OWON VDS1022I Quick Teardown (versus the Hantek 6022be)

[balnazzar]

This is sounding like it's gaining a little traction. Indman raised it on Florent's github page a couple of months ago but I just noticed that Florent closed it as being a bit too broad a definition. Maybe you could add some additional ideas or detail...   https://github.com/florentbr/OWON-VDS1022/issues/57

Jason is right in saying that the API exposes a good autoset function. The problem is that you can't call it every time the AWG takes a step forward, unless the awg does a very, very slow sweep, and you make sure to sync properly. E.g., in pseudocode:

establish frq_bins width (steps)
  for each frq_bin
    measure frq&phase, then store;
    keep polling and wait for the next frq_bin threshold;
    as it hits that threshold, autoset(); then go to the next iteration
  endfor
end
   
Note that as you call autoset() and wait for the scope to adjust, the awg should temporarily stop the sweep. That's why, probably, Florent implemented that FRA example snippet without resorting to any autoset() call..
But then I'm not understanding the logic behind his implementation. Not much surprisingly, that code doesn't work out of the box, no matter how slowly the awg performs the sweep. Maybe you guys could chime in here.

If you have time and will, please test the code by yourself..

Thanks!

P.S.: I'll try and open an issue (which of course is not an issue) on the repo, but my impression is that the good Florent, while actively engaged in improving the vds1022i software, has nevertheless no time to take requests from multiple users into account.
 

[Anthocyanina]

having taken this oscilloscope with me a few times to other places, it has suffered some eye damage, requiring emergency surgery in one ocasion
So i'm wondering if anyone has found a suitable hard carry case for it? searching on amazon, hard carry cases for electronics are either too large or too small for it. Does anyone have any recommendations? thank you!

Edit: in line image attachments never work for me!

[balnazzar]

having taken this oscilloscope with me a few times to other places, it has suffered some eye damage, requiring emergency surgery in one ocasion
So i'm wondering if anyone has found a suitable hard carry case for it? searching on amazon, hard carry cases for electronics are either too large or too small for it. Does anyone have any recommendations? thank you!

Edit: in line image attachments never work for me!

See the link below. You can arrange the separators so that it doesn't move. I use it to carry the 1022i and my multimeter.

https://www.amazon.it/gp/product/B08LPFGVQG/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

[VEGETA]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

[Anthocyanina]

having taken this oscilloscope with me a few times to other places, it has suffered some eye damage, requiring emergency surgery in one ocasion
So i'm wondering if anyone has found a suitable hard carry case for it? searching on amazon, hard carry cases for electronics are either too large or too small for it. Does anyone have any recommendations? thank you!

Edit: in line image attachments never work for me!

See the link below. You can arrange the separators so that it doesn't move. I use it to carry the 1022i and my multimeter.

https://www.amazon.it/gp/product/B08LPFGVQG/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

Thanks! that seems to be not super large, could you please share a picture of the scope in it? I would like to be able to see how it would fit in that case, thanks again!

[tautech]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

Probably only when set to high sensitivity and it's mains frequency or some local noisy SMPS.
Just normal shit you need to work with or ignore.

[VEGETA]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

Probably only when set to high sensitivity and it's mains frequency or some local noisy SMPS.
Just normal shit you need to work with or ignore.

the problem is that I need to measure power supply noise and ripple to below 10 mv which is why I need to null this.

so there is no solution or workaround?

[tautech]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

Probably only when set to high sensitivity and it's mains frequency or some local noisy SMPS.
Just normal shit you need to work with or ignore.

the problem is that I need to measure power supply noise and ripple to below 10 mv which is why I need to null this.

so there is no solution or workaround?

Well we are hardly clairvoyant so without a screenshot showing settings and connections or not we have no idea what you see so can't advise with any degree of accuracy.

[VEGETA]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

Probably only when set to high sensitivity and it's mains frequency or some local noisy SMPS.
Just normal shit you need to work with or ignore.

the problem is that I need to measure power supply noise and ripple to below 10 mv which is why I need to null this.

so there is no solution or workaround?

Well we are hardly clairvoyant so without a screenshot showing settings and connections or not we have no idea what you see so can't advise with any degree of accuracy.

here are the pictures taken: https://www.eevblog.com/forum/projects/my-psu-design-ripple-and-noise-with-picture-measurements/msg4520213/#msg4520213

one of the images shows the loose probe and gnd readings

[Anthocyanina]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

Probably only when set to high sensitivity and it's mains frequency or some local noisy SMPS.
Just normal shit you need to work with or ignore.

the problem is that I need to measure power supply noise and ripple to below 10 mv which is why I need to null this.

so there is no solution or workaround?

Well we are hardly clairvoyant so without a screenshot showing settings and connections or not we have no idea what you see so can't advise with any degree of accuracy.

here are the pictures taken: https://www.eevblog.com/forum/projects/my-psu-design-ripple-and-noise-with-picture-measurements/msg4520213/#msg4520213

one of the images shows the loose probe and gnd readings

this is a 18mvpp square wave at 50hz, i tried to match what i saw in the first picture, and it doesn't look as noisy as yours, this of course not being probing a power supply but connecting a coaxial cable directly to a waveform generator, and the other image is what it looks like when a probe is connected straight to ground through the ground clip of the probe, so perhaps the noise really is there?

[VEGETA]

anyone knows how to cancel out the noise of the probe before using it? I mean when the probe is not connected to anything... it still displays some waveform which I think affect real measurements.

Probably only when set to high sensitivity and it's mains frequency or some local noisy SMPS.
Just normal shit you need to work with or ignore.

the problem is that I need to measure power supply noise and ripple to below 10 mv which is why I need to null this.

so there is no solution or workaround?

Well we are hardly clairvoyant so without a screenshot showing settings and connections or not we have no idea what you see so can't advise with any degree of accuracy.

here are the pictures taken: https://www.eevblog.com/forum/projects/my-psu-design-ripple-and-noise-with-picture-measurements/msg4520213/#msg4520213

one of the images shows the loose probe and gnd readings

this is a 18mvpp square wave at 50hz, i tried to match what i saw in the first picture, and it doesn't look as noisy as yours, this of course not being probing a power supply but connecting a coaxial cable directly to a waveform generator, and the other image is what it looks like when a probe is connected straight to ground through the ground clip of the probe, so perhaps the noise really is there?

well you setup is very clean then, of course you will get clean results. mine is messy and the nature of the design also noisy to some degree... but even if the probe is not connected to anything it will still show noise as seen in pictures.

i am using the alligator clip ground not the spring one... i will try using the spring one to get cleaner results soon. what do you use and recommend?

[Anthocyanina]

this is a 18mvpp square wave at 50hz, i tried to match what i saw in the first picture, and it doesn't look as noisy as yours, this of course not being probing a power supply but connecting a coaxial cable directly to a waveform generator, and the other image is what it looks like when a probe is connected straight to ground through the ground clip of the probe, so perhaps the noise really is there?

well you setup is very clean then, of course you will get clean results. mine is messy and the nature of the design also noisy to some degree... but even if the probe is not connected to anything it will still show noise as seen in pictures.

i am using the alligator clip ground not the spring one... i will try using the spring one to get cleaner results soon. what do you use and recommend?

yeah, that's preciesly my point, that perhaps it's not that the oscilloscope is erroneously showing you a nosiy trace, but maybe all that noise is actually part of the signal you're looking at? this because my tests show the 1022i is actually capable of displaying those cleaner traces. Have you unplugged everything else around it for the grounding test? for the picture of it grounded with the probe, i also used the long aligator clip on the probe tip to ground it, and it wasn't that noisy

[VEGETA]

this is a 18mvpp square wave at 50hz, i tried to match what i saw in the first picture, and it doesn't look as noisy as yours, this of course not being probing a power supply but connecting a coaxial cable directly to a waveform generator, and the other image is what it looks like when a probe is connected straight to ground through the ground clip of the probe, so perhaps the noise really is there?

well you setup is very clean then, of course you will get clean results. mine is messy and the nature of the design also noisy to some degree... but even if the probe is not connected to anything it will still show noise as seen in pictures.

i am using the alligator clip ground not the spring one... i will try using the spring one to get cleaner results soon. what do you use and recommend?

yeah, that's preciesly my point, that perhaps it's not that the oscilloscope is erroneously showing you a nosiy trace, but maybe all that noise is actually part of the signal you're looking at? this because my tests show the 1022i is actually capable of displaying those cleaner traces. Have you unplugged everything else around it for the grounding test? for the picture of it grounded with the probe, i also used the long aligator clip on the probe tip to ground it, and it wasn't that noisy

i have lots of devices connected to mains.

scope is connected to laptop via usb of course... and my circuit is a laptop charger 12v feeding my designed switching psu. i tried to separate everything to have it best as possible.

i will see if i can separate them more + post pics.

[markone]

I have question about this little toy :

is it still lacking the capability to trigger in single shot with time base over a certain values ?

I'm talking about the official Windows  software available for download at Owon site.

[Gyro]

I suspect that it is still missing, below (I think) 50ms/div, where it switches to roll mode. That's unless they've copied florentbr's software, which allows single trigger right down to the slowest timebase settings. It's quite a long time since I've used the stock OWON s/w though, so I can't say for certain.

Irrc you can probably download the s/w without having the scope. It should display a message when you set the timebase to 50ms/div or slower.

[markone]

I suspect that it is still missing, below (I think) 50ms/div, where it switches to roll mode. That's unless they've copied florentbr's software, which allows single trigger right down to the slowest timebase settings. It's quite a long time since I've used the stock OWON s/w though, so I can't say for certain.

Irrc you can probably download the s/w without having the scope. It should display a message when you set the timebase to 50ms/div or slower.

Hi Gyro, thank you very much for your prompt answer, i'll try that path.

[sx57]

Hello friends.
Can you kindly tell me if it is possible to convert vds 1022 into the isolated version using this?
https://www.walmart.com/ip/Adum3160-Digital-Signal-Audio-Power-Isolator-Usb-To-Usb-Digital-Isolator/1704951055?wmlspartner=wlpa&selectedSellerId=101260823
I know some users here have used something like this but not this exact product.
Is there any disadvantages or problems with using the usb isolator?
Will i need to change any parts on it's board?
Thank you.

[Gyro]

The VDS1022 will work with USB isolators containing the ADUM3106, that's what is included inside the I version. Unfortunately though, the Walmart isolator only has a 1W rated  isolated power converter (the one inside the VDS1022I is rated at 2W), so it won't be able to supply enough current to power it.

You could maybe arrange something with a USB 'Y' cable to supply extra power to the VDS after the isolator, maybe from a USB power bank, but you would be better looking for a USB isolator with at least 2W power capability.

[sx57]

The VDS1022 will work with USB isolators containing the ADUM3106, that's what is included inside the I version. Unfortunately though, the Walmart isolator only has a 1W rated  isolated power converter (the one inside the VDS1022I is rated at 2W), so it won't be able to supply enough current to power it.

You could maybe arrange something with a USB 'Y' cable to supply extra power to the VDS after the isolator, maybe from a USB power bank, but you would be better looking for a USB isolator with at least 2W power capability.

very informing answer mate.thanks alot

[sx57]

AHello .
i noticed that when i set my time division for more than 2us/div,the sampling rate goes lower probably because of the 5k memory.i need to work with 50s/div so i have a sampling rate of 5KS/s! Now if i divide 5000by 10 i get just 500hz max bandwith.is this right or i am hopefully very wrong?
I am worried and have serious buyer remorse now. I live in iran and bought it for 150 dollars because of sanctions and couldn't afford to buy better one.i need 50ms/div to see what is going on for one second but now i have just 500hz .i use it to measure the ripple and noise of the phone chargers that use smps. i could just use my pc soundcard  and get better results if i am understanding it right.please help . I hope i am wrong about 500hz bandwidth. i am really confused

[Anthocyanina]

with a sample rate of 5k samples per second, theoretically you could see unaliased waveforms up to 2500hz, but the software for the owon doesn't have anything but linear interpolation, so you see just the 2 samples per period and that looks like a triangle wave, like in the first picture.
The other pictures show how it would look like with a 2500hz sine wave at 50ms/, 500hz at 50ms/ and zoomed in, and 50hz at 50ms. At such slow timebase, why would you want to look at frequencies above 50hz? that would provide you nothing that you could work with visually. ripple is repetitive, so you could set the timebase such that it matches the frequency of interest, and if there is a high frequency riding on a much lower frequency ripple, it being repetitive, you could get the information you need by measuring each separately and not in the same acquisition.

[Gyro]

AHello .
i noticed that when i set my time division for more than 2us/div,the sampling rate goes lower probably because of the 5k memory.i need to work with 50s/div so i have a sampling rate of 5KS/s! Now if i divide 5000by 10 i get just 500hz max bandwith.is this right or i am hopefully very wrong?
I am worried and have serious buyer remorse now. I live in iran and bought it for 150 dollars because of sanctions and couldn't afford to buy better one.i need 50ms/div to see what is going on for one second but now i have just 500hz .i use it to measure the ripple and noise of the phone chargers that use smps. i could just use my pc soundcard  and get better results if i am understanding it right.please help . I hope i am wrong about 500hz bandwidth. i am really confused

Don't you think it's rather early to be getting serious buyer's remorse? It sounds as if you need to brush up on some scope fundamentals first [ Edit: https://www.eevblog.com/forum/beginners/oscilloscope-training-class-(long)/ ], particularly as you are planning to show meaningful results on your website. SMPS chargers run at 10s of kHz, so you're probably going to be looking in the 100us/div or faster. Their ripple and noise doesn't extend into the seconds timeframe. 50s/div is extremely slow (chart recorder) speed, so of course you're not going to be able to resolve high frequencies as that that sort of timebase. If, for example, you are looking for output voltage variation with load, you would be better with a DMM so that you can note the voltage reading as you adjust the load current, it's a different test from ripple and noise though.

You are working at a very low price point with difficult availability ( https://www.eevblog.com/forum/beginners/need-an-oscilloscope-on-a-very-low-budget/ ).  I remeber that even the VDS1022 was a financial stretch above your original goal, in fact you couldn't stretch to the galvanically isolated 'I' version (from PM). You can't expect multi Mega-samples of buffer at that price point. You do have an adequate tool for your stated (and general) purpose though.


P.S. The inputs of PC soundcards are AC coupled, there's no sound (just atmospheric pressure) at DC.

[sx57]

Hello and thank you both gyro and Anthocyanina for your helpful and informative answers.
Yes friends having buying remorse was wrong .after I worked more with this really worthy device, i saw that my time frame was not a good example but I needed it to see a wider picture.it seems that even If I had giga samples, the wide time frame still made it difficult to see the higher bandwidth anomalies and in the end I had to zoom and go into smaller timeframe anyway. unfortunately there is still limitations In most of the timeframes but i think the device is well balanced enough because the sample rates that are pre configured related to the selected timeframe, still show the would be visible to the eye bandwidth signals anyway. for example If I use a gan charger with 2 mhz switching frequency, to see ripple/noise I have to go into the filed of us/div any way right? please correct me if I am wrong.

 a complete sine wave needs two times sample rate but square wave 10 times right?
do I understand the samples formula correctly?
in measuring ripples, it is more important to see VRMS and VPP than the full shape of the anomalies.
as a test to understand the relation of sample rate and bandwidth,i used my ipod to output signals from 50hz to 22khz and limited the sampling to 250s/s through timeframe.it still detected the frequency of 22khz and showed the frequency right if I put the trigger button on the section of the screen where the signal was shown. how with 250s/s the device still shows the frequency and amplitude of 22khz with good precision ?
Shouldn’t the 22khz frequency need at least 44k/s sampling?
What am I understanding wrong?
Considering the above results are right , will the device still show the VRMS and VPP of 5 to 10mhz even when I use 50ms/div ? I will lose the shape but who will see it any way in such a big time frame . the ac voltage output is the most important factor. We just want to know if it is beyond a threshold or not.it seems it will still show and detect it with good precision right?

it’s ripple and noise measurement just to review not to make an adapter in the end.
BTW, Do you know of anyway to use pc memory for more sampling rate the way hantek 6022be  seem able to? i know hantek sucks in every other way ofcourse .
there was hantek 6074bd also with way better speces that I could somehow stretch myself into buying but i was so doubtful of the brand and had serious doubts about it’s noise floor that I just forgot it. i saw people here don’t trust hantek at all for very good and justified reasons.
looking forward to your informative answers. thanks again.

[Anthocyanina]

for example If I use a gan charger with 2 mhz switching frequency, to see ripple/noise I have to go into the filed of us/div any way right? please correct me if I am wrong.

To see things at 2MHz, yes, you would want to be in the 1us/ or faster (500ns/ 200ns/...) setting.

a complete sine wave needs two times sample rate but square wave 10 times right?
do I understand the samples formula correctly?

10 samples per period, for a square wave, would let you see the general shape of the square wave, yes

in measuring ripples, it is more important to see VRMS and VPP than the full shape of the anomalies.

I think both the shape, amplitude and frequency of the waveform will give you important information.

as a test to understand the relation of sample rate and bandwidth,i used my ipod to output signals from 50hz to 22khz and limited the sampling to 250s/s through timeframe.it still detected the frequency of 22khz and showed the frequency right if I put the trigger button on the section of the screen where the signal was shown. how with 250s/s the device still shows the frequency and amplitude of 22khz with good precision ?
Shouldn’t the 22khz frequency need at least 44k/s sampling?
What am I understanding wrong?

This i think it's the frequency counter being somewhat separate from the acquisition system? purely a guess, might be completely wrong about this, but i have other scopes that do mention working this way, so maybe the owon does too.

Considering the above results are right , will the device still show the VRMS and VPP of 5 to 10mhz even when I use 50ms/div ? I will lose the shape but who will see it any way in such a big time frame . the ac voltage output is the most important factor. We just want to know if it is beyond a threshold or not.it seems it will still show and detect it with good precision right?

It will likely not show the proper amplitude because it will definitely alias and you don't know where in the period the samples will be taken.

it’s ripple and noise measurement just to review not to make an adapter in the end.
BTW, Do you know of anyway to use pc memory for more sampling rate the way hantek 6022be  seem able to? i know hantek sucks in every other way ofcourse .
there was hantek 6074bd also with way better speces that I could somehow stretch myself into buying but i was so doubtful of the brand and had serious doubts about it’s noise floor that I just forgot it. i saw people here don’t trust hantek at all for very good and justified reasons.
looking forward to your informative answers. thanks again.

I don't think there's a way to get it to use more memory, I think all the processing is done in fpga and it just streams the processed acquisition to the computer for display, but also might be wrong about this. maybe with recording? maybe with florentbr's python software?

[sx57]

Thanks a lot my friend you helped me alot.