2
\$\begingroup\$

I'm using the following function generator: http://www.farnell.com/datasheets/499058.pdf?_ga=1.129729036.1263433037.1467971694 and a Tektronix TDS 1002B scope for some measurments. Most of the time I use the function generator's square wave output(at %50 duty-cycle) as an input to a circuit.

The issue is that I see a lightning-like random flickering noise when I observe the rising or falling edges of the squarewave above 200Hz. It is hard to describe what I see so I uploaded a very short record of the scope screen. Here is how it looks like: https://youtu.be/hOy5-SOk2mg

I would like to describe the setup briefly to elimitae some possible causes. To observe the signal from the function generator, I first hook up the function generator's squarewave output to the scope's channel (I also use a 50ohm terminator to eliminate ringing). Below is how the BNC is connected to 50ohm output of the function generator:

enter image description here

And here below is how BNC is connected to the scope channel(If I don't use 50ohm terminator I still have the noise plus ringing):

enter image description here

I first thought the noise could be coming from the scope itself so I hooked up the scope probe to its own builtin 1kHz output. But the edges were very clean without any weird interference. I also tried the setup with different BNC coax cables and still having the same issue. I also don't observe this noise on the scope screen only if the function generator's output is set below 200Hz.

Do you have any idea what could be the issue here? And could it be fixed with a different setup.

Edit:

Below is a photo from the scope screen. I had to take tens of photos to catch when the noise hits. Below when the noise strikes:

enter image description here

As I wrote this happens randomly all of a sudden. So the noise is not continuously there that was the reason why I had uploaded a video.

Another clearer photo when the noise appears:

enter image description here

Edit 2:

Some new observations:

1-) I used 2-prong power plugs and the issue still existed. So it is not related to ground loops

2-) I cannot sample fast enough 10ns signals with my DAQ device

3-) If I use 20MHz bandwidth instead of 60MHz in settings the noise becomes better

4-) Noise only exists for the square wave output not for sinusidal or triangle

And I guess the most important observation is the following:

When I set the function generator main output's Vpp anything less than 2V the noise disappears totally. And I hear a 'tick' sound the momemnt Vpp is adjusted below 2V (a tick sound similar to when you hear in CC limiting in a power supply). There is obviously some relation with function generator's pulse output amplitude but have no idea what it could be. But under 2Vpp if I play with duty cycle noise appears again for different duty cycles..

\$\endgroup\$
  • 2
    \$\begingroup\$ Not watching video. Ain't gonna happen. Post a picture or scope snapshot of what you are asking about. \$\endgroup\$ – Olin Lathrop Jul 8 '16 at 10:41
  • 1
    \$\begingroup\$ I'm inclined to believe an analog scope that slow may simply be hiding the problem. Mine is 400MHz... If that photo shows 10ns/division then the potential problem lies outside its bandwidth. \$\endgroup\$ – Brian Drummond Jul 8 '16 at 11:50
  • 1
    \$\begingroup\$ Same results without T-connector or different BNC cables. \$\endgroup\$ – user16307 Jul 8 '16 at 14:10
  • 1
    \$\begingroup\$ I can see why this was about to get closed before the edits. Now it turned in to a good question IMO. \$\endgroup\$ – pipe Jul 8 '16 at 19:17
  • 1
    \$\begingroup\$ Guessing: high frequency ringing is always there and you are seeing it only when the scope samples at the right moment. Tek makes great scopes, but they are digital devices. If your ringing frequency exceeds the nyquist criteria for the scope, you can have this kind of artifact when the scampes line up with the peaks of the ringing. \$\endgroup\$ – John Birckhead Jul 8 '16 at 23:03
-1
\$\begingroup\$

This would not be an answer, but reading through question and comment stream I have a suggestion: reach generator's manufacturer, and report the problem. This way they would be able to know about your issue, prove or disprove it. There're could be other people who may suffer from the same issue, everyone will thank you for reporting. Additionally, you will be able to test manufacturer's support to see and feel if you still want to use their products :)

One more quick thought: coaxial cables and probes may have attenuation switch in them, and you can change their bandwidth. Did you play with it? What I see on your "clearer photo" looks very familiar... kind of impedance mismatch with (rising or falling) edge shooting back into the wire.

\$\endgroup\$
  • \$\begingroup\$ You really shouldn't use answers to make comments - there is a reason for the way the reputation system is set up. \$\endgroup\$ – Sean Houlihane Oct 1 '16 at 8:57
  • \$\begingroup\$ Reputation system is crap. Helping and guiding people solving their problems is the value. The question already has 26 comments to it, which means it is a time to aggregate knowledge and give conclusion or direction. \$\endgroup\$ – Anonymous Oct 1 '16 at 9:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.