[sdiy] In search of a reasonably precise comparator output

[Guy McCusker]

Aha, thank you Lloyd. It looks like it is time to shift from thinking
to doing and see what happens when I build one or two of these
approaches.



On Thu, Jun 21, 2018 at 9:22 AM, Lloyd Abbott <lloydabbott94 at gmail.com> wrote:
> If you're worried about the LM339's V_OL, the datasheet mentions, "The V_OL
> is resistive and scales with the output current." The 150mV to 400mV
> specified is measured with I_OL = 4mA, if you can happily reduce the output
> current down an order of magnitude or two from 4mA, then you're dealing with
> much lower offset.
>
> On Wed, Jun 20, 2018 at 10:30 PM, Guy McCusker <guy.mccusker at gmail.com>
> wrote:
>>
>> Thanks everybody for the suggestions. I am sort of relieved that there
>> is not one completely obvious way to do this but a range of
>> possibilities -- I guess I just have to choose one, or experiment a
>> bit.
>>
>> The only one I won't make use of in this case is the idea of doing
>> this digitally, but only for the not very good reason that the overall
>> circuit I am aiming for would be soooo much easier to do digitally
>> (and better too I imagine) but I am trying to create an analog
>> solution for the fun and pain of it.
>>
>> Thanks again!
>>
>>
>>
>>
>>
>> On Wed, Jun 20, 2018 at 8:58 PM,  <d1ffeq at teknik.io> wrote:
>> > So it's a voltage modulo function. All I can think of is a comparator
>> > controlling an analog switch that switches between original and -5V biased
>> > inputs. But this solution is not what I'd call simple. Maybe some sort of
>> > A/D converter would be simpler.
>> >
>> > June 20, 2018 8:27 PM, "Guy McCusker" <guy.mccusker at gmail.com> wrote:
>> >
>> >> It's an analog thing: what I really want to do is take an analog
>> >> voltage X and return X-5 if if X > 5, and just X otherwise.
>> >>
>> >> I did mention this in the first message, but right at the end, and in
>> >> parentheses :)
>> >>
>> >> On Wed, Jun 20, 2018 at 5:34 PM, Scott Gravenhorst
>> >> <music.maker at gte.net> wrote:
>> >>
>> >>> It would be helpful if we knew what was being driven by the
>> >>> comparator. Logic inputs
>> >>> don't require absolute 0.000 volts for a low logic level nor absolute
>> >>> Vcc for a high logic
>> >>> level. Each logic type will have it's own specification for that. Is
>> >>> this a zero volt
>> >>> and five volt source for some analog process - or is it digital?
>> >>>
>> >>> Guy McCusker <guy.mccusker at gmail.com> wrote:
>> >>>> --===============1895041468437579696==
>> >>>>
>> >>>> Thank you.
>> >>>>
>> >>>> I can readily stick a solid 5V reference on the pull-up of an
>> >>>> open-collector or open-drain comparator, but I am feeling stupid
>> >>>> about what
>> >>>> happens for the low output voltage. The LM339 data sheet lists a V_OL
>> >>>> of
>> >>>> 150 to 400mV. That's a reasonable way above 0V.
>> >>>>
>> >>>> Should I just live with it and trim it somewhere else in the circuit?
>> >>>>
>> >>>> Guy.
>> >>>>
>> >>>> On Wed, 20 Jun 2018 at 17:02, Pete Hartman <pete.hartman at gmail.com>
>> >>>> wrote:
>> >>>>
>> >>>> http://pi3c.lan/cgi/mailbody.cgi?MD4696.MSG=+Get+&MAILBOX=chordman&DATE=20_06_2018_04_00_20
>> >>>
>> >>> On Wed, Jun 20, 2018 at 10:30 AM, Lloyd Abbott
>> >>> <lloydabbott94 at gmail.com>
>> >>> wrote:
>> >>>
>> >>> What rails do you have available ? Have you considered a comparator IC
>> >>> (e.g. LM339) rather than using general purpose op-amps ?
>> >>>
>> >>> If you have 0V, +5V and +15V rails, with an LM339 you can reference
>> >>> the
>> >>> comparator input side to the 0V and +15V rails and have the output
>> >>> side
>> >>> referenced to 0V and +5V. The datasheet has an example circuit using
>> >>> one
>> >>> pullup resistor and one cap per comparator.
>> >>>
>> >>> I was going to suggest this as well. The 339 and its family are all
>> >>> open
>> >>> collector outputs, so you can use a pull up to any voltage you like, a
>> >>> rail
>> >>> or a reference voltage or whatever to get the swing you need and if
>> >>> necessary convert levels.
>> >>>
>> >>>>
>> >>>
>> >>> -- ScottG
>> >>>
>> >>> ________________________________________________________________________
>> >>> -- Scott Gravenhorst
>> >>> -- http://scott.joviansynth.com
>> >>> -- When the going gets tough, the tough use the command line.
>> >>> -- Matt 21:22
>> >>>
>> >>> _______________________________________________
>> >>> Synth-diy mailing list
>> >>> Synth-diy at synth-diy.org
>> >>> http://synth-diy.org/mailman/listinfo/synth-diy
>> >>
>> >> _______________________________________________
>> >> Synth-diy mailing list
>> >> Synth-diy at synth-diy.org
>> >> http://synth-diy.org/mailman/listinfo/synth-diy
>> >
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy
>
>