Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
Michael Karas
11/08/11 05:11
  11/08/11 05:12

Read: 393 times
Beaverton Or

#184584 - Some things to consider....
Responding to: ???'s previous message
John M said:
I planned on scaling a 0-5V Analog signal to a 0-3.3 volt signal for an ADC to read. I was going to use a couple resistors. The thing is, the Voltage signal is at a high frequency (~4kHZ) and will be sampled at around 10-12kHZ. Will I lose any resolution or add any noise to the circuit by doing this, or would I be better off getting an ADC with a 5V VREF?

Edit: BTW - You really should have started a new thread for a new subject.

You may not lose any resolution doing the scaling with a couple of resistors. However the two resistors will add to your error budget because they will have value tolerance and may exhibit some value fluctuation due to environmental factors such as temperature.

If the source impedance is not accounted for in the selection of the two resistors you scaling divider could end up compressing the signal voltage range such that 5V in does not equal 3.3V out. This would make it not possible to use the whole measurement range of the 3.3V A/D converter and thus some loss of resolution.

The discussion thus far assumes that the number of bits used for a 5V ADC would be the same as a 3.3V ADC.

Measuring a signal in a 3.3V range in a system with a given amount of noise will lead to more uncertainty in the ADC outputs than those from a measurement of a 5.0V range signal as signal to noise ratio becomes lower.

Adding the resistors increases the source impedance seen by the A/D converter. This could add errors due to down wind capacitance depending of the values of the resistors and stray capacitance in the circuit relative to the frequency of the signal. The increase of the source impedance could also affect the performance of the 3.3V A/D converter if it uses a channel selection MUX and/or has a relatively low input impedance.

Michael Karas

List of 13 messages in thread
Looking for 8051 that supports high speed I2C      John M      11/02/11 15:22      
   SPI      Per Westermark      11/02/11 16:06      
      gotcha      John M      11/02/11 21:03      
         3.4 MHz I2C      Andy Peters      11/04/11 13:13      
            RE: SPI is simpler      Andy Neil      11/04/11 14:03      
               One other quick question for all you smart people.      John M      11/07/11 20:24      
                  Some things to consider....      Michael Karas      11/08/11 05:11      
                     Full range and clipping      Per Westermark      11/08/11 05:24      
                        Impedance      John M      11/08/11 06:04      
                           You sure?      Per Westermark      11/08/11 06:18      
                           Impedance With Divider      Michael Karas      11/08/11 09:00      
                           Impedance matching      Andy Neil      11/08/11 15:54      
   if I recall      Erik Malund      11/03/11 07:13      

Back to Subject List