Noob measuring T/S parameters

[mikee55]

Hi all, I'm just about ready to start measuring the Thiele Small parameters of a driver. I've built plenty of cabs in my time, but as far as accuracy goes, the only accurate part being my woodwork skills. I'm refering to Rod Elliotts article on his site. My first question is regards to:Quote

The loudspeaker driver should be suspended in free space, with no obstructions or interfering surfaces nearby. Any boundary closer than around 600mm (about 2ft) will affect the accuracy of the measurements

Can I suspend the driver using a thin cable, or does it need to be fixed? I had in mind a pole across a pair of chairs and using a bit of wire to hang the driver. Me thinks oscillating the cone might cause the driver to swing, would this effect the readings?


More to follow....

Cheers mikee55

[richie00boy]

As long as the driver does not start moving about thus creating slight offsets due to the modulation of it swinging about, it should be ok.

[AndrewT]

Hi,
T/S parameters are always done at very low signal levels, It's inherent in the definition of T/S.

I use the constant current method i.e. large resistor in series with the voice coil fed from a nearly constant voltage source (oscillator).

The cone can just be heard to produce a tone, but the signal level is hundreds of uW, not mW nor W.

This level of signal will not cause untoward swinging of the driver. But vibration coming from external sources will upset your voltage readings. I suspect the fan noise in my lab supply probably causes errors in my readings. I measure the voltage across the voice coil not the voltage across the 1k0 series resistor.

[pinkmouse]

Originally posted by AndrewT
...But vibration coming from external sources will upset your voltage readings...

Things not to do with Speaker Workshop

[Vikash]

You should mount the driver vertically (ie. not on a table top) and fairly rigidly. Things like room temeperature also have an effect.

http://www.diyaudio.com/forums/showt...113#post744113

I wouldn't mount the drivers with any kind of cable! This is what I use: http://www.vikash.info/audio/measure...lamp/clamp.asp

[mikee55]

Hi, thanks for the feedback. I've just read this after the test. I repeated it a couple of times. I used a digital AC mV meter first, then my Farnell analouge meter and got the same results. I know the resonant freq is stated appx 60hz of this driver (an Audax fibreglass 6") and my results gave 62.4hz. Im not sure if the spreadsheet is right? Have a look, please.
Source current of 13.79 Amps???

I've stopped at resonant freq test, and come here to post.

As for other techniques, I don't want to take on board too much, this is frying my brain already.

Thankyou. mikee55

[richie00boy]

I think there's a typo on the spreadsheet, it should read 13.79 mA not A.

[mikee55]

Ok, thanks richie00boy, err, now what do I do??


Speaker current Im = Vm / Rs
Resonance impedance Rm = (Vs - Vm) / Im
r0 (reference value) r0 = Is / Im
-6dB current Ir = �ã(Im * Is)
-6dB voltage Vr = Ir * Rs


Grrr I hate maths...
Help, Pretty Please

Thanks mikee55

[richie00boy]

Not sure what you're getting at with those formulas. The procedure is on the ESP site.

Although it looks like you have made error somewhere, as the Fs sanity check failed and Qts is stupidly high.

The problem might lay in that you have used 15 ohms for Rs when I think Rod says to use like 0.1 ohm?

[mikee55]

Hi richie00boy, basicaly I'm Copy/Pasting from Rod's article(Sorry Rod), to get the text right rather than type it,(lazy me),

An accurate test resistor (any value, although I suggest 10 ohms) A ½W component will be quite sufficient

This is what I read from the list of required parts from the test rig.From the page
http://sound.westhost.com/tsp.htm

Now I'm stummped

Cheers mikee55