24VDC Serial Logic Level Shifter: Give me your thoughts!

[Str1ker]

Hey everyone,

I recently acquired an old Flip-Dot destination sign from a bus. After doing some research, I found out that it uses the IBIS standard, which is a hardware-modified serial interface, to communicate with the main controller that is operated by the bus driver. The only difference from a normal serial interface is that it uses a 24V logic level for the communication.

I came across some old (hobbyist) level shifter schematics designed for this specific IBIS purpose. However, I thought I could come up with a better solution myself (where's the fun in using someone else's work, right?).

My goal was to design a solution that works with 3.3V and 5V logic levels, providing a universal solution for different control strategies building on this work.

I've attached my solution to this problem (T2 in the schematic needs to be replaced by a 2N2222). I'm eager to hear what you think of it and if you have any thoughts on how to improve this circuit.

(Side note: R5 and R6 have such a low resistance because the TX signal must provide at least 100mA on the output side)

By the way, I simulated my approach in Falstad's circuit simulator and tested it on a breadboard, and everything seemed to work properly.

Looking forward to your suggestions!

[BennoG]

I did a design that uses the AEIC-7272-S16 shifter to get to a decent 24V push pull output.
For receiving you can do a simple resistor divider (with some protection)

Benno

[Phil1977]

There are many ways to solve this problem, and if you are satisfied with your design - then go for it!

Just the points that could be improved:
- The "double-pull-down" design pulls a lot of idle power, especially with 24V. Why not use a push/pull driver?
- The receiving path can be realized just with a voltage divider + a zener. Is there any reason to use transistors?
- It´s recommended to user buffer/decoupling caps on level converters for all voltage levels.

[Str1ker]

I did a design that uses the AEIC-7272-S16 shifter to get to a decent 24V push pull output.
For receiving you can do a simple resistor divider (with some protection)

Benno

Thanks for the suggestion! I wasn't aware that such an IC exists. I might keep an eye out for one that reliably operates on 3.3V.

[Str1ker]

There are many ways to solve this problem, and if you are satisfied with your design - then go for it!

Just the points that could be improved:
- The "double-pull-down" design pulls a lot of idle power, especially with 24V. Why not use a push/pull driver?
- The receiving path can be realized just with a voltage divider + a zener. Is there any reason to use transistors?
- It´s recommended to user buffer/decoupling caps on level converters for all voltage levels.

I really appreciate your help!

- The "double-pull-down" is usually off, as the serial interface is in a high state when idle. The signal must supply at least 100mA (at least that's whats specified in the standard)... I guess it has something to do with multiple and "oldish" receivers.

- Simplifying the receiving part might work with the divider and a zener but I was not sure if a 5V receiver works with a maximum 3.3V high level that the zener creates (It shall work with 3.3 and 5V logic levels). I will give this idea a try in the upcoming days!

- Caps are also a good hint!

[BrianHG]

Have you considered a mosfet gate driver IC?

3.3v logic inputs.
Enable inputs.
Available with multiple drivers per IC.
Outputs push-pull at 2+ amps peak, 100ma continuous wont be a problem.
Example devices which can support 25v:
https://www.infineon.com/cms/en/product/power/gate-driver-ics/irs4427s/

[BrianHG]

Another dual gate driver, the IXYS IX4424NG claims 30v support, again multiple amps push-pull drive output.

[Benta]

Just us a SPDT analog switch for the 3.3/5 -> 24 V translation.
The 24 -> 3.3/5 V translation can be done with a simple voltage divider.

[Str1ker]

Thank you all for your replies!! Your recommendations will help me in future projects.