Worse yet I'm not at all certain what they are using to connect the CLK (GPIO6) and CS (GPIO7) lines (see C). Can anybody provide a clarifying schematic using standard symbols?
Worse yet I'm not at all certain what they are using to connect the CLK (GPIO6) and CS (GPIO7) lines (see C). Can anybody provide a clarifying schematic using standard symbols?
Hi Dave,
Looks like two ordinary NAND gates to me. The one connected to GPIO6 acts as an inverter and then "NANDS" with GPIO7 forcing the module to RESET when (and only when) GPIO7 (CS\) is high and GPIO6 (CLK) is LOW.
At least that is how I read it.
/Henrik.
Thanks, Henrik. I thought about the same but I'm not very strong on logic gates.
Isn't the GA1000 going to be reset on every clock pulse whenever CS\ is high?
Hi Dave,
Yes it is. But if the CLK line only goes to the GA1000 then there's no reason for any pulses to be present on the CLK line when the module isn't selected and I suspect they are thinking you'll have a dedicated SPI clock between the EM500 and the GA1000...
If you need to share the CLK signal between peripherals then a dedicated I/O for the RST-line would be required, if it can't be omitted all together and reset thryu hardware only. (?)
Again, that's just my understanding of it, take it for what it is.
Exactly, but they don't mention you can't share the CLK line with others on the bus (if you go with the NAND gate aproach) which means you need a second CLK line and the pin you gained by using the NAND gates is gone. So, if you have more than one SPI device on the same CLK (like that memory device) it seems to me that it would be best not to go with the NAND gate aproach.
/Henrik.
I already knew I could neither share the CLK nor DI/DO - they've hard coded all of this which sort of defeats the whole concept of SPI.
Bookmarks