Original question:
Hello,
When I capture data with Get_Capture_Data_32bits function from automation dll, I get the raw data around 36000. However, when I save it with Save_Raw_Data_As_CSV function, I get good converted raw data with small numbers, and I see it is much more better for plotting, and analyzing the results. I want to convert the data from Get_Capture_Data_32bits to data like that. Can I know which formula is used for this converted data in csv?
Thanks for quick response.
Here is the screenshot of the csv that I have saved with numpy after I collect the data with Get_Capture_Data_32bits.
And here is the csv that I have saved directly with the Save_Raw_Data_As_CSV function from HSDC Automation DLL.
As you can see the second csv is much better than first one. That is because it converts the raw data inside the dll that I don't know. My question is to find out that function for converting the data.
Hi Farid,
Import the attached file into HSDC Pro. ADCdataToImport.csv
To import this in HSDC Pro, click File -> Import Data File. Choose 2 channels, set the resolution to 16 bits, and make sure 2s complement is checked. (see image below)
Next, run the python script.
# -*- coding: utf-8 -*-
"""
Created on Mon Apr 11 10:30:34 2022
@author: CW
"""
def loadDLL():
"""This function loads the HSDC Pro 64-bit DLL.
\nArgs: {None}
\nReturn: None
"""
if not 'HSDCProDLL' in locals():
loadedDLL = cdll.LoadLibrary("C:/Program Files (x86)/Texas Instruments/High Speed Data Converter Pro/HSDCPro Automation DLL/64Bit DLL/HSDCProAutomation_64Bit.dll")
return loadedDLL
else:
pass
if __name__ == '__main__':
import ctypes
from ctypes import cdll
import numpy as np
default_timeout = 10000
NumChannels = 2
Number_Of_Samples_Per_Channel = 4096
HSDCProDLL = loadDLL()
HSDCProDLL.ADC_Analysis_Window_Length(ctypes.c_ulong(Number_Of_Samples_Per_Channel),default_timeout)
# Save data using CaptureData_32Bits approach:
Capture_Data_Array_Len = NumChannels * Number_Of_Samples_Per_Channel
CaptureData_32bits = (ctypes.c_ulonglong*Capture_Data_Array_Len)()
HSDCProDLL.Get_Capture_Data_32bits(Number_Of_Samples_Per_Channel,0,CaptureData_32bits,Capture_Data_Array_Len,default_timeout)
data = np.frombuffer(CaptureData_32bits, dtype = np.int32)
data_from_CaptureData_32bits = data.reshape(Number_Of_Samples_Per_Channel*2,2)[:][:Number_Of_Samples_Per_Channel]
data_from_CaptureData_32bits = data_from_CaptureData_32bits -(2**(16-1)) # convert from unsigned to signed
# Save data using the CSV save approach:
SaveCSVFilePathWithName = b"C:/Users/Public/Documents/Texas Instruments/High Speed Data Converter Pro/Example Files/Python Files/ADCdataNEW.csv"
HSDCProDLL.Save_Raw_Data_As_CSV(SaveCSVFilePathWithName,default_timeout)
#load CSV data to np array
import pandas as pd
csv_data = pd.read_csv(SaveCSVFilePathWithName.decode("utf-8"),header=None).to_numpy()
#compare results:
print(np.array_equal(data_from_CaptureData_32bits, csv_data))
This python script will save the data using both approaches. After subtracting the 2^15 offset, the data from the CaptureData_32bits reads the same as from the CSV approach. To download the python script, click on the upper right corner of the embedded code viewer.
Thanks, Chase