Author Topic: Is my jitter measurement method correct?  (Read 1291 times)

0 Members and 1 Guest are viewing this topic.

Offline drummerdimitriTopic starter

  • Frequent Contributor
  • **
  • Posts: 532
  • Country: lb
Is my jitter measurement method correct?
« on: March 26, 2019, 05:09:26 pm »
I would like to compare the jitter measurement of my dac on TOSLINK vs USB input using my oscilloscope.

Played a 20 KHz sine wave and hooked up my scope probes to the output XLR pins and got the pure signal.

I then adjusted my vertical scale to the lowest possible value, set the display to infinite persistence and measured the signal's horizontal sway and got about 40 ns of jitter.

Not sure if this is the right approach but the measurment seems to be in the realm of expectation. :-//

Any way I could improve this measurement?
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Is my jitter measurement method correct?
« Reply #1 on: March 26, 2019, 06:47:00 pm »
I think you are just measuring the jitter of the oscilloscope's trigger which should be zero.  I am not sure why it is not.

Jitter measurements assume trigger points on two different signals or two points on the same signal.  I think what you want to do is trigger in the audio output, and then use the oscilloscope's delay function to measure the jitter in an edge later in the signal.  If that is the next edge, then it is cycle-to-cycle jitter which is a rather demanding measurement and I think it will be more relevant to measure a later edge which is produced from a separate USB packet of data.
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: Is my jitter measurement method correct?
« Reply #2 on: March 26, 2019, 06:51:17 pm »
Sorry, not even close. Scope with it's 8bit ADC  is useless for audio signal quality measurements.

20KHz is way too high freq. You shall use 10KHz or 12KHz, hi-end ADC, spectrum analyzer or audio analyzer or another audio analyzer. Example of jitter estimation in audio domain:

https://www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-cypher-labs-theorem-720-and-fiio-q1-dac-and-headphone-amp.3324/

With scope you may try to catch DAC clock jitter - by probing digital clock signal and using jitter analysis function of scope (if any). More about jitter measurements:

http://literature.cdn.keysight.com/litweb/pdf/5991-4000EN.pdf

I think you are just measuring the jitter of the oscilloscope's trigger which should be zero.  I am not sure why it is not.

It's noise. Pay attention to vertical resolution which is 500uV/div.
« Last Edit: March 26, 2019, 07:03:06 pm by ogden »
 

Offline bson

  • Supporter
  • ****
  • Posts: 2497
  • Country: us
Re: Is my jitter measurement method correct?
« Reply #3 on: March 27, 2019, 08:55:35 am »
RMS stochastic jitter is the standard deviation of the period.  So set the scope to measure the period and collect statistics for the measurement.
« Last Edit: March 27, 2019, 08:57:42 am by bson »
 
The following users thanked this post: colorado.rob


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf