Yet another Planar Magnetic Line Source, the SMAPPP

Yes, one should always ask one more question:

  • The mass per unit area of 3M 56 (~73.5 g/m²) is approximately 3 times higher than that of 3M 74 (~24.9 g/m²).
This significant increase in mass will result in:

  • Lower resonance frequency of the diaphragm, thus reducing efficiency at higher frequencies.
  • A general roll-off in frequency response at higher frequencies, potentially reducing detail, clarity, and sensitivity in the treble region.

Consideration of Trade-offs:​

However, the higher stiffness and lower distortion of 3M 56 (due to greater film thickness and stronger adhesive bond) might partially counteract the negative mass impact, slightly preserving some high-frequency performance through improved linearity at moderate amplitudes.

  • At lower amplitudes, increased stiffness can help preserve transient response.
  • At higher amplitudes, the increased mass will dominate, causing more noticeable loss in treble response and reduced transient detail.

Practical Recommendation:​

  • If preserving maximum high-frequency extension and transient response is the main priority, 3M 74 would be advantageous due to its significantly lower mass.
  • If improved rigidity and lower distortion at moderate amplitude are more critical—and some high-frequency trade-off is acceptable—3M 56 can still offer benefits.

In summary:
You are correct: The significantly higher mass of the 3M 56 tape generally will reduce the diaphragm’s high-frequency response. This represents a fundamental acoustic trade-off between mass, stiffness, and distortion.

I still gotta test it.
 
  • Like
Reactions: esl 63
then again above 10khz there is less and less content. so not sure if that extra 10db is a problem. i did notice foils with an aluminium tape can handle quite some power because the adhesive kind of isolates the alu from the mylar. the laminates i have here really have a problem with power handling because the glue layer is different and very thin. if you sweep them at high volume you can see the foil heatup at high frequencies and see it is losing tension haha
then go back to normal tension after it cools 🙂
 
I'm really not worried about the heat. The adhesive, a thermosetting rubber, and the polyester are designed for continuous operation not exceeding 130°C.
And I guess, that an EQ of 10dB isn't really that much to worry about as you say.
But, as the EQ is in the digital realm, I cannot boost the HF levels; I must lower the frequencies below by 10dB.
Hopefully I have enough headroom for it in the various components so the SPL will have a decent level in the end.
 
Mid membrane 14 µm aluminium traces with 56 film (red) and 74 film (purple):
Mid 56 or 74 film SPL.jpg

Not much difference in SPL.

THD:
Mid 56 or 74 film Dist THD.jpg

Not much difference here either.

But 3HD shows some difference:
Mid 56 or 74 film Dist 3HD.jpg

So the 56 film for the mid membrane is the winner.

For the tweeter I used old 6 µm aluminium traces 74 film as reference.
New measurements were made with 14 µm aluminium traces.
SPL:
Tweeter 56 or 74 film SPL.jpg

The 6µm membrane is still the winner.

THD Distortion:
Tweeter 56 or 74 film dist.jpg

More or less the same.

3HD:
Tweeter 56 and 74 film 3HD.jpg

The 56 film has much lower 3HD below 1.3 kHz. Too bad it is outside the tweeter's intended range.
 
The higher tempertures affected the PET film, it wiggled a bit, so I've thrown them away as they were useless.
Perhaps 120°C is too much on the safe side; I guess that a rabbit hole is coming up (down?)!

Was the only problem when baking at 140 degrees that the traces came loose or was there a negative impact on the mylar also? Since my adhesive should handle higher temperatures I'm thinking if I should go with 140 degrees for the baking.
 
Well, 140°C and 10 minutes was too much.
Two adjacent traces that should be 0.5mm apart weren't after baking, they were actually shorted a couple of cm.
The PET hadn't shrunk; it still the correct width.
But it seems that one trace had slipped down.
IMG_8698.JPG


I guess baking with the steel cylinder lying down is better, letting gravity work on the flat membrane.
And also lower the temperature to 130°C.
 
Last edited:
I'm having great trouble making the 2.5m long membranes. The root cause is that the Silhouette Cameo 5 isn't as accurate as I thought.
This is due to the fact that between each 110mm segment, the machine does a re-homing of the tool. That is in the X direction.
This re-homing causes sometimes the cut traces to be off by a fraction of a millimeter but enough to maka weeding difficult, especially for the 0.5mm bare parts.
The reason for having the shorter segments is that otherwise the cuts will be along the whole membrane to and fro and thus gets misaligned in the Y direction.

I can:
Buy a better cutter (that will cost a lot),
Build my own cutter (that will take a long long time),
Make many membranes and select the best ones (that will take some time),
Let someone else make the membranes to my specification (that will be costly and take some time)
Change the membrane layout (that might affect the performance).

Regarding the latter, I can at least for the tweeter membrane change the layout so that the traces are wider than the perfect 3mm.
This might work due to the fact that at the intented frequency range for the tweeter, it doesn't move that much.
At 3kHz, 1/100 movement is required of the movement at 300Hz to get the same Xmax, so it is worth a try:

Middle trace is 8mm instead of 3mm:
MWSnap909 2025-04-04, 09_19_30.jpg

With 6µm aluminium it will still be around 6Ω.

For the mid membrane I haven't tested the 56 film without the fillers, so that is worth a try as well.