Need to shift 3.3v to 5v TTL and space is a min. I'm using the TXB0102 to do this. I already have the 5v for Vccb, any pitfalls to be aware of using a V-divider to power the Vcca 3.3v side or should I add a 3.3v voltage regulator?

 

Use a regulator.

The voltage output from the divider will change every time the load changes.

Voltage dividers are used to provide a reverence voltage to a high impedance input, and do not perform properly when having to drive a load.

 

I did not look to see how much power the IC uses but definitely more than the 10k can supply.
Why don't you have 3.3V?

 

Voltage dividers are used to provide a reverence voltage to a high impedance input, and do not perform properly when having to drive a load.

I kinda figured but was hoping to get away with it on a small load.

Why don't you have 3.3V?

It's a small board and space is a premium, was hoping to get away from adding a second regulator. The screen is 5v/5v TTL and the micro controller is 5v Vcc, but it's TX/RX is 3.3v level.

I'm using the OKI-78SR 5V to power the screen and the Micro as it doesn't require any additional components, so no 3.3v available. There is a different Micro controller with the same footprint I can use that is 5v TTL so I will probably switch to that.


Thanks

 

It is not clear how much current draw on pin 3. It depends on the load. And data speed.
R1=1k gives you 1.7mA max to IC1. R2 change to a 3.3V Zener. C2= large bypass cap.

 

It's always been my impression that 10mA is the typical minimum current needed to maintain the Zener voltage. (VZ)

I could be wrong...it wouldn't be the first time.

Personally, I would use the minimum current in the data sheet and add the load current then calculate the resistor from that.

 

It's always been my impression that 10mA is the typical minimum current needed to maintain the Zener voltage. (

I picked a 100mW Zener at random. At 3.3V they are not good as can be seen by the curves. The curves are flat above 5V. This diode was tested at the factory at 5mA. Most ate tested at 10mA as you know. The high current diodes are tested at more current. I think it takes 2mA to get 3.0V from this diode. I agree the 1.7mA I proposed is low. 470 ohm resistor will move the current to 3.4mA. & 3.2V

The IC probably does not care exactly what the voltage is.

 

I always learn a lot on this site - thanks!

 

How about using two silicon diodes in series instead or the 10K resistor?? The voltage drop is much less dependent on the load current.