I'm trying to calculate maximum input voltage that can be applied to a voltage divider based ADC circuit attached in the picture below along with converting the ADC raw value to actual voltage.

In the above attached image, P8 is a terminal screw block connector where the external analog voltage is applied. And as you see, the resistors form the voltage divider circuit. AN0-AN7 go to the ADC pins of a microcontroller. This is a 5 volt system and so I believe the reference voltage should be taken as 5 volts. I am trying to figure out what would be the maximum voltage that I can apply to the pins on P8 connector and also how to convert the raw ADC values to actual analog voltage that was applied. So, based on the circuit, 5 = (V x 10k)/30k -> V = 15 volts. [Here 5 volts was taken as the reference voltage since it is a 5 volt based system.] So the maximum voltage that can be applied is 15 volts. The resolution of the microcontroller is 12 bits. So, to calculate the actual analog voltage from raw ADC values, Actual Voltage = (Raw ADC value)*15/4095. However with this calculation I'm not getting the right voltage that was applied. Can anyone guide me regarding where I am going wrong in this whole calculation ?

EDIT: Circuit 2

In this above circuit, when the dip switch S2 is turned ON, if I am correct the 1k resistors come in parallel with the 20k resistors. So, the new calculation would be 5=(20k||1k)X V/((20k||1k)+10k), which would turn out to be V being equal to 58 volts approximate. Is this correct ?

• 5 = (V x 10k)/30k - that's wrong. It should be 5 = (V x 20k)/30k, yielding 7.5V max. – Eugene Sh. Jul 28 '16 at 19:32
• @EugeneSh.Thanks for replying back and correcting me. So, to get actual analog voltage applied, I use the formula, Actual Voltage = (Raw ADC value)*7.5/4095 . Is this right ? – bobdxcool Jul 28 '16 at 19:34
• Yeah. But make sure you are working with floats rather than ints.. – Eugene Sh. Jul 28 '16 at 19:35
• @EugeneSh.Thanks for replying back. I have added another circuit in the main post. It would be helpful for me if you can let me know if my calculation is right in this case. – bobdxcool Jul 28 '16 at 19:55
• 58V?? Where did you get it from. That's correct it will become parallel with 20K though.. – Eugene Sh. Jul 28 '16 at 20:27

What you're describing is a passive attenuation circuit as a stage prior to your ADC, probably because you expect there to be some voltage exceeding the accepted range of the ADC and you want to either (1) protect it from damage, or (2) scale the data so that you can still perform ADC measurements on large voltages.

For aim (1), this is not sufficient. It is very important that you have TVS (transient voltage suppression) and/or over-voltage zener protection on every input line to your system. If it is a 5V-tolerant input at the MCU, you can connect every input line after your 10k resistor bank to a Shottky diode, each connected to 5V. This is what I consider a bare minimum for circuit protection.

For (2) (scaling your data), it seems you have two attenuation modes - the first, where you have a 20k/10k divider scaling to an equivalent 66%, and the second where you have a (20k||1k)/10k divider scaling to an equivalent 8.7%.

In the first mode, your maximum input voltage would be 7.5V. In the second mode, your maximum input voltage would be 57.5V.

You probably also want to consider stability - get some idea of how fast you expect your input to vary, and then add stabilization capacitors to all of your inputs, likely across your 20k resistors, to filter out noise.

Your pictures did not come through.

Your title does not look like a question but I will interpret it as asking how to do voltage divider caculations.

Here is the answer: Voltage Divider Calculator

This link will let you figure out the values then it shows you the correct calculations to use in your software to figure out the applied voltage from the presented (reduced) voltage.

You will sooner or later figur your dividers. What i don't see here is any capacitor. You must add some for a whole bunch of reasons. I don't want to mention antialiasing, but at least you must be able to hold your voltage constant while ADC is sampling, which requires some current. So without capacitors you will not see what you expect, you will see lower and noisier voltage.