On Sun, 25 Aug 2002, Jim wrote:

>   "If one takes a 12C508A  f.ex. and uses internal
>    RC then what clock frequency does it run on for
>    FCC purposes ?"
>
>Gee, I can see major spurs (spurious) being created
>at the 'clock' frequency (what ever f the proc core is
>clocked at), determined by whatever the RC combo works
>that out to be, and if you *fail* to think that just

Of course one has proper decoupling. I was unable to measure noise at the
suspected clock frequency with a scope. Maybe there is enough signal to
look with a spectrum analyzer. I was looking at the Vcc node with
capacitive coupling on the lowest scale my scope has (5mV/div) with 1:1
probe. Decoupling was 10uF tantalum + 0.1uF chip ceramic on very short
leads. I tried to put a choke in series with the supply with the same
result. Pretty good imho. The PIC was running a simple program that
outputs 50kHz signals to drive a pair of mosfets. These were not connected
when I did the measurement. Tiny 50kHz 'blips' were present corresponding
to the flanks of the 50kHz signals but no trace of 4MHz or 1MHz could be
seen.
>
>- the PIC with everything on-board has GOT to be the one
>of the cleanest uPs on the market. I have a Z80 design
>that spews garbage EVERYWHERE - but it's not marketed
>or sold so I can get away with it. It creates noticable
>'spurs' (carriers) throughout the HF spectrum and even
>causes noticable herring bone patters on VHF-Low TV
>channels ... the Z80 design uses every peripheral IC in
>the Z80 family plus RAM and EPROM IC. The board design
>is only two-sided  w/no Ground plane and the traces
>are run nearly everywhere!

I once built one of those 16F84+LCD counters (I still have it and use it
sometimes). It uses the resistor gate 'switch' scheme with two sections of
a HC00 biased in the linear region as preamp. With this I can see hash on
UHF TV when the box is near the TV cable. The hash occurs when the PIC
clocks out the prescaler. It cannot be produced by the HC00 (whose input
is shorted at the time of measurement with a 50 ohm plug). The circuit is
in a plastic box and is breadboarded. The fast edges on the PIC output
clocking the prescaler are sneaking out ... there is no trace of
oscillator noise/hash/moire (the osc is built tight and has a wire loop as
guard around its parts on the circuit side, the crystal can is grounded).

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads