Shop OBEX P1 Docs P2 Docs Learn Events
Now this is weird - Ideas? — Parallax Forums

Now this is weird - Ideas?

CannibalRoboticsCannibalRobotics Posts: 535
edited 2010-04-28 21:09 in Propeller 1
I'm working with a CrystalFontz display & button panel. Right now, it's winning. It requires 19200 baud, 1 start, 1 stop and no parity. I'm using a Max 3232 for the RS-232 conversion. I can not get the prop-max combination to work reliably. So I compared the working PC-USB generated output to the Prop-Max output.
Taking samples of the same 4 bytes of data from two different sources I see what is shown in the picture attached. Odd thing is that the yellow - the prop/max combination does not work. The blue from the PC-USB dongle does. They are both triggered identically on the scope and the settings on the scope channels are the same.
The only real difference is the p-p which is ~15v on the blue and ~12v on the yellow. In all of my years working with RS-232 I have never seen sensitivity like this.
Is there another source of the problem I am missing here?
Thanks,
Jim-

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Signature space for rent!
Send $1 to CannibalRobotics.com.
ScopeMerge.jpg
341 x 296 - 32K

Comments

  • MagIO2MagIO2 Posts: 2,243
    edited 2010-04-28 18:54
    And you are sure that the display want's RS232 serial and not logic lever serial? I just had a quick glance at one crystalfontz display where it says that the module want's logic level and you can get full swing RS 232 with a daughter board.

    Only asking for the obvious which tends to be forgotten ;o)

    Post Edited (MagIO2) : 4/28/2010 6:59:14 PM GMT
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-04-28 20:09
    Did you try measuring with a load? I believe the nominal RS232 load value is 3K to ground.

    -Phil
  • CannibalRoboticsCannibalRobotics Posts: 535
    edited 2010-04-28 20:40
    @MagIO2: Good question, the board comes in three flavors - Rs-232, 5v serial and I2c. The models I have are the RS-232 or 'KS' variant. The obvious is what I must be missing.
    @PhiPi: The measurements are being taken with the boards connected - I'm sampling them with a breakout in between. I'll try the load resistor. I'm also going to try some blocking caps - maybe it's a sneak current I cant see? I'll let you know.
    Thanks
    J-

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Signature space for rent!
    Send $1 to CannibalRobotics.com.
  • CannibalRoboticsCannibalRobotics Posts: 535
    edited 2010-04-28 21:09
    OK,
    Problem solved.
    It's power supply to the CrystalFontz units. If they are more than about 0.3 volts off of 5vDC - they go flakey.
    I found this just as a random stab in the dark after trying the caps and resistors with no luck. I decided to power the CF unit separately from the prop source and it worked.
    The prop source was at about 5.5 volts.
    When I hooked it back up and nailed it down to 5.0volts - works every time. When they say 5 volts, they mean 5.00 volts.
    Thanks,
    Jim-

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Signature space for rent!
    Send $1 to CannibalRobotics.com.
Sign In or Register to comment.