Posted: Sat Jun 18, 2016 11:07 am Post subject:
electro-trautonium Subject description: building a DIY ribbon controller
This is the prototype of my ribbon controller for an electro-trautonium. Just like the original instrument it includes an envelope generator and a filter. The sound is still a bit rough and not tuned properly yet.
The ribbon controller is completely do-it-yourself. I am using a resistive foil with a length of 29cm and a total resistance of 70K Ohms at the bottom layer.
The spacers are made from 0.4mm thick self-adhesive rigid-PVC foil. The contacts are made by copper foil in the top layer. I am still experimenting with the materials.
Excellent, though I don't remember the life-like sparrow chirps as being in the original instrument, but maybe it works as a historical detail to remind us of a late spring day in 1929.
I look forward to the schematic and especially the random chirp generator.
The second prototype of the ribbon controller is complete. I believe that the crackles in the first one were primarily because of contact issues. As a remedy I sacrificed one of my strips and used it to polish the surface of the already mounted strip. The same goes for the copper foil. Contacts are much better now without dropouts during slides.
The total length of the strip is 40 centimeters, providing two octaves over a length of 30 centimeters. The total resistance of this strip is 117kOhm. I modified the sandwich structure slightly, closing the gap between strip and copper foil a bit to about 0.33 millimeters. It is easier to play than the last one, but still needs a bit more force than I like. Think 'press' - not 'touch'.
I couldn't adjust my linear to log conversion to give me three octaves - no matter what I tried. So I guess I have to be happy with two octaves plus a bit for now. The note range is from C3 minus 2 semitones to C5 plus three semitones with pretty close equal spacing between the tones. Good enough for me. I will focus on the circuit now.
Sorry piedwagtail, no secret random chirps schematic. That was the audience in my garden during the recording
This is the circuit I am using. When a contact is made at the ribbon, U1A produces a gate signal. This creates a pulse through C5 that loads C1 for the envelope, and a pulse through C3 for the filter. U1D and U1C form a VCO that generates a triangle wave. The voltage from the ribbon determines the frequency of the VCO. Rlog makes the voltage/cm response curve slightly logarithmic, so that the positions for the tones on the ribbon controller have equal spacing.
The sound reminds me to a plucked string with a range of overtones at the beginning that disappear as the tone fades slowly out. There is no sample and hold in this simple circuit, which means that the tone immediately stops when you lift your finger from the ribbon.
Oops, just noticed that there is no CMOS logic in the circuit this time - sorry if this is the wrong forum.
Trautonium.JPG
Description:
Filesize:
153.56 KB
Viewed:
616 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
Success! The third prototype works as desired. It now reacts to touch. No need to press hard on the ribbon any more. The trick was to make it wider. Instead of 3cm wide I made it 5cm wide. The copper tap line now has a 1cm gap towards the spacers. Thus less force is needed to make contact with the resistor foil. Vibratos on a tone are also easier to produce.
I haven't change the circuit yet. I tried to add a track&hold, but that didn't work. The circuit can hold the tone fine, but there is no way to open the switch to the hold capacitor before the gate signal goes down. The problem is that the gate signal itself is derived from the ribbon voltage, and if that already went down, it is too late to stop the tracking. Last edited by synaesthesia on Fri Jul 08, 2016 1:19 pm; edited 3 times in total
Posted: Sat Jul 02, 2016 12:53 pm Post subject:
few more details...?
Synaesthesia,
That sounds interestigly enough to make me write some lines, since i follow your experiences in this case for my own interest nearly daily.
Therefore, allow me some further questions:
As far i know, the sensitive strip is about +/-2cm wide, so what do you mean by "wider-up to 5cm"? 2-3 strips in parallel?
And which "1cm gap below/above" would that be?
In 3d-thinking i would assume that you now have to press down the upper foil by a 1cm-distance to achieve contact which doesn't really sound like "sensor-feeling", due all respect....
Or maybe i miss understood your kind of sketchy description, while also yet believing in your success in which i'm urgently interested.
Folks keep telling us that we supposedly live in "information-ages", which is not really proved by describing a technically breaktrough in 2 lines...
Can you help by 2-3 more describing lines or even a photo?
A picture says more than 2-3 lines of text, so I drew a picture. The layers are not at scale, in particular the copper foil and the resistive foil would actually be much thinner than depicted. The original resistive foil strips are 40cm x 1cm in size. They come from heating mats and need to be very carefully removed from the lamination to use them. I split one strip into two halves and used a 30cm by 0,5cm piece in the latest prototype. This increases the resistance/cm almost by a factor of two and in my experience improves the linearity a bit. The copper foil used is standard self adhesive copper foil that is 1cm wide. Again, I split that into 0,5cm wide strips for the contacts.
On top of the base plate I put spacers at the sides and one layer in the middle carrying the resistive foil. I used double-side self-adhesive rigid-PVC foil for that (ASLAN foil). Only for the reason that I can use it to attach the resistive foil to the base plate without glue. The two ends of the resistive strip are connected using thin copper foil. Contact is made only by pressing the conductive side onto the resistive strip and fixing it between the layers. I then ran one copper foil line along the length of the resistive strip to have all three contacts at one side in the end. This is depicted by the smaller copper foil piece in the picture.
The third layer is rigid-PVC as well and covers the whole softpot area like a lid. I placed the copper foil onto the adhesive side of the PVC foil before attaching it. The copper foils leave the whole assembly by a few millimeters and have wires soldered to it.
The latest improvement comes from using a wider base plate. It measures 5x35cm instead of 3x35cm. This puts the outer spacers wider apart and allows to make contact between the top copper layer and the resistive strip with less force. I did not polish the copper nor resistive foil this time, but tried very hard to keep them clean during assembly.
Hope this is a bit clearer now.
FOLIE.JPG
Description:
Filesize:
46.75 KB
Viewed:
11746 Time(s)
Softpot.JPG
Description:
Filesize:
97.22 KB
Viewed:
419 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
More pictures showing all assembly steps. The hatched areas of the copper tape denote the adhesive sides. The copper band is turned 90 degrees and folded onto itself, so that the conductive side touches the resistive strip.
was curious about the resistive material you are using for your ribbon controller. You mentioned it coming out of heating pads? Could you let me know what the product was and where I might find some please?
I am going through the ribbon controller design process myself but only have SVHS and VHSc video tape to play with which both have huge resistances. For a 60 cm controller the SVHS tape gives 1.2M with which I have to use a 820K resistor to improve linearity and the VHSc tape gives 655K for 60 cm and using a 270K for linearity. I am using a current source to drive these. The thread is:
Hi Werebear, I am using resistive carbon stripes that I got from heating mats. Search the web for "INFRARED ELECTRIC FLOOR HEATING FILM". The really difficult part is to extract the carbon stripes from the lamination. First cut out the stripes with a cutting maschine. Cut very slightly into the carbon material when doing that. Then peel off one corner of the lamination for about 1 cm. Now the tricky part: hold one end down firmly and in a single swipe pull the two transparent layers apart. Don't stop while pulling, or you will create gaps in the resistive material. With a bit of exercise you will have about 80% gain.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum