MSP430F5310: MSP low-power microcontroller forum

Jonathan Tshimama-Kabangu

Part Number: MSP430F5310

Tool/software:

Using 2% in the error equation does not give you the equivalent of a 2% change in the voltage divider,
of the two method which one is correct?

my question is a follow up from my previous thread:

https://e2e.ti.com/support/microcontrollers/msp-low-power-microcontrollers-group/msp430/f/msp-low-power-microcontroller-forum/1525487/msp430f5310-adc-reading-worst-case-error-calculation

4 months ago

0 Dennis Lehman 4 months ago

TI__Guru 71216 points

Hi Jonathan,

Let me look into this and I'll work out an example.

0 David Schultz 4 months ago

Guru 25225 points

I can't tell if you considered the case where one resistor had +1% error while the other was -1%.

0 Dennis Lehman 4 months ago in reply to David Schultz

TI__Guru 71216 points

Hi Jonathan,

You are correct and I was seriously wrong. My apologies. Thank you for catching this and I will update the original posting.

So yes, when changing only one resistor at a time by +/-1% in the voltage divider, the resulting vout will not change by +/-1%, but rather roughly +/- 0.5%. And of course this would change if the two resistors where different values to begin with, like a 330K and 100K.

Again sorry for the confusion on this.

0 Jonathan Tshimama-Kabangu 4 months ago in reply to Dennis Lehman

Intellectual 280 points

Hi,

Thank you, I'm looking forward to the response.

Please show how the +/-1% in the voltage divider will result in the Vout change roughly of +/- 0.5%

Thanks

0 Dennis Lehman 4 months ago in reply to Jonathan Tshimama-Kabangu

TI__Guru 71216 points

I will post shortly.

0 Dennis Lehman 4 months ago in reply to Dennis Lehman

TI__Guru 71216 points

Hi Jonathan,

Ok, here is screen shot of calculations (spreadsheet attached)

When R1 = R1 + 1%, R2 = R2, Vout changes by -0.498% (0.00498).

When R1 = R1 + 1%, R2 = R2 - 1%, Vout changes by -1.0%.

Does this help?

R_divider_percent_error_calc.xlsx

0 Jonathan Tshimama-Kabangu 4 months ago in reply to Dennis Lehman

Intellectual 280 points

Hi Dennis,

based on the attached spreadsheet, is it correct to conclude that you must always calculate voltage divider percentage error?

for example if R1 and R2 are not equal ,the error for 1% resistors does not equal a 1% Voltage divider percentage error

Can you provide an updated total ADC error using Vout change % for resistor tolerance at 3% ,

taking into account Vref error of 1.5%,2 LSB error for 10 bit ,vout change % due to resistor tolerances and any other eror that i may have missed

7416.R_divider_percent_error_calc 20250820.xlsx

Regards

Jonathan Kabangu

0 Dennis Lehman 4 months ago in reply to Jonathan Tshimama-Kabangu

TI__Guru 71216 points

Hi Jonathan,

Jonathan Tshimama-Kabangu said:
is it correct to conclude that you must always calculate voltage divider percentage error?

I would say if you want a true worse case error for your measurements, then yes, you must include all sources of error. ADC error, Vref error, and R divider error.

Ignoring the error in the voltage divider I would think is important to consider. I see from earlier post you are measuring 12 v or 24 v. How important is it to accurately measure 12v or 24v? What precision is required?

I don't know what you ultimately do with the measurement but would a 3% error cause your system to now work?

If you agree with how the errors are calculated, and using the calculations you added to the spreadsheet (BTW thanks for the contribution) then calculating the measurement error due to 3% error in the resistors, I leave this exercise to you. ;)

**Attention** This is a public forum

MSP low-power microcontrollers

MSP low-power microcontroller forum

MSP430F5310: MSP low-power microcontroller forum