Crosstalk in AT45DB642 Flash memory

[LunaS25]

A capacitive waveform synchronized at the same voltage (3.6V) and frequency of a CS signal (13.5Khz) appears over the SO (Serial Output) signal of the memory. SO is used as a data transmission pin (Rx_Flash) and is connected to a Rabbit 3000 micro controller. CS is also generated by the micro.

I have a ground plane covering both lines on the other side of the PCB and 100K pull ups (3.3V) connected to CS and SI (Serial Input) which is the other data line (Tx_Flash) on the memory.

Any ideas on how to eliminate that kind of interference ?

 

[felixc]

A capacitive coupling that gives 3.6 volts? There must be something else that goes wrong. Put a 1K resistor load at this signal. Is the signal still there? If yes it is probably not a capacitive coupling.

Have you verified that if Vcc is connected correctly to your flash chip? It sound stupid, but it is one of the things that I can think of, for an output that follows the input.

 

[IRstuff]

There is really no way to get this level of interference from a correctly connected set of devices.

There has to be a wiring error of some sort.

TTFN

 

[melone]

Or a sneak path. Are the CS and SO signals reference to the exact same supply? Are there other circuits connected to the SO line?