Transmission Line Modelling Software

[PB2]

Yes, please go back to Re.
If you've added a check box you could have one also for the industry standard voltage sensitivity of 2.83 but all it does is save typing it in.

[schmeet]

Originally Posted by bjorno

Suggestion. Copy this approach:

Hornresp

b

Well eg is the voltage that we are talking about which is derived from sqrt(Power * Resistance). We are discussing what value to take from resistance Re or Znom.

I think I am going to stick with Re as that is what is used by Thiele and Small in their papers.

[NEO Dan]

Originally Posted by PB2

Yes, please go back to Re.
If you've added a check box you could have one also for the industry standard voltage sensitivity of 2.83 but all it does is save typing it in.

Sigh
Why would you want to base power calculations on Re Pete?

[NEO Dan]

Originally Posted by schmeet

Well eg is the voltage that we are talking about which is derived from sqrt(Power * Resistance). We are discussing what value to take from resistance Re or Znom.

I think I am going to stick with Re as that is what is used by Thiele and Small in their papers.

We drive speakers with voltage from constant voltage sources AKA amplifiers. Because the impedance of the loudspeaker varies greatly with frequency loudspeakers are generally rated with a nominal impedance. 2v/4Ω=1w 2.83v/8Ω=1w and so on... This makes comparisons of speakers based on nominal power based voltage sensitivity very straight forward. If you were to see the power dissipated by a loudspeaker in real time you would likely be very surprised. What sort of music do you listen to?

[Keriwena]

Just tried version 2.5.1 and compared it to Augspurger's TLwrx. The results were very, very close, likely within expected differences of scaling and rounding off.

Quibbles:
The scale of some of the parameter inputs is odd. MilliHenries is more common, as is liters for Vas.

The 'always on top' behavior of the auxiliary windows is annoying.

The default amp power should probably be one watt. (since I used a rather efficient speaker to test, with 200 watts the system output was literally 'off the chart'! )


Still, as I say, "quibbles", and you've created an easy to use program with some clever options and abilities that TLwrx lacks. Bravo!

[PB2]

Originally Posted by NEO Dan

Sigh
Why would you want to base power calculations on Re Pete?

Hi Dan, mind if I ask are you Dan Wiggins?

I'm sure you know this but I'll just state it so people know where I'm
coming from:
The whole power vs. voltage issues in speaker analysis and
simulation probably bothered me for about 5 years or so, now I'm used
to it. We either use voltage, or we use a voltage stated as power into
some resistive value, that being Re as defined by T&S. I suggest Re
just to be consistent with the papers and the majority of the other
simulators.

The theory is all based on the idea and most common use
where we drive speakers with voltage sources. I believe that the only
use of the actual calculated power in most of these simulators is to
determine if the driver is thermally or excursion limited and using Re is
obviously a gross approximation but thermal ratings are not generally
accurate anyway and I view T&S analysis as mainly for small signal.

There are a few simulators that calculate thermal characteristics in order
to determine thermal compression curves these must use the actual
impedance versus frequency. This was not part of T&S analysis.

[Silent Screamer]

If you want a hand with resizing the controls in the group boxes when you go to maximise let me know...

Edit:
I am just blind? you don't seem to have a "New Project" button to create a new project, to get started.

[Keriwena]

Would an 'undo' button be possible? Sometimes, when something doesn't work as expected, it would be nice to back up to the numbers I had before.

[PB2]

Pete, wondering about your resistance/stuffing model, is it constant versus frequency or did you build in any frequency dependency?

[Keriwena]

Originally Posted by PB2

Pete, wondering about your resistance/stuffing model, is it constant versus frequency or did you build in any frequency dependency?

Very good question.


More work for you:

The most common use of modeling programs is to try a design, then tweak it. Some way of comparing two (or more) variations at the same time (short of opening another copy) would be a wonderful thing.