ADC128S102: ADC128S102 reading problem when switching between channels (for me too)

Hi,

Many of us are out of the office this week and next, so I thought I'd jump in and take a look at your post.

First, I can't think of a reason why the ADC would give bad values by swapping the inputs of the ADC. The device should still be reading out channels in the same way, so the master shouldn't have problems with the readout of the ADC. With the 11:1 voltage divider with the resistors, the ADC outputs seem reasonable.

However, there are still some things to check. I would like to see a schematic of the circuit. Perhaps there's something unusual in the connections to the remaining part of the board. Are you able to drive the inputs with a variable voltage? If so, you may want to sweep the input voltage of either channel from 0V to 24V (before the voltage divider) just to check if there's anything else going on. I would highly recommend trying this just to see that there isn't a problem in the setup. If you can post the results of the test. Again just changing the inputs, you shouldn't see any changes in operation, and you should only get the proper changes in inputs.

Other than swapping the input voltages, are you changing anything else in the setup or software?

Joseph Wu

Dear Joseph,

Thank you for your reply.

I'm sending the related part of the schematic.

During the swap, I don't change anything on the setup. Only I swap the voltages.

But, let me say other thing.  This situation can be realized not only in the 24V-5V duo, but in many small and large values.(etc. 3V-5V)

Metin

proj.pdf

I use 2 channels in ADC128S102 to measure each voltage as (+) and (-) over 2 opamps. (you can see in block schematic).
When I have connected the positive voltages to ANALOG VOLT IN-1 and ANALOG VOLT IN-2, I measure the from IN1 and IN3 of ADC128S102.
When I have connected the negative voltages to ANALOG VOLT IN-1 and ANALOG VOLT IN-2, I measure the from IN2 and IN4 of ADC128S102.

For ANALOG VOLT IN-1 = (between 0 and +33V), ANALOG VOLT IN-2 = (between 0 and +33V)
---------------------------------------------------------------------------------------------------------------------
When I removed R6(IN2) and R8(IN4) resistors from circuit, The voltages in IN1 and IN3 has been reading clearly and accurately. doesn't matter voltage values in this situation.
But if I do not remove these resistors, the negative voltage at the input IN2 and IN4 of the ADC128 prevents me from measuring the value at IN1 and IN3.

I hope this explanation better explains my problem.

Block.pdf

Hello,

I would suggest using a known DC input at the op amp input and then probing down the circuit to make sure that each node is at the expected voltage.

The ADC will measure what it is inputted, and it seems the issue is somewhere in the analog input circuit,

I am reaching out to my colleague who support op amp to take a deeper look at this.

Regards

Cynthia

Hello,

I would like to ask to you simple question.

does the negative voltage at one of the inputs of your IC cause the voltages at the other inputs to be measured incorrectly?  

if it is yes, Why and what is the solution? 

I took a deeper look at your circuit and consulted with the op amp team

and they caught something I simply overlooked.

This ADC does not accept negative voltages, the ADC can only measure positive voltages.

The design is inverting the positive voltage to a negative voltage by referencing the second op amp to ground, this is violating the abs max of the device.

the input circuit will need to be changed if you wish to use this ADC.

Also, this ADC is a singled ended device, thus it cannot measure the difference between the negative and positive input voltages.

We have created a series of circuits that to cover topics such as this, called Analog Circuits within this, there is a Level Translation Input Drive Circuits, that I think directly relates to what you are trying to achieve here.

I suggest looking specifically at Single-ended to differential using a two op-amp circuit. note that in this design the second op amp has a reference voltage at the positive terminal at mid scale. this shift the output voltage. this is what you will need to do as well to move the "negative" voltage to within the ADC input scale range

Also, if you decide to move to a differential ADC, we can help find the correct device.