Simple as pie.

Use a small hammer or whack the cup with the handle of a screwdriver.

It shouldn't take much effort and just a few taps around the perimeter

should remove it. Mine took about 3 taps.

You can use the plastic shipping cover to help in the process of altering it to not damage the dome

while handling. I used that, and faced it into a roll of masking tape that will also be required.

Take 1 or 2 of the felt rings in the cup and keep that ready per tweeter.

Carefully insert a cotton ball into the pole-vent. Mine was placed where you didn't see the

cotton above the felt ring under the dome, but pretty much even in height. It should fill the entire

pole vent and be close to flush with the backplate.

Weigh out 1/8 oz of Ultratouch (or Meniscus Econocore), and fashion it to fit the chamber. It should

dome up a little when sitting in it. If it is not domed up, it won't sandwich right inside when assembled.

Place a felt ring from the cup over the 6 holes on the backplate, and the UT should keep the ring in place as you

snug the chamber back on the magnet and wrap with masking tape to secure it temporarily. Do not glue the ring in place.

I personally would measure with masking tape around the chamber on the magnet to make sure the effect

is as desired before using any adhesive. FWIW, I used E6000 adhesive on mine.

Remove masking tape if desired effect is achieved, and then use an adhesive around the groove the chamber

attached to over the backplate, resnug the chamber, and retape it to hold until cured. I used the masking tape roll

as a stand to face the tweeter down and keep the adhesive from moving around inside while the glue dried.

Retest the tweeter. You'll notice some variance in the results from before sealing it up, but they are usually minor and better than before.

Mine both measured out to 497 Hz Fs after this mod, with the energy storage bumps being reduced in frequency and magnitude, as well as the Fs

being less prominent. One tweeter had a dual Fs bump before the mod, and it was all but eliminated afterwards.

I will say It does VERY well xover'd about 2kHz, and it's a little different in terms of response. I used a parallel notch

(placed in series) to fix the bump in the high treble, along with a fairly easy 12dB electrical filter. My placement on

the baffle was also used to provide a peak in the response where a dip formerly

resided to minimize response variations. It seemed to work very well.

I won't say it'll xover at 1.5kHz, but I don't know how low it'll likely go without issues as it is only a $30 tweeter.