[Test-Equipment] How to do Time interval on freq. Counter

[John Kolb]

Paul Cianciolo wrote:
> Folks,
> 
> I have a freq counter that works great as a freq. counter but I would like
> to try the Time interval function.
> 
> I have 2 10Mhz osc. that have a difference of about 20 Hz.  When connected
> to the HP 5326A and I look at the TI between them
> it shows 14 nanosec???

The TI function should be measuring the time between the rising edge 
zero crossing of the A channel compared to the rising edge zero crossing
of the B channel. The edge polarity and trigger point may be adjustable.
Thus if measuring two signals of slightly different freqs, the TI should 
be displaying changing numbers from 0 to 100 nS for 10 MHz inputs. A 
fixed 14 nS reading sounds like something broken or counter not adjusted 
correctly.

> when I input A & B from the same 10 Mhz clock it still shows 14 Nanosecs.

If the counter is working correctly, a 14 nS reading could indicate
a trigger level adjustment not identical for the two channels, or a
delay difference for the two inputs, a 14 foot difference in input
cable length for example ( signals on a PCB go about 1 nS per foot as
I recall). For 0 phase difference, use a "T" connector and identical
cables from the two sides of the "T" to the A and B inputs.

> The TI number differs on when the freq of the oscillator is changed.
> 
> Do I not understand what is going on?   tThe assumption here is that if both
> signals are same freq. the diff should be zero??
> 
No, see above. The phase difference between two different oscillators
of the exact same freq could be any number.

John

> The period works ok and is on the money.
> 
> Can anyone point me to a refence on how this is supposed to work?
> 
> Thank you
> 
> PaulC
> 
> W1VLF