New Episode:OWON Show #3 XDS 12-bits resolution

[OWON]

Hi, guys

It's a good summer here in China.  Hope everything goes well in your side as well.

We still keep on video shoting to introduce our product for you. It's also grateful to get your advice and know what aspect you are interested in (such as Wifi connection, intensity gradation).

OK. Go back on video, in this video, we will compare the oscilloscopes from 8-bits to 12-bits. You could see the difference between them, and know why you should purchase a 12-bits oscilloscope. However, as the content is too much to cut into one video, so we take it apart into 2 episode. This time, pls see the episode 1 of 12-bits introduction.

https://youtu.be/-f5M4fc6bV4

Little market news, OWON will attend International Microwave Symposium in San Francisco,California,U.S.A during May 22~27, our booth is at No.2619 in Floor 2. Welcome if you have interest in our product. Here is the official website of this exhibition:http://www.ims2016.org/

[OWON]

New episode (part 2 of 12-bits oscilloscope) will release soon within these days.

[rf-design]

Hi producer,

I am interested in 12bit scopes but found up to know that most of them lack analog accuracy. My application have wide voltage range transitions and it is not only important to have a high resolution for signal detail. But accuracy of the measurement at the detail seems to be distorted by amplifier effects. We also have some 30 year old Tek 11300/11400 here because they made a very precise 11A33 which must complement our measurement because it is unsurpast by any scope.

So could you provide some information about your frontend amplifier?

1. At what fullscale factor related to the ADC the amplifier get clipped?
2. What is the analog offset range for the most sensitive input range?
3. If I apply a pulse what is the voltage droop after initial settling?
4. If I overdrive the input amplifier how the recovery transient is looking?

Excuse for the detailed questions but I need the accuracy, not only the resolution.

At high resolution I listen music, for measurement I must trust accuracy.

[pascal_sweden]

For my understanding: Isn't a low background noise level in the scope, and a high ADC resolution (12-bit) enough, if the ADC does not use interleaving?

Or are you refering to the linearity of the amplifiers in the front-end stage?

[nctnico]

Or are you refering to the linearity of the amplifiers in the front-end stage?

When it comes to accuracy the first thing you'll want is an extremely very flat frequency response. Unfortunately 0.1dB is already an error of about 1% so you'll see that it is almost impossible to build an oscilloscope which is accurate over a wide frequency range. 12 bits is nice to look at small wrinkles in a signal by using the vertical zoom but it won't be very useful for absolute accuracy. If you need to measure the amplitude of a signal very accurately you need different kind of equipment.

[rf-loop]

Or are you refering to the linearity of the amplifiers in the front-end stage?

When it comes to accuracy the first thing you'll want is an extremely very flat frequency response. Unfortunately 0.1dB is already an error of about 1% so you'll see that it is almost impossible to build an oscilloscope which is accurate over a wide frequency range. 12 bits is nice to look at small wrinkles in a signal by using the vertical zoom but it won't be very useful for absolute accuracy. If you need to measure the amplitude of a signal very accurately you need different kind of equipment.

Yes. If think 12 bit, resolution is 1:4096. I can not see any kind of possible to make oscilloscope, say example 0-100MHz so that true accuracy is 12bit over this frequency band. "Mission impossible".  I can see it is possible in theory but if transfer this theory to practice it is so expensive that no one have money to buy it. And why even try? For what? 
Because we need then ask mandatory question. How you connect DUT so that you do not destroy this accuracy. Even with quite low 100MHz freq. Of course if we just use one spot frequency, case is much more easy, but do we than anymore talk about common oscilloscope.   I can tell that full 8bit accuracy over 0-100MHz low frequency band is - more than just difficult. (btw, I have never seen this miracle)

If need measure some level more accurate it really need different tools as @nctnico told.

[pascal_sweden]

Please define "different tools"

[rf-design]

For my understanding: Isn't a low background noise level in the scope, and a high ADC resolution (12-bit) enough, if the ADC does not use interleaving?

Interleaving is not interesting now because at frequencies of of interest most scopes are doing interleave with spur levels many 10dB's worse than a medium class spectrum analyser. Also because of the bad oscillators in scopes the inverse mixing give high close in noise floors. I expect that it would take a decade before scopes frequency performance get closer to an analyzer.

Or are you refering to the linearity of the amplifiers in the front-end stage?

I give you an example: If I want to measure the power device switch performance going from some hundred volt to some hundred millivolt I see with most scope either the overdrive recovery transient or do not have enough dynamic without overdrive. The switch on tail from the device is sometimes hidden from the scope artifact. The Tek 11A33 excels here because you overdrive in off-condition and could see the tail with very little artifact.

[testarea]

As far as I know, very few of manufacturers can make 12 bit resolution oscilloscope, I just saw Lecroy advertised it on magazines, but the prices are expensive. If the price is reasonable from OWON 12 bit oscilloscope, that is good choice 

[SW_design]

  sounds good, it might be useful to observe small signal, but I wonder

1. what is the max record length ?
2. 12bits high resolution is realized by hardware or software ?
3.  And is it expensive

[rf-loop]

sounds good, it might be useful to observe small signal, but I wonder

1. what is the max record length ?
2. 12bits high resolution is realized by hardware or software ?
3.  And is it expensive

If you read even one minute this EEVblog and using search you may find all answers and more.

Example this: https://www.eevblog.com/forum/testgear/new-owon-xds-oscilloscopes/msg770683/#msg770683


Shortly
1. 40M
2. Hardware.
3. http://www.saelig.com/product/xds3102a.htm

[pxl]

sounds good, it might be useful to observe small signal, but I wonder

1. what is the max record length ?
2. 12bits high resolution is realized by hardware or software ?
3.  And is it expensive

If you read even one minute this EEVblog and using search you may find all answers and more.

Example this: https://www.eevblog.com/forum/testgear/new-owon-xds-oscilloscopes/msg770683/#msg770683


Shortly
1. 40M
2. Hardware.
3. http://www.saelig.com/product/xds3102a.htm

I still not understand this product all: what is the 40 M for, if there is no search func., no serial decoding. Moreover no stats, no advanced math (except 8192 pts FFT), no nothing. It is basically just able to show a waveform and that's all. (Probably I missed all these in the manual).

[rf-loop]

sounds good, it might be useful to observe small signal, but I wonder

1. what is the max record length ?
2. 12bits high resolution is realized by hardware or software ?
3.  And is it expensive

If you read even one minute this EEVblog and using search you may find all answers and more.

Example this: https://www.eevblog.com/forum/testgear/new-owon-xds-oscilloscopes/msg770683/#msg770683


Shortly
1. 40M
2. Hardware.
3. http://www.saelig.com/product/xds3102a.htm

I still not understand this product all: what is the 40 M for, if there is no search func., no serial decoding. Moreover no stats, no advanced math (except 8192 pts FFT), no nothing. It is basically just able to show a waveform and that's all. (Probably I missed all these in the manual).

There is many reasons for memory.

One and important is for keep samplerate high with also slower timebases.
With this memory it can keep full 1GSa/s down to 2ms/div for one channel (40M) and full 500MSa/s down to 2ms/div for 2 channels. (20M + 20M). So, it have maximum 40ms waveform with full sample rate.

No serial decoding? Of course there is serial triggering standard  and (Option) serial decoding. IIC, SPI, UART/RS232 and  sub Option for decoding there can add also CAN.

40M is not much!  And for two channels 20M+20M just barely enough but still "small" memory today.

(Of course it do not have as rich decoding as example in ZLG what have CAN, LIN, I2C, SPI, UART, USB, PS/2, DALI, Wiegand, 1-Wire, DS18B20, HDQ, SD_SPI, SD_SD, IrDA, Manchester, Differential Manchester, Miller, DHT11, SHT11 , NEC, RC5, RC6, FlexRay, CAN FD).

40ms with full samplerate is some times nice.
Here 50Hz square cycle to cycle time jitter measurement. Trigger edge and next edge in same capture and zoomed for resolution to see jitter. (here used infinite persistence)  Of course this measurement can do with long trigger delay but here can also see whole cycle and trigger edge is not out from image.  (and this can also use for this kind of measurement because clock do not drift everywhere - because this model reference is in 1ppm class unlike many other cheap scopes. (and it is also tested using High-End reference and long time period.)

Here one trivial basic example. Do this with small memory so that all is also visible on the screen.
(note that 1ppm from 20ms period is 20ns.  In image p-p jitter is under 500ps. It is under 25ppb p-p from cycle period and persistence data collected ~ 15min.  (Honestly I do not know which one here in this "test" is DUT (siggen or scope))

[babybaza]

Look nice! Expect more good products from OWON.

[EEVblog]

sounds good, it might be useful to observe small signal, but I wonder
1. what is the max record length ?
2. 12bits high resolution is realized by hardware or software ?
3.  And is it expensive

MODERATOR MESSAGE:
The user SW_design is OWON, it's a sockpuppet account.
Congratulations  OWON, you just got yourself banned from the forum. Sockpuppet accounts are not allowed.

EDIT: Geeze, the following accounts are also OWON sockpuppet accounts!
babybaza
testarea
baoblackcoal
Ross Wassie

[Muttley Snickers]

Shifty Buggers.   

[GEuser]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

Thanks for the update dave , it's a good thing that you got the Gonads .

[tautech]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

Thanks for the update dave , it's a good thing that you got the Gonads .

They're no special mates of mine. (if you mean OWON)

Dave acted appropriately. 
Last I looked he hadn't labeled each as "user banned".

[GEuser]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

Thanks for the update dave , it's a good thing that you got the Gonads .

They're no special mates of mine. (if you mean OWON)

Dave acted appropriately. 
Last I looked he hadn't labeled each as "user banned".

http://www.thefreedictionary.com/hobbyist , read that "Carefully" and and then adjust your signature i suggest .

Shame on Dave that he does not work fast enough for you too!

[tautech]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

Thanks for the update dave , it's a good thing that you got the Gonads .

They're no special mates of mine. (if you mean OWON)

Dave acted appropriately. 
Last I looked he hadn't labeled each as "user banned".

http://www.thefreedictionary.com/hobbyist , read that "Carefully" and and then adjust your signature i suggest .

Shame on Dave that he does not work fast enough for you too!

Na.....old, blind and stupid. 
PM me the link.

[nctnico]

http://www.thefreedictionary.com/hobbyist , read that "Carefully" and and then adjust your signature i suggest .

Shame on Dave that he does not work fast enough for you too!

OMG!

[CustomEngineerer]

Its all fun and games until someone gets banned for using sockpuppet accounts. Then its freaking hilarious.

After seeing babybaza's comment I was starting to think something looked fishy, then next post, banned. 

Thanks and nice work Dave! 

Edit: Typo

[EEVblog]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

I don't know Tautech, and IIRC there have been no complaints against him, and there are no sockpuppet account issues.
Do you have a point?

[ebastler]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

http://www.thefreedictionary.com/hobbyist , read that "Carefully" and and then adjust your signature i suggest .
Shame on Dave that he does not work fast enough for you too!

GEuser -- what are you getting at here? I don't see how being a Siglent distributor makes tautech a "mate" of Owon; and I don't see why it should be a contradiction to be an electronics hobbyist and be a Siglent distributor at the same time.  I have read many balanced and well-informed posts from tautech on this forum -- plus the occasional post which stresses the benefits of Siglent gear, which is perfectly fine with me given that tautech discloses his affiliation in his signature.

GEuser, I think a clarification or an apology is in order here.

[GEuser]

What you think about your mate now tautech , that mob has always been dodgy , my condolences .

Thanks for the update dave , it's a good thing that you got the Gonads .

They're no special mates of mine. (if you mean OWON)

Dave acted appropriately. 
Last I looked he hadn't labeled each as "user banned".

http://www.thefreedictionary.com/hobbyist , read that "Carefully" and and then adjust your signature i suggest .

Shame on Dave that he does not work fast enough for you too!

Na.....old, blind and stupid. 
PM me the link.

I'm glad at least you do understand what i mean going by that response .