Random digital potentiometer driving

[bigkim100]

Can anyone think up a way, using analogue electronics to randomly change the value of a digital potentiometer, and stop that value for a predetermined time. once that time is off it changes to another random value, and so on and so on...
I was thinking of using the output of a 4017, driven by a 557 type of thing, but cant wrap my head around it.
Can anyone help???
Thanks in advance for any help.
Kim

 

[kellys_eye]

using analogue electronics

I was thinking of using the output of a 4017,

errr...... so, what's it to be? Digital or analog?

 

[bigkim100]

I just meant no microcontrollers

 

[bertus]

Hello,

The 557 must be a typo.
I know the 555 Single timer, 556 dual timer, 565 PLL and 567 tone decoder.

What digital potmeter do you intend to use?
How is the pot controlled?
I know versions that use SPI, that must use a microcontroller to set it.
I also know versions that has up/down and clock pins.

Bertus

 

[CircutScoper]

I also know versions that has up/down and clock pins.

Bertus

How about a 16bit feedback shift register generating a 65535 bit long pseudorandom input to the up/down pin of the pot, the pot clocked 180 out of phase with the shift register clock, and some means to pause the process after every 256 or so cycles?

https://upload.wikimedia.org/wikipe...G16.svg/786px-LFSR-G16.svg.png?20170616112515

 

[bertus]

Hello,

@CircutScoper , Direct linking to the wiki does not work.
The forum shows broken images.

Here is the complete article from the wiki:
https://en.wikipedia.org/wiki/Linear-feedback_shift_register

Bertus

 

[Harald Kapp]

• What's the purpose?
• How many different settings would you need?
• Does it have to be a digital potentiometer or would another circuit be acceptable?
• How "random" does it have to be? Using counters ad stuff like that it is hard to generate a truly random signal (one would have to somehow include a source of truly random noise). But is is comparatively easy to generate so called pseudo random sequences. Depending on the length of the sequences these will repeat sooner or later.

 

[kellys_eye]

An R-2-R ladder used as the 'pot' and clock-driven by a psuedo-random gated clock signal would work.

 

[Harald Kapp]

An R-2-R ladder used as the 'pot'

Excellent. What I had in mind. That's why I asked about the number of different settings.

 

[bigkim100]

Hello,

The 557 must be a typo.
I know the 555 Single timer, 556 dual timer, 565 PLL and 567 tone decoder.

What digital potmeter do you intend to use?
How is the pot controlled?
I know versions that use SPI, that must use a microcontroller to set it.
I also know versions that has up/down and clock pins.

Bertus

 

[bigkim100]

-correct, its a typo
-I intend to use the pot that is suggested, except to the resistance required by me
-The pot would be controlled by the circuit that would be suggested by the reader
-Maybe the one that has up down pins would be easiest??
Thank you

 

[bigkim100]

How about a 16bit feedback shift register generating a 65535 bit long pseudorandom input to the up/down pin of the pot, the pot clocked 180 out of phase with the shift register clock, and some means to pause the process after every 256 or so cycles?

https://upload.wikimedia.org/wikipe...G16.svg/786px-LFSR-G16.svg.png?20170616112515

-EGADS!!!..I dont even know what that means, but thank you.
Its well over my head

 

[bigkim100]

How about a 16bit feedback shift register generating a 65535 bit long pseudorandom input to the up/down pin of the pot, the pot clocked 180 out of phase with the shift register clock, and some means to pause the process after every 256 or so cycles?

https://upload.wikimedia.org/wikipe...G16.svg/786px-LFSR-G16.svg.png?20170616112515

sounds good...now how is it done??

 

[bigkim100]

An R-2-R ladder used as the 'pot' and clock-driven by a psuedo-random gated clock signal would work.

-That sounds best, and simple...any ideas on how its done??

 

[kellys_eye]

Use a 74HC595 to drive the R2R ladder and clock it up/down through a NAND being driven by a slow pseudo random clock and a fast version (fast being the clock, slow being the 'stop' - both being 'random' makes for an effective result). You could even use another slow P-R clock on the 595's up/down pin.

Google "R2R using shift register" to ge the basics of R2R operation and the datasheet for the 595 to see how to implement it. P-R signals are relatively easy to derive using EXOR/EXNOR gates - again, easily found using Goog.

 

[bigkim100]

-I have been thinking of two 555s driving two 4017s, each of the 4017s outputs having a resistor.
-When each of the two 4017s stops, a relatively random resistance is generated from the 2 resistances totalled together.
-Relatively low tolerances on all the components may add to the overall random nature of the project (??)

Whatcha think???...I think its my simplified version of what Kellys Eye has suggested

 

[crutschow]

stop that value for a predetermined time

Is that time to be fixed or random also?

 

[bigkim100]

Is that time to be fixed or random also?

Fixed-thanks

 

[kellys_eye]

The R2R/LS595 version gives 128 distinct steps. Not sure what your requirement is.

 

[crutschow]

What is the type of input and number of bits for the digital pot?