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.
Hello,
I don't see in the data sheet anything about the ADS1118 on-board temperature sensor and data sheet. I have been sampling it with the 250 sps rate just as with my analog inputs. It is very clean. I was writing code to support a TMP006 yesterday and couldn't help but notice it has essentially the same charts, suggesting to me that perhaps the on-board temperature sensor is separate from the A/D? If so, what rate is best to sample the temperature sensor, or can I get readings back from it immediately without a delay?
Thanks
James
Hello James,
The ADS1118 on board temperature sensor uses a diode based temperature sensor. The accuracy guaranteed for the on-board temperature sensor is 0.2C typical (+/-0.5max) from 0C to 70C; and 0.4C typical (+/-1C) from -40C to 125C as shown in the table. The resolution (1LSB) of the on-board temperature sensor is 0.03125 C/bit. The data rate of the temperature sensor measurement is the same as the one programmed to read the analog inputs. The accuracy of the on-board temperature sensor is not affected by the data rate.
When looking into the variation of the on-board temperature measurements with different data rates using the ADS1118EVM, I see a very steady temperature measurement that may only move 1 LSB or 0.03125C (or not move at all). Since the LSB weight is only 0.03125C; this is much smaller than the guaranteed accuracy; therefore you should be able to read the on-board temperature measurement at the same chosen data rate of your analog inputs without worrying about degradation in accuracy of the on-board temp sensor reading.
Thank you and Regards,
Luis
Thanks. I will move it up to 860 sps then.
Do you know if perhaps this is because the temperature sensor is a separate block with a separate A/D? The similarities on the TMP006 data sheet made me wonder this. I am using timing to determine when to requery, and I am curious if it would be possible to query even faster than 860, which would be logical if this were true and it is just a unified interface between these two chip functions. Admittedly, 3 channels at 250 sps and temperature at 860 being 76 Hz, there's only another 7 Hz to be gained (at most) if this were possible.
Hello James,
The internal temperature sensor and the internal PGA of the ADS1118 share same ADC, therefore the maximum rate is 860SPS. The user may select one measurement at a time, either the PGA or the internal temperature sensor for each measurement. One method to perform thermocouple measurements is to interleave the thermocouple input reading with one internal temperature sensor measurement. In applications/environments where the temperature of the cold junction is expected to be very stable, a longer period between each temperature sensor reading (cold junction reading) may be performed.
Noise may be picked up and injected at the inputs of the ADS1118 PGA; specially when long length thermocouple connections are used or the thermocouple is in a noisy environment. The user must account for the differential noise picked up by the thermocouple, try to eliminate sources of noise and include the appropiate anti aliasing filter. The ADS1118 datasheet figure 43 provides some recommendations for a first order RC filter at the inputs of the PGA. The user will typically select the optimal RC filter corner and sampling rate according to the environment and application needs.
Best Regards,
Luis