Author Topic: Trap for Young Players: HF Reject  (Read 1887 times)

0 Members and 1 Guest are viewing this topic.

Offline PrehistoricmanTopic starter

  • Regular Contributor
  • *
  • Posts: 216
  • Country: gb
Trap for Young Players: HF Reject
« on: December 17, 2019, 01:16:20 am »
I was at my university's electronics lab today working on some reverse-engineering/glitching attack when I realised that my scope was triggering in completely the wrong place.
i had an microcontroller set up to deliver a clock burst (green) and before sending it, it would set an IO line high (my trigger). However, what I realised was that I was triggering way after the burst.



Turns out this HF reject option delays the trigger significantly.
Here is with HF reject turned off:



Why do you think Keysight didn't adjust the trigger in software to account for the filter delay?

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Trap for Young Players: HF Reject
« Reply #1 on: December 17, 2019, 04:06:25 am »
Why do you think Keysight didn't adjust the trigger in software to account for the filter delay?

I suppose they could since a given filter has a fixed group delay but analog oscilloscopes do the same thing when high frequency reject is used.  When high frequency reject is used appropriately on low frequency signals, the delay is insignificant compared to the time/div.
« Last Edit: December 17, 2019, 04:07:58 am by David Hess »
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Trap for Young Players: HF Reject
« Reply #2 on: December 17, 2019, 04:08:13 am »
What delay?  If it's triggering on Ch1, it looks to be triggering on the step change, rounded over slightly as you would expect given the bandwidth specified for "HF reject".

It would be more obvious if they had a "trig view" function, though you may have a software filter function that can approximate such a display.  If so, try it with different signals, combinations of transients/glitches, pulses, bursts and steps, and get a feel for what it's doing. :)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Trap for Young Players: HF Reject
« Reply #3 on: December 17, 2019, 04:35:31 am »
What delay?  If it's triggering on Ch1, it looks to be triggering on the step change, rounded over slightly as you would expect given the bandwidth specified for "HF reject".

But the trigger point is misaligned by about 600 nanoseconds.

On an analog oscilloscope if you make that mistake, it often becomes very apparent because the displayed jitter is much higher.
 

Offline jklasdf

  • Regular Contributor
  • *
  • Posts: 79
  • Country: us
Re: Trap for Young Players: HF Reject
« Reply #4 on: December 17, 2019, 06:28:10 am »
It looks like you're triggering on the Ext. input, I'd be interested in seeing what it looks like.
« Last Edit: December 17, 2019, 06:30:06 am by jklasdf »
 
The following users thanked this post: SparkyFX

Offline PrehistoricmanTopic starter

  • Regular Contributor
  • *
  • Posts: 216
  • Country: gb
Re: Trap for Young Players: HF Reject
« Reply #5 on: December 17, 2019, 02:22:47 pm »
It looks like you're triggering on the Ext. input, I'd be interested in seeing what it looks like.

Correct. You can see a little disruption in the two displayed signals at the point of the 3.3V GPIO switching that goes to the external trigger.
It should just be a normal transition  so I didn't bother taking a look. If I take my project back to the lab I'll do it.
« Last Edit: December 17, 2019, 02:26:28 pm by Prehistoricman »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Trap for Young Players: HF Reject
« Reply #6 on: December 17, 2019, 03:31:47 pm »
Oh, so the delay was about 2.2 microseconds.
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2302
  • Country: us
Re: Trap for Young Players: HF Reject
« Reply #7 on: December 17, 2019, 04:16:02 pm »
The HF Reject adds a 50kHz low pass filter to the trigger path which is going to add quite a slope to a fast edge.  On a MSOX3104A, I can get it to trigger up to 20us after the actual transition, depending on the trigger level.  And there is a ton of jitter.

Wrong trigger option.  Simply don't use it for fast signals.

If you are using it to counteract a lot of noise, consider using the Noise Reject option which increases the hysteresis, or improve the probing to get a better edge on the external input.

Or cook up some other trigger configurations for burst or pattern to trigger on the two displayed signals.  They look pretty clean, relatively speaking.
 
The following users thanked this post: Someone


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf