New toy(?) scope, DSO154pro, 1ch, claimed 40MS/s

[Fungus]

In this video @ 15:40 the bandwidth of the 154Pro is tested and does just fine @ 18MHz.  No where near -3dB.

Yes, the analogue bandwidth of the front end is greater than 18MHz.

Yep. I've got one, if I put in a 22Mhz sine wave I see a perfect 18MHz sine wave on screen (as predicted).

You need to go up to maybe 28Mhz before you see much attenuation.

Those things are nice little gadgets but aliasing is a big problem even at "Arduino" frequencies.

[Dave_g8]

Hi Bill,
If you inject a 1MHz square wave with a faster edge, you will see more of a difference.
The top trace is a Rigol sampling at 1GSa/s and there is no ripple.

The DSO154Pro is fine, but you may not be seeing a true representation of the signal.

[BillyO]

Hi Bill,
If you inject a 1MHz square wave with a faster edge, you will see more of a difference.
The top trace is a Rigol sampling at 1GSa/s and there is no ripple.

The DSO154Pro is fine, but you may not be seeing a true representation of the signal.

(Attachment Link)

You are seeing exactly what you should be seeing for a scope with 18MHz bandwidth.  Harmonics above 18MHz will begin to be attenuated.  Your image is showing good contributions from the first 16 or so harmonics.  That is 100% expected for an 18MHz scope.  I am willing to bet the other image comes from a scope with far greater bandwidth.  In this case it has nothing to do with sample rate.  Just bandwidth.

[Dave_g8]

Hi Bill.
If you put a 1MHz square wave into a 20MHz analogue oscilloscope, you will not see the ripple as seen on the DSO154Pro.

[BillyO]

Hi Bill.
If you put a 1MHz square wave into a 20MHz analogue oscilloscope, you will not see the ripple as seen on the DSO154Pro.

Well, you will but it will be harder to see because of the response of an analog scope.

Here is an image of a square wave composed of the first 16 harmonics.  Does it look familliar?

[BillyO]

If you want to play around with it the tool is here: https://www.ultracad.com/square.htm

[Dave_g8]

Hi Bill.
See below a simulation of a 1MHz square wave with a 100ps rise/fall time passed through a 20MHz low pass filter.

[BillyO]

Hi Bill.
See below a simulation of a 1MHz square wave with a 100ps rise/fall time passed through a 20MHz low pass filter.

(Attachment Link)

Correct, and that would be the response of an analog scope.  I have no problem with that.

Oh, okay.  I get what you are saying.  Yes, the 40MHz sample rate will have an effect as harmonics beyond the 16th or so will not even be detected.

Sorry for the misunderstanding.  My mind was in a different place.

[Dave_g8]

Hi Bill,
No problem.
It is the sin(x)/x reconstruction without any anti-aliasing filter feeding the ADC that screws things up when injecting a square wave with fast edges.
That's why the square wave from the DSO154Pro test generator look reasonable, because it is effectively band limited.
Regards, Dave

[BillyO]

Hi Bill,
No problem.
It is the sin(x)/x reconstruction without any anti-aliasing filter feeding the ADC that screws things up when injecting a square wave with fast edges.
That's why the square wave from the DSO154Pro test generator look reasonable, because it is effectively band limited.
Regards, Dave

Yes, agreed.

[Fungus]

The DSO154Pro is fine, but you may not be seeing a true representation of the signal.

Square waves are the sum of a lot of sine waves.

You're seeing the sum of the sine waves that an 18MHz 'scope can see.

At higher frequencies the same thing will happen on your Rigol.

[Dave_g8]

Hi,
The analogue bandwidth of the DSO154 will be much greater than 18MHz.

Attached is a capture on the DSO154Pro with a 1Hz (2V Pk-Pk) square wave input from a signal generator (fast edge speed).



You can see the same ringing before and after the transition which is a result of the display reconstruction.

I don’t want to repeat all the discussions on sin(x)/x, Gibbs, but in my opinion the displayed waveform is not always representative of the input signal.

If I sample a 1Hz square wave at 40MSa/s (25ns) and then join the dots, I will not see the pre and post edge ringing.
For completeness, the Excel graph below shows an example of sampling a square wave.
The dotted yellow line is joining the dots between samples and the red line is the sin(x)/x reconstruction. The frequency of the ringing is half of the sampling frequency.

[sizziff]

I don’t want to repeat all the discussions on sin(x)/x, Gibbs, but in my opinion the displayed waveform is not always representative of the input signal.

I also hold the opinion that the wave at the top of the pulse is the result of the Sinx/x interpolation
On the Hantek 2d72 oscilloscope (250Ms), it displays a square wave of more than 10 MHz very similar.
The straight line of linear interpolation is replaced by sinx/x arcs, which makes it possible to plausibly restore the sinusoidal signal, but gives such artifacts on the square wave signal.

[Dave_g8]

Hi,
The attached link provides a summary of the pros and cons for different interpolation methods.
https://saving.em.keysight.com/en/used/knowledge/glossary/oscilloscopes/what-are-interpolation-methods

The sin(x)/x is great for band limited signals, but not so good for signals with fast transitions.

[Dave_g8]

Hi,
For general information I noticed an undocumented feature in the DSO154Pro.
Since the feature is not covered in the documentation, it may be firmware/hardware dependant and not fully tested.
Whilst in the “Disp” Menu, press the “Stop” button.


Two options are available “F_Meter” and “Counter” as shown below.
Note: The options are mutually exclusive.


Enable the “F_Meter” by moving to the “ON” icon and pressing the “OK” button.


The “F_Meter” works irrespective of the time-base setting as shown below, unlike the normal frequency reading used in “Meas” options which is based on the waveform shown on the screen.


Notes
- The “F_Meter” seems to use its own auto trigger level setting, irrespective of the setting in the trigger menu.
- The “F_Meter” reading is updated whilst in run mode and keeps the previous value when the “Stop” button is pressed.
- The “F_Meter” seems to work correctly from around 30Hz up to 12.5MHz. The resolution is much better than the normal frequency reading used in “Meas” options.
     Below 30Hz the reading is erratic, perhaps due to the auto threshold detection.
     Above 12.5MHz the reading does not seem to be valid.
- The “F_Meter” will not operate correctly if the oscilloscope voltage range is set to high.

Enable the “Counter” by moving to the “ON” and pressing the “OK” button.


Notes
- The “Counter” seems to count the number of rising edges crossing the trigger level, irrespective of the rise/fall setting in the trigger menu.
- The “Counter” seems to use the same trigger level setting as the oscilloscope (Auto/Manual).
- The “Counter” increments in run mode and stops when the “Stop” button is pressed.
    Whilst in stop mode, pressing the “Auto” button does not restart the “Counter”.
- To reset the count to “0”, reselect the “Counter” from the options menu.
- The “Counter” will not operate correctly if the oscilloscope voltage range is set to high.

[Dave_g8]

Interesting, there is a recent firmware upgrade for the DSO3D12 which adds a frequency meter and pulse counter.

This appears to be a similar specification to the feature found on the DSO154Pro (HW V1.4). The DSO3D12 seems to have a higher upper frequency limit.

It may be that the feature on the DSO154Pro is also HW/FW dependant, since this seems to be the case for the DSO3D12.

[Dave_g8]

For general information, using a tinySA signal source (sine wave) terminated with 50Ω, the approximate analogue -3dB bandwidth of the DSO154Pro front end (using the supplied probe, x10 mode) is 32MHz.




At test frequencies above 20MHz, the alias frequency will be shown on the DSO154Pro, but the amplitude will be representative of the actual sampled values at the ADC.

[sims070]

I would say even buy the 2ch one for 6 bucks more with a choice day coupon.

https://nl.aliexpress.com/item/1005005953868224.html?spm=a2g0o.productlist.main.1.53bcd9c5aXJmQZ&algo_pvid=22987cd2-ff81-4434-8f3a-a8f9ae8a8f3b&algo_exp_id=22987cd2-ff81-4434-8f3a-a8f9ae8a8f3b-0&pdp_npi=4%40dis%21EUR%21174.44%2178.50%21%21%211332.59%21599.67%21%402101e64117120085714392277e985c%2112000035009715553%21sea%21BE%212723124816%21&curPageLogUid=rT1nw5orPbVX&utparam-url=scene%3Asearch%7Cquery_from%3A