VOLTCRAFT DOV series

[Martin72]

As it stands, the Scope isn't all that bad.
I wouldn't recommend it at the original price, though; all things considered, the Rigol DHO800 has the edge.
It's definitely better than a Hantek or Fnirsi, though.
At the moment, it looks like the “main issue” lies with the OS, i.e., the software.
Given its current configuration and assuming that no firmware updates will be released that could change this, I think the current price is a good one.
This also means that the “original,” i.e., the Owon, is too expensive.

But I'm not done testing it yet.
Besides, I'm tempted to make the scope a little “better,” so I'll probably keep it for now.
What I’ll do after the final tests is take it apart and take some good pictures of it.
Then I’d probably swap out the fan for a quieter or smoother-running one and finally look into the cause of the interference in the 200 mV range and the ripple in the calibration signal.
It could turn out to be a fun little project.


Additional tests I’d like to perform:
- Two-tone test
- Test all decoder types using the demo board
- Test the trigger modes available on the scope.
- FFT of various signals, mostly sine waves with and without significant interference.
- And as announced, screenshots of the individual menus, which will be done via the web server, since there is no dedicated screenshot button on the scope; you have to navigate through a menu, and then the previous screen is gone, of course.
- Possibly a short video of the startup and the fan noise.

[egonotto]

Hello,

The new images showing multiple channels indicate that ch1 has an issue that the others do not have. Could you please repeat the noise measurement using ch3.

In my opinion, ch3 has the lowest noise at 100 mV/div.

Best regards,
egonotto

[Aldo22]

The new images showing multiple channels indicate that ch1 has an issue that the others do not have. Could you please repeat the noise measurement using ch3.

I don't think you can tell that from the images.
In the image with 4 channels, Ch1 is set to 200 mV/div, and the others to 1 V/div.
Martin did say that there’s a ripple at  200 mV/div or 2 V/div:
https://www.eevblog.com/forum/testgear/voltcraft-dov-series/msg6266078/#msg6266078

The question is whether this is a system issue or just specific to Martin’s unit.

EDIT
I took a look at a video of the 2ch OWON.
It shows the calibration signal at 2V/div.
Unfortunately, it's a bit blurry, but I can't see any signs of ripple.
Neither from the calibration signal nor due to the 2V/div setting.

[tunk]

The DO704 is also sold here for 217€.
Ships to most of Europe for 7-20€.
https://www.sotel.de/en/Tools-DIY/Elektrowerkzeuge/Zubehoer/VOLTCRAFT-DOV704-Digital-Oszilloskop-70MHz-4-channel-1-25-GSa-s-50-Mpts-12-Bit-Digital-Speicher-DSO.html

[egonotto]

Hello,

Channel 1 has 200 mV/div, and channels 2, 3, and 4 have 100 mV/div (10.00X).

And we know from a previous image that ch1 has a 7.5 kHz interference.

So yes, ch2, ch3, and ch4 must have better noise (+interference) values.

Best regards,
egonotto

[Martin72]

Hi,

Two Tone Test

2-channel function generator SDG6052X; instead of using Wavecombine, the two channels are connected via a splitter, with 50-ohm termination.
10 kHz, 100 kHz, 1 MHz, 10 MHz, 50 MHz, 100 MHz.

Back to the interference issue:
All channels active simultaneously; as expected, all exhibit the same phenomenon at 200 mV/div (1:1).
At 100 mV/div, there is no sign of this.
I'll just accept that for now and take a closer look at it later to see what might be causing it.
It's not a deal-breaker for further testing at the moment.

Martin

[KungFuJosh]

Are they using a glitchy trace smoothing to compete with Fnirsi's schemes?

[Performa01]

Hi,

Two Tone Test

2-channel function generator SDG6052X; instead of using Wavecombine, the two channels are connected via a splitter, with 50-ohm termination.
10 kHz, 100 kHz, 1 MHz, 10 MHz, 50 MHz, 100 MHz.

If I read the screenshots correctly, that doesn't look promising, at least for frequencies above 100 kHz.
All the folks who think "yay, we get 12 bit for high dynamic at very low cost" should be disappointed.

[Aldo22]

If I read the screenshots correctly, that doesn't look promising, at least for frequencies above 100 kHz.
All the folks who think "yay, we get 12 bit for high dynamic at very low cost" should be disappointed.

Could it be that this is because Martin selected the —on his device— rippled 200 mV/div for the two tone test?
Could that affect the measurement?

[Martin72]

Regardless, I obviously shouldn't choose the “faulty” measurement range.
In other words, I'll try again...

[2N3055]

If I read the screenshots correctly, that doesn't look promising, at least for frequencies above 100 kHz.
All the folks who think "yay, we get 12 bit for high dynamic at very low cost" should be disappointed.

Could it be that this is because Martin selected the —on his device— rippled 200 mV/div for the two tone test?
Could that affect the measurement?

To be honest, fact that device has faulty measurement range should be immediate pack it and send it back.
It means device is defective.. Simple as that.

It has to be investigated further to prove what is going on.

When testing, you find worst case scenario and that is the guaranteed performance.
If that is good enough for requirements, than instrument passes. Otherwise it is not good for purpose.

[Aldo22]

To be honest, fact that device has faulty measurement range should be immediate pack it and send it back.
It means device is defective.. Simple as that.

I'd do the same.
Before that, I might try a different USB-C power adapter and plug it in somewhere other than the “Lab.”

[2N3055]

To be honest, fact that device has faulty measurement range should be immediate pack it and send it back.
It means device is defective.. Simple as that.

I'd do the same.
Before that, I might try a different USB-C power adapter and plug it in somewhere other than the “Lab.”

Yes, no need to be trigger happy. It should be investigated. But if proven as having such problem, that would make it basically defective.

[Performa01]

If I read the screenshots correctly, that doesn't look promising, at least for frequencies above 100 kHz.
All the folks who think "yay, we get 12 bit for high dynamic at very low cost" should be disappointed.

Could it be that this is because Martin selected the —on his device— rippled 200 mV/div for the two tone test?
Could that affect the measurement?

I don't think so. The two tone test reveals the intermodulation-free dynamic range - basically it's just a special form of linearity test. It is preferred over THD(+N) measurement, because it can be measured within a narrow bandwidth and doesn't require the source signals to be ultra clean low-harmonics sine waves.

It's hard to believe that any interference signals could affect the linearity of the frontend + ADC - as long as the signal level isn't that strong to drive the active fontend circuitry into saturation. These interference signals would just show up in the spectrum somewhere and can be easily distinguished from the intermodulation signals we are looking for in this test, because they are completely unrelated to the input signals.

[Aldo22]

I'm not entirely sure where to place this scope.
At the standard Owon ADS800A price point, it competes with the Rigol DHO800, so you'd expect it to perform well.
As the Voltcraft DOV700, it competes with the FNIRSI 1014D, and in that regard, it's already a winner, no matter how many quirks it has. 

[Martin72]

I'm just going to go ahead and send the device back, so the final tests will be put on hold until I receive a replacement.
If the replacement has the same issue, it's a system-wide problem, which means the original Owon models have the same issue as well.

[Aldo22]

I'm just going to go ahead and send the device back, so the final tests will be put on hold until I receive a replacement.
If the replacement has the same issue, it's a system-wide problem, which means the original Owon models have the same issue as well.

I just ordered the DOV702.
We'll see if and when it arrives—probably not on Whit Monday. 
Maybe I'll get it before you get your replacement? 

[Martin72]

Why just a 2-channel model?

[Aldo22]

Why just a 2-channel model?

Because I’m not going to pay 47% more for a feature I’ve never needed before and probably never will.
If I do, I also have my Hantek, which I can probably trigger-synchronize with the Voltcraft—I’ll test that out.
Remember? https://www.eevblog.com/forum/testgear/voltcraft-dov-series/msg6264318/#msg6264318

Besides, I don’t really need a new oscilloscope; I’m buying it because of the low price (CHF 149) and the features to play around with.
In 1-channel mode, it’s exactly the same as the DOV704. 

[Martin72]

Because I’m not going to pay 47% more for a feature I’ve never needed before and probably never will.

OK..

The scope has already been shipped; just to be safe, I wrote down the serial number—you never know...
If the second one behaves the same way, I might keep it anyway to see where the problem is coming from.
But I'll decide that on a whim.

[tooki]

First, check if they ship the oscilloscope to your country.

I tried to buy it, but it seems that both Voelkner and Conrad.de only ship to Germany.

That may be true for Conrad.de, but may not be true for Conrad.com, with the correct country selected.

(Since you don’t have a country set on your forum profile, I can’t verify…)

[adam4521]

I'm curious about the signal mixing test that Martin demonstrated. Is it ok to mix the two outputs just relying on their source impedance (up to a certain level) or do you need an additional impedance (eg pass through 50 ohm) before the 'mix'? And why is that setup preferable to the internal combine in the generator?

[2N3055]

I'm curious about the signal mixing test that Martin demonstrated. Is it ok to mix the two outputs just relying on their source impedance (up to a certain level) or do you need an additional impedance (eg pass through 50 ohm) before the 'mix'? And why is that setup preferable to the internal combine in the generator?

What do you mean? Those are proper 50 Ω impedance outputs on the gen, going into proper 50 Ω combiner (example MiniCircuits ZFRSC-2050+).
Difference to internal WaveCombine is fact that internal combine is in digital domain.
Some people are skeptical about it, So Martin wanted to do it "old school" so there are no doubts.
Although internal digital combine works really well and you would not see any difference with such results for this scope.

[adam4521]

Thank you! I was missing a product reference for that. And it is also useful to know that the internal/digital combine is expected to offer similar results.

[Martin72]

Some people are skeptical about it, So Martin wanted to do it "old school" so there are no doubts.

This.
There should be no doubt about the functionality of the SDG6000X’s Wave Combine feature, but I chose to take this approach anyway to prevent any doubts from arising in the first place.

I'm not entirely sure where to place this scope.

I do.
7-inch display, 12-bit native resolution, affordable price....
I've already had a few options on the table, so I know what I'm talking about.
Let's not kid ourselves: Rigol and Siglent are the top dogs in this field, period.
And of the two, the SDS800X HD is the gold standard, no question.
I’ve had both of them here—I bought them myself—the DHO804 and the SDS804.
The latter is sort of a “mini-me” version of the larger Siglent touchscreen oscilloscopes.
Although it has (in some cases intentionally) fewer features, it still delivers impressive performance.
The Rigol DHO800 is positioned below the SDS800 X HD.
So, while Rigol and Siglent are in the lower price range, anyone who wants or needs something even cheaper is venturing into territory that tends to be characterized by random highlights.
That's why I'm curious about the Voltcraft, which is currently on sale at a very low price.
As things stand, it clearly lags behind the Rigol, not to mention the Siglent.
That’s fine—but only at the current price.
If you consider the actual price, or rather the price Owon is still asking for it, it would be a loser compared to the models from Rigol and Siglent.
That needs to be stated clearly, because it’s so obvious.
And it’s mainly due to the software, the OS.
So I'd say, yes, for €200, this is a good to very good device.
But basically, I would recommend that everyone wait until they’ve saved up enough money to buy the current gold standard in small 12-bit oscilloscopes.
However, I won't make my final assessment—including the pros and cons—until I've received another Voltcraft device.