• State
  • Replies 3 replies
  • Subscribers 46 subscribers
  • Views 47 views
  • Users 0 members are here
Support feedback
Options
Options
Related

DAC63002: Using an external reference with the DAC63002

Yeongbin Ahn
Prodigy 40 points
Part Number: DAC63002
Other Parts Discussed in Thread: MSP430F5659

I previously posted a similar question but didn’t receive a response, so I’m posting again.
I’m currently using the DAC63002 with an external reference voltage connected to the VREF pin. The DAC receives data via SPI communication from an MSP430F5659 MCU.
Under normal communication conditions, the VREF pin draws almost no current. However, when a communication error occurs, a large current flows into the VREF pin, causing the reference voltage to drop.


Here’s what I’ve observed:

Normal operation:
The MCU sends 3 sets of 8-bit data (total 24 bits) via SPI, and the SYNC signal is closed immediately after transmission. In this case, the reference voltage remains stable, and no significant current flows into the VREF pin.

image.png


Error condition:
The transmission is interrupted after only one 8-bit data packet is sent, meaning the full data is not delivered to the DAC. During this state, the SYNC signal remains open, and a large current flows into the VREF pin, causing the reference voltage to drop. The voltage only recovers after a normal signal is received again.

image.png


My questions are:

Why does current flow into the VREF pin when a communication error occurs and the SYNC signal remains open?
Is there a way to prevent the reference voltage from dropping even if a communication error happens?

Of course, the ideal solution would be to eliminate communication errors entirely, but despite various attempts, they still occur intermittently. So I’m looking for a workaround or protective measure to ensure the VREF voltage remains stable even during such errors.
Any suggestions or insights would be greatly appreciated.
Thank you.

  • 0
    TI__Guru 54673 points

    Hi Yeongbin,

    Can you add some additional information?

    1. Can you share the schematic for your configuration? i am wondering if the reference circuit is shared with anything else.  

    2. Does this issue only happen when the initial 8-bits are addressing a specific register? For example, address 0x1F? I am wondering if something is causing the internal reference to turn on.

    3. Does this happen in multiple systems or boards?

    4. Can you extend or intentionally reproduce this issue? For example, can you hold SYNC low for 20ms to see what the final voltage of the VREF settles to?

  • 0 in reply to Paul_Frost
    Prodigy 40 points

    Hi Paul,

    Thank you for your response.

    1. Regarding the schematic for our configuration:
      The reference voltage is shared across multiple components, including the MCU (used as a reference voltage) and several sensors that require a stable voltage supply.

    2. About the initial 8-bit address:
      Currently, we are consistently using address 0x1C. Could this specific address potentially be causing the internal reference to turn on?

    3. Regarding multiple systems:
      We have several PCBs with the same circuit configuration, and all of them exhibit the same issue.

    4. Reproducing the issue:
      As shown in the figure below, we held the SYNC signal low for an extended period without providing any Data or Clock signals. We observed that the 2.5V VREF eventually dropped to approximately 2.3V. During this time, the current flowing into the DAC’s VREF pin was measured to be around 408uA (Under normal operating conditions, the current is approximately 14 µA). Could this behavior be caused by the prolonged low state of the SYNC signal?

    Thank you again for your support. If there’s anything else you recommend I check, please let me know.

  • 0 in reply to Yeongbin Ahn
    TI__Guru 54673 points

    Hi Yeongbin,

    1. This is interesting, can we just for confirmation remove the DAC from the PCB and test this again? I am wondering if you are experiencing some other parasitic load that is collapsing the supply.  We dont have any mechanism in the device to cause this from our understanding.  I would be good to have this datapoint.

    2. can you connect SDI and CLK to GND and try again? This would help determine if the register or the SPI interface could be contributing to this.  

    3. Maybe, does the reference have stronger drive capability?  If you short the reference to a supply voltage that has strong current drive, does it draw any more current?