• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 4 answers
  • Subscribers 45 subscribers
  • Views 1965 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ADS1232 Reference

Rich Morris
Prodigy 210 points
Other Parts Discussed in Thread: ADS1232, TS5A3160

I'm doing a wireless weigh scale design using the ADS1232. Everything is working great but every once in a while, I experience a very large offset that doesn't seem to go away after battery replacement/reset/etc. Eventually it does, but I'm a bit baffled. I'm wondering if I'm treating Vref poorly and causing the problem. Here are some details that might pertain.

1) I do a power on offset calibration

2) REFN is ground, REFP is hooked Battery via a TS5A3160 analog switch. I'm applying power 500mS before sampling the ADC and then removing it until the next cycle in a few minutes. (I do this to share the e3xact power with the load cell. Maybe I should now switch the REF?)

3) I have just modified the software to perform an offset calibration each cycle before sampling which can be an hour apart. One thing that worries me is that the part comes back as ready in 400 mS instead of the 800 mS as indicated on the data sheet. (I'm sending 32 clock pulses)

The part seems very simple to use, so it must be something stupid on my part...

  • 0
    TI__Guru** 113765 points

    Hi Rich,

    First, can you tell me what the VBAT voltage is that you are using? 

    Here are a couple of things I noticed.  One is that you are only using a gain of 2 based on schematic configuration.  You would be much better served with a gain of 128 for bridge measurements to get the lowest input referred noise.  Perhaps part of the problem you are seeing is related to the low gain.

    Another consideration is related to analog settling of the reference.  It shouldn't take very long based on the values you are using, but as the offset calibration takes several cycles to complete, the reference may not be stable if it is still settling to the final value.  Speaking of analog settling, I see that you have no input filtering for anti-aliasing or for eliminating EMI/RFI.  This could be a potential problem later on.

    Lastly, 400ms is the timing coming out of standby mode, so it may be possible that the offset calibration is not being issued at all.  If you are placing the device in standby mode by holding SCLK high, bringing SCLK low restarts the device and conversion is ready after 400ms.  It would be better to issue the offset cal after that time.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 210 points
    VBAT is 3.2 volts (CR2032).

    These bridge components are great. The highest gain I can run without saturation is 2. When things are working, they work very well. Noise does not seem to be a problem (I'm surprised actually).

    I'm actually not going into standby mode, SCLK is remaining low (I probably should implement that....) The reverence seems stable, but I give it 100 mSecs just in case before I do the calibration.

    Then, like I said, it only take 400 mSecs, instead of the 800.

    Thanks fore the prompt reply.
  • 0 in reply to Rich Morris
    Prodigy 210 points
    Actually, a bit more info... I have a device where this is happening when I insert the battery. The values are frozen and it never seems to come out of reset. If I short out VBAT and allow it to reboot, chances are everything will be OK (but not always). Can it get latched up during power up? I figured if I cycled PWRDN* it would come back to life. I don't purposely bring it high, but it does go high during reboot, along with everything else.
  • 0 in reply to Rich Morris
    Prodigy 210 points
    One more bit of info... I did not realize that I had to wait 400 mSec after the PWRDN* line was brought hig prior to doing the calibration. I think now it is operating as I expected. I will test more and confirm.
  • 0 in reply to Rich Morris
    TI__Guru** 113765 points

    Hi Rich,

    It is possible that the ADS1232 may not come out of reset based on the startup conditions where the PDWN pin goes high too quickly.  See page 23 of the datasheet and Figure 39.  One way to prevent an issue is to keep the PDWN low using a pull-down resistor to ground, and then making sure that at least 10us has gone by before the micro places the PDWN pin to a high state.

    You might want to check the section on page 22 of the datasheet where you can shorten the on time when exiting Standby Mode, where the offset calibration takes place when SCLK goes low taking the device out of Standby.  Standby Mode differs from Power-Down Mode.  Standby uses more power, so it may not be desirable. In the Power-Down Mode, it takes about 8us for the device to wake-up (Figure 41) using the internal clock. If you attempt to communicate with the device before the internal clock has started, the communication will be ignored.  If you were violating that timing previously, the offset calibration would not have started in any case.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 210 points
    Thanks Bob, As long as I don't get stuck in an unrecoverable state, I'm fine with taking a few extra mSeconds rather than add more parts. I've implemented on 3 more with good results, so maybe I'm in the clear.

    Do you know if it is possible to get hung up in an unrecoverable state? It does seem like it should, but I* worry....
  • 0 in reply to Rich Morris
    TI__Guru** 113765 points

    Rich,

    Based on your current design and explanation I would make sure that I pulsed the PDWN pin after power up to ensure that the ADS1232 has been reset to the proper conditions.  This will clear any issues from possible startup glitches that may occur on the PDWN pin when power is first applied.

    The state where you say the device is frozen, have you double checked to make sure that the micro is working correctly?  One of the issues with inserting a battery is the amount of transients that can take place that can prevent proper operation of the devices in your system including the ADS1232, the micro or the switches routing the power to the bridges.  I would think that resetting the micro should straighten things out assuming there is a proper pulse on the PDWN pin of the ADS1232.  PDWN must have a minimum low time as stated in the datasheet as t14 in Figure 41 on page 23 for the device to properly reset.

    Best regards,

    Bob B