Dual comparator closest to LM311

[devinw1]

There are 2 different DG403 in TH (on mouser anyway). There's the regular Vishay one and then Maxim's versions. The regular one is ok, but it has pretty high charge injection (60pC). The Maxim one is crazy expensive ($11) but looks improved on the specs. I haven't used this one.

If you're ok with +/-8V or +16/0V supply, the MAX4053 I mentioned is pretty solid. And you get 3 instead of 2 switches per package for about $4.

[guest]

i like the HC4053, it has slightly better specs than the CD4053, but at the cost of decreased rails (+/-5V typically). but, compared to the 4066 and other switches, it has a digital ground, so you dont have to level shift your logic.

im guessing you already considered this, but the 555 has two comparators and a flipflop inside. the LMC555 is a bit faster than the LM version, but i think the delay times are around the same as an LM339 for the LM555.

[devinw1]

Yeah, 555 is cool, but it uses dividers from the supply rails for the trigger points. Kind of defeats some of the purpose for a precision shcmitt trigger in this case without being able to have a stable 5V or whatever reference IMO. Other than that though, yes 555s are pretty awesome!

Will check out the HC4053. Regular CD4053 has logic level conversion btw...only the 4066 doesn't I think.

[elektrouwe]

HC4053 alternative DG4053E : <0.3 pC charge injection.

[bgreeves]

As someone who has only ever done TL07x and said "good enough," I'm curious for what applications you would need a dedicated "comparator" op-amp for. My hunch based on what I'm reading here is that they have faster slew rates / can switch faster? For what types of applications does this matter enough to the point where TL07x becomes not enough?

[guest]

the LM311 can also be wired so its output rails are the digital rails, and not the analog rails. this makes it easier to interface to logic and switches and such. the output also has more defined saturation points that are closer to the digital rails. but, if those arent required in a design, then the speed is the useful bit. in this case its for a VCO, since delays in the reset times can cause frequency shifts. there are also applications where you get glitches if the comparator isnt fast enough - like on an optional inverter or waveshapers. the comparator determines when its time to invert, but this is slightly delayed, and you get a small bump because the input signal wasnt exactly at 0 anymore when that happened.

[devinw1]

Right, as guest said, the speed is the huge thing. Comparators are designed to do one thing and they do it very very fast. An op amp can be run open loop and act like a comparator, but it will never be as fast as a comparator. And then, especially with models like the LM311, the output is very flexible. It has not only an open-collector output, but the ground of the output stage is brought out as a pin as well. So you can configure the output to do a lot of different things and interface it with a lot of different things (CMOS, TTL, etcc).