Woofer Tester II

[spkrman57]

From helpful e-mail from forum member (WTPRO), I found out you have to be sure the audio/sound for your pc does not default to the WT2, if it does, change the audio/sound default to the sound card in your pc.



Many thanks to "WTPRO" who gave me the info I could not find elsewhere!!!

Ron

[WTPRO]

Good Morning All

I just thought I would mention that as questions come in, Brian and I (the other woofer tester guy) are adding those answers to the www.woofertester.com FAQ page. The FAQ is not too flashy at this point, but hopefully this will change soon.

Another tip is that another test code version will probably post soon. For example, and Giskard may want to chime in as it was him that wanted to know if the WT could generate an accurate TS model for a mid or tweet, is a new method (still under consideration) that will back out TS params from an efficiency rating and piston diameter. You may recall the normal method of Vas testing uses a test mass or box. Well, if you dig into the equations you will find that if you know piston size and efficiency, you can generate the other parameters.

Making a 'guess' like this may not be the best option, but for something like a mid or tweet, it more or less works. More importantly, and the original intent of the effort, is to create an electrical model that works with a crossover design package.

Best regards
Keith

[Zilch]

Yo, KEITH!

How come I consistently get negative Le's on compression drivers?

Am I not calibrated correctly or something?

[WTPRO]

Hi Giskard

More than likely the phase of your compression driver is negative at 1Khz. The issue is that the convention for 'Le' has always been to measure at 1Khz. Hm... in this case Le would be more like Ce! This Le convention never was any good IMO.

The *better* solution is to export an ASCII log file and look at the impedance, phase, resistance, inductance and capacitance columns. From this you can figure out what the actual load will be for a particular test frequency.

An even better solution, especially since I know you have the upgraded SW, and if your crossover design SW can handle it, is to use the frequency dependent Rem and Lem series reactance values computed from the parameters Exm, Kxm, Erm and Krm. This goes in series with the mechanical tank circuit.

1) Rem=Krm*(2*pi*f)^Erm
2) Xem=Kxm*(2*pi*f)^Exm
3) Lem=Kxm*(2*pi*f)^(Exm-1)

An electrical equivalent 'tank circuit' for the Fms peak is also given in a message box at the end of the Q/Fs test or when copying the test data into the simulator. Again, this is a new version 1.02 feature. Keep in mind that at the moment the 'tank info' message box only pops up during the test or during a test value to simulator copy. This is because the Ces, Les, Res tank is really only valid for free air Fms, which as far as I can tell is of no practical value since I suppose most people will want to put their driver in a box!

This is where (IMO) the also new to V1.02 simulator comes in handy. The bottom line is that if you tinker with the box volume and tuning you can get the impedance humps to line up with the measured box test or arbitrary sweep values. If you then adjust Rem, which ends up being the same as box loss, you can also get the peak magnitudes to line up. If you then look at the response, you should be pretty close to reality.

This you cant do right now, but assuming this last 'alignment' has gone well, you may still want to modify the box. Therefor it might make sense to export the *simulated* box Z/P so you can work on a crossover *before* committing to a new box. Usefull? If 'Yes' I might just have to consider adding this.

Oh... I almost forgot... that early version of SW that I made specifically for you might be a bit goofy *if* the impedance does not sufficiently rising at the high sweep frequency. You may recall that the new method I added for you was testing the high frequencies first and then backing up to calculate the lows. This gives Lem Rem before the Fs/Q sweep begins. That in turn becomes important when finding the other points in the graph that would have otherwise failed in the old Q/Fs method. As an example, look at the phase of your compression driver at Fms. It may not even fall below 0' degress, which makes it pretty hard to find a zero phase crossing. By backing out Rem/Xem first, this is brought back to a zero reference and the test completes.

Cheers ,
Keith
www.woofertester.com

[Zilch]

Well, I'm not Giskard, but I do kinda understand your response.

Thank you, Keith....

[We DO play the same tunes, sometimes, tho....]

[WTPRO]

Sorry 'bout that Zilch

Doa! Is this me?->

(If the emoticon fails to show, thankfully Im not yet quite like Homer Simpson)

Thats what happens when I am flipping screens and not using the 'quote' feature. As an FYI, what I was referring to was a special build I did for Giskard where he wanted to create an impedance acurate TS parameter list for mids and tweets. What I came up with was a back annotation for the Vas derived parameters (BL etc...) using efficiency and piston area. From a pure electrical model standpoint this worked pretty darn well if I dont say so myself.

Though this would not be that usefull in the initial Vas calculations (what goes wrong if Sd or No are slightly off?), it *can* be used in successive runs as a fast way of computing a full TS parameter list from only the Q/Fs data. This would for example be usefull as a driver heats up under stiff usage.

If you want to give this new method a try, I can send you a special version like I did for Giskard. The bottom line is that I would like to know if this would be a usefull feature for the next full release.

Back to mids and tweets. Assuming the mid/tweet is a normal 'piston' this should not only work for generating electrical impedance and phase, but quite possibly also produce some usefull acoustic impedance and phase. Maybe you can see where I am going with this with regards to a multiple modeling of a woof, mid and tweet. I suspect however there will be a problem as soon as the driver gets sustantially above Fms, in which case the piston rarely acts like an ideal piston. What happens with a horn is beyond my thinking at the moment... unless I use actual measured in-air phase and magnitudes.

Best regards
Keith

[spkrman57]

Is is necessary to limit the current settings? And if so, to what current limitations should I follow to be safe?

Sorry if this sounds like a stupid question, but I thought I would ask before I might accidentally cause harm to a driver.

Regards, Ron