• Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 2020 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.

ADS8344 NonLinearity

Tim Riley1
Prodigy 110 points
Other Parts Discussed in Thread: ADS8344, REF02

I am having issue with a 4-20 ma measurement loop. The circuit is designed as follows, with the output connected to a 16-bit ATD (ADS8344). Not my selection, it is standard on the micronontroller board.

I am attempting to inject a signal with a 4-20 ma calibrator to verify operation, but am getting the following output:

PART_CNT1 Current (mA) ATD Counts1 Vin @24.9ohm Scope Volts* Scope Volts** ATD Counts2
0 236 0 0.121 0.157 256
2 8095 0.0498 0.675 0.684 8000
4 12799 0.0996 1.11 1.13 13167
6 16703 0.1494 1.63 1.67 18751
8 19807 0.1992 2.1 2.1 24047
10 23135 0.249 2.57 2.61 27551
12 26496 0.2988 3.1 3.1 32127
14 29951 0.3486 3.52 3.55 37055
16 33631 0.3984 4.06 4.08 41855
18 37759 0.4482 4.5 4.51 46591
20 65407 0.498 4.97 5 65279

Where: Current - the value injected at the amplifier inpu to produce the output response

           Vin @24.9ohm - input voltage measured across the 24.9 ohm resistor

           ATDCounts1 - The ADC reading with no scope attached to the AD627 output

           ScopeVolts* - Voltage measured at AD627 output with ADC OFF.

           ScopeVolts** - Voltage measured at AD627 output with ADC ON and acquiring data.

           ATDCounts2 - ADC reading taken in conjunction with ScopeVolts**.

(SEE PLOTS BELOW)

OK, this is what I did, I connected the current source input of the AD627 AMP test and connected a scope probe to the output of the AD627.

1)Test circuit with no scope attached to amp output. Read ADC value from microcontroller. (See ATDCounts1)

2) Repeated test with scope attached to output of AD627. I ran the test first with the ADC OFF (see Scope Volts*) Then I repeated the test with the ADC ON (See Scope Volts** and ATDCounts2).

I plotted these valuse as shown above.

So, the question is, since the amplifier response seems very linear over the entire range of values for both cases, scope attached and not attached.

What is causing the large noticable change in slope in the ADC response?

The design is such that the amplifier is driving the ADC input directly. So, DC coupled, and also no matching or compensation networks.

Any suggestions?

Thanks

 

  • TI__Mastermind 43915 points

    Hi Tim,

    Could you send the schematic portion showing the reference voltage. May be that there is not enough drive strength for the reference or there is not a 1-10uF capacitor on the reference pin to help. Note that with the ADS8344 being a SAR converter, the reference is sampled may times through each ADC conversion period. If there is not sufficient reference drive strength, you may see droop in the ref line causing errors in the output result. These may appear as a gain error. You also may want to consider a well placed RC filter at the front end of the input to help with charge distribution as the SAR samples. What data rate are you using with the SAR?

    Regards,

    Tony Calabria

  • in reply to Tony Calabria
    Prodigy 110 points

    Tony,

    Sorry for the delayed response, but I did not get a notification that I has a reply.

    Anyway, the ADC interface looks like the picture below. The only difference being the fact that there is an analog devices REF02 5v reference chip connected to the pin labelled VREF on the bottom left side of the sketch. So it looks like there is probably 100 nF on that pin internal to the board. with no additional capacitance on the external interface board.

    The SAR is running at 100kHz, internal clock mode.

    Also, the sensor response is pretty slow changing so there is not a high data rate to worry about. Having said that, I assume that a passive LPF would be satisfactory. Do you know of a suitable Amplifier to ADC filters design guide?

     

     

    Thanks for the initial reply.

    Tim

     

     

  • in reply to Tim Riley1
    TI__Guru**** 182421 points

    Hi Tim,

    In your original post, you mentioned that the ADS8344 came 'standard on the micro-controller board'.   Is this some sort of commercially available eval setup?  Can you let us know more about it? 

  • in reply to Tom Hendrick
    Prodigy 110 points

    Tom,

    Again sorry for the delayed response, but I don't seem to be getting email indicators that I have had a reply. Yes the board that I am using is an Elmicro S12_Compact Board. See link (http://elmicro.com/en/s12compact.html). I have not modified the board at all so the circuit as indicated on the schematic is what I am using. As a side note, I have other devices connected to other channels on this chip that seem to perform much better than what I have indicated for this circuit.

    Regards,

    Tim Riley