Connect with us

PC Oscilloscope ... getting better

Discussion in 'Electronic Design' started by Francesco Poderico, Mar 31, 2013.

Scroll to continue with content
  1. Hi all,
    a few months ago I started to design and build my own oscilloscope...and in this groups I had a lot of help in solving some problem...

    I'm getting so excited now... and surprise at same time... how things are getting better and better.
    The oscilloscope is now very useful indeed, I have added the capability to see the spectrum, and it works really well.

    I still have a long way to go... but it looks promising. please see the picture on my blog.

    I'm starting thinking at the next prototype now...
    shall I go for a 200 MHz + 64 KRAM ( which I can put on the market for less that 200 pounds cheap oscilloscope but does what it says...) or shall I prototype a 1GSPS? and have a larger margin?
    Any suggestions is welcome.

  2. brent

    brent Guest

    You are doing great work!
  3. Hey Jan,

    Looks like you need to get a bigger PIC! :)
  4. Thanks,
    now works at 200 MSPS! :)
  5. Yes, yes, nice... but Francesco is already doing 100MSPS and making good
    progress towards more it looks like.
    I remember doing that, sort of, when I was a teenager with my first
    scope. Not very good though, did not know how to do the syncs properly,
    had to just adjust the frequency and hope it locked somehow.

    Was still cool to see!
    The only thing comparable to your scope recently was a lashed-up 120MHz
    STM32F205 microcontroller running a 2MHz 12 bit external ADC.

    During testing I captured the sample buffer with gdb, dumped it through
    a little python script I found that fed it into gnuplot. Ended up with
    some really nice, true 12 bit high-resolution plots of various
    waveforms. Could make a beautiful low frequency scope like that, maybe
    with a really high resolution screen like the high end ultrabooks
    use. Job for my pi perhaps?
  6. rickman

    rickman Guest

    You seem to be doing good work. Is your spec of 5 mV/div really valid?
    That would be pretty nice. Most of the low cost scopes don't do so
    well on sensitivity.
  7. josephkk

    josephkk Guest

    Maybe 500Ms/s and 1 Ms ram. Really long record lengths have some specific

  8. rickman

    rickman Guest

    Can you even buy 64 kword RAMs these days? A 32 MB SDRAM chip will do
    all that is needed I believe. Either run it at 100 MHz with twice the
    width of the storage needed or I expect it won't be hard to find SDRAM
    that can be clocked at 200 MHz. This wouldn't add much to the cost of
    the device, but would greatly improve functionality.
  9. josephkk

    josephkk Guest

    The 64k is a ram block inside the FPGA. Completely different situation.

  10. rickman

    rickman Guest

    Of course... I'm used to using the poor stepchild end of the FPGA
    families with only a few kB of RAM. Still, I think a RAM chip should be
    on the board. It seems silly to hobble what seems to be a good design.
    I know the extra storage would help the work I do.

    I think I've offered my help before, but I'll make the same offer. I
    have some experience with memory interfaces.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day