🚀 go-pugleaf

"David L. Jones" <tronnort_@yahoo.com> wrote in message
news:4004fa1c.596911@News.CIS.DFN.DE...
> On Wed, 14 Jan 2004 11:22:20 +1000, "LBNote" <lbonnell@ozemail.com.ua>
> wrote:
>
> Your problem is most likely one of physical construction and wiring.
> Your LED will be pulling significant current each time it switches on
> (2mA?, 20mA? depends on your dropper resistor). If you supply your UHF
> module via the same ground connection as your LED then the resistance
> of your ground connection will induce a small voltage drop, enough to
> upset your UHF module. The longer your ground connection the greater
> the potential problem becomes.
>
> Good practice would be to wire seperate ground connections to your UHF
> module and your PIC/LED circuit, from your power supply. This is
> called a "star ground" connection. Also you would have seperate bypass
> capacitors on each as well (right across the powerpins of each
> device). 100nF and 10uF values are common, but their value is not
> critical. 22uF (not mf unless you are an oldie!) is fine.
>
> Regards
> Dave :)
> ---------------------------
> (remove the "_" from my email address to reply)

Yes indeed, I'm with you on this one Dave.

I had a design using the same TX/RX pair.
The receiver would be 'upset' by an RS485 device transmitting.
What a pain that turned out to be... Seems that with all the gain of
the super-regen RX, the slightest power supply ripple would be
seen on the data output. (probably mostly to do with the slicer really)
Decoupling, decoupling and then a bit more decoupling made things
a lot better but still not perfect. The grounding was to a bottom layer
plane in this case and still the noise got through....

The end soloution was to not use the AM stuff.(crap)
Although, someone did mention that applying an offset DC bias to the
analog output (~1Meg -> Vcc) would shift the slicer decision point
to somewhere less sensitive and also stop the noise chatter on the
digital output. I haven't yet tried this but it sounds like a good idea and
would at least give the uart a break... (pun?)

An observation about the TXers.  I experienced a bit of difficulty with
the start-up time of the oscillator when operating it from 5 volts. An 8V
regulator was later fitted to supply the TX.

HTHs
Greg the Grog