Shop OBEX P1 Docs P2 Docs Learn Events
White Noise - AD Transformer — Parallax Forums

White Noise - AD Transformer

heroldherold Posts: 66
edited 2005-04-29 00:51 in BASIC Stamp
Hello,

I am experimenting with a white noise generator and I have the basic 2 stamp on the education board.

I would like to sample data from the white noise spectrum and send it via serial port. Some bit tests worked fine but I need much faster sampling. I probably need an Analog Digital decoder chip to do this.

Any ideas on that?

·

Comments

  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2005-04-28 14:28
    Hello,

    ·· Are you trying to generate·White Noise with the Stamp module, or sampling it?· What is the PC doing?· Is that where the serial port data is going?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
    csavage@parallax.com
  • heroldherold Posts: 66
    edited 2005-04-28 17:26
    I am generating the white noise with 2 RF transisitors on the bread board. I would like to sample the data and than get it out over serial into my PC. In the PC I have a Realbasic program that will analyzse and read the data.

    So what I need is a fast AD converter chip for on the board that samples the white noise to a max frequency.

    Thank you for thr fast reply

    Thomas
  • allanlane5allanlane5 Posts: 3,815
    edited 2005-04-28 17:34
    What is 'fast' here? 10 Mhz? 2 Khz?

    And what is your goal? To verify if the white noise is indeed 'white'? (Equal amplitude at each frequency, as opposed to 'pink' noise which has equal power at each frequency?)

    To generate a good 'random number'?

    The BS2 could sample a 2 Khz signal, maybe. It definitely can't sample a 10 Mhz signal in a meaningful way -- unless you just want a 'random' number.
  • heroldherold Posts: 66
    edited 2005-04-28 17:46
    2 kHz is not bad, whatever is the max. that the stamp can do. Yes, it will produce random numbers. But I want to be able to split the frequency band into several parts and than anlayze the numbers for each band. This could be done in the PC. So the higher the frequency I can sample the more bandwidth I get for this.
  • allanlane5allanlane5 Posts: 3,815
    edited 2005-04-28 19:03
    There might be programs which would let you feed the 'white noise' signal (with appropriate voltage level conversion) directly into your PC's sound card microphone input, and digitize the signal at some really useful rate like 45 K-samples/second. Check out some of the no-hardware logic analyzer programs, for instance.

    You can run a fast-fourier transform on the PC to find the frequency components of the resulting samples. If you continue to try to use the BS2 -- well, its sample frequency compared to your sound card is way low. And THEN you need to communicate the samples to the PC over a 9600 baud serial link. I think this might lower the usefullness of any additional processing you might do with the PC.

    Just be aware.
  • heroldherold Posts: 66
    edited 2005-04-28 21:07
    Thank you very much for this information. I think you got me on the right track.
  • Vern GranerVern Graner Posts: 337
    edited 2005-04-29 00:51
    If you're looking for a way to analyze the signal with a PC, you might try:

    Free oscilloscope (WINSCOPE) INFO:

    http://polly.phys.msu.su/~zeld/oscill.html

    Same program, different site:

    http://www.geocities.com/nlradiofm/winscope.htm

    And, if FFT is more your need, here's a free FFT frequency analyzer:

    http://www.relisoft.com/freeware/

    Hope this information is useful. [noparse]:)[/noparse]

    Vern
Sign In or Register to comment.