AnsweredAssumed Answered

synchronizing two oscillosopes (DSO5034A)

Question asked by klw001 on Aug 26, 2009
Latest reply on Aug 29, 2009 by alexw

Like • Show 0 Likes0
Comment • 4

Hi,

I am wondering if anyone could recommend the best method to synchronize two DSO series oscilloscopes which are located 10km apart.

The two scopes are to sample the signal every minute.

Currently, the data acquisition is automated using Labview 8.0 and we have two units of GPS receivers from Garmin.

We know that there are other hardwares that we could acquire to perform this task accurately but we preferred to use these two scopes instead.

We could not figure out how to do the synchronization properly with minimum time mismatch.

Any suggestion will be highly appreciated!

Alan

No one else has this question

Outcomes

4 Replies

klw001 Aug 27, 2009 5:01 PM
Mark Correct
Correct Answer
Thank you for your reply.

I am wondering if it is possible to connect the PPS output of a GPS receiver to the external trigger port on DSO5034A to synchronize the two scopes?

Has anyone try this before? This will overcome the software delay.
Like • Show 0 Likes0
Actions
alexw Aug 27, 2009 8:44 PM
Mark Correct
Correct Answer
We need to know more to give a detailed answer. What hardware are you using, what kind of signal are you trying to measure, and most importantly, what do you mean by "synchronize"? What kind of precision? Is the whole concept of "simultaneously" even well-defined for the kind of signal you are trying to measure in two places 10km apart?

A little googling on your hint about the Garmin pps output suggests that something is possible. At least on some models, the pps wire, used in concert with the serial/USB/NMEA output, gives an alignment signal just /before/ the textual data that is accurate to within ~20 microseconds. The pps signal is 1Hz at "logic level" (a few V) into 50 ohms, pulse width variable and software-defined. You could certainly trigger off that, though you may need to solder up a connector.

-Alex
Like • Show 0 Likes0
Actions
klw001 Aug 27, 2009 8:58 PM
Mark Correct
Correct Answer
Dear Alex,

The current hardware setup that we have is:
Agilent DSO5034A oscilloscope connected to 4 capacitive couplers (1 coupler per channel)

The idea is to calculate the time-of-flight for a 10MHz pulse travelling on power distribution line (length~10km). To achieve this, we will need to set up two monitoring stations at both ends of the line.

By synchronizing the two scopes, we could extract the time delay information and in turn calculate the original location of the pulse.

There are specialised equipment to achieve this but usually these equipment are very costly.

Any suggestion? I think the GPS PPS signal has a accuracy of 100ns as specified in many sites.

Alan
Like • Show 0 Likes0
Actions
alexw Aug 29, 2009 12:06 AM
Mark Correct
Correct Answer
"
Dear Alex,
The current hardware setup that we have is:
Agilent DSO5034A oscilloscope connected to 4 capacitive couplers (1 coupler per channel)
The idea is to calculate the time-of-flight for a 10MHz pulse travelling on power distribution line (length~10km). To achieve this, we will need to set up two monitoring stations at both ends of the line.
By synchronizing the two scopes, we could extract the time delay information and in turn calculate the original location of the pulse.
There are specialised equipment to achieve this but usually these equipment are very costly.
Any suggestion? I think the GPS PPS signal has a accuracy of 100ns as specified in many sites.
Alan
"

    10000m /3e8 m/s = 33 microseconds (lower bound; your signal's group velocity may be somewhat slower).

    http://tf.nist.gov/service/gpscal.htm
     ... has some bad things to say about the short-term stability of the pps signal output by many receivers, even if the time signal is quite good when averaged over a day.

    I suspect that the Agilent scope is not your limiting factor; any scope should be able to trigger off the pps signal if you solder up an appropriate connector. The questionable part is the quality of behavior of your GPS receiver.

    I would be inclined to make a looped measurement back to the source and divide by two. Could you reflect the signal back along the line (use an open instead of termination at the far end (and what is the impedence of a power line at 10MHz anyways?)? Or, could you put a pulse generator at the far end, trigger off the incoming pulse, and just send a return pulse back to the source? You need only one scope in that situation, though to correct your time it may be more than divide-by-two; depending on the pulse generator's post-trigger delay you may have to subtract that out.

    Note that the details of this problem may be departing from the subject of Agilent scopes.

-Alex
Like • Show 0 Likes0
Actions

synchronizing two oscillosopes (DSO5034A)

Outcomes

Related Content

Recommended Content