Hi Everyone,

I'm using the ADC with a sample-and-hold circuit. There is no buffer to the ADC input, just a 0.01uF capacitor. I have set SHTIM and STARTUP to the correct values. Every conversion yields the correct result EXCEPT for the first conversion after a reset. I have looked at the ADC input on a scope during the first conversion and have seen that the analog voltage level shifts positively during the conversion. It almost looks like the ADC pin is either sourcing current or dumping charge into the capacitor. Again, this only happens on the first conversion. I am using a software trigger to start the conversion and waiting for the EOC bit. I've seen this with a SAM7SE256 and a SAM7SE512. Has anyone seen anything like this?

Thanks,

Pat

Hello,

How soon are you performing the first conversion out of reset?

The pull ups on the IO lines are ON until you reconfigure them (ie. set up for ADC).

You may be seeing the residual pullup voltage on the line for your ADC channel.

You could try setting the ADC line as input and disabling the pullups then give a slight delay before configuring the ADC channel (i.e. enough time that the pullup voltage on the line has had time to discharge, TC).

Or you could just throw away the first couple of readings...

Regards,

Thanks for the suggestion, Duane. Actually, the AD pin I'm using is not multiplexed with the PIO controller; it is solely an AD pin. I have no idea what could be causing this, and I've noticed that, depending on the physical microcontroller sample, the first conversion can be "a lot" wrong or just "a little" wrong. I think your second suggestion will have to be the solution - just ignore the first conversion!

Thanks,

Pat