Sampling in RasPi3 and ADS1115 using Sample count or time.time()

Question

I'm using Raspberry Pi 3 and ADS1115, my project requires me to get evenly spaced samples so as to plot and analyse. The other posts were about achieving 10k and 50k sps but I only require 500SPS and that isn't working either. Is there a way to run my code for 120 seconds with 500 sps and get 60,000 samples in the end from both A0 and A1 channel at the same time? I have attached the code for reference. Thanks in advance

from Adafruit_ADS1x15 import ADS1x15
import time
import numpy as np

pga = 2/3                  # Set full-scale range of programmable gain  
                            # amplifier (page 13 of data sheet), change 
                            #depending on the input voltage range
ADS1115 = 0x01              # Specify that the device being used is the 
                            #  ADS1115, for the ADS1015 used 0x00
adc = Adafruit_ADS1x15.ADS1015()   # Create instance of the class ADS1x15

# Function to print sampled values to the terminal
def logdata():

    print "sps value should be one of: 8, 16, 32, 64, 128, 250, 475, 860, 
    otherwise the value will default to 250"

    frequency = input("Input sampling frequency (Hz):     ")    # Get 
                                               #sampling frequency from user
    sps = input("Input sps (Hz) :     ")                        # Get 
                                           # ads1115 sps value from the user
    time1 = input("Input sample time (seconds):     ")          # Get how 
                                           #long to sample for from the user

    period = 1.0 / frequency        # Calculate sampling period

    datapoints = int(time1*frequency)       # Datapoints is the total number 
                               #of samples to take, which must be an integer

    startTime=time.time()                   # Time of first sample
    t1=startTime                            # T1 is last sample time
    t2=t1                                   # T2 is current time

    for x in range (0,datapoints) :     # Loop in which data is sampled

            while (t2-t1 < period) :        # Check if t2-t1 is less then 
                                     #sample period, if it is then update t2
                t2=time.time()              # and check again       
            t1+=period                      # Update last sample time by the 
                                            #  sampling period

            print adc.read_adc(0, pga, data_rate=sps), "mV      ", ("%.2f" % 
(t2-startTime)) , "s"      # Print sampled value and time to the terminal

# Call to logdata function
logdata()

Answer 1

1) Are you using the ADS1115 ???

adc = Adafruit_ADS1x15.ADS1015()   # should be adc = Adafruit_ADS1x15.ADS1115() 

2) you can't read two or more single-ended channels at the same time. In differential mode one can compare two channels to yield one value. To read a value from channel 1, besides channel 0 you would have to add another call in your loop:

print adc.read_adc(0, pga, data_rate=sps) ..... # original call for channel 0
print adc.read_adc(1, pga, data_rate=sps) ..... # new call for channel 1

3) before a value can be read, the ADS must be configured with several parameters, like channel, rate, gain etc. After some time, which is needed for doing an analog/digital conversion, values can be read, once, or over and over again, when in continuous mode. In the original ADAFruit library this time is calculated from the data rate (insecure) and in the recent port to CircuitPython this time will most often be around 0.01 s, because the conversion will most probably not have finished directly after configuration (check the _read method).

4) reading at 500SPS is about the fastest the ADS1115 can read. The reference states that conversion at 860SPS takes 1.2 ms. Adding time for configuration and reading you will not be able to read two or more values continuosly every 0.002s, even if you were receiving notifications for conversion, like I described on my homepage, instead of waiting for a fixed period of time.

5) I think the closest you can get with Python is to run 2 daisy-chained ADS1115 in continuous mode with GPIO notifications, but I have no experience with that.