2014-12-16 05:16 AM
I have a problem of crosstalk between sequence channels scan automaticly on the ADC3,2 of the MCU. It is look like it is depend on the sequnce of the sample becase the voltage changes on the first scan channel influence the second one scan but not vice versa. So it is look like the internal sample cap didn't discharge between the sampling. So I prolong the sampling time for each of them from 3cycles to the maximum of 480cycles and also lower the ADC clk from dev2 to div8 but didn't help.
1. Is someone see that phenomena and get to solve it?2. Is there anther setting can reduce this influence?ThanksBar. #crosstalk-stm32f205-adc2014-12-16 05:31 AM
So it is look like the internal sample cap didn't discharge between the sampling.
Quite possible. But this is no crosstalk, but a common component - there is one S&H capacitor for all channels of one ADC.So I prolong the sampling time for each of them from 3cycles to the maximum of 480cycles and also lower the ADC clk from dev2 to div8 but didn't help. If longer sample times do not help, your input impedance is far too high, at least for the affected channels. Use buffer amplifiers (e.g. opamp in voltage follower mode).2014-12-16 06:40 AM
2014-12-16 07:42 AM
Can the source in question supply enough current to charge/discharge the capacitor? Do you need a buffer?
Nevermind, missed the earlier responses...