cancel
Showing results for 
Search instead for 
Did you mean: 

V3MODS Voltage Translator Questions

debugging
Lead

Schematic 

1. Most documents refer to add a PU on UART TX  (target side), why is there a PU on UART RX and not TX, but there is a PU on VUART TX and not on VUART RX ? Seems inconsistent.

2. For consistency or guideline, should the UART PU be on the board or debugger side ?

3. I2C PU is 560 Ohm, normally I2C PUs are 4K7, why this strong PU ?

4. I plan to use the V3MODS w/o translators and just 3.3V targets (STM32MP1/ F and L family MCU's) running at 3V3 I/O. any risks/concerns ?

5. what does the red cross indicate in the schematic ?

6. why are there PU on  I2C, SPI and GPIO pins, nRESET and nGND Detect on the MCU side before the translator ? are the V3MODS outputs configured as  OD ? Can this be omitted and configured as PU using any software ?

7. Why is there a PU on CAN RX and not TX ?

8. This application is just for internal development, not final product integration.  Are 47R series resistor needs w/o using voltage translator, That is directly connect to the target. or are these just for EMI concerns in final products ?

1 ACCEPTED SOLUTION

Accepted Solutions

Hello,

Here are the comments from the designer:

  1. Most documents refer to add a PU on UART TX  (target side), why is there a PU on UART RX and not TX, but there is a PU on VUART TX and not on VUART RX ? Seems inconsistent.

=> All inputs on level shifters (look at the signal direction) are connected to a PU/PD, this is not a question of TX or RX. Outputs are push pulled so they don’t require PU/PD

  1. For consistency or guideline, should the UART PU be on the board or debugger side ?

=> see above, no need to add anything

  1. I2C PU is 560 Ohm, normally I2C PUs are 4K7, why this strong PU ?

=> Dimensioned according to I2C speed, can be up to 4Mbps. You can change to fit your needs if you don't go so fast

  1. I plan to use the V3MODS w/o translators and just 3.3V targets (STM32MP1/ F and L family MCU's) running at 3V3 I/O. any risks/concerns ?

=> No

  1. what does the red cross indicate in the schematic ?

=> Components not fitted. It is there just for reference.

  1. why are there PU on  I2C, SPI and GPIO pins, nRESET and nGND Detect on the MCU side before the translator ? are the V3MODS outputs configured as  OD ? Can this be omitted and configured as PU using any software ?

=> PU required when OE_CTRL is high and output in tri-state

  1. Why is there a PU on CAN RX and not TX ?

=> Same reason given in 1

  1. This application is just for internal development, not final product integration.  Are 47R series resistor needs w/o using voltage translator, That is directly connect to the target. or are these just for EMI concerns in final products ?

=> 47R not needed without the translator.

View solution in original post

5 REPLIES 5
MHoll.2
Senior

Hi,
If You are referencing to this schematic:
https://www.st.com/resource/en/hw_model/stlink-v3mods_voltage-translator_reference-design.pdf
Then I don't think You have to use it if your target is 3.3V, the V3MODSMCU is powered by 3.3V.

The Pull up's are mostly there to prevent problems and are not strictly necessary.
But to me they seem to be where they belong (to pull up eventually high impedance lines).

To the I2C:

There is non "standard" pull up value for I2C, the value has to be designed according to:
- Voltage
- Bus speed
- Bus capacitance
- Current sink capacity of all connected IC's
The value of 560R is probable calculated for 1.8V and a Bus speeds up to fast-mode plus (1MHz).


The red cross means “not mounted” (the schematics was made with Altium designer).

debugging
Lead

Many thanks, Yes, that link is the same as provided in my OP.

I do wonder though about the choice made  for PU for item [1] & [7], any ideas ?

Hello,

Here are the comments from the designer:

  1. Most documents refer to add a PU on UART TX  (target side), why is there a PU on UART RX and not TX, but there is a PU on VUART TX and not on VUART RX ? Seems inconsistent.

=> All inputs on level shifters (look at the signal direction) are connected to a PU/PD, this is not a question of TX or RX. Outputs are push pulled so they don’t require PU/PD

  1. For consistency or guideline, should the UART PU be on the board or debugger side ?

=> see above, no need to add anything

  1. I2C PU is 560 Ohm, normally I2C PUs are 4K7, why this strong PU ?

=> Dimensioned according to I2C speed, can be up to 4Mbps. You can change to fit your needs if you don't go so fast

  1. I plan to use the V3MODS w/o translators and just 3.3V targets (STM32MP1/ F and L family MCU's) running at 3V3 I/O. any risks/concerns ?

=> No

  1. what does the red cross indicate in the schematic ?

=> Components not fitted. It is there just for reference.

  1. why are there PU on  I2C, SPI and GPIO pins, nRESET and nGND Detect on the MCU side before the translator ? are the V3MODS outputs configured as  OD ? Can this be omitted and configured as PU using any software ?

=> PU required when OE_CTRL is high and output in tri-state

  1. Why is there a PU on CAN RX and not TX ?

=> Same reason given in 1

  1. This application is just for internal development, not final product integration.  Are 47R series resistor needs w/o using voltage translator, That is directly connect to the target. or are these just for EMI concerns in final products ?

=> 47R not needed without the translator.

Thank you for the answers and explanations This is very helpful.

As a side question I wish to access SPI/I2C devices on the bridge from linux using command line scripts. is that possible ? (and the GPIO pins as well) is there an API, SDK for it, or I even could use USB commands to send to the device if there is a USB command document ?

Hi again,

thank you for the feedback.

The following may be what you're looking for:

ST-Link V3 bridge API .