Measuring Output Ripple voltage in Scope

Thread Starter

Electronic_Maniac

Joined Oct 26, 2017
253
I have a DC-DC converter which converts 16V to 23V. I have an output E-Cap at the end of the converter. I need to measure the ripple voltage at the E-Cap. I am using a 300MHz scope.

I connect my probe to the positive terminal of E-cap and ground the scope ground with my board ground. I am using the scope in Auto Mode.

I have sent the horizontal scale at 100mV/div and Vertical scale at 2us/div.

I see the voltage signal continuously running and at certain moments when I see the peak values, I stop the reading. I get different values at different instances. How to measure the ripple voltage now?

And how will I set the trigger point if I dont know the max ripple voltage?

Can you please help me?
 

danadak

Joined Mar 10, 2018
4,057
Best way is put the scope on infinite persistence which will integrate the
noise on the display so you get pk-pk noise.. Let a number of sweeps
accumulate so you capture the statistical distribution. If scope is DSO
or analog storage you can do this.

Regards, Dana.
 

Danko

Joined Nov 22, 2017
1,829
How to put in infinite persistance?
Check your oscilloscope documentation.
For example - from DSO OWON VDS3102 documentation:
Persistence
When the Persistence function is used, the persistence display effect of the picture tube oscilloscope can be simulated. The reserved original data is displayed in fade color and the new data is in bright color.
Different persistence time can be chosen: Off, 0.5 second, 1second, 2seconds, 5seconds and Infinite.
When the "Infinite" option is set, the measuring points will be stored till the controlling value is changed.
Click "Clear" button, the persistence will be cleared.
Note:
If the time base, voltage division, deep memory is changed, or the channel is turned off/on, the persistence will be cleared automatically and record the updated waveform.
upload_2019-1-30_19-36-5.png
 

Reloadron

Joined Jan 15, 2015
7,501
I have sent the horizontal scale at 100mV/div and Vertical scale at 2us/div.
I believe you have that backwards but we get the idea.

You don't mention the scope make and model? We also don't know the switching rate of the DC/DC converter. I would just choose a sweep speed to start which provides a decent display and likewise the vertical gain with AC coupling. I would use a Normal trigger mode and try and get a stable triggered display. Then use the sweep time to actually see what you have.

How to put in infinite persistence?
That should be a function of the type scope you have. It may or may not be possible with your specific scope.

I see where Danko has it covered as I typed. :)

Ron
 

Thread Starter

Electronic_Maniac

Joined Oct 26, 2017
253
Thank you. My scope is Agilent technologies MSO6034A 300MHz. I don't think it has the persistance option.

And apart from this, I noticed this while measuring the frequency of a the timing capacitor pin of NCV3063 IC, According to my design, the frequency should Lie between 110-125kHz. But, while measuring, when I change the time division in the scope, the frequency of the signal also changes. How is this possible? The frequency of the signal should be constant irrespective of the scope adjustments, right? Am I missing something or understanding it incorrectly? Please help. Thanks
 

Danko

Joined Nov 22, 2017
1,829
My scope is Agilent technologies MSO6034A 300MHz. I don't think it has the persistance option.
MSO6034A has persistence mode:
upload_2019-1-30_21-38-40.png
See Agilent-MSO6034A-Datasheet, page 22, Display system.
when I change the time division in the scope, the frequency of the signal also changes.
I think you are seeing time base frequency (scan rate) instead of signal frequency.
Agilent_6000_Series_Oscilloscope_Users_Guide.pdf, page 238, Frequency.
 
Last edited:

Wuerstchenhund

Joined Aug 31, 2017
189
Thank you. My scope is Agilent technologies MSO6034A 300MHz. I don't think it has the persistance option.
It does (press the DISPLAY key, then 'oo PERSIST'.

However, for looking at ripple, I'd use the Trigger capabilities of your MSO6034A instead of Persistence mode.

So lets look at the ripple first:

- Put the scope input in AC mode, so that the DC part of your signal is removed. What's left is the non-DC part, i.e. ripple (and glitches, if there are any).
- Increase the vertical resolution so you your waveform takes as much screen space as possible while not vertically exceeding the grid area
- Set the Trigger to NORMAL and AC, and set the Trigger Level so that you get a stable waveform
- Press QUICK MEAS, select the right channel and the parameter you want to measure. Add more as required


Now let's look at glitches (if there are any, which we'll find out), and here we can use Persistence mode:

- Set the scope to Peak Detect: press 'ACQUIRE', then 'ACQ MODE' until 'PEAK DETECT' is shown
- Enable Persistence mode (see bove)
- The scope will show glitches, if there are any.
- To characterize a glitch, go into Zoom mode ('MAIN/DELAYED' key, then 'ZOOM' softkey)
- To get a better view, expand the timebase, and use the Horizontal Position knob to pan through the captured glitches.
- Use 'QUICK MEAS' (see above) to measure the glitch parameters.

And apart from this, I noticed this while measuring the frequency of a the timing capacitor pin of NCV3063 IC, According to my design, the frequency should Lie between 110-125kHz. But, while measuring, when I change the time division in the scope, the frequency of the signal also changes. How is this possible? The frequency of the signal should be constant irrespective of the scope adjustments, right? Am I missing something or understanding it incorrectly? Please help. Thanks
You didn't describe your probe solution or how you measure the frequency so it's hard to tell. But measurement functions on a scope work better if the signal is as big as possible, i.e. it takes up most of the vertical area. If the signal is small (less than a division) then the accuracy of the measurement functions drop.
 

Thread Starter

Electronic_Maniac

Joined Oct 26, 2017
253
It does (press the DISPLAY key, then 'oo PERSIST'.

However, for looking at ripple, I'd use the Trigger capabilities of your MSO6034A instead of Persistence mode.

So lets look at the ripple first:

- Put the scope input in AC mode, so that the DC part of your signal is removed. What's left is the non-DC part, i.e. ripple (and glitches, if there are any).
- Increase the vertical resolution so you your waveform takes as much screen space as possible while not vertically exceeding the grid area
- Set the Trigger to NORMAL and AC, and set the Trigger Level so that you get a stable waveform
- Press QUICK MEAS, select the right channel and the parameter you want to measure. Add more as required


Now let's look at glitches (if there are any, which we'll find out), and here we can use Persistence mode:

- Set the scope to Peak Detect: press 'ACQUIRE', then 'ACQ MODE' until 'PEAK DETECT' is shown
- Enable Persistence mode (see bove)
- The scope will show glitches, if there are any.
- To characterize a glitch, go into Zoom mode ('MAIN/DELAYED' key, then 'ZOOM' softkey)
- To get a better view, expand the timebase, and use the Horizontal Position knob to pan through the captured glitches.
- Use 'QUICK MEAS' (see above) to measure the glitch parameters.



You didn't describe your probe solution or how you measure the frequency so it's hard to tell. But measurement functions on a scope work better if the signal is as big as possible, i.e. it takes up most of the vertical area. If the signal is small (less than a division) then the accuracy of the measurement functions drop.
When I put the mode in AC coupling and normal trigger(also tried auto mode) and measuring 23V at the output E-Cap, only the ripple voltage around 23V must come,right?
But I am getting the DC offset of 23V also. Any mistake I am making? Or is this correct?
 

Reloadron

Joined Jan 15, 2015
7,501
When I put the mode in AC coupling and normal trigger(also tried auto mode) and measuring 23V at the output E-Cap, only the ripple voltage around 23V must come,right?
But I am getting the DC offset of 23V also. Any mistake I am making? Or is this correct?
Then I am missing something. When you AC couple the vertical input the DC component is blocked. You should only see the AC component which would be the ripple. The DC level is blocked which is the whole idea of AC coupling a signal. Also, what exactly is an E-Cap? Would it be the DC/DC converter output filter capacitor?

Ron
 

Thread Starter

Electronic_Maniac

Joined Oct 26, 2017
253
Then I am missing something. When you AC couple the vertical input the DC component is blocked. You should only see the AC component which would be the ripple. The DC level is blocked which is the whole idea of AC coupling a signal. Also, what exactly is an E-Cap? Would it be the DC/DC converter output filter capacitor?

Ron
Yes. the E-Cap is at the output of the DC-DC converter
 

Danko

Joined Nov 22, 2017
1,829
But I am getting the DC offset of 23V also. Any mistake I am making? Or is this correct?
It is correct if oscilloscope is in DC coupling state.

TS's experiment simulation:
V2 frequency = 150kHz.
Sine wave - L1 parasitic oscillation.
Red plot - oscilloscope picture at AC coupling.
upload_2019-2-1_2-11-42.png

Ripple on C1 (DC coupling):
upload_2019-2-1_3-4-55.png
 

Attachments

Last edited:

Thread Starter

Electronic_Maniac

Joined Oct 26, 2017
253
So again, using AC coupling the DC component should be blocked. Can you provide a screen shot of what your scope is showing and with that the scope settings?

Ron
Hi,

I faced the AC-DC Coupling problem with the agilent scope only. With the keysight scope, I am able to get only the AC ripples of the 23V. Thank you
 
Top