One big potential issue with a single sourced part like the GreenPak devices is the potential for non-delivery. I had a part become unobtanium when the factory burned down. Now a $3 part is $100. Yup. It happens.
A part like the SN74CBTD3384CPW quickswitch device is multi-sourced and made in many factories and is quite cheap without custom programming. 10 bits of bi-directional level limiting for $0.25. Hard to beat.
I would call this part "level clamping", It is good for 5 V -> 3.3 Volt. But 3.3 V -> 5 Volt can not be done with it.
Yes, but TTL is happy receiving 3.3V CMOS levels, so there is no problem. Also, most TTL drivers only have weak pullups, so the CMOS inputs can usually just rely on their clamping diodes.
John
But, 5V TTL does not need 5V logic levels, only 3V or so; it's only 5V CMOS that benefits from higher Vin(high). 3.3V CMOS levels are just dandy as TTL inputs, as long as the current sinking is sufficient. Mostly, that means 74S gates are not compatible...
If you are driving CMOS levels, that is true, but no one uses 5V CMOS. TTL was the standard indicated above which 3.3V CMOS drives quite well. These parts prevent the 5V input on the TTL part from pulling up the 3.3V signal. So unless there is a requirement to drive 5V CMOS which was not stated, these parts work well. I'm using them in a 13 year old design of which I am working on delivering another 10,000 units.
The real beauty of these parts is they are inherently bidirectional (like a wire), so no need to worry with direction controls or even which direction the signals pass. They also work well interfacing 3.3V and 5V I2C signals.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required