Author Topic: Level Shifting: practical differences between these two components  (Read 975 times)

0 Members and 1 Guest are viewing this topic.

Offline vmalletTopic starter

  • Regular Contributor
  • *
  • Posts: 87
  • Country: us
I'm exploring the world of level shifting to interface 5v systems with 3.3v or vice versa, and I'm wondering what the practical differences are between these two components:
I can read the datasheets and see some differences but I'm lacking the basic knowledge of why would one use one vs. the other depending on applications. What are the fundamental differences?

Could someone shed some light?
 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 11906
  • Country: us
    • Personal site
Re: Level Shifting: practical differences between these two components
« Reply #1 on: January 09, 2020, 07:23:21 am »
The major difference is that SN74LVC2T45 is not bi-directional. There is a DIR input that controls which side is input and which side is output.

TXS0108E8 is automatic. You put it between the rails and it does the shifting.
Alex
 

Offline vmalletTopic starter

  • Regular Contributor
  • *
  • Posts: 87
  • Country: us
Re: Level Shifting: practical differences between these two components
« Reply #2 on: January 09, 2020, 05:40:11 pm »
The major difference is that SN74LVC2T45 is not bi-directional. There is a DIR input that controls which side is input and which side is output.

TXS0108E8 is automatic. You put it between the rails and it does the shifting.

Thanks Alex. So in that case, why would one chose to use the SN74LVC2T45 when the TXS0108E8 does everything on its own? Is it a bandwidth difference? 
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Level Shifting: practical differences between these two components
« Reply #3 on: January 09, 2020, 05:44:09 pm »
The "automatic" parts are notorious for glitching.  The traditional directional bus driver is used when you have the requisite bus control signals available and require provable and repeatable functionality.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 11906
  • Country: us
    • Personal site
Re: Level Shifting: practical differences between these two components
« Reply #4 on: January 09, 2020, 05:45:00 pm »
Thanks Alex. So in that case, why would one chose to use the SN74LVC2T45 when the TXS0108E8 does everything on its own? Is it a bandwidth difference? 

Their architecture matters a lot. SN74LVC2T45 has a real push-pull output, so it also acts as a buffer for the signals.

TXS0108E8 does not really buffer the signals. It essentially does what a typical MOSFET-based level shifier does.

I assume price is also a factor, but I can't even find the price for TXS0108E8.
Alex
 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 11906
  • Country: us
    • Personal site
Re: Level Shifting: practical differences between these two components
« Reply #5 on: January 09, 2020, 05:47:33 pm »
The "automatic" parts are notorious for glitching.
TXS0108E8 is interesting in this respect. It looks like it is just a bunch of MOSFET level shifters, so it won't really glitch as long as both sides do not drive at the same time.
Alex
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Level Shifting: practical differences between these two components
« Reply #6 on: January 09, 2020, 06:11:36 pm »
At DC, yeah, but the "accelerator" thing is where the quirks happen.

AFAIK, they usually work alright with normal, hard driven, logic level signals, but rapidly start to fall apart when long traces or analog levels are involved.  Because they really aren't digital inputs or outputs, as such, they're some weird reactive thing.

The bus gate has traditional logic inputs, so can be (ab)used the same as any other; they have the expected behavior.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf