Author Topic: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)  (Read 234890 times)

0 Members and 3 Guests are viewing this topic.

Offline CatWhisker

  • Contributor
  • Posts: 13
  • Country: gb
    • My Github
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #225 on: March 16, 2018, 12:29:13 am »
Just for fun: OS X updates

https://github.com/tinwhisker/Owon-VDS1022

Busy year, so this got neglected. But I had a fresh install of 'macOS' (10.13.3) and thought I'd plug the scope in and fix up the scripts.

Fixes:
- Dumb typo - 'Owen'  :wtf:
- Latest 1.0.29 release.
- Added missing usbLib files.
- Changed the bootstrap script from pure 'bash' to AppleScript & 'bash' to simplify the driver install*.

(* I think later OS X builds are stricter for arbitrary folder access, so we need to elevate the privileges when copying the usblib files. Using just bash would never show the sudo password prompt, so instead it'll prompt to run the script instead.)

Remember to run a calibration.
 

Offline CatWhisker

  • Contributor
  • Posts: 13
  • Country: gb
    • My Github
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #226 on: March 16, 2018, 12:34:22 am »
I spotted 1.0.23 -> 1.0.29 that 'com.google.gson_2.1.0.v201203072145.jar' has appeared.
Are they calling home now?

I hadn't seen/used any build between those versions, so dont know when it was added.
 

Offline piotr99

  • Newbie
  • Posts: 2
  • Country: pl
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #227 on: March 18, 2018, 01:49:17 pm »
Just for fun: OS X updates
https://github.com/tinwhisker/Owon-VDS1022
Busy year, so this got neglected. But I had a fresh install of 'macOS' (10.13.3) and thought I'd plug the scope in and fix up the scripts.

CatWhisker many thanks for the upgrade.

I don't have VDS1022 yet but I plan to buy it. I have just installed the application on my MacBook Air according to your description.
It seems to be working properly, but I had to change the folder as below:

OSX:
Copy libusbpp-legacy-0.1.4.dylib to /opt/local/lib/libusb-legacy/ (Note: On OS X, these folders may need creating)
Copy libusb-0.1.4.dylib to /usr/local/lib
 

Offline The_Boots

  • Contributor
  • Posts: 35
  • Country: us
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #228 on: March 18, 2018, 07:57:29 pm »
Question for any 1022 owners: I just got one of these, but it periodically pops an error message up and had to reconnect. I'm assuming that's not normal, but thought I'd check if anyone else has this issue.
 

Offline CatWhisker

  • Contributor
  • Posts: 13
  • Country: gb
    • My Github
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #229 on: March 18, 2018, 08:13:24 pm »
It seems to be working properly, but I had to change the folder as below:

Copy libusb-0.1.4.dylib to /usr/local/lib

That should only apply only to the manual install guide. Thanks for spotting that.

Most OS X users should just grab the Owon-VDS-OSX.zip and use that.
 

Offline GyroTopic starter

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #230 on: March 18, 2018, 08:27:53 pm »
Question for any 1022 owners: I just got one of these, but it periodically pops an error message up and had to reconnect. I'm assuming that's not normal, but thought I'd check if anyone else has this issue.

No, not a common problem I don't think. The 1022 is very undemanding in terms of USB bandwidth etc.

A couple of points that come to mind...

- USB port, try another one. Maybe the socket has got loose and is glitching the power. Does wiggling the USB connector provoke anything?  Power requirement is under 400mA so shouldn't be an issue, even for older laptops.

- Assuming that you do not have the USB isolated 1022I, do you have a ground loop that is injecting current via the USB ground. Does it coincide with attaching probe ground clips, does it still happen with probes isolated.

- What os version and S/W version are you running? Does anything in particular provoke the problem?

That's about all I can think of, off hand.
« Last Edit: March 18, 2018, 08:30:01 pm by Gyro »
Best Regards, Chris
 
The following users thanked this post: The_Boots

Offline VitaliyKrym

  • Newbie
  • Posts: 6
  • Country: ca
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #231 on: March 18, 2018, 10:40:49 pm »
I had same issue with periodical reconnecting. After I converted it to "I" this issue disappeared. 
 

Offline The_Boots

  • Contributor
  • Posts: 35
  • Country: us
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #232 on: March 18, 2018, 11:07:39 pm »
Yes, it happens with other USB ports, no it doesn't happen when I jiggle the USB cable.

I'm not entirely surprised. I got it used directly from Amazon ("Like new"). I think it was returned, and I may have an explanation why.  :)
So much for getting a good deal, now I need to decide what I'm going to do for a working Oscilloscope...
 

Offline The_Boots

  • Contributor
  • Posts: 35
  • Country: us
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #233 on: March 19, 2018, 12:47:39 am »
Good to know. Due to various circumstances, Amazon offered to send me a non-used one, and they said since the shipping on a 1022 put it over the price of an I, they ended up sending me the isolated version. Can't complain. Hopefully it solves my problem, too!
 

Offline GyroTopic starter

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #234 on: March 19, 2018, 10:16:38 am »
Quote
... and they said since the shipping on a 1022 put it over the price of an I, they ended up sending me the isolated version.

That sounds like a win!  :-+

Ground loops can be a real issue for any non-isolated usb connected equipment - up to and including total destruction of the scope and PC (see Reply #225 as just one extreme example).

 You may want to investigate a little further - try putting a meter between a connector shield on your computer and to point where you were connecting your ground clip and see if there is any voltage present (DC and AC). It may give an insight into the possible ground loop path. It's a worthwhile exercise (educational at least) in case you pick up a low cost logic analyser or other equipment. I suspect that in your case, the voltage will be very low, but the loop resistance is also low, so surprising currents can still flow, enough to disrupt USB comms.
« Last Edit: March 19, 2018, 10:24:24 am by Gyro »
Best Regards, Chris
 
The following users thanked this post: The_Boots

Offline CatWhisker

  • Contributor
  • Posts: 13
  • Country: gb
    • My Github
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #235 on: March 19, 2018, 03:56:26 pm »
Question for any 1022 owners: I just got one of these, but it periodically pops an error message up and had to reconnect. I'm assuming that's not normal, but thought I'd check if anyone else has this issue.
No, not a common problem I don't think. The 1022 is very undemanding in terms of USB bandwidth etc.

I second this. I had mine running for the last ~4 hours to check if it happened to me - still stable.

Rather than another USB port, I suggest another PC if possible - drivers, internal hubs, shared chipset - can be a pig if the device is easily startled.
Especially if you have any funky USB stuff installed (USB sniffers, Wireshark USB tools, VM tools and the like).

FYI: Using 1.0.29 build on OS X.
 
The following users thanked this post: The_Boots

Offline Nullarbor

  • Contributor
  • Posts: 20
  • Country: 00
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #236 on: March 19, 2018, 08:11:55 pm »
...
- Latest 1.0.29 release.

Remember to run a calibration. ...

I don't have apple stuff, but nonetheless thank you CatWhisker for the info on the latest release AND ESPECIALLY the reminder to recalibrate.

I was having problems with my OWON 1022:

- DC levels were off by a long shot (showing say 9 volts instead of 5 volts)
- When in AC mode, the NULL level of the traces shifted up in steps when changing
  sensitivity (50mV, 100mV, etc). This repeated itself starting with 1V again from NULL. That was strange, so I decided to look here.

I thought my Owon hat gotten defective and was contemplating calling the vendor.
I installed the 1.0.29 software and went into the calibration mode.
It got stuck there on about 20% of the bar. I left it on for an hour.

Frustrated, I re-booted and erased all traces of the OWON software on the Netbook and Registry (it's windows 7).
In hindsight: this Netbook has no access to the internet, so it might have to do with this, since you mentioned the new version may want to "phone home".
Well spotted if I may say!

I installed version 28 and went into calibration mode. It went well and the DC levels
are correct again and the the AC levels at NULL are solid now too when switching through the voltage range.

I was very relieved about this, but have no clue as to what had gone wrong.

If something like this happens again I will film it.
 

Offline Nullarbor

  • Contributor
  • Posts: 20
  • Country: 00
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #237 on: March 19, 2018, 08:16:44 pm »
Question for any 1022 owners: I just got one of these, but it periodically pops an error message up and had to reconnect. I'm assuming that's not normal, but thought I'd check if anyone else has this issue.

I have this problem, but rarely. I just unplug and plug in again.
 

Offline Nullarbor

  • Contributor
  • Posts: 20
  • Country: 00
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #238 on: March 19, 2018, 08:25:27 pm »
Quote
... and they said since the shipping on a 1022 put it over the price of an I, they ended up sending me the isolated version.

That sounds like a win!  :-+

Ground loops can be a real issue for any non-isolated usb connected equipment - up to and including total destruction of the scope and PC (see Reply #225 as just one extreme example).

 You may want to investigate a little further - try putting a meter between a connector shield on your computer and to point where you were connecting your ground clip and see if there is any voltage present (DC and AC). It may give an insight into the possible ground loop path. It's a worthwhile exercise (educational at least) in case you pick up a low cost logic analyser or other equipment. I suspect that in your case, the voltage will be very low, but the loop resistance is also low, so surprising currents can still flow, enough to disrupt USB comms.

Interesting!
I just measured this with my simple means and get the following values:
32mV (from USB metal to probe GND)
2.4mA (from USB metal to probe GND)

Hope I measured this right.
 

Offline GyroTopic starter

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #239 on: March 19, 2018, 09:16:45 pm »
Interesting, as you say. The 32mV measurement confirms that there are no floating supplies, Y cap leakages etc. involved, ie. everything is grounded.

The fact that you were able to measure 2.4mA though a DMM current range (with its reasonably high value internal shunt resistor) implies that with a low impedance connection through the ground clip and PC ground the current could be significantly higher. It shows that 'Ground' really never is, when you consider voltage drops across current carrying conductors. There might also be some higher frequency currents involved that your meter wouldn't be able to detect.

This sort of low voltage, low impedance, hash is just the sort of thing that USB isolation is useful for, both to prevent circulating ground currents and reducing measurement noise.
Best Regards, Chris
 

Offline CatWhisker

  • Contributor
  • Posts: 13
  • Country: gb
    • My Github
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #240 on: March 28, 2018, 11:32:40 pm »
In hindsight: this Netbook has no access to the internet, so it might have to do with this, since you mentioned the new version may want to "phone home".

Poking around at the source, there is no 'phone home' feature - the devs are just leveraging GSON library as a better/cleaner mechanism for storing settings.

That's not to say the Google library isn't itself phoning Google, I havn't looked there yet, but at least Owon are behaving here.


In other news, for OS X users: If anyone wants to try a 'native' style package, grab the WIP archive from https://github.com/tinwhisker/Owon-VDS1022
Notes:
Some features (like 'save to bitmap') doesn't work yet.
Driver install isn't done.
'Help' works (re-packed as a PDF, rather than CHM).
'Update' button removed.
 

Offline Nullarbor

  • Contributor
  • Posts: 20
  • Country: 00
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #241 on: March 31, 2018, 02:44:22 pm »
There seems to be a problem with the 1.0.29 Software.

I installed it and tried calibrating again. It gets stuck at the beginning now.
There is no change in behaviour when the netbook is connected to the internet,
so it has nothing to do with my netbook not being online before.

When clicking "cancel" and then "yes" to the stuck calibration procedure, nothing happens. I had to kill it with the task manager.

I deleted all files of the Owon and installed Version 28 again.
The calibration works just fine again in this version.

Any ideas?

Happy Easter!
 

Offline Nullarbor

  • Contributor
  • Posts: 20
  • Country: 00
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #242 on: March 31, 2018, 02:51:22 pm »
Interesting, as you say. The 32mV measurement confirms that there are no floating supplies, Y cap leakages etc. involved, ie. everything is grounded.

The fact that you were able to measure 2.4mA though a DMM current range (with its reasonably high value internal shunt resistor) implies that with a low impedance connection through the ground clip and PC ground the current could be significantly higher. It shows that 'Ground' really never is, when you consider voltage drops across current carrying conductors. There might also be some higher frequency currents involved that your meter wouldn't be able to detect.

This sort of low voltage, low impedance, hash is just the sort of thing that USB isolation is useful for, both to prevent circulating ground currents and reducing measurement noise.

Thank you Gyro. I will keep an eye on that, hopefully learning more.

Does what you're saying mean that all "normal" NON-ISOLATED oscilloscopes
are somewhat inferior to the isolated ones (which are quite rare)?

Thank You!
 

Offline GyroTopic starter

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #243 on: March 31, 2018, 04:58:48 pm »
Haha, well I think you'd get some get some indignation from folks with "normal" scopes that their units were "somewhat inferior"  ;)

However, yes, any system which involves ground loops, particularly where that loop path passes through mains ground is subject to issues. Audio systems, for instance, tend to suffer from hum problems if audio ground connects to mains ground in both 'sources' and amplifiers. What it does mean is that you need to always take ground loops into account when measuring - one of the reasons why many Bench PSUs have floating output and separate mains ground terminal. The ideal is to have the 'system' reference to mains ground at a single point, often not possible in the real world with several pieces of grounded test equipment and debugging on a piece of equipment that is also mains ground referenced. Not a show stopper, but something certainly to be aware of.

The problem is particularly bad with USB instruments though. Computer USB ports are fairly fragile, the USB spec specifies that the outer braid and connector shells should only be grounded at the PC end so, for instruments that follow the spec, the only ground return path is the little black wire in the USB cable. In fault situations where significant current flows through the earth clip, the black wire fries, followed almost instantly by the USB 5V and data signals! It also means that noise levels can be higher than integrated bench type scopes. That makes the USB isolation in the Owon (I) versions highly desirable, as far as I'm aware they are the only low cost USB scopes that can do this, Hanteks (not sure about the Siglent Analog discovery products I haven't looked into them) and the like like rely on lots of USB bandwidth - beyond the reach of current USB isolator chips.

Sorry, I'm rambling a bit. In terms of noise, it isn't too critical for an 8 bit scope like the VDS1022(I), but the isolation does give some benefit and most importantly in protecting the PC from damage. Any situation where a low voltage low impedance current could potentially flow through the ground loop will benefit from isolation.

Once you get into the realms of sensitive or precision measurements are involved, then various measures become applicable, I mentioned single point system grounding, there are also differential probes for common mode rejection, decent bench meters have floating inputs with a separate guard terminal, both screened from mains ground, etc.

Thanks for flagging the bug in the 1.0.29 S/W it adds to the group inteligence in lieu of release notes! - Owon do seem pretty accident prone. You may have seen a few posts back, that I uploaded my collection of previous S/W versions  (I see it still has a week to run). If you have any problems with 1.0.28, then I used 1.0.23 for a long time without noticing any issues... https://uploadfiles.io/70a4d (appropriate virus check cautions apply!).
« Last Edit: March 31, 2018, 05:27:39 pm by Gyro »
Best Regards, Chris
 
The following users thanked this post: cosmin1

Offline Riaj159

  • Newbie
  • Posts: 9
  • Country: bd
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #244 on: July 11, 2018, 06:49:25 pm »
recently i have opened up my OWON VDS1022(non isolation version) as there was nothing left in my home for teardown & discovered that copper strip is not making any contact with the outer shell that should act as a Electromagnetic shield. i have fixed that issue,all owner of OWON VDS1022/OWON VDS1022i should check for it.also discovered  that lowest volts/div(5mv) without any probe the amount of average noise vpp&vrms depends on the computer/laptop usb ports. some ports induce less noise,some ports induce more noise,you have to discover which usb port is best for your lowest volts/div measurement.some channel are also by born more noise free then other channel,for example my channel 1 with settings(5mv volts/div,1ms,DC,without any probe,persistence: infinite & make sure that other channel is off.if you turn on other channel or turn on other channel and make it off again in that case you will not get much lower sensitivity until you power off the device and turn on with only that by born less noisy channel) i can get vpp: .600mv vrms: .138mv where my 2nd channel never reach close to that level what ever i do.Someone have success with owon VDS1022 software to run on OSX & Debian, i have tried on raspbian but no success(software works but device not found),is there any one who have tried owon VDS1022 on Raspberry Pi??? it will be then great DIY  handheld oscilloscope with display & touch interface!
 

Offline GyroTopic starter

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #245 on: July 12, 2018, 06:18:48 pm »
Interesting input Riaj159,  I guess that copper strip is easily damaged by the sharp front edge of the Aluminium case shell extrusion whenever the board is inserted.

My early revision VDS1022I (photos on page 1 of this thread) didn't have the copper strip fitted, it was a later addition (it was there by the revision that Mark_O purchased). I checked today, and my case is reliably floating relative to the front panel ground.

Your post reminded me that I need to do some noise comparisons on my 'I' version to see what effect grounding the case to input ground has at low levels. On the one hand, the floating case makes accidents less likely, on the other, it may increase noise and if it does suffer insulation breakdown then I'm uncertain what part of the board would be 'struck' first. Not a problem in practice as it would be very bad to float the ground leads on any scope to anywhere near enough voltage to cause insulation breakdown, even a well isolated one.

I'm also curious about the progress on other platforms, it's all gone a bit quiet. I'm now a Raspberry Pi owner too and I agree, it would make a very nice portable touch screen setup.
Best Regards, Chris
 

Offline Riaj159

  • Newbie
  • Posts: 9
  • Country: bd
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #246 on: July 13, 2018, 04:50:44 am »
tnx gyro for reply,I am also interested about the effect of grounding the case on the isolated version.what if i replace the on board 100mhz clock with 125mhz clock for the sake of overclocking the ADC??? my unit have the MXT2088(equivalent to AD9288 ). hantek dso5102p oscilloscope actually use same ADC but let them run at 125mhz by overclocking so that they can achieve 1gsps with 8 ADC(physically 4 AD9288 ).Though 125mhz might not be guaranteed to achievable due to stability in every case & what will be the consequence on the FPGA site is unknown,also owon software might be another issue related to time base(maybe there is hope as owon software released with code? ).currently don't have any 125mhz near clock source,but it’s just a thought...or i'm too greedy for extra bandwidth  ::)  >:D .
 

Offline GyroTopic starter

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #247 on: July 13, 2018, 08:05:56 pm »
Haha, yes, I think you are too greedy for extra bandwidth - actually you mean sample rate, the bandwidth is a measure of the analogue input stage frequency response and rise time.

I don't know for sure, but I think overclocking at 125MHz without changes to the FPGA code or PC software would lead to some strange results. At minimum I think all the measurements and timebases would be off. I don't think anyone has reverse engineered the FPGA code.
« Last Edit: July 13, 2018, 08:07:27 pm by Gyro »
Best Regards, Chris
 

Offline Riaj159

  • Newbie
  • Posts: 9
  • Country: bd
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #248 on: July 17, 2018, 11:43:09 am »
tnx for sharing your opinion gyro :) :) :)
 

Offline noopwafel

  • Newbie
  • Posts: 2
  • Country: nl
Re: OWON VDS1022I Quick Teardown (versus the Hantek 6022be)
« Reply #249 on: August 21, 2018, 08:06:54 am »
I got a VDS1022I very cheaply from aliexpress recently. It's a nice toy and really portable which is great; thanks for all the commentary in this thread, I appreciated it. Not yet sure if it has actual value as a scope but we'll see.. :)

To talk to it over USB, you just need to upload the bitstream and then send a bunch of pretty simple commands. I started work on a sigrok driver a few days ago; I plan to finish+upstream it, but just in case I get bored first then I'll keep the unfinished code on my github for anyone who comes after me (right now it's a mess which just hard-codes all the settings, but hey, the signal *looks* OK in pulseview as long as you don't look at the scale).

It looks like the VDS1022(I), the VDS2052 and the RDS1021 are all very similar ('tiny' models), with this 5K internal buffer on an internal MCU. They seem to need slightly different commands but hopefully adding support for the other models should be simple enough if anyone ever wants this. :)

The higher-end models ('smart' models: VDS 2062,2064,3102,3104) seem to be very different, and they talk SCPI instead.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf