1
\$\begingroup\$

I think some logic IC input can handle a range of lower voltage levels but will output at supply voltage...

What is the best way to achieve an input that can handle higher levels but output at a lower level for the next stage? Some logic ICs or circuits that can do this?

not dealing with high speeds, 5kHz max.

5v logic with 5-10V input range

What about a zener on the input?

\$\endgroup\$
11
  • 1
    \$\begingroup\$ What voltages specifically do you need? Logic that can be powered by 3.3 (or maybe 2.5 or even 1.8 V) with 5-V-tolerant inputs is fairly common. \$\endgroup\$
    – The Photon
    Apr 13, 2020 at 22:31
  • \$\begingroup\$ @ThePhoton 5V output and up to 10V tolerant input \$\endgroup\$
    – Jay
    Apr 13, 2020 at 22:37
  • 3
    \$\begingroup\$ Specify the maximum voltage to be treated as a LOW, the minimum voltage to be treated as a HIGH, and specify similarly for the output along with how many loads must support at those specs. What is the input impedance to be, also? \$\endgroup\$
    – jonk
    Apr 13, 2020 at 22:38
  • \$\begingroup\$ Also does it need to have hysteresis \$\endgroup\$
    – Spehro Pefhany
    Apr 13, 2020 at 22:44
  • \$\begingroup\$ The simplest IC of this kind is called resistor divider. \$\endgroup\$
    – Marko Buršič
    Apr 13, 2020 at 22:48

2 Answers 2

Reset to default
0
\$\begingroup\$

There are several options that could work

  • Use a voltage divider, and receive the signal using TTL-compatible inputs (i.e. HCT family parts) to allow for low \$V_{IH}\$ levels when the incoming signal is at 5 V levels.

  • Use an open-drain output on the high-voltage side, with pull-ups to the receiver's Vdd.

  • Use a diode connection as shown in the answer by Janka.

  • Use a zener-diode limiter to ensure the receiver's input isn't driven above, say, 4.5 V, even when the sender is driving 10 V.

  • ...

The best choice depends on unstated requirements like

  • data rate

  • propagation delay requirements

  • need to allow for even higher over-voltages (could the 10 V occasionally be 15 or 25 V momentarily?)

  • power consumption requirements

  • cost

  • ...

\$\endgroup\$
2
  • \$\begingroup\$ Is there any reason no one other than you has suggested or agreed with the idea of using a zener, as it seems to be ideal? \$\endgroup\$
    – Jay
    Apr 14, 2020 at 23:41
  • \$\begingroup\$ @Jay There are even more options than I've seen listed here. Some of those in my head, not present here, I might even prefer over any of these so far. But there just isn't enough context for me to know what to offer. You are in a much better position. If you like a zener, then a zener it is. \$\endgroup\$
    – jonk
    Apr 15, 2020 at 4:48
2
\$\begingroup\$

How to make a low-speed digital input tolerant against voltages > Vdd:

schematic

simulate this circuit – Schematic created using CircuitLab

Ignore anything but 0V on the input. That's what the diode does. Pullup to +Vdd if not 0V. If the IC has a built-in pullup (e.g. TTL behaves this way, and also many µC) you can leave out the external one.

\$\endgroup\$
4
  • \$\begingroup\$ Wouldnt the high threshold be practically nonexistent? For a static voltage a zener diode seems like it could achieve this without affecting threshold of the input, not sure how it would behave with a switching input. \$\endgroup\$
    – Jay
    Apr 14, 2020 at 2:08
  • \$\begingroup\$ For TTL, input levels between 0.8V and 2.7V are forbidden. So there isn't a high threshold it could affect, only that "forbidden zone". For CMOS, input voltages below 20V will be shorted by the high-side protection diode on the input, and those above will punch though the gate oxide and fry the chip. A Zener has too much capacitance to be any help. (This is for logic ICs, for higher integrated CMOS parts the limit is usually 8…10V.) \$\endgroup\$
    – Janka
    Apr 15, 2020 at 4:11
  • \$\begingroup\$ too much capacitance for 5kHz? seems to be in ballpark of 100pf, with 10k current limiter thats 150kHz cutoff. Everyone seems reluctant to even suggest the potential problems that may arise from this in general logic applications without having the exact details of my particular application but after some reading I think maybe the concerns here are delay and dull edges? neither should matter here. \$\endgroup\$
    – Jay
    Apr 15, 2020 at 23:42
  • \$\begingroup\$ It's not the 5kHz that matters but the rise time. That one may be in the µs range. But as stated before, your zener is useless anyway as there is a protection diode inside the chip that shorts to +Vdd much faster than the Zener could react. As long the input voltage is below roughly 20V for CMOS logic chips or 8..10V for CMOS µCs. And above, the gate oxide is toast before the zener reacts. \$\endgroup\$
    – Janka
    Apr 16, 2020 at 9:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.