New 2ch pocket DSO+SG - Zeeweii DSO2512G

[Aldo22]

so I got a 200 euro scope for 164 euros ;-)

Which one?

[RAPo]

The TO1112D.

[Aldo22]

Now I happen to have a very simple one because there is a 30MHz low pass filter in the RF test board (for Nanovna).

Thanks, I forgot that!

Yes, the RF test board is pretty nice for the price.
You can see/test a lot with it.
Here the 30MHz LPF in the NanoVNA-H and a filtered 5MHz square wave in the TinySA

[ptluis]

• Stop does not stop:

In normal trigger mode:
After pressing stop, waveform can't be moved with cursor keys.
If new trigger condition arrives, image is updated (even in STOP!) and scrolling becomes possible.
This is only noticeable with non periodic signals, because with periodic signals the second trigger after stop happens all the time.

Hello, can you make a video showing this bug?

[visenri]

I've recorded a quick and dirty video (sorry for the quality).

The test signal is a signal pulse with a period of 5s.

Current behavior
Case 1 - Waveform not moving until next trigger in RUN mode:
0s to 8s in the video.

Trigger in NORMAL, running.
A signal is present in the screen, moving with the arrows moves the trigger point but not the signal.
This is no big deal, and maybe other devices do it like this too.

Case 2 - Normal mode goes to WAIT when pressing STOP button
8s to 13s in the video.

Stop button is pressed.
Trigger changes from NORMAL to WAIT.
Moving with the arrows, moves the trigger point and some part of the signal, but the pulse remains in the same place.
When another pulse arrives, waveform is updated with new samples, trigger changes to STOP and the waveform can be finally moved correctly.

The desired behavior:
Case 1:
In my opinion, the arrows should move both the trigger point and the waveform (Like in STOP mode).
But this one is not a big problem, I can live without it.

Case 2:
When Stop button is pressed.
Trigger must change to STOP and KEEP the signal already acquired.
So the last signal recorded before the stop can be inspected with the cursors or zoom.


Why fixing case 2 could this be useful?:
Leaving the scope in RUN mode (NORMAL trigger), until I see an interesting event, pressing the STOP allows me to inspect the event, but ONLY if behaves as requested.

Sorry once more for the video quality, but I think it's good enough to show the problem.

[visenri]

I just did a quick test with my Rigol, and both cases behave as I describe in the "desired behavior".

[Aldo22]

I've recorded a quick and dirty video (sorry for the quality).

Where is the video?

[ptluis]

I've recorded a quick and dirty video (sorry for the quality).

post the video link so that i can send it to zeeweii. thank you

[visenri]

I've recorded a quick and dirty video (sorry for the quality).

post the video link so that i can send it to zeeweii. thank you

I swear, I added it as an attachment, but somehow, it got lost.

I've attached it again in my previous post.

[ptluis]

I've recorded a quick and dirty video (sorry for the quality).

post the video link so that i can send it to zeeweii. thank you

I swear, I added it as an attachment, but somehow, it got lost.

I've attached it again in my previous post.

Thanks, i never notice that because i mostly use single mode to catch and stop signals. The reason you can't move the waveform in normal triggering is because the waveform is not stored on memory just showed on screen and the scope is still triggering. when you press cursor keys the screen is not refreshed causing some confusion. it's somehow mentioned in the user manual. it's not well implemented of course. video was sent. let's see what kind of modifications they will do on this device. They told me it will take some time, so we just have to wait and see.

[Aldo22]

I am still experimenting with low pass filters.
Since I find them too expensive to buy and I don't know exactly which one is needed, I soldered one myself.
I have never done this before, but it seems to be quite useful.

The ripple on the Zeeweii caused by "nervous" square wave generators like Si5351 or MS5351M is greatly reduced.
There are relatively few side effects (Below 60MHz). At 20MHz the square wave is still visible.
It is a very simple PI LC LPF (220nH, 2 x 47pF).

[ptluis]

I am still experimenting with low pass filters.
Since I find them too expensive to buy and I don't know exactly which one is needed, I soldered one myself.
I have never done this before, but it seems to be quite useful.

The ripple on the Zeeweii caused by "nervous" square wave generators like Si5351 or MS5351M is greatly reduced.
There are relatively few side effects (Below 60MHz). At 20MHz the square wave is still visible.
It is a very simple PI LC LPF (220nH, 2 x 47pF).

you are measuring a 200mvpp square wave right?

[Aldo22]

you are measuring a 200mvpp square wave right?

Yes, roughly.
In the screenshots you see the output of NanoVNA-H with the MS5351M.
(Coax with 50 Ohm load resistor)
It's less about the amplitude in this case.

[ptluis]

you are measuring a 200mvpp square wave right?

Yes, roughly.
In the screenshots you see the output of NanoVNA-H with the MS5351M.
(Coax with 50 Ohm load resistor)
It's less about the amplitude in this case.

I think the ripple is due to the fact that the dso2512g is operating in full bandwidth mode. One of my requests is too include a 20MHz bandwidth limit option. In noisy environments  this scope is too sensitive and capture everything compromising the readings. The use of an external LPF is something I think about but honestly never tried yet, but could be a solution.

[tautech]

In noisy environments  this scope is too sensitive and capture everything compromising the readings.

And you call 10mV/div sensitive ?
Good DSO's offer 10+x better sensitivity.

[Aldo22]

In noisy environments  this scope is too sensitive and capture everything compromising the readings.

And you call 10mV/div sensitive ?
Good DSO's offer 10+x better sensitivity.

We're talking about an $80 2-channel scope here, not about a "good DSO". 
It is not bad for the price.

Actually, the topic has already been discussed here.
If you use the scope with the included probes in 10x mode, it works quite well.

If you connect it directly with coax you can have two problems:
1: It does not like higher voltage at the input (discussed ad nauseam above).
2: The low pass filter effect of the probes is lost.

With an external LPF and attenuator you can correct this a bit.
Since I didn't want to spend much for this experiment, I soldered one myself.
In the photo you see 6 MHz square wave of the Si5351 with LPF and attenuator (20dB).

Playing! It is good for that!

[tautech]

If you use the scope with the included probes in 10x mode, it works quite well.

And get just 100mV/div sensitivity, no thanks as that immediately excludes it for any even half serious work.

[Aldo22]

If you use the scope with the included probes in 10x mode, it works quite well.

And get just 100mV/div sensitivity, no thanks as that immediately excludes it for any even half serious work.

Yes, it is a cheap scope. It's easy to find flaws. So what?

The attenuator is only needed at higher frequencies if you want to see a "nice, calm" signal.
At lower frequencies it is not needed. (attached screenshot).

[ptluis]

In noisy environments  this scope is too sensitive and capture everything compromising the readings.

And you call 10mV/div sensitive ?
Good DSO's offer 10+x better sensitivity.

ok what I mean about sensitive is not referring to V/div of the scope but to electric interferences on the environment or the power line that contaminates the readings,  BUT let me put it this way, yesterday when testing some ECU PWM signals on the bench i couldn't get a clean signal  on the scope, lots of glitches, ripple and other interferences.  after debugging I've found the cause, it was a cell phone charger. i remove it and everything is back to normal. THEN i decided to use the Siglent scope and the fnirsi dpox180h (yes i know, i couldn't resist and try one), with those two it was enough to select the 20MHz bandwidth limit option and bingo! a clean signal even with the cell phone charger connected. I think an external LPF could be a solution for this zeeweii scope since I don't believe they can do it in software.

[Aldo22]

I think an external LPF could be a solution for this zeeweii scope since I don't believe they can do it in software.

That is probably true. I've had quite a bit of success with my LPF experiment.

Next week I will get a new soldering iron and try some other cutoff frequencies.

[Aldo22]

I just wanted to check how well the DSO2512G and the An870 match on the RMS measurement.
While for a square wave (500Hz, 50%, ~1V) the agreement is quite exact (1.00V vs. 1.0017V), it is not so exact for a sine wave (0.721V vs 0.7413V).
Why is this and which device measures more correctly?

Thank you!

[ptluis]

I just wanted to check how well the DSO2512G and the An870 match on the RMS measurement.
While for a square wave (500Hz, 50%, ~1V) the agreement is quite exact (1.00V vs. 1.0017V), it is not so exact for a sine wave (0.721V vs 0.7413V).
Why is this and which device measures more correctly?

Thank you!

maybe the an870  is using some kind of average calculations i think the dso2512g don't use average aquire, that could be one of the reasons, i don't know. I will have to do some testing with my scopes to see the results.

[Aldo22]

I just wanted to check how well the DSO2512G and the An870 match on the RMS measurement.
While for a square wave (500Hz, 50%, ~1V) the agreement is quite exact (1.00V vs. 1.0017V), it is not so exact for a sine wave (0.721V vs 0.7413V).
Why is this and which device measures more correctly?

Thank you!

maybe the an870  is using some kind of average calculations i think the dso2512g don't use average aquire, that could be one of the reasons, i don't know. I will have to do some testing with my scopes to see the results.

Thank you.
It is not that I would expect more accuracy from the oscilloscope.
I was just wondering about the difference between sine and square, if there is something systematic behind it.

[Veteran68]

Oscilloscopes are not accurate voltage instruments, and seldom compare to even the cheapest DMMs these days. That's not their purpose.

[Aldo22]

Oscilloscopes are not accurate voltage instruments, and seldom compare to even the cheapest DMMs these days. That's not their purpose.

Thank you.

That is not my point.
I'm wondering why the match is so amazingly accurate for the square wave but not for the sine wave under exactly the same conditions.
I suspect a systematic reason behind it and that would interest me.

Otherwise you are right of course, BUT We are talking about AC measurements here.
The ±(0.3%+3) Accuracy for the DMM are only given in a very narrow frequency range (unfortunately this is not declared in the specifications).
From 2kHz the AN870 becomes inaccurate. At 3kHz it already reaches -3dB.

The DSO2512G shows the same RMS value up to several MHz with +- 2% precision.
So it's not that simple.