pseudo random bit generator not working

[Estebandito]

Hi all,
I lifted the pseudo-random bit generator from the CMOS cookbook, breadboarded it and it worked. Had a pcb made, and now something’s wrong. I checked all connections, replaced every chip, and still, this happens: the circuit only runs when I touch or nearly touch the top half of the first IC in the circuit.

Working with cmos chips and Lunetta circuits is often so straightforward it makes me forget my basic theoretical knowledge of electronics is really rather poor. So when something like this happens, I am at a loss. Can someone point me in the right direction?

thanks!

---

IMG_3313.mov
Description:		Download (listen)
Filename:	IMG_3313.mov
Filesize:	112.34 MB
Downloaded:	168 Time(s)

---

[PHOBoS]

I'd need a circuit diagram to have any idea what's going on but by the looks of it you have 1 or more pins floating.
Likely it's the chip in the top left but it could also be a pcb trace going to another chip that's picking up the noise.

[blue hell]

Hmm .. it is touchy eh ... do you maybe have any unconnected (as in: hanging loose; not connected to anything) inputs?

[Estebandito]

I used this circuit. Instead of 4013s I used 4015s, which was suggested as an option in the text. This means the outputs bar two on each 4015 are floating, but there aren’t any floating inputs.

---

IMG_3318.jpeg
Description:
Filesize:	78.28 KB
Viewed:	118 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[PHOBoS]

what are pin 6,14 of the 4015 connected to ? and how is the clk input connected ?

[Estebandito]

This is how I hooked up the circuit

---

Screenshot 2023-06-22 at 08.53.40.png
Description:
Filesize:	184.54 KB
Viewed:	146 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[Estebandito]

The clock is a 555 timer oscillator

[PHOBoS]

so far so good, though 1M (SR2) could be too high to keep the data input stable.
I am actually wondering what the reason is behind having is connected through a capacitor, usually it's just connected directly.
I guess it might have something to do with that "optional starting circuit". You could try bridging that cap to see if that helps.
In the video it looks more like an issue with the CLK though, unless it's running at a high frequency.

Also, does IC4 have any unused XOR gates and if so did you ground the inputs of those as well ?

[Estebandito]

The clock is running at a high frequency (101/102 cap and 1m pot). As for the 1uf cap 1m resistor circuit:

---

IMG_3408.png
Description:
Filesize:	785.57 KB
Viewed:	102 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[Estebandito]

The other XOR inputs aren’t floating either, they are used in another circuit:

---

Screenshot 2023-06-24 at 08.43.26.png
Description:
Filesize:	308.81 KB
Viewed:	123 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[Estebandito]

The second circuit is this one: https://makezine.com/projects/really-really-random-number-generator/

---

IMG_3410.jpeg
Description:
Filesize:	61.23 KB
Viewed:	116 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[zaphod betamax]

Did you check for cold solder joints?
You said it would work if you touched the chip.

[noto]

The shift register needs to be "seeded" to start. I noticed the capacitor polarity is reversed in your drawing. You might also want to buffer the output depending on its load

[zaphod betamax]

I designed and breadboard this and it works well.

---

rough noise generator for evaluation Model (6).pdf
Description:		Download (listen)
Filename:	rough noise generator for evaluation Model (6).pdf
Filesize:	18.84 KB
Downloaded:	96 Time(s)

---