• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 47 subscribers
  • Views 934 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PGA309: Design for Weigh Scale

PaulieT
Prodigy 40 points
Part Number: PGA309

Hello,

I plan to use the PGA309 for a number of Weigh Scale applications in a new product line and have a few questions. 

1. Is the PGA309 recommended for new designs?

2. Do I need to use the One-Wire interface to write / read the internal PGA309 registers or can I do this with I2C?

3. If I connect an EEPROM to the PGA309 can I also have the system micro talk directly to the PGA309 or EEPROM using the I2C lines?

The system could have as many as 8 PGA309 & EEPROM pairs, so I need to keep them all isolated from each other; I plan to use a TCA9617 so when the system micro wants to access the EEPROM it enables the TCA9617 first.  Do I need to use the TEST pin to accomplish this?

4. The Load Cells are 350 ohm, which is too much load for the EEXC output. Is there anything wrong with putting a buffer IC on this line to drive the load cells?

Thanks for your help.

Paul

  • 0
    TI__Mastermind 23050 points

    Hi Paul,

    1. Yes the PGA309 is active and thus recommended for new designs.

    2. The internal registers can be configured by either the One-Wire interface, or I2C. However since the external EEPROM is usually connected to the I2C lines and communications between the PGA309 and EEPROM can be initiated through the One-Wire interface, it is generally more straightforward to program through the One-Wire interface.

    3. Yes, you can have the micro communicate over the I2C line shared by the EEPROM. The PGA309 acts as a master during communication with the EEPROM but it will relinquish bus control for 1.3 seconds after a successful acknowledge whenever it is addressed. Please refer to section 4.2 and 4.9 in the following user's guide for a more detailed explanation: www.ti.com/.../sbou024b.pdf

    I do not believe you need to use the TEST pin to free up the bus, you should be able to accomplish that by adressing the PGA as mentioned above. Our resident PGA expert is out of office right now, but I will confirm the best approach to this with him when he returns.

    The method you are proposing should work provided you abide by all the communication requirements, but I believe this could be accomplished more easily by reserving the I2C for the EEPROM on each PGA and using a multiplexed one-wire interface to read/write to the desired device.

    4. I do not see anything wrong with doing this, however it may introduce additional offset and offset temperature drift error if you are attempting to externally sense the bridge temperature as described in section 2.7.2. Alternatively, you could add some resistance to either side of your bridge to reduce the current drawn, but this also reduces the voltage available to the bridge.

  • 0 in reply to Zak Kaye
    TI__Genius 11970 points
    Hi Paul,

    3. As Zak said, it should be possible to use the I2C bus to handle communication with the PGA309. Troubleshooting the interactions of an I2C bus multiplexed between 8 PGA309s and their external EEPROMs with a single host may end up being more complex than simply multiplexing the PGA309's one wire busses to the microcontroller, however.

    The TEST pin is used to disable the PGA309's internal state machine, preventing the PGA309 from updating its register settings from the EEPROM (and stopping I2C reads from the EEPROM in the process). This can be useful for calibration, as it allows you to change settings in the PGA309's registers without having them overwritten by an EEPROM read. As Zak said above, it shouldn't be necessary to use the TEST pin to free the I2C bus.

    4. Buffering the Vexc output should be okay, and initial calibration should be able to account for any (fixed) offset in the buffer, but offset drift errors the buffer introduces may compromise both the bridge temperature measurement and possibly the bridge linearization. I'd recommend picking a low drift amplifier to minimize its impact.

    That being said, the PGA309 datasheet (page 13, table 1) indicates that it should be able to handle bridge impedances down to 200Ω without issue, so an external buffer may not be required at all.