[time-nuts] 53132A triggering

[Scott Stobbe]

Well I have looked into this before for the 53131A, and the way signal peak
volts is implemented is bizarre. I say bizarre because for low frequency,
low duty cycle signals (like PPS) the 53131A will at times report the lower
peak as being higher than the upper peak...

>From the 53131A Manual:
     Peak volts (53131A, 53132A, 53181A) Measurement is specified on
Channels 1 and 2 for dc signals; or for ac signals of frequencies between
100 Hz and 30 MHz with peak-to-peak amplitude greater than 100 mV.

     Results range –5.1 V to +5.1 V
     Resolution 10 mV Peak volts systematic
     uncertainty for ac signals: 25 mV + 10% of V
     for dc signals: 25 mV + 2% of V

Well, I read the specification as being guaranteed for any DC coupled
signal, in my experience even DC signals with a frequency less than 100 Hz
fail to register correctly.

As an example at 1 kHz (10% duty cycle) the channel 1 VPP (DC, 50 Ohm),
behaves as expected:
 +0.00V +1.58V
 +0.01V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V
 +0.01V +1.58V
 +0.00V +1.58V
 +0.00V +1.58V

For 1 Hz (10% duty cycle) the channel 1 VPP (DC, 50 Ohm), behaves
unexpectedly:
 +0.05V +0.05V
 +0.05V +0.05V
 +0.06V +1.28V
 +0.05V +0.05V
 +0.05V +0.05V
 +0.05V +0.08V
 +1.28V +0.05V  <- Lower peak greater than upper peak
 +0.05V +0.05V
 +0.05V +0.05V
 +0.05V +1.54V
 +0.05V +0.05V
 +0.05V +0.05V
 +0.05V +0.05V
 +1.24V +0.05V <- Lower peak greater than upper peak
 +0.05V +0.05V
 +0.05V +0.05V
 +0.05V +0.64V
 +0.05V +0.05V
 +0.05V +0.05V
 +0.05V +0.05V
 +0.06V +1.28V


So the reason auto-trigger fails, is the instrument does not set an
appropriate trigger level, since the VPP reading fail to have any integrity
below 100 Hz.

I would be interested to see what you get on the 53132, I suspect its
implementation of VPP to have the same issues.


On Sat, Sep 17, 2016 at 10:28 AM, Bob Camp <kb8tq at n1k.org> wrote:

> Hi
>
> Set it up on your 53132 and see what happens …
>
> Bob
>
> > On Sep 16, 2016, at 8:35 PM, Scott Stobbe <scott.j.stobbe at gmail.com>
> wrote:
> >
> > For a low duty cycle pulse, the ac coupled signal will be approximately
> the
> > same as if it were dc coupled. Not sure I follow what you mean. There
> will
> > be only one rising edge for a narrow pulse ac coupled, as the falling
> edge
> > occurs much quicker than the HPF time constant.
> >
> > On Friday, 16 September 2016, Bob kb8tq <kb8tq at n1k.org> wrote:
> >
> >> Hi
> >>
> >> Most PPS signals these days are very low duty cycle. If you AC couple
> >> them, you can easily be triggering on the wrong edge. With the narrow
> pulse
> >> it may not be very obvious.
> >>
> >> Bob
> >>
> >>> On Sep 16, 2016, at 5:46 PM, Charles Steinmetz <csteinmetz at yandex.com
> >> <javascript:;>> wrote:
> >>>
> >>> Bob wrote:
> >>>
> >>>> Set it to:
> >>>>
> >>>> 1) DC coupled (AC does not go low enough)
> >>>> 2) 50 ohms if your driving source will tolerate it, otherwise 1 meg
> ohm.
> >>>> 3) Manual trigger mode (Auto is to fast and it forgets where the
> >> trigger should be)
> >>>> 4) Trigger level around 1/2 the PPS P-P voltage
> >>>
> >>> I would just add the following:
> >>>
> >>> 1)  I'd be very surprised if AC coupling wouldn't work fine with a
> >> typical PPS pulse, which has very fast edges (low nS).  No LF response
> is
> >> required.  Indeed, AC coupling will keep any LF noise out (not that we
> >> expect much in this application).  This is true even if the PPS is a 50%
> >> duty-cycle square wave -- the spikes that get through every 500mS,
> >> alternating positive and negative, will have fast, accurate leading
> edges
> >> and will be way longer than necessary for proper triggering.
> >>>
> >>> 2)  If your source will not tolerate a 50 ohm load, buffer it.  Any
> >> significant length of cable between the source and a 1M termination will
> >> just slaughter your pulse.
> >>>
> >>> 4)  The relevant peak voltage is the actual voltage at the counter
> input
> >> connector -- which may be only 1/2, or possibly even less, of the
> nominal
> >> logic level, depending on the source impedance.
> >>>
> >>> Best regards,
> >>>
> >>> Charles
> >>>
> >>>
> >>> _______________________________________________
> >>> time-nuts mailing list -- time-nuts at febo.com <javascript:;>
> >>> To unsubscribe, go to https://www.febo.com/cgi-bin/
> >> mailman/listinfo/time-nuts
> >>> and follow the instructions there.
> >> _______________________________________________
> >> time-nuts mailing list -- time-nuts at febo.com <javascript:;>
> >> To unsubscribe, go to https://www.febo.com/cgi-bin/
> >> mailman/listinfo/time-nuts
> >> and follow the instructions there.
> >>
> > _______________________________________________
> > 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.
>
> _______________________________________________
> 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.
>