Talk:Integrating ADC

Enhanced dual-slope
Is there a source for the so called "Enhanced dual-slope" ? I don't think this modification is really important and may be more confusing than helping. The separate resistors are a 2 sided thing. It can be slightly faster, but is less accurate because of the resistor ratio. Normally having the same resistor for run-up and run-down is more of an advantage of the classical dual slope ADC. --Ulrich67 (talk) 21:50, 12 January 2019 (UTC)
 * I found a reference to the Enhanced dual-slope in the 2nd reference ( HP Journal, 40 (2): 8–15). However I this was so far the only place to find it. I would see this more as a first step to explain the multi-slope ADC, not as a practical enhancement, that would be worth using. There is a practical and common step in this direction, by having the full scale at twice the reference voltage. So unless someone finds a good reason to keep this section, I suggest removing it all together.--Ulrich67 (talk) 15:10, 23 February 2019 (UTC)

Enhancements section issues
In the section titled "Multi-slope run-up", we read "For example, assume that we are capable of measuring the charge on the integrator during the run-down phase to a granularity of 1 coulomb." Subsequent sentences use, as an example, a total charge capacity of 32 coulombs.

32 coulombs is an absurdly large quantity here. Given a typical fully charged voltage of between 1V and 10V, that suggests a capacitance of 3F to 32F, absurdly large capacitances. I'm guessing actual capacitances in the range of nF, so something like 1E8 to 1E9 difference.

I get that the writer was trying to use simple numbers, but these are so far out as to distract from the point. It would be like discussing microscope measurements and giving "one mile" as an example resolution.

A further issue is that C is used for the Capacitor symbol (and presumably value), but in the narrative, C is sometimes used for capacitance, and sometimes for charge, which is highly confusing. Instead, charge should be represented by some other letter; Q is customary. Gwideman (talk) 04:24, 17 February 2020 (UTC)

·The example is indeed off by several orders of magnitude. In addition the resolution for the charge for the run-down is not fixed. A main point of the multi-slope run-up is that a smaller capacitor can be used and thus the resultion for the charge improves. The input current to the integrator is more fixed and not going up. Ulrich67 (talk) 11:49, 22 November 2021 (UTC)