Something about Digital to Analog Converter

IanP said:

I can see two possible reasons for not having 0V at the outputs:

- zero-scale error --> ( fig.18 in the data sheet ) gives you the output voltage error of ≥0.5mV, depending on temperature ..

- "Ground Reference Point for All Circuitry on the Part", which is pin labelled GND;
this pin has to be used as 0V reference for all outputs, and any measurement should be refered to this pin ..

So in conclusion, if you have an error of ≈1mV, or less, you will have to leave with it as it looks it will be always there ..

Regards,
IanP

Click to expand...

cyw1984 said:

IanP said:

I can see two possible reasons for not having 0V at the outputs:

- zero-scale error --> ( fig.18 in the data sheet ) gives you the output voltage error of ≥0.5mV, depending on temperature ..

- "Ground Reference Point for All Circuitry on the Part", which is pin labelled GND;
this pin has to be used as 0V reference for all outputs, and any measurement should be refered to this pin ..

So in conclusion, if you have an error of ≈1mV, or less, you will have to leave with it as it looks it will be always there ..

Regards,
IanP

Click to expand...

Yes,
I knew that the IC has a zero code error,the data sheet is said about 0.5mVpp
but, I had an error over then 2mV

Besides,the reference gound is common with the all circuity ground
How to solve it~~~Please~~~

Click to expand...

IanP said:

Try to use CLR (pin 9) with Clear Code Register bits CR1 and CR0 ( Table 13. and 14. ) to find out whether this error is present for 0000h (CR1=0, CR0=0), 8000h (CR1=0, CR0=1) and FFFFh (CR1=1, CR0=0), and if it is, you will need to add -2mV offset ..

Regards,
IanP

Click to expand...