggest you start a new thread about using one of them for your woofer.

You wouldn't have a closet door you can mount this on, would you? That is would be a great way to use these.

Another alternative would be to mount in a large sealed box, which would give you a large hump in the response. Then apply something called a Linkwitz Transform to take the hump out. The Transform circuit is something that many people are building here also.

Those seem to be your choices: Put into the door of your closet or on a wall to another room, build a folded diple enclosure, or use a Linkwitz Trasform circuit. I have never used a Linkwitz Transform circuit, but I know they are capable of taking a box with a speaker of too high a Qts and making it work.

[zuki]

any other info ?

[kelticwizard]

Working on it. Will post either on this thread or a new one to get the opinions of the board on a lesser-used enclosure type.