Author Topic: AND gate with schmitt-trigger inputs  (Read 3161 times)

0 Members and 1 Guest are viewing this topic.

Offline HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1607
  • Country: gb
AND gate with schmitt-trigger inputs
« on: September 05, 2020, 07:45:32 am »
I've been looking for single-channel 2-input AND gates that have schmitt-trigger inputs, but haven't come up with much. They seem to be very thin on the ground.

My requirements are 5V operation, preferably in a SOT23 package. I need schmitt-trigger inputs because I need to AND a microcontroller output and a user-toggled switch input (for a safeguard), but the switch will have RC debouncing so will have slow rise time.

So far I've only found the 74AHC1G08 from Diodes Inc, and the SN74AHC1G09 from TI. The latter is open-drain output, but that's fine because I'm only switching an NPN transistor.

Anybody know of any others?
 

Offline exmadscientist

  • Frequent Contributor
  • **
  • Posts: 404
  • Country: us
  • Technically A Professional
Re: AND gate with schmitt-trigger inputs
« Reply #1 on: September 05, 2020, 08:32:57 am »
Alas, there's no standard logic gate that does exactly what you want.

The LVC and lower-voltage families typically include Schmitt trigger action on their inputs, but you should be warned that it's not a "real" Schmitt trigger: the hysteresis is deliberately set to a small value in comparison to a full-fat '14 or '17 gate. This usually works out well in most cases, but if you are depending on it to clean up a very slow edge, you will be disappointed. There will be a specification in the datasheet (or, less commonly these days, in the family databook); for whatever reason it's typically listed as a "maximum rise rate". For the SN74LVC1G08 at 5V, that's 5 ns/V, or if you invert that to get the more common "slew rate", it's a minimum of 200 V/us. Either way, it's an edge time of no more than 25 ns! Like I said, there's not much actual Schmitt triggering on that input! In comparison, the SN74HCS08, quad AND from TI's new HCS family with true Schmitt triggers on all inputs, has a rise rate of "Unlimited". It doesn't get much clearer than that!

Anyway, if you're satisfied with its edge rate, the 74LVC1G08 from your favorite vendor is a fantastic choice for you. If not... none of your options are great, I'm afraid.

First up, obviously there's the larger packages, which would get you the previously-mentioned 'HCS08. Or there's sticking a 1G14 or 1G17 true Schmitt trigger in front of any old gate or GPIO. (This can usually also be arranged to get you level translation if you need it.) If you can make do with a NAND gate instead of an AND gate, you're in luck! The '132 is a standard part available in the usual families; it's a single NAND with true Schmitt triggers on its inputs. SN74LVC1G132 is one example part, but the usual variants (vendor, family, gate count, etc) exist as well.

If you don't like any of those options, I'm afraid I don't know of anything else to offer :(
 

Offline HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1607
  • Country: gb
Re: AND gate with schmitt-trigger inputs
« Reply #2 on: September 05, 2020, 10:13:34 am »
Yes, I'm dealing with an edge rate of approx. 300,000 ns/V due to the filtering on the switch input (10K, 100nF). Although, I suppose the cap could be swapped out for a smaller value, but can't go too far without impacting the debouncing, so we're still talking ns/V in the thousands.

I just looked in more detail at the datasheets for both Diodes Inc. 74AHC1G08 and TI SN74AHC1G09, and while they both say "Schmitt-trigger action at all inputs", they don't appear to mean 'real' schmitt trigger, as the stated maximum rise rate is still only 20 ns/V! So they're out the window. Can't really trust parametric search results at face value... |O
 

Offline HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1607
  • Country: gb
Re: AND gate with schmitt-trigger inputs
« Reply #3 on: September 05, 2020, 10:38:58 am »
I found a possibility I hadn't considered: a '1G57 or '1G97 configurable multi-function gate. Will work as an AND gate by simply tying one of the three inputs high.

They're also described as having schmitt inputs, but having looked at the recommended operating conditions, I was dismayed to find some datasheets state max. transition rate of 5 ns/V. However, they also give input hysteresis figures of around 0.6V - just like a '1G17 buffer. So, it would appear that this is actually 'true' schmitt-trigger functionality, right?
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13218
Re: AND gate with schmitt-trigger inputs
« Reply #4 on: September 05, 2020, 10:47:18 am »
Why not use a SN74LVC2G132 dual Schmitt NAND, with the second gate to de-invert the first.  Just strap all the signal pins on one side of the package and you've got your real Schmitt AND gate.  In the DCU package its the same length as a SOT23-6 and only 0.3mm wider lead tips to lead tips.
 
The following users thanked this post: edavid

Offline HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1607
  • Country: gb
Re: AND gate with schmitt-trigger inputs
« Reply #5 on: September 05, 2020, 12:19:34 pm »
Ah, yes, that's also a possibility. Thanks for the suggestion.
 

Offline hli

  • Frequent Contributor
  • **
  • Posts: 261
  • Country: de
Re: AND gate with schmitt-trigger inputs
« Reply #6 on: September 05, 2020, 02:34:00 pm »
I second the 1G57 - Schmitt-Trigger-Inputs, Vcc up to 5.5V, can be configured as 2-input AND gate. Available at least from TI and Nexperia (formerly NXP). Both do _not_ specify a max. rise time for the inputs. (They even accept 5V inputs with 3.3V Vcc)
 

Offline HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1607
  • Country: gb
Re: AND gate with schmitt-trigger inputs
« Reply #7 on: September 05, 2020, 03:15:19 pm »
Both do _not_ specify a max. rise time for the inputs.

It's weird that the Diodes Inc version does, which was the first one I looked at, and made me think "oh, these aren't proper schmitt inputs either!". But then I saw the next page had hysteresis values.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf