Test and Measurement Forum

Measuring 48 Khz square wave

Which product is right for your exact requirements

Measuring 48 Khz square wave

Postby Janne » Mon Dec 10, 2012 1:30 pm

Hello!

I am new to oscilloscopes. I have to measure with great precision the rising and falling edge of a 48 Khz square wave and looking for a pico to match my needs.
The signal is 48/96kHz Wordclock that is used in pro audio applications for synchronizing equipment. Typical rise-times for the pulse is 3-10ns, but i have a faulty splitter that is much greater than that (130 ns) and there is some strange things going on. Even if the 48kHz pulse frequency is not high, the sharp edge makes some demands. Voltage range is 0-7V.
I would like to have 2-4 inputs in use for measurement.

What is the right product for me?
Are the inputs terminated (50/75 ohm)?

The other question is about the demo-feature and triggers in the software. I am not able to "lock" to the frequency that i gernerate for measurement. I have tried adjusting the trigger settings (and using auto) but the displayed curve shifts along the x-axis, making the persistence mode unusable. What am i missing?

Thanks, Janne
Janne
Newbie
 
Posts: 0
Joined: Mon Dec 10, 2012 12:07 am

Re: Measuring 48 Khz square wave

Postby ziko » Tue Dec 11, 2012 12:03 pm

Hi Janne,

To measure a signal that has a typical rise time of 3-10ns you need to have a bandwidth that is high enough, the equations goes like this:

Bandwidth = 0.35/ Rise time.

Therefore

Choosing the worst case scenario of 3ns, the bandwidth would have to be at least 117MHz. You would also need to sample at a higher rate than the bandwidth maybe 10 times higher so you would need maybe at least 1GS/s.

Something like the 3206A/B would be able to achieve this for 1 channel.

http://www.picotech.com/picoscope-specifications.html

For 4 channels you need to go with a device like 6000 series , which can have sample rates of 1.25GS/s for 4 channels and have the sufficient bandwidth.:

http://www.picotech.com/picoscope6000.html

The scopes can measure up to +/-20V and have a switchable input impedance of 50 Ohms (6402A/B,6403A/B, 6404A/B)

The 6407 is a digitizer and whilst it can do the same as the other 6000 devices it only has one voltage range which is +/-100mV, so the other 60000 series devices would be more cost effective.

This is all assuming that you want to see a 3ns rise time, but in any case this gives you the tools to understanding which product is correct for you.

Could you explain more about the problems you are having with the demo software, what settings have you chosen? what version of the software are you using as well?
Hope this helps.
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Measuring 48 Khz square wave

Postby Janne » Tue Dec 11, 2012 12:15 pm

Hi!

Thanks for your answer.
Software version 6.6.37.1
I am using Auto Setup and Trigger Auto but it does not work. And after that i tried to fiddle with manual settings that are visible on the software (timing and triggers) without help. I have not tried to go into any other application settings so it should be a pretty vanilla install.
Janne
Newbie
 
Posts: 0
Joined: Mon Dec 10, 2012 12:07 am

Re: Measuring 48 Khz square wave

Postby ziko » Tue Dec 11, 2012 4:51 pm

Hi I cannot recreate this, you could take a video capture using jing, it records up to 5 mins, you can then upload the file to your account which can then be posted as a link:

http://www.techsmith.com/jing.html

In terms of products depending on how you are doing your measurements the 3400 series could also prove to be useful as they have sufficient bandwidth and also offer multiple channels.

Regards
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Measuring 48 Khz square wave

Postby Janne » Mon Mar 18, 2013 8:50 pm

Hello.
Bump.

Now I'm back to try to get this to work!
http://screencast.com/t/hvh9jzDOfht3
Here is a video with my silly voice.
How do i get a nice "locked on" wave display as i can do with my analog scope?

Best,
Janne
Janne
Newbie
 
Posts: 0
Joined: Mon Dec 10, 2012 12:07 am

Re: Measuring 48 Khz square wave

Postby ziko » Tue Mar 19, 2013 9:25 am

Hi,

If a picture is a thousand words and a video running at 25 frames per second then a video is worth 25,000 words a second!

Ignore me, been doing benchmark testing on my PC and graphics card!

I tried to do this on a number of software versions and don't encounter this problem. Are you using VMware or something?

Ziko
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Measuring 48 Khz square wave

Postby Janne » Tue Mar 19, 2013 10:02 am

Parallels desktop.. should i go bootcamp?

/Janne
Janne
Newbie
 
Posts: 0
Joined: Mon Dec 10, 2012 12:07 am

Re: Measuring 48 Khz square wave

Postby ziko » Tue Mar 19, 2013 10:09 am

Yeah give that a go.
Ziko

Technical Specialist
ziko
Zen Master
Zen Master
 
Posts: 1705
Joined: Fri Dec 01, 2006 10:03 am
Location: St Neots

Re: Measuring 48 Khz square wave

Postby Janne » Fri Apr 05, 2013 6:50 am

Hi. It worked. Bootcamp is fine.

I have currently on my desk a 3026B that i'm testing with the 14day return policy (wich is soon ending!) so quick answers are appreciated :)
I have the following problems/questions:

- In Scope mode i can see approximately what i want, but in persistence mode I can only get to 1 µs, and at 500 ns the display blacks out. examples here: http://imgur.com/a/JETdr#0
Will i need a faster scope? is this an oversampling/buffer thing?

Rise Time Measurement dislay is changing too fast. how can i make the display smoother so that i can get a good reading?
Janne
Newbie
 
Posts: 0
Joined: Mon Dec 10, 2012 12:07 am

Re: Measuring 48 Khz square wave

Postby Martyn » Fri Apr 05, 2013 7:58 am

As a guide if you look at the properties pane as soon as you get a number in brackets for "No. of Samples" it indicates that you have Sin(x)/x interpolation enabled. Turn this off in Tools->Preferences->Sampling and the trace reappears.

You need to use "Fall Time" on that trace to get a measurement.

To get a smoother trace goto channel options (Letter with Arrow) and enable Resolution Enhancement or Filtering.
Martyn
Technical Specialist
Martyn
Site Admin
Site Admin
 
Posts: 2291
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots


Return to Pre-Sales Advice

Who is online

Users browsing this forum: No registered users and 1 guest