65C22 at 3V3 but with 5V inputs

[cbmeeks]

It seems my understanding of input voltages of the 65C22 was not correct. I had falsely assumed that because the 65C22 can run at 5V or 3V3, then it could take either-or without issue.

For example, I thought I could run VCC at 3V3 and the inputs would accept 5V. But it seems I didn't "RTFM" close enough. LOL (no harm done, at least).

Instead, the maximum voltage is VCC + 0.3V.

If my board is mostly 3V3 but I want to interface to the outside world (via the two ports) at 5V, what is the best way to handle that? The 75LVC245 is common but I would have to set the direction and considering each pin of a port could be either input or output, I don't think that would work.

Should I just look into some bi-directional level converter and put between the ports and the headers?

Thanks!

[barnacle]

Your problem is the body diode between I/O pins and the substrate. Overdriving the inputs is unlikely to damage the chip, particularly if current limited, but will power the chip rather unexpectedly - and potentially the rest of the circuit to which its Vcc is attached.

Not good. It happens sometimes with 'TTL level' RS232 signals, which idle high, and can power, or prevent proper reset behaviour, entire CMOS circuits.

If you're not going to be driving stuff at MHz rates, the approach here https://learn.sparkfun.com/tutorials/bi ... -guide/all might be worth a read.

Neil

[cbmeeks]

I plan on running between 1-2 MHz normally with 6-8 MHz being top end.

What about these?

https://www.mouser.com/ProductDetail/Te ... tYOw%3D%3D

[barnacle]

If you can live with the packaging - TSSOP is the largest - then they look ideal. I've considered them myself for the same application but eventually didn't need them (I changed everything to 3v3) so can't tell you how they work, but the spec sheet looks good - either for a 5v IO part interfacing to a 3v3 bus, or for 5v I/O from a 3v3 part.

Neil

[Rob Finch]

Can you use a simple resistor divider?

[BigDumbDinosaur]

Rob Finch wrote:

Can you use a simple resistor divider?

Possibly, if you can tolerate the resistors’ effects on signal rise and fall times.

[GARTHWILSON]

This topic is relevant:
viewtopic.php?f=4&t=7119