Author |
Message |
aerogramma
Joined: Feb 27, 2008 Posts: 156 Location: Roma, Italy - London, UK
Audio files: 13
|
Posted: Thu Jun 25, 2009 2:53 am Post subject:
Can you do white/pink noise with a CMOS/lunetta? Subject description: to the noise lords out there! |
|
|
hello,
i was asked by a friend to fill an exhibition space with sound and i thought it'd be a good idea to fill it with noise
i could use 1000's different ways to do it but i really like the idea to do it in real time with a 9v powered lunetta circuit
i had a look around but haven't found anything that would use a CMOS as the heart of the noise producing device
and ideally i should be able to modulate the noise with some sort of lfo etc..
any suggestion?
thanks a lot!
aero
|
|
Back to top
|
|
|
Uncle Krunkus
Moderator
Joined: Jul 11, 2005 Posts: 4761 Location: Sydney, Australia
Audio files: 52
G2 patch files: 1
|
Posted: Thu Jun 25, 2009 5:10 am Post subject:
|
|
|
I might be wrong, but I think digital noise generating circuitry is always going to result in noise which has an inordinately large amount of high frequencies, coupled with very specific lower frequency bands. It's a type of noise to be sure, but nowhere near white noise.
For real white noise, the best easy option is something which "listens" to a reverse biased PN junction. In other words,..... a noisy transistor through an amp. _________________ What makes a space ours, is what we put there, and what we do there. |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sun Jul 05, 2009 9:12 am Post subject:
|
|
|
I've done a fair amout of playing with digital LFSR noise generators. I've found that the more stages one uses, the more white it sounds. Filtering can then be used to produce what sounds pink. I would use no less than 32 flipflops in the shift register. One thing I tried was to clock the LFSR from a VCO - this changes the band in which the noise exists, the higher the clock rate, the higher in frequency the band of noise goes.
Unfortunately, I've not found much on the spectral characteristics of LFSR noise - however, it sounds "white" to my ears.
I think that it's worth a try to throw a circuit together and see if the character of the noise is what you need.
EDIT ADD: I did better Google searches and found several sites that said an LFSR can be used to generate an approximation of white noise. My guess is that the more stages and the faster you run the clock, the better quality (i.e. closer to real white) you will get. I've found this to be quite true while experimenting with one of my FPGA-based synthesizers in which I use a 64 bit shift register. I found that fairly simple digital filters can be used to change the spectrum character quite radically.
The main advantage to this kind of noise generator is the consistency from built unit to built unit and output voltage is high, so an amplifier is not needed as is the case with diode noise generators. There is no "pick the best component" stuff to horse with either. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Fri Aug 14, 2009 11:50 am Post subject:
|
|
|
Please call me daft if I'm missing something, but - couldn't u use the (cough cough) 'standard' 4069 amplifier setup (or 2 or more in series) to amplify the reversed PN junction? That'd be no more'n a tranny and a single chip op'ing off of a 9V battery.
Tranny - 100k - 4069 with 1M feedback - 100k - 4068 with 1M feedback - nonpolar cap - output.
?? Actually - I could bread board this and hear how it sounds.
Edit: (A couple of hours later) I just found this circuit here. It's using the 74ALS04 (Low Power Schottkey Inverter) as the amplifiers, as well as the 2nd transistor. Guess I'll have to try out my own version of it, with a CD4069 and see how it goes. From there, other uses can be determined. |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Fri Aug 14, 2009 4:18 pm Post subject:
|
|
|
The 4069UB used in linear amplifier mode soft clips, that is, there is a moderate size linear gain region centered at (Vdd-Vss). Above and below that, gain fades out. I'm sure this would color the noise you'd generate unless you limit the amplifier's input to stay within the linear region. I've used this fact to make a simple tri to sine waveshaper.
A 4007 might be interesting to try as an amplifier, it's got several MOS transistors that you could use separately. (I like to think of the 4007 as an MOS transistor array...). _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Fri Aug 14, 2009 4:34 pm Post subject:
|
|
|
ScottG wrote: | The 4069UB used in linear amplifier mode soft clips, that is, there is a moderate size linear gain region centered at (Vdd-Vss). Above and below that, gain fades out. I'm sure this would color the noise you'd generate unless you limit the amplifier's input to stay within the linear region. I've used this fact to make a simple tri to sine waveshaper.
A 4007 might be interesting to try as an amplifier, it's got several MOS transistors that you could use separately. (I like to think of the 4007 as an MOS transistor array...). |
Ooooooh! Now, that's for that tip Scott! Would you've any URL's with examples of suchness around? If not, I could go and search for them myself for examples. If I'm recalling correct, I have 1 or 2 x 4007's that I'd bought for some other purpose which I didn't produce and ...... can't remember. I could use one of them for this.
Also, thank you for cluing me in on the 4069 I'd much rather have as much of the tranny's audio range as possible, for later processing. |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Fri Aug 14, 2009 5:14 pm Post subject:
|
|
|
I don't have examples of using a single 4007 transistor as an amplifier, but since it's just an MOS transistor, then some sort of generic schematic ought to be around (perhaps in some MOS transistor datasheet).
If you look at the data sheet for the 4007, you can see that the only gate dedicated to being an "inverter" is the far right one (pins 9,11 & 12) and even here, at least one of the transistors could be used as a "loner". The other two offer the use of each of the P and N MOS transistors as loners. One should be able to use at least 3 separate transistors in the 4007 package (the gates are tied together for each complementary pair). One must be careful, however, to keep in mind the fact that there are parasitic diodes. As such you may or may not be able to use both P and N MOS transistors in a given inverter location at the same time depending on the exact application. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
|