Noise in Lunetta system

[ianbax]

I've been looking at making a noise module which will suit my lunetta's digital nature not the traditional sample and hold random voltages kind.

One constraint is I'm working at 5v single sided supply, so I've gone for digital noise generated by a LFSR - I could roll my own but I've got hold of one of Electric Druid's noise chips which doesn't repeat for 60 years or something crazy.

My plan is to extract random bits from this chip (using a CD4094 shift register with latch I think) which can be made available at patch points then drive triggers or the binary inputs to a 4051 to select random notes say.

But on to my question ! I was looking at filtering the noise to other colours. I'll probably have these available as audio output but I was wondering if there was any value in routing them to the shift register - random gates circuit. I know in traditional noise modules like the Source of Uncertainty, the colour of the source noise is important to the distribution of voltages it produces. Will it make any (noticeable) difference to the gate pattern produced if different coloured noises are used? I'm not sure how to simulate it.

[zaphod betamax]

Personally I use an analog brown noise to gate my CD4517BE SR.
I never got along with the CD4094 SR.

---

313970217_10160988144018923_1025194382687683899_n.jpg
Description:
Filesize:	368.94 KB
Viewed:	109 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[ianbax]

Thanks - I'll take a look. I didn't think it was possible to get transistor noise with just 5v (all the circuits I've seen online are 12v).

Does brown noise make difference to the gating?

Interesting about the 4094 - I was having a play this week and the strobe pin doesn't work as I thought it might. I've introduced some d flip flops (40174) to capture the random 0s and 1s at set intervals.

I might be over complicating things but having fun.

[zaphod betamax]

This circuit works fine all the time at 5V, it might even work at 3.3, but I never tried.
To simplifiy you can use a 10k,10k voltage divider on pin 3 of the op am and it works fine as well.
This triggers shift registers fine as it is biased near the trigger point.

[zaphod betamax]

Here is my take on a random trigger, as we seem to use some of thee same chips.
This one is a 5V 4 channel random trigger

---

325367366_535978838489455_2099049742297576002_n.jpg
Description:
Filesize:	442.48 KB
Viewed:	94 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[zaphod betamax]

I find the LFSR a whole lot of chips that can be done analog with just some caps, resistors a TL431 and a MCP6002.
Analog brown noise has enough random to trigger a CD4517BE.

[Top Top]

My contribution.

Basically just a white noise generator with an amplifier stage to square off and output random pulses. If it is not outputting strong enough signal from one CD4069 stage as shown, it can be followed by additional identical stages to further square the signal up. The necessity of additional stages may depend on how noisy your transistors are.

Following it with a divider, flip flop, shift register, etc, you can bring it down into lower frequencies or "sampled and held" results for use as a random logic source for various things.

I have used it as a clock source into a CD4017 to output random results. High logic into the "enable" pin on the CD4017 (or using a switch to connect/disconnect to the CD4017 clock pin) will freeze it on one of the output pins, basically acting like rolling "electronic dice." Setting the reset on the CD4017 from one of the counter output pins, you can set how many possible output states you would like.

---

NOISE_CLOCK_AMP.png
Description:
Filesize:	63.3 KB
Viewed:	88 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[zaphod betamax]

May I suggest the CD4517BE rather than the CD4094.
In my 3 years of using these as S&H they sound more musical
than the ramdomality of the CD4094 in my opinion.

My 4094s sit unused whilst I have a dozen 4517s standing by!