I am following a schematic which implements a basic 8-bit register. The schematic is pictured below:

enter image description here

According to the datasheet of the 74LS173 chip, The high-level output voltage is 2.4V with testing conditions assuming the minimum input voltage (although my case Vcc is 5V). The current configuration is set to output ~2.4V at each output pin on the 74LS173. I can confirm that the output is indeed ~2.4V. It appears that this chip is functioning perfectly well.

The chip receiving this 2.4V potential is the SN74HC245. This chip is also fed 5V Vcc. The datasheet only states two input states (high + low) high being >3.15V at 4.5V Vcc and low being <1.35V at 4.5V Vcc. It is wired such that Ax is the input and Bx is the output. When I provide 5V to an input, I have 5V present at the cooresponding output. This is as we would expect. However, I need the 2.4V from the 74LS173 output to be present at the output of the 74LS245. In my head, I assume that I cannot simply pass 2V from the input of the transceiver since this is neither high or low. However, in the above schematic 2V is present in his outputs of the transceiver.

This is demonstrated here: https://youtu.be/9WE3Obdjtv0?t=193

In my testing, when 2.4V is present at any A input, it does not exist at the corresponding B output. Whereas the same test with 5V input is indeed present at the output. In the datasheet for the SN74HC245 it appears that lowering the supply voltage lowers the minimum high-level input voltage, but in his case this was not necessary. How is this possible?


1 Answer 1


It is not entirely clear what you are asking here, or what you say happens. Sometimes things outside the limits of the data sheet will work - or at least seem to for a while.

The output voltage of your 74LS173 is so low because you are heavily loading it with the LEDs, and because LS logic doesn't have all that strong output drive to begin with. The best solution would be to use a CMOS inverter to drive the LED cathodes instead of the anodes, and to isolate driving them from other consumers of the logic signals being displayed. But you could substantially improve the situation without changing the topology by using a 74HC173 or 74HCT173 which will drive much closer to the rail even when loaded, and/or by enlarging the LED resistors so they draw less current.

With regard to the 74xx245, if you use specifically 74HCT245 (note the "T") that has an input high threshold of 2 volts or less, even when powered at 5v. As such, it has a good chance of helping you by doing what all digital logic is designed to: restore the "noise margins" by taking any acceptable input, and turning it into a definitive output as close to the rail as its output structure can accomplish against a given load.

  • \$\begingroup\$ Indeed in general, LS outputs are not directly compatible with HC inputs but using pull-ups makes the logic high voltage compatible. But pull-ups don't help in this case as the LEDs load the output too much. \$\endgroup\$
    – Justme
    Sep 7, 2019 at 7:57
  • \$\begingroup\$ This is exactly what I need to know, thank you! Solved. \$\endgroup\$ Sep 15, 2019 at 0:08

Your Answer

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

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