LTC1298 Reports Raw Readings That Are Offset by 25-27 Counts
JavelinUser
Posts: 1
Hi Everyone,
We have been working with the LTC-1298 A\D Converter using the standard parallax's java classes (appnote 8 and appnote 6) for some time, and had not noticed a problem when·running the program in such a way that it reports only the·individual channel readings.· However,·we·tried·running the program in differential mode by using the built in classes,·and it appears that the voltage difference in raw counts is always offset by 25-27 counts or·approximately 0.032 volts from the correct value.
Let me give you an example.· When·running the·VoltageComparer program that comes·with appnote·8,·setting the potentiometers so that both channels read the same voltages, i.e.·1.224 volts, the counts at·that point or lastRaw value should be 0·on channel 3 (which is difference of ch1 from ch0), but is really reporting about 25 counts instead.
Another scenario is·when the·voltage in channel 0 =·3.340 and channel 1 =·2.340·(1 volt difference),·the difference is channel 3 (differential channel) should be 819 counts but 819 - 25·=·794 counts is reported and thus·about 0.960v are reported in the difference instead of 1.00v.· You can see that the offset is about·0.032v or 32 milli-Volts which is equivalent (converting to counts) 0.032·V·*·819 count·/ 1 V = 26 counts.· I am ruling out my LTC1298 chip
being bad since I have tried 2 already.
I have also notice this behavior only when the·voltages on both channels is greater than zero.
When one of the channels is·0 and the other is greater than·0·volts, it·does report the·correct answer.·
I just can't figure out what is going on.· I would really appreciate any input with this problem.
Thank You
Milton
·
We have been working with the LTC-1298 A\D Converter using the standard parallax's java classes (appnote 8 and appnote 6) for some time, and had not noticed a problem when·running the program in such a way that it reports only the·individual channel readings.· However,·we·tried·running the program in differential mode by using the built in classes,·and it appears that the voltage difference in raw counts is always offset by 25-27 counts or·approximately 0.032 volts from the correct value.
Let me give you an example.· When·running the·VoltageComparer program that comes·with appnote·8,·setting the potentiometers so that both channels read the same voltages, i.e.·1.224 volts, the counts at·that point or lastRaw value should be 0·on channel 3 (which is difference of ch1 from ch0), but is really reporting about 25 counts instead.
Another scenario is·when the·voltage in channel 0 =·3.340 and channel 1 =·2.340·(1 volt difference),·the difference is channel 3 (differential channel) should be 819 counts but 819 - 25·=·794 counts is reported and thus·about 0.960v are reported in the difference instead of 1.00v.· You can see that the offset is about·0.032v or 32 milli-Volts which is equivalent (converting to counts) 0.032·V·*·819 count·/ 1 V = 26 counts.· I am ruling out my LTC1298 chip
being bad since I have tried 2 already.
I have also notice this behavior only when the·voltages on both channels is greater than zero.
When one of the channels is·0 and the other is greater than·0·volts, it·does report the·correct answer.·
I just can't figure out what is going on.· I would really appreciate any input with this problem.
Thank You
Milton
·