Hi,
I am a little confused in the exact relationship between an ADC's clock and it's sampling rate. According to what I understand, an ADC's clock determines the amount of time (in cycles) it takes to start a conversion i.e. just enough time to get the ADC's capacitor charged up and running.
Hence, its clock would define the minimum time between two successive conversions. This is where I get a little lost. Two successive conversions would mean two successive sample acquisitions and isn't the time between consecutive samples is what we know as the sampling time? I am surely missing something here.
To provide some context - the ARM7 LPC2148 processor user manual says the 10-bit ADC clock can provide a clock speed of 4.5 MHz max. advising to choose the divider in such a way so as to keep the clock as close to 4.5 MHz as possible for better conversions.
Somewhere else, it mentions the time taken for a 10-bit conversion as >= 2.4 us. I understand the correlation between these pieces of data (1/4.5 ~= 2.4) but I am unable to arrive at my desired sampling rate (8k samples/sec) using it.
I'd be glad if I could get some help understanding this.
I am a little confused in the exact relationship between an ADC's clock and it's sampling rate. According to what I understand, an ADC's clock determines the amount of time (in cycles) it takes to start a conversion i.e. just enough time to get the ADC's capacitor charged up and running.
Hence, its clock would define the minimum time between two successive conversions. This is where I get a little lost. Two successive conversions would mean two successive sample acquisitions and isn't the time between consecutive samples is what we know as the sampling time? I am surely missing something here.
To provide some context - the ARM7 LPC2148 processor user manual says the 10-bit ADC clock can provide a clock speed of 4.5 MHz max. advising to choose the divider in such a way so as to keep the clock as close to 4.5 MHz as possible for better conversions.
Somewhere else, it mentions the time taken for a 10-bit conversion as >= 2.4 us. I understand the correlation between these pieces of data (1/4.5 ~= 2.4) but I am unable to arrive at my desired sampling rate (8k samples/sec) using it.
I'd be glad if I could get some help understanding this.