# 4017 Counter LED Shuffle By Touch

Hi all,

I found the following circuit on the internet and made it on the back of a business card as a fun project. The idea is that you ask a question, touch the naked wire, the lights shuffle and you end up with an answer highlighted by the one of the LED's. A poor mans magic 8 ball.

I've just got a couple of questions concerning the circuit:

• What exactly is happening at pin 14? I know that pin 14 cycles through the output pins when the power changes high to low or vice versa, but how is touching the wire achieving this?
• Why does the circuit only work when connected to actual ground or, oddly, when it's not connected to actual ground but near something electrical?
• Finally, is there a way to get this circuit to work without connecting to ground? Or would I need to up the complexity and connect some sort of oscillator like a 555 timer and capacitors?

Thanks for the help.

• A link to the circuit would be best. Jun 28, 2016 at 1:48
• The circuit comes from this guys website. Here's a video of him making it link Jun 28, 2016 at 1:57
• Deleted my answer as obviously it's incorrect. Not sure why you need to add an external ground for pin 14's floating pin to work that way. Interesting Jun 28, 2016 at 2:23

The 4017 IC is a decade counter - each clock cycle causes it to shift the positive level to the next output. Pin 14 is the clock input.

Fingers are complex and multifunctional. Like any other part of the body, a finger can serve as a weak signal source. I slowed the video and counted between 10 and 20 switches in a second (it is hard to tell if certain diodes are lighting up or merely reflecting others on at the same time). This is faster than I would expect from the 4-5 rising slopes found in 70-beat-per-minute electrocardiogram, so other noise is probably responsible. There are dozens of other distinct signals and noise sources in the 1's and 10's of Hz. As the blinking shows, the "clock" signal is irregular, which is how it can produce a pseudo-random signal despite operating at near-human speeds where careful timing might otherwise allow for manipulation.

The body is a weak signal source, but the circuit will float with it if nothing (ground or some sort of induced voltage from "something electrical") holds it back. This would prevent the voltage between the clock and ground pins from crossing the threshold at which a rising edge would be detected.

If you have ever touched the input to an amplifier or a piece of test gear like an oscilloscope or even a DMM, you will note that your body is a pretty good receiving antenna for whatever kind of RFI (Radio-Frequency Interference) and other types of radiated garbage is just floating around in the air in most civilized parts of the world.

The suggestion about "near something electrical" is a strong clue that the circuit is working by picking up the "hum" from nearby power mains (60Hz in North America, and 50Hz in most of the rest of the world.)

Just in case you want to 'up the complexity' here's an alternative clock circuit.

Pin 4 is the reset input which is normally connected to +V, this is pulled to ground by R3. R1,C2 form an astable circuit with the 555. R2 prevents (pitting) damage to SW1 by limiting the charging current for C1. C1 and R3 keep the astable going for a short period after the switch has been released making it difficult to judge exactly when the final count will happen.