ldr Audio ITC with Light-Dependent-Resistors

[Andres Ramos]

I started a new series with experiments with LDR's (Light dependent resistors.

These are semiconductor components that are changing their impedance if exposed to light. I had the vague feeling they might produce usable noise for ITC under certain conditions. My investigations are not finished yet and currently I am unsure if this is yet-another-noise-source (YANS) or if there is more in it. Results are ambigous so far.

In my current setup I use a yellow LED with a low current of 0,5mA that projects it's light onto an LDR. Both are sealed in a black plastic tube to protect the LDR against ambient light. The voltage across the LDR is pre-amplified and recorded. It is pink noise so far.

I combined the noise output with Michael Lees ML application. From a sequence of roughly 4 Minutes I could distill 45 audio samples which are pretty goos. Most of them in german but also some in english.

Andrés: Andrés.mp3

This audio was nice. I got a "hangup" in the ML algorithm that cause a series of phonem repititions. Directly after that incident I got this message. "Es hakt" -> "It hangs":Es hakt.mp3

"Massen Krieg" -> "Masses war": Massen Krieg.mp3

"She's a christ": Shes a Christ.mp3

"Sie war der Anschiss" -> "She was the scold": Sie war der Anschiss.mp3

"Streber" -> "Nerd":Streber.mp3

"Our answers are keen": Our answers are keen.mp3

"Er verwandelt sich" -> "He changes": Er verwandelt sich.mp3

When my experiments are finished I will put a detailed documentation into the experiment section. Moreover I am planning to write a study paper about using an LDR for ITC provided the results are encouraging.

 

[Andres Ramos]

Hi everyone,

I made some progress with the LDR-ITC receiver. This is how it looks like now.

The LDR is contained in an aluminium cube that serves as a base and faraday cage. I use a rock crystal as an diffuser.

I packed my proceedings in the format of a small podcast.

LDR-ITC Proceedings 2021-03-25.mp3

[Andres Ramos]

Thanks Jeff! You're ideas are quite right. In fact I've been doing  what you proposed in my first experiments. I glued the LDR together with an LED to the ends of a small black plastic tube.

The original idea was to measure the noise level of the LDR depending on the LED current. Maybe it turns out that this is the ideal setup especially for optical feedback. The only reason I'm currently using an open loop for feedback is that the experimenter can add influence in the kind of controlled interference like CFLs or smartphone displays. We'll see what comes out but it's interesting you had the same idea that I had. I think this is community spirit!

The pulsating LED is something mandatory to investigate in my eyes. I don't know if you had the same experience, it's somehow more a gut feeling but I noticed that the best voices came through when something "changes" in the receiving system or environment. This also matches my observations that most voices come through directly after starting the recording session. Maybe it's that "change" that punches a temporary hole into the interdimensional borders.

[Andres Ramos]

Thats amazing! Especially the "There build aural optical" is very clear. Keith said they often talk about photons as carriers of information.

In the meantime I did the LED pulse experiment. This is how a cycle looks like.

The first third is the sugnal in the ON phase of the LED. It's a swing in dominated by a sharp hissing sound.
On phase LED:On-Phase.mp3

The second big impulse indicates the OFF phase of the LED. The LDR impedance is running up it's curve here. It shows a short period dominated by spikes. In this sequence you can hear voices after paulstretching. However in the following sample is not processed. Thus you can hear just the spikes. Then you can see that the signal level is rising again linearity. This is not the LDR but the automatic gain control I implemented.

OFF phase LED: Off-Phase.mp3

Complete cycle:Pulse cycle.mp3

Off-Phase.mp3 Pulse cycle.mp3

So basically our thoughts were right. However the results are not very impressive so far.
I now will let the device run by night. Results generally sre better then

 

 

[Andres Ramos]

Just took the time to listen. There are really more voices in it that I thought it would. As always I hear the voices a little different. I hear "Für Andrés da" -> "Here for Andrés". The second clip in the sample is thrilling. Sounds like a very low bumping voice overlayed by screaming voices.