Hi to all...
Note: if not interested in why I want to buid a SPI sniffer skip to
where say: *I HAVE TRIED THIS:*
I'm a ham radio (CO7WT) and I have two fine transceivers that are dead
for the MCU of the radio... (Yaesu FT-747GX) all the other electronicas
are just fine... but the "brain" is dead
I have one working radio of that and I use it for experimenting/hacking,
please not that in Cuba there is not a free/open market for ham radios,
so by another is not an option...
And hacking is a very pleasant/instructive work :-)
Yes I have writed (email) to Yaesu and the former manufacturer of the
7925B IC: dead end.
I have hacked (reverse engineering) all the electronics and signalling
but the way the MCU speaks with the two PLL... (one for big steps an the
other for small steps...)
I need the datasheet or any form of how to speak with the
7925B IC (Japanese PLL IC)...
From the radio's tech manual and with a lot of math I have deduced the
math involved to calc the data I need to send to the TWO PLL ICs but I
can't talk the 7925B dialect !!!
The internet was not of help on find any clue about the 7925B as it's
discontinued part a long ago...
So I have decided to sniff the SPI code... but, ho boy!!!, it's harder
than I think in the first place...
I have a PICKIT2 clone and I tried to use the debug function: no joy...
the bus appears to run at 1Mhz... I know that it's about 40 bits... to
each IC... I get a random number of bits on each try...
*I HAVE TRIED THIS:*
PIC for the job is a PIC18F25K22 from a jalunio bee board... full 64 Mhz
with x4 PLL, it's running at 16Mhz real clock (64/4)
1 - Detect the CLK change and the read the data and two latches, then
send it by serial at full speed (115,200 kbps)
No joy, the serial Tx is blocking or overriding the detection, just
detected about 8 to 10 bits per Tx from the MCU...
2 - Detect the CLK change and the read the data and two latches, store
in a I2C EERPOM (CLK+DAT+L1+L2 = 4 bits), so I split each byte in a 4
bit positions... Then when 24LC32 is full dump it by serial...
No joy... about 5 bits per tx, the write in the I2C eeprom is too SLOW
for the job..
3 - Detect the CLK change and the read the data and two latches, store
it on a memory array, when full dump by serie...
better but no joy yet... about 15 to 20 bits per tx... it appears the TX
is very fast for the sensing-store procedure...
The Original MCU is clocked at 4Mhz by a ceramic resonator... I guess
they are running at 1Mhz and sending the SPI at full speed.
Any ideas for this hacking project about revive and old and good radio
with a new "PIC brain?"
Cheers from Pavel, CO7WT.
--
You received this message because you are subscribed to the Google Groups
"jallib" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/jallib.
For more options, visit https://groups.google.com/d/optout.
