Alright, I'm throwing in the towel and asking ya'll.
What is the deal with planar tweeters. I know the supposed to be superior because they can go higher but no matter what I do in Xsim they always behave badly. I usually can run one large woofer, a mid to fill in the mid range, and a tweeter up top to get a fully flat range. Every time I try to design something with a planar tweeter the high end goes nuts.
Am I doing something wrong?
Am I doing something wrong?
How is your baffle set up in the model? Reference axis, driver sizes, driver locations?
Are you using data you measured or someone else's files?
What does the planar frequency response look like on its own, with the microphone lined up with that driver in the model?
Are you using data you measured or someone else's files?
What does the planar frequency response look like on its own, with the microphone lined up with that driver in the model?
To make this all as simple as possible I am using all Dayton drivers with their respective FRDs from the site. One 6" reference woofer, one 4" full range, and their dayton PT2 planar tweeter.
Is S1 the tweeter? If so, why is there an inductor in series with it? That might explain the high frequency rolloff in the sim...
To highpass the tweeter you should use (at least) a cap in series with the driver.
To highpass the tweeter you should use (at least) a cap in series with the driver.
I don't know what simulation you say you are showing here, but the exclamation mark on each speaker shows that the specific files are not loaded and a simulation was made with generic files.
You have no crossover, so the midrange is canceling the tweeter output and causing that comb filter effect. 1) Load the .zma files in for each driver. 2) drag a HighPass2ndOrderQ from the CircuitBlocks menu for the tweeter. 3) Drag a Lowpass and Highpass 2nd OrderQ circuit block filter for the midrange. Once you have crossover filters in there so the drivers don't overlap, you will stop seeing the cancelations. Look on youTube for videos about using Xsim. A typical 3-way crossover looks like this one I developed for the Magnepan 3.7i. All ribbons and no problems. Unlike the stock crossover the response is now very flat, and the impedance stays at about 4 ohms or higher. Oh, by the way. You are a superior human being, just learning to use Xsim. You will figure it out and soon will be designing and building amazing speakers.
Last edited:
There doesn't appear to be crossover filter components for each driver. The summed response of the 3 drivers together will be weird if there is no filter network for each driver.