[EE] Logic level conversion and power consumption
Sean Breheny <[email protected]> Wed, 29 Apr 2026 23:43:36 -0400
| Newsgroups | gmane.comp.hardware.microcontrollers.pic |
|---|---|
| Message-ID | <CAE8uMMr2HG7Qz-3dC87MKomt-UTU2kT2oMsToG8aMcvehujQdA@mail.gmail.com> |
Hi all, I encountered a problem today that I was not previously aware of. We all know that applying an input voltage which is close to the threshold to a CMOS logic gate often results in cross-conduction of at least one internal CMOS pair which greatly increases the supply current. What I did NOT know is that this also applies to some logic families which are rated for reduced input logic levels. I have some 74LVC logic which is rated for up to 5.5V supply voltage but is still supposed to be able to accept 3.3V logic level inputs. It does and it works properly, but the power supply current climbs from microamps to almost 10 mA per gate which has a 3.3V output driving an input. 74HCT is also rated to perform this level transition task (accept 3.3V inputs with a 5V supply) but it seems to have a much less severe version of this problem. The supply current rises to at most 0.5mA (according to the datasheet - I haven't tried measuring it yet). The 74LVC parts' datasheet is a bit misleading because it gives a spec on the increase in supply current (about 300uA) but it only gives it for a logic high which is "Vcc-0.6V", which is much higher than 3.3V in my case (4.4V). The same spec on the HCT datasheeet is given for "Vcc-2V" which is much more reasonable. Unfortunately the chips I was using in the 74LVC family have no drop-in equivalent in HCT logic so I need to add an additional level conversion stage. Anyone run into this before? Sean