• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 45 subscribers
  • Views 583 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.

ADS1274 single-ended input mode

Ronaldo Viera
Prodigy 100 points
Other Parts Discussed in Thread: ADS1274, ADCPRO

Hi community!

We are trying to make some measurements on our ADS1x7x (equipped with a ADS1274)eval board connecting a DC source direct to its analog inputs (bypassing the input buffers of that board). To negative analog input we fixed the voltage to 2.5V, and we want to change the positive input from 0V to 5V. But when we set the voltage to 0.1V we measure 0.7V-2.5V=-1.8V, using just one channel, and 1,4V-2,5V=-1,1V when using four channels. 

I left the positive input floating and the negative input connected to 2.5V, then I used a voltmeter to measure the voltage at the negative input, and for my surprise i got 1.068V at that pin. All voltages measured are related to ground (AGND).

We connected the AGND to the DC sources negative outputs and to the common ground pin of the source. The experiment uses the MMB0 and ADCPro as software.

Is that voltage on the negative input of ads1274 expected? Should the device capable to measure from 0V to 5V fixing one of its input to 2,5V?

We want to use the ADS1274 to measure 4 channels simultaneously, signals frequencies are very near to DC (about hundreds of hertz).

  • 0
    TI__Guru** 113765 points

    Hi Ronaldo,

    We made a couple of different versions of the EVM.  Does yours have red switches or silver looking switches?  Can you be a little more specific about where you are actually making the connections on the board?

    Thanks,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 100 points

    Thank you for your interest Bob.

    Then, we are working with the revision E of ADS1x7x daughter board, and Revision D of MMB0. The switches (S4, S5 ...) on ADS1x7x board are silver.

    The connections are made on this way: (The source is set to Output 1: 2.5V and Output 2: Variable from 0.1V to 4.9V)

    DC workbench source output 1 + --> All channels negative input of ADS1274 (bypassing the buffer)
    DC workbench source output 1 - --> ADS1x7x AGND pin (J3-19)

    DC workbench source output 2 + --> All channels positive input of ADS1274 (bypassing the buffer)
    DC workbench source output 2 - --> ADS1x7x AGND pin (J3-17)

    Using this set, the minimun voltage that we read at the ADCPro is 1.4V-2.5V = -1.1V. When we connect just channel 1 (or 2 or 3 or 4, but just one at a time) the minimum voltage measurable is 0.7V-2.5V=-1.8V. We don't expect a smaller input range when fixing 2.5V to the negative input. I belive that the range shouldn't change (0V to 5V on positive input).

    Please, any help will be welcome.

    Thanks in advance,

    Ronaldo

  • 0 in reply to Ronaldo Viera
    TI__Guru** 113765 points

    Ronaldo,

    I just tried duplicating your setup and had good results.  I'm suspecting that your supply is not low impedance output.  One test to try is to see if using low-speed mode improves the readings.  Low speed mode improves the input impedance of the ADC.  Another test would be to see if using the buffer improves your readings.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 100 points

    Hello Bob,

    I changed the source as you suggested (now I'm equiped with a signal generator with output impedance of 50 Ohm and capable to generate low offset and low noise DC signal generator). The results are described on the following table (please consider the comma as decimal separator). Again, I bypassed the buffers, no anti-alias filters are connected.

    Measured Code

    Measured Code converted to Voltage(V)

    Vp

    Vn

    Vp - Vn

    CH1

    CH2

    CH3

    CH4

    CH1

    CH2

    CH3

    CH4

    0,25

    2,5

    -2,25

    -6102584,95

    -6104495,50

    -6102341,90

    -6103602,30

    -1,81871

    -1,81928

    -1,81864

    -1,81902

    0,50

    2,5

    -2,00

    -5425496,01

    -5427428,65

    -5425317,60

    -5426601,34

    -1,61692

    -1,6175

    -1,61687

    -1,61725

    0,75

    2,5

    -1,75

    -4747641,06

    -4749574,12

    -4747518,64

    -4748820,99

    -1,41491

    -1,41548

    -1,41487

    -1,41526

    1,00

    2,5

    -1,50

    -4067925,73

    -4069879,99

    -4067876,67

    -4069197,52

    -1,21234

    -1,21292

    -1,21232

    -1,21272

    1,25

    2,5

    -1,25

    -3389774,04

    -3391781,60

    -3389800,78

    -3391160,16

    -1,01023

    -1,01083

    -1,01024

    -1,01064

    1,50

    2,5

    -1,00

    -2711483,44

    -2713529,33

    -2711590,57

    -2712573,43

    -0,80809

    -0,80869

    -0,80812

    -0,80841

    1,75

    2,5

    -0,75

    -2032413,41

    -2034500,51

    -2032594,55

    -2034016,27

    -0,60571

    -0,60633

    -0,60576

    -0,60618

    2,00

    2,5

    -0,50

    -1355720,75

    -1357899,51

    -1355988,42

    -1357476,13

    -0,40404

    -0,40469

    -0,40412

    -0,40456

    2,10

    2,5

    -0,40

    -1081326,31

    -1083518,62

    -1081623,62

    -1083125,64

    -0,32226

    -0,32291

    -0,32235

    -0,3228

    2,20

    2,5

    -0,30

    -813816,48

    -815974,40

    -814129,62

    -815623,91

    -0,24254

    -0,24318

    -0,24263

    -0,24307

    2,30

    2,5

    -0,20

    -547229,22

    -549387,81

    -547561,53

    -549064,13

    -0,16309

    -0,16373

    -0,16319

    -0,16363

    2,40

    2,5

    -0,10

    -278265,65

    -280418,38

    -278615,38

    -280142,29

    -0,08293

    -0,08357

    -0,08303

    -0,08349

    2,50

    2,5

    0,00

    -5124,92

    -7291,94

    -5500,49

    -7735,16

    -0,00153

    -0,00217

    -0,00164

    -0,00231

    2,60

    2,5

    0,10

    268031,69

    265849,70

    267618,18

    266077,77

    0,07988

    0,079229

    0,079756

    0,079297

    2,70

    2,5

    0,20

    546171,17

    543942,12

    545721,70

    544156,51

    0,162772

    0,162107

    0,162638

    0,162171

    2,80

    2,5

    0,30

    812186,19

    809937,27

    811707,60

    810130,46

    0,24205

    0,24138

    0,241908

    0,241438

    2,90

    2,5

    0,40

    1079794,44

    1077503,87

    1079264,32

    1077667,01

    0,321804

    0,321121

    0,321646

    0,32117

    3,00

    2,5

    0,50

    1350943,62

    1348607,21

    1350371,73

    1348756,96

    0,402613

    0,401916

    0,402442

    0,401961

    3,25

    2,5

    0,75

    2040196,73

    2037945,15

    2039600,20

    2038014,83

    0,608026

    0,607355

    0,607848

    0,607376

    3,50

    2,5

    1,00

    2711859,07

    2709500,61

    2711170,46

    2709507,88

    0,808197

    0,807494

    0,807992

    0,807496

    3,75

    2,5

    1,25

    3394555,08

    3392115,94

    3393782,23

    3392059,41

    1,011656

    1,010929

    1,011426

    1,010912

    4,00

    2,5

    1,50

    4076018,18

    4073530,24

    4075175,38

    4073404,34

    1,214748

    1,214007

    1,214497

    1,213969

    4,25

    2,5

    1,75

    4754649,59

    4752118,54

    4753717,68

    4751900,86

    1,416996

    1,416242

    1,416718

    1,416177

    4,50

    2,5

    2,00

    5442889,12

    5440208,33

    5441880,89

    5440007,08

    1,622107

    1,621308

    1,621807

    1,621248

    4,75

    2,5

    2,25

    6113234,48

    6110576,03

    6112139,25

    6110221,18

    1,821886

    1,821094

    1,821559

    1,820988

    As you can note, voltages near to the end of range are far from the expected value. Is that behavior expected?

    Thanks for your  help,

    Ronaldo

  • 0 in reply to Ronaldo Viera
    TI__Guru** 113765 points

    Ronaldo,

    Your measured codes are in fractions.  Are you averaging?  Is it possible that there is a numeric processing error?  Any way that I can see a series of raw codes for a single channel over a period of time?  I don't see this behaviour at all.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 100 points

    Bob,

    The error were at my sources. They were not "sourcing" the exact DC level that they must to, I made the measuremnts again, considering the real DC level at source output, and got the (Vp-Vn) exactly as expected. I should consider a Voltage Reference for this kind of experiment, not a DC source.

    Thank you very much for you help,

    Ronaldo