Electronic – How to chose the right decoupling/bypass capacitor variant for comparator power rails

An LM358 will function fine in this circuit, however the output rise and fall times will be limited by the slew rate of the op-amp, typical more than 15usec. That will limit the frequency response, and not all counters will be happy with such long rise and fall times in a clock input. For example, the 74HC4040 specified maximum is 139ns/volt, which is way (more than 20x) faster than the 0.3V/us typical slew rate of an LM358. It could be cleaned up with a Schmitt trigger gate if required, however frequency response is unlikely to be more than a few 10s of kHz.

Further limiting the input frequency response with an RC on the non-inverting input could be desirable in some cases (maybe ~10kHz or less cutoff). The reason for considering this is that the positive feedback is delayed by the glacial slew rate of the output, and in the meantime the comparator is vulnerable to fast variations in the input. Here is the response near switching of the comparator with about 0.23V of hysteresis, a slow input ramp with a 10kHz 0.2Vp-p square wave superimposed:

And the same with a 10K/1nF filter ahead of the comparator (16kHz cutoff).

As Nick Johnson says, an op-amp as a comparator is sometimes possible (some op-amps are even worse suited) but usually not desirable except in very special circumstance such as where precision is more important than speed.

Op-amps have front ends that may be ill-suited for large differential voltages (the LM358 is okay), and most are internally compensated for stability at unity gain so they behave quite sluggishly. In the old days we would sometimes use uncompensated op-amps as comparators, but there are much better parts available these days.

Let me preface this with: decoupling is generally something you do as a rough judgement call based on theory principles; I haven't ever done it by analysis at a board level (versus IC level, where I have done back-of-the-envelope time-constant analysis). If you later find that you used too many caps (brings cost up) or not enough/not the right values (performance problems), then you'll adjust the caps experimentally—it's good to leave extra pads on PCBs for additional parallel caps you think you might need if things go wrong, at least on prototypes, sometimes even on a final PCB.

To answer your questions:

1. It depends on transient currents and distance (or more to the point, trace inductances). For 10μF particularly but 1μF too, you can probably get away with a few mm or low cm of traces and use only a single capacitor, and I would probably go with only one on a prototype if the final space/BOM cost is an issue. Upgrade to the two caps if you find one only one causes issues, or if space permits leave some pads for two caps closer to the pins. The 0.1μF will be much more sensitive to distance so keep that one closer to the chip's actual pins.

2. Depends on distances and the amount of transient currents the cap has to filter. If your voltage regulator output is right beside this chip (and the transient currents aren't terrible—if they are maybe consider increasing the capacitance), you can just use one 10μF for the entire thing. If your voltage regulator is a handful of centimetres away, you'd want to put in a 1μF or 10μF closer to your STP24DP05, along with the one near the voltage regulator.

3. Excessive number of capacitors, no, except it costs more and takes more space! However, trying to put big capacitors instead of smaller ones can cause issues—large capacitance values help filter bigger transient currents which is good, but physically larger capacitors cannot handle fast current transients as well. Capacitors are less effective as frequency goes up, and physically big capacitors can't go as high in frequency due to higher equivalent series inductance; see this other question. This is why we put parallel capacitors of different values in the first place.

Someone once told me that, as a general rule, using parallel capacitors two decades (factor of 100) apart provides good coverage of frequency range. So 10μF and 0.1μF, or 1μF and 0.01μF.