Programmable S2C : ADC not accurate

Question

Programmable S2C TH : stabile vref 2500 at pin 14 / hardware circuit with precission resistances 7k5 and 15k that give us voltages at pin 17/18/20.
The resistance-measurements (avg of 10 adc-measurements) :
- 7k5 : 7382 / 7390 / 7374 ohm
- 15k : 14453 / 14467 ohm
Why this is so low and not closer to 7k5 and 15k?
If we measure the voltage on the xbee-pin (pin 17/18/20), we get voltage values that do corresponded to the 7k5 and 15k.
So, the voltage outside measured on the pins does not correspond exactly to the adc-value? it is approximately correct, but you should expect better. So, i think, maybe we do something wrong?

Answer 1

According to the manual, the max voltage is 1.2V. Not 2.5.

Comments

If i check the userGuide for S2C, then there is a topic 'hardware specification for the programmable variant' on page 20. The last line in the table says -> Vref Range : 1.8 VDC to Vcc. So, that's why we used 2.5V. Please advise.

---

The Programmable variant refers to the version of the product that comes with a 2nd processor that you write code and port that code to. If you check your part number, a programmable version will have a part number that ends with something like B004 where by a Non Programmable would be -004. Notice the B where the - is. That tells you which one it is.

---

Yes, I am talking about the programmable S2C TH (clearly stated in title and question)
The part-number ends with ...WITB003.
In the mean-time i checked with VREF 1250 and 2500 and various input voltages. Both give a quasi-linear ADC progress (so it can be calibrated by a gain/ofset). But, as i stated in my question : the adc-value that is returned deviates 10-20 units according to input voltage (measured on the xbee-pin)  ...

---

Mark, please pardon me double checking.  You don't know how many people call the Standard module a Programmable module so I had to verify.

Now that we have verified that, the ADC function then falls to the Freescale processor and not the RF processor. The RF processor still only supports up to 1.2V. So if you are going to use a high voltage, then the ADC function on the RF processor needs to be disabled and you need it to add it to the Code on the FreeScale processor.

---

Ok, that's clear.
I checked the configuration of the Xbee itself in XCTU and all ADC-related settings seems to be disabled by default. That's why it worked for me with VREF 1250 and 2500 (using already the ADC of the FreeScale).
Are the following XB settings ok : PR=E81 / PD=1FFF?
To summarize : i use the FreeScale-ADC which can have higher voltage VREF (than 1.2V). Stays with my initial question : why this ADC seems not 100% accurate to the voltage measured on the pins. (i measure less). Or can this be related to the PR/PD settings ? Some small voltage-drop on a resistance?