1
\$\begingroup\$

A force traducer has a resolution of 0.0625N. And the analog output sensitivity is 32N/V. So I calculate the voltage output resolution as 0.0625/32 which is around 2mV.

If this signal now is sampled by an ADC with an effective resolution of 0.5mV or 4mV what would be the benefits or disadvantages? What would be the advantage of going better ADC resolution or disadvantage going lower ADC resolution?

(I'm neglecting any other external noise)

\$\endgroup\$
  • \$\begingroup\$ Going towards lower resolution ADC can save you a lot of power. It depends on your application, really. Are you looking to capture the smallest (0.0625N) change in force ? \$\endgroup\$ – shreyas Apr 23 at 8:53
  • \$\begingroup\$ Yes I wonder how to choose the adequate ADC resolution for that change in force \$\endgroup\$ – atmnt Apr 23 at 9:03
  • \$\begingroup\$ @atmnt This sounds an awful lot like a hypothetical. Which means probably some homework essay to show how much you might remember from what's been presented to you, instead of a real project ahead. Your use of "resolution" and "sensitivity" also seem a bit misguided to me. For example, a transducer resolution specification without a bandwidth specification pretty much isn't useful. So the first thing we'd normally want to know is the datasheet for your transducer and the bandwidth you need. But I suspect this is all a hypothetical question and you are supposed to write a bit of an essay. \$\endgroup\$ – jonk Apr 23 at 18:22
  • \$\begingroup\$ @jonk Not an homework or essay really. The thing I wonder(was a discussion with a friend) is a general question and is easier to ask with an example fx: If the minimum variation(min step or call it resolution) of the analog input signal is lets say 10mV, what should be our ADC's resolution limit? Would 5mV ADC resolution suffice?(Im neglecting any other noise or assuming random noise is around 1mV) \$\endgroup\$ – atmnt Apr 23 at 22:19
  • \$\begingroup\$ @atmnt Questions can be simplified just so much. If over-simplified, all you get are nonsense answers to a problem not characterized, but instead caricatured. Like cartoon people are to real people. So answers lose meaning. Take your example and tell me what the raw resolution is and the inherent bandwidth limitations, or if you just mean the static (near DC) resolution. Then tell me how often you need a measurement. That will help some. Also, is the noise white? If white and if you are allowed to average, and if you want increased precision, then you get different answers than otherwise. \$\endgroup\$ – jonk Apr 23 at 22:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.