Author Topic: UART level translator for FT2232H  (Read 1815 times)

0 Members and 1 Guest are viewing this topic.

Offline muthukural001Topic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: in
UART level translator for FT2232H
« on: May 06, 2020, 04:36:30 pm »
Hi,

I want to use TXB0304 level translator with FT2232H for UART level translation. The TXB0304 having output drive strength of 20uA. I am seeing FT22332H input leakage current Iin as 45uA (typical). It means that I can not use TXB0304 with FT2232H since its drive strength is 20uA and can not drive 45uA load. Attached images are TXB0304 drive strenght and FT2232H input leakage current details. Could anyone suggest?


Thanks and Regards,
Muthu
« Last Edit: May 06, 2020, 04:38:21 pm by muthukural001 »
 

Offline ebclr

  • Super Contributor
  • ***
  • Posts: 2328
  • Country: 00
Re: UART level translator for FT2232H
« Reply #1 on: May 06, 2020, 08:27:46 pm »
Test Condition does not mean max condition
 

Offline TomS_

  • Frequent Contributor
  • **
  • Posts: 834
  • Country: gb
Re: UART level translator for FT2232H
« Reply #2 on: May 07, 2020, 08:16:07 am »
The TXB0304 datasheet suggests +/- 50mA with a maximum of +/- 100mA per power pin the way I'm reading it.
 

Offline Ian.M

  • Super Contributor
  • ***
  • Posts: 12807
Re: UART level translator for FT2232H
« Reply #3 on: May 07, 2020, 10:21:17 am »
Using a weak autosensing bidirectional level translator for a UART is moronic*.  Why not simply use unidirectional  level shifting buffers?

If you don't need to handle logic levels above 5.5V or below 1.65V Vcc, 74LVC1T45 is suitable, with its direction pin tied high or low to make it unidirectional  (or 74LVC2T45 if you need two lines in each direction).  They are good for +/-24mA drive @3.3V, so it would probably be a good idea to add series 330R resistors on the target side if its a  MCU under development to limit the current in the case of bus contention due to improper I/O initialisation.  They also go tristate on partial power down, so if you tap target side power from the target, you don't have to worry about what happens if the target looses power while the FT2232H is powered by USB or visa versa.

If you need to go below 1.65V, I'm sure someone else will suggest an alternative.

* Even for a USART in synchronous mode its fairly dumb if there is any possibility of providing control lines for data direction and clock as master (out) or slave (in).
 
The following users thanked this post: muthukural001

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 14309
  • Country: fr
Re: UART level translator for FT2232H
« Reply #4 on: May 07, 2020, 02:48:55 pm »
Agreed. Were the OP using a bidirectional 8-bit parallel bus mode, a bidirectional level shifter could have been useful (but they can be finicky too), but for simply UART? You have two data lines with a fixed direction. Don't bother.

For alternatives to the 74LVC1T45 working below 1.65V, there is the NLSV1T34 (that I've used in a JTAG probe circuit) which works down to 0.9V.
 
The following users thanked this post: oPossum, muthukural001

Offline muthukural001Topic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: in
Re: UART level translator for FT2232H
« Reply #5 on: May 08, 2020, 11:18:27 am »

Thanks, I will go with direction controlled level translator.

Hi Toms,

Please see IOH and IOL spec.

Thanks,
Muthu
« Last Edit: May 08, 2020, 11:20:19 am by muthukural001 »
 

Online NorthGuy

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: UART level translator for FT2232H
« Reply #6 on: May 09, 2020, 04:13:47 am »
For alternatives to the 74LVC1T45 working below 1.65V, there is the NLSV1T34 (that I've used in a JTAG probe circuit) which works down to 0.9V.

Also newer 74AVC series (e.g. sn74AVC2T45 or sn74AVC8T245), which can go down to 1.2V. They're also faster than 74LVC. I used them for JTAG at 1.5V with FT232H at 30 MHz and they worked well.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26757
  • Country: nl
    • NCT Developments
Re: UART level translator for FT2232H
« Reply #7 on: May 09, 2020, 09:08:12 am »
Using a weak autosensing bidirectional level translator for a UART is moronic*.  Why not simply use unidirectional  level shifting buffers?
I wouldn't put it that strong. It will work; a unidirectional translator usually only goes one way so you'd need 2 of them. I have had good results with the bi-directional level translators from TI. I'd be careful to use the ones from a low-cost chip supplier.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online NorthGuy

  • Super Contributor
  • ***
  • Posts: 3137
  • Country: ca
Re: UART level translator for FT2232H
« Reply #8 on: May 09, 2020, 02:45:03 pm »
Using a weak autosensing bidirectional level translator for a UART is moronic*.  Why not simply use unidirectional  level shifting buffers?
I wouldn't put it that strong. It will work; a unidirectional translator usually only goes one way so you'd need 2 of them. I have had good results with the bi-directional level translators from TI. I'd be careful to use the ones from a low-cost chip supplier.

All signals in UART are unidirectional.

The bidirectinal level translators are typically two unidirectional buffers in opposite directions furnished with output resistors and connected in parallel. For some reason, they use very weak resistors. Unless decreasing power consumption is the main goal, you will get better results if you build it by yourself using stronger resistors.

The best way, of course, is to use switches instead of resistors, but this will only work if you can produce a "direction" signal which selects the direction.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 14309
  • Country: fr
Re: UART level translator for FT2232H
« Reply #9 on: May 09, 2020, 02:51:19 pm »
Yep. I don't see a point using bidirectional level shifters for unidirectional signals, unless maybe you were already using them (with the same part) in other sections of your circuit and you wanted to limit the number of different parts in your BOM.

For UART signals, I agree bidirectional level shifters shouldn't be problem, at least with conventional UART signaling, in which TX/RX signals idle at VH and are constantly driven. But unless for reducing part count, why bother? (Can be a valid point in some cases though!)

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf