Hi, I am thinking about making a 4x8 soundplane.
But I don't yet understand yet how the Max patch works (and what do the externals objects do) and how to tweak that to suit a 4x8 surface. Please help.
Here are some questions regarding building:
1) In your paper you have a table of "Physical Layers of the Sensor". I see aluminum foil on top of polyethylene playing surface, is that a mistake? I guess you would want the polyethylene on top of everything because that's what player touches?
2) And bump layer is above dielectric, is that right? I guess a plywood is hard and it would prevent dielectric layer to be squeezed. Or I misunderstand it? I watched the video of the G-Tar project, and they don't use any bump layer in the middle of carrier and pickup.Also they didn't use any GND strips between the signal strips.
3) Can I use 2 pieces of 3mm thick acrylic boards as the bump and spacer layers?
4) On carrier and pickup layers, are the GND strips connnected all together and into all GND of input/output on the interface? Or I have to have all the GND of each input/output connected to the audio interface separately?
By the way, here is a pic of the materials I plan to use, please check if there's something wrong:
(the black strips are connected to ground;
the parts of the tapes that're sticking out of the base is for soldering wires)
HI Meng Qi,
First, I will give some really general advice: to approach the build incrementally and experiment. The material choices are many, and lots of things you have access to may work. But the geometry (thicknesses) might be changed from what is in the paper. I would build a 1x1 system first, which is a z-only controller. You can do this in Max by sending and receiving a single sine wave. Any frequency from around 5 kHz-50kHz should be OK.
Then with only 2x2 strips you can make a small xyz controller. The math to get the x and y positions is given in the paper, or you can probably figure it out with a little algebra. Again this will not require any custom max objects because it is just 2 frequencies, 2 outputs and 2 inputs. Use either FFT or bandpass filters to compare the signal strengths at each frequency.
Some of the layers in the paper are shielding for the case. Sorry if this is not well explained. The layers in the sensor itself with no case would be:
playing surface (polyethylene for example)
carrier antennas (foil)
elastic (many kinds of rubber will work, but try open-cell foam)
pickups (foil, or you can make a circuit board)
Over a large area, it is hard to make all the layers lie flat mechanically and return to the proper positions. This is why the "bump" was there , to make tension holding things in place. The Soundplane uses a different technique.
Today I would not use GND strips in between the signal strips. A small gap of >2mm is all that is necessary to prevent most interference between strips.
Please try a 1x1 or 2x2 setup to get some feel for your materials. On one of these setups you can also experiment easily with different foams and spacings.
I have not tried the Max objects for years so I apologize if they are out of date. If so, please let me know and I will try to fix them.
I hope to see details of your project!
Thank you very much for the detailed reply.
Now I've found out that magic trackpad is a multitouch surface which is also pressure sensitive. Tho it is much slower @ 125Hz scanning rate, but it is already much better than using keyboard.
I just want a portable solution that I can play aalto in the way I want. As well as messing around with the various possiblities with a multitouch controller. So magic trackpad is pretty enough for now.
The problem for now is this : http://madronalabs.com/topics/2866-aalto-cpu-hog-in-max-4-different-pitchbend-in-1-instance
Very useful to hear about the ground foil; that makes the build much easier! I was going to etch a PCB to do the antennae on the base too, so it's nice to see that thinking confirmed. I've just ordered some 1.5mm EPDM foam to move this thing on. Looking forward to progressing this.
can't you just format the trackpads output to fit the T3D format?
an interesting thing to see would be if you could use multiple trackpads to make one bigger surface.