• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 46 subscribers
  • Views 436 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.

ADS1282EVM and BiPloar Supplies

Paul Atkins
Prodigy 60 points
Other Parts Discussed in Thread: ADS1282, PGA280

Hi,

I have been using the ADS1282EVM (and MMB0 Rev D) evaluation kit for a while, in a single ended (+5V) mode (using the supplied power pack), with good results (and no power supply related problems).

However, I now have to extend the input range to cover greater than 5V (about 7.5V), so I am trying to use a bipolar power supply (+/-5V) with some problems.  I am actually using a battery pack, but I don't think the actual power source matters greatly.

Initially, after reading some other posts to work out how to connect the dual supply (on J14) and set the jumper (on J13) on the MMB0, I successfully connected the +5V to the +5VD connection (and +5VA line, using the J13 +5V/+5VA jumper) and the -5V supply to the -5VA connection.  I had not connected to +/-10V supplies at this stage, as I wanted to be sure I had properly connected the +5V analog and digital power.  All seemed OK at this stage (I also had not plugged the ADS1282EVM daughter board in just yet either).  I connected the MMB0 Gnd to the earth on the battery pack. Voltages were correct and no evidence of excess current drain etc, so next step was to plug in the ADS1282EVM board (still without connecting the +/-VA supplies).

This is where the fun starts.  Within probably 10-20 seconds of turning on just the +/-5V (and earth), I could smell an brown burning smell.  I disconnected the power and hooked up a voltmeter and (somewhat gingerly) reconnected the +/-5V supply.  Again I quickly got a burning smell but had the chance to notice the -5V supply was dropping rapidly – it went down to about –2V and I had to pull the power).  One last round of testing confirmed the +5V supply seemed to be holding OK, but for some reason, the -5V supply is being "shorted" by something on the ADS1282EVM board.

I'm embarrassed to mention that the last time, I also saw a wisp of smoke from one of the smaller ICs (but couldn't say which one), as well as the burning smell.  Something, somewhere ion the ADS1282EVM is decidedly unhappy about how I’m providing the –5V supply.

Both the +5V and-5V supply provided by fixed voltage regulators (7805 and 7905) and should have ample current to supply the ADS1282EVM (and there was no indication of a hot/blown regulator).

Is there something else that needs to be done to use a -5V supply or do you have any suggestions as to why I am cooking some IC on the ADS1282EVM board, when there was no problem with the -5VA supply when just using the MMB0 board on its own?  I'd love to try some more troubleshooting, but I don't think I can chance blowing up the ADS1282EVM (if I haven't already fried something).

Best Regards

Paul Atkins

  • 0
    TI__Mastermind 44680 points

    Hi Paul,

    Your procedure looks mostly correct, (see page 12 in the ADS1282EVM User's Guide http://www.ti.com/lit/ug/sbau144c/sbau144c.pdf). Did you remove the J12 jumper on the MMB0? If you connected the +/-5V to the correct input terminals, I don't see a cause for this failure. Might you perhaps have a picture of the connection and jumpers?

    FYI: You will be limited to a +/-2.5V supply on the ADS1282, however (using the regulators on the ADS1282EVM-PDK). The ADS1282 is not intended to be used with a supply greater than AVDD-AVSS =+5.5V. "GPIO3" (pin 12 on the serial header) will need to be pulled to ground to enable the bipolar supplies on the ADS1282EVM.

    I also suggest looking at page 29 in the same document for a schematic of the ADS1282EVM and checking out ftp://ftp.ti.com/pub/data_acquisition/ADCPro/Support/MMB0_Sch_RevD.PDF for a schematic of the MMB0 (rev D).

    Regards,
    Chris

  • 0 in reply to Christopher Hall
    Prodigy 60 points

    Hi Chris,

     Thanks for the fast reply.  I think my basic problem was misunderstanding the EVM and power supply and thinking I could use a +/-5V supply (I’d overlooked the AVDD-AVSS limit of 5.5V).  For the record, I had remembered to disconnect the J12 jumper (to disconnect the 5V power pack supply, but it wasn’t plugged in anyway); although its an easy thing to miss.  I hadn’t realised that GPIO3 also needed to be grounded.  Fortunately, and despite a bit of burning and smoke, the EVM still seems to be working (from a single ended, +5V, supply).

     I now realise that I won’t be able to use the ADS1282EVM for a measurement range greater than +/-2.5V (due to the onboard regulators).

     However, I also just noticed that the EVM manual says for calibration a zero signal should be applied for the offset and a maximum signal for the gain value.  Is a zero signal actually meant to be a  –2.5V input or a 0V input and similarly, is the maximum signal actually meant to be a +2.5V input or a 5V input?  I ask, since the inputs are still effectively a differential signal (at least I don’t think the input is referenced to the ADS1282 Analog Ground, but I could be wrong), but I’m just not sure whether the input range is effectively –2.5V to + 2.5V or 0V to +5V (I suspect its –2.5V to + 2.5V but just want to be sure if this is what is needed for the “zero” and maximum” signals for the calibration).

     Sorry to get a bit off the original topic, but my original attempt to use a +/-5V input range was partly driven by ease of calibration.

     Best regards

     Paul

  • 0 in reply to Paul Atkins
    TI__Mastermind 44680 points

    Hi Paul,

    For a greater input range you would need to attenuate the input. We do have PGAs with 1/4 and 1/8 V/V gains (such as the PGA280) to accommodate these kinds of cases. You can also use resistor dividers, but that would introduce some additional noise and component tolerances may introduce some gain error. (The resistor divider noise can be reduced by adding some input capacitance, but this creates a trade-off between noise and input settling).

    Regarding your questions on calibration:

    • The "zero signal" is a differential voltage. It is done by shorting the two differential inputs to a common-mode voltage source (within the common-mode range of the inputs, from AVDD - 1.25 to AVSS + 0.7 volts).
    • The maximum signal is equal to the full-scale input (i.e. +Vref/(2xPGA) or 2.5V for Vref=5V and PGA=1). This should also be a very accurate differential input voltage that meets the common-mode input range requirement. Use a precision reference to generate this voltage, as this will affect gain error.

    I hope that helps!

    Regards,
    Chris

  • 0 in reply to Christopher Hall
    Prodigy 60 points

    Hi Chris,

    Many thanks again for the rapid turnaround.

    Both answers are extremely (and I DO mean extremely) useful - I know the info is probably somewhere in the manual and/or data sheet, but your clarfications provide great insight.

    Thanks again and regards

    Paul