[time-nuts] Simple AC mains zero-cross detector

[Chris Albertson]

There is no need for a transformer here.  Yes  you want isolation but
an optocoupler in a 8-pin DIP package can do that job better.

You can reduce the 120 VAC volts down to anything you like by clipping
it with diodes then send the clipped AC to the coupler and

The usual method if edge dietitian is to simply connect the 60Hz
signal to the DTR line of a serial port on the computer.  The computer
then make a time take file.  Accuracy of this is roughly a couple
uSec.

On Sun, Dec 21, 2014 at 10:04 PM, nuts <nuts at lazygranch.com> wrote:
> On Sun, 21 Dec 2014 17:37:07 -0500
> Charles Steinmetz <csteinmetz at yandex.com> wrote:
>
>> Ed wrote:
>>
>> >It seems to me that a low voltage secondary should be OK by using a
>> >fast comparator IC rather than a transistor to decide - the gain of
>> >the IC allows for much smaller detection levels, so the equivalent
>> >zero-crossing velocity could be the same. An IC tripping in a 10 mV
>> >band should provide the same effective ZC velocity at 12 V input as
>> >a transistor working around 100 mV with 120 V input. Or am I missing
>> >something?
>>
>> When the switching band gets that small, device noise, input offset
>> voltage drift, and other errors have a proportionally greater
>> effect.  I actually built a similar circuit with a 12v transformer
>> and an LT1720 comparator, and it had worse jitter than the
>> two-transistor circuit with a 120v feed.  In this case, there is no
>> substitute for starting with a higher-slew-rate signal.  (Yes, the
>> LT1720 did marginally better than the two-transistor circuit when
>> both were fed from 120v -- but the fussiness of working with a fast
>> comparator and the small gain over the two-transistor circuit made
>> the latter the better choice, particularly in a design being put "out
>> there" for others to build.)
>>
>> Best regards,
>>
>> Charles
>>
>
> Looking at the data sheet of the LT1720, 1mv would have about 8ns
> delay. Call it 10ns. A Vp of 29 volts should be sufficient to put the
> delay around 90ns, making 100ns error or target percent of the 1uS
> target.
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.



-- 

Chris Albertson
Redondo Beach, California