Author Topic: At last, Siglent's SDS5054X touchscreen DSO...  (Read 23033 times)

0 Members and 1 Guest are viewing this topic.

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17663
  • Country: nl
    • NCT Developments
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #325 on: September 15, 2019, 08:56:57 pm »
Still I'm wondering why you'd want to synchronise an oscilloscope with a reference clock. Every measurement is related to the trigger point anyway and with a trigger jitter of 100ps for the SDS5000X you lose any advantage of having an external clock anyway. For comparison: the good old Agilent 54845A has a trigger jitter of 8ps but it doesn't have an external clock input.

Actually... while studying the specs more closely I notice the 1GHz version has a shortest time/div setting of 200ps/div. But if the trigger jitter is 100ps then the signal would jump left/right for half a division.
« Last Edit: September 15, 2019, 09:01:15 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 16013
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #326 on: September 15, 2019, 09:06:37 pm »
Clock source indication is provided by the arrow in the circle in the bottom right symbol above the time.
Arrow pointing inward indicates an external Ref clock while arrow pointing out indicates the internal clock.


This from beta FW FW 0.8.7***
« Last Edit: September 15, 2019, 09:10:35 pm by tautech »
Avid Rabid Hobbyist
 

Offline randomOracle

  • Newbie
  • Posts: 3
  • Country: de
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #327 on: September 15, 2019, 09:09:25 pm »
There has been a change in the clock Ref scheme for SDG2000X in very recent firmware and I suggest you install it. [...]

VERY nice!! Thank you for this great advice! Now it is actually usable! Indeed, once the clock is set to external on the AWG and the scope is turned on, the icon changes.


Also SDG5000X beta firmware versions are now much progressed from that in units released to the marketplace and we hope for a public release soon.
I have this beta firmware so I can check for this issue you report but only with the lesser SDG1032X but it has the new clock scheme also.

Can you confirm you are using V0.8.2R1 firmware in SDS5000X ?

I can confirm I have 0.8.2R1 in the SDS5000X. Looking forward to the new firmware!

Clock source indication is provided by the arrow in the circle in the bottom right symbol above the time.
Arrow pointing inward indicates an external Ref clock while arrow pointing out indicates the internal clock.

Yes, but it shows the arrow even if there is no external clock; at least with my firmware version (there is not X whatsoever to indicate that the external clock could not be detected).

BTW, do you also happen to know a good compatible high-speed differential probe for the SDS5000X? (specs could/should be similar to the LeCroy AP033 which is 500 MHz, 5V CMVR, 50 ohms terminated, input capacitance is low).
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 16013
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #328 on: September 15, 2019, 09:25:46 pm »
BTW, do you also happen to know a good compatible high-speed differential probe for the SDS5000X? (specs could/should be similar to the LeCroy AP033 which is 500 MHz, 5V CMVR, 50 ohms terminated, input capacitance is low).
Sorry I don't as Siglent's offerings of differential probes are abismall for use with a scope like the 5kX.  :(

I expect they are working on something but I don't know for sure however this is a very good question and I've endeavour to find out something later today when the factory is on deck.
Meanwhile I'll check if the SAP1000 active probe offers differential operation or if the ref connection is BNC ground referenced.
Avid Rabid Hobbyist
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 2087
  • Country: hr
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #329 on: September 15, 2019, 09:30:42 pm »
BTW, do you also happen to know a good compatible high-speed differential probe for the SDS5000X? (specs could/should be similar to the LeCroy AP033 which is 500 MHz, 5V CMVR, 50 ohms terminated, input capacitance is low).
Sorry I don't as Siglent's offerings of differential probes are abismall for use with a scope like the 5kX.  :(

I expect they are working on something but I don't know for sure however this is a very good question and I've endeavour to find out something later today when the factory is on deck.
Meanwhile I'll check if the SAP1000 active probe offers differential operation or if the ref connection is BNC ground referenced.

It is classic single ended active probe. It is ground referenced.
 
The following users thanked this post: tautech

Offline tautech

  • Super Contributor
  • ***
  • Posts: 16013
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #330 on: September 15, 2019, 09:32:34 pm »
BTW, do you also happen to know a good compatible high-speed differential probe for the SDS5000X? (specs could/should be similar to the LeCroy AP033 which is 500 MHz, 5V CMVR, 50 ohms terminated, input capacitance is low).
Sorry I don't as Siglent's offerings of differential probes are abismall for use with a scope like the 5kX.  :(

I expect they are working on something but I don't know for sure however this is a very good question and I've endeavour to find out something later today when the factory is on deck.
Meanwhile I'll check if the SAP1000 active probe offers differential operation or if the ref connection is BNC ground referenced.

It is classic single ended active probe. It is ground referenced.
OK, thanks. Saved me getting mine out.  :)
Avid Rabid Hobbyist
 
The following users thanked this post: 2N3055

Offline tautech

  • Super Contributor
  • ***
  • Posts: 16013
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #331 on: September 16, 2019, 01:29:02 am »
BTW, do you also happen to know a good compatible high-speed differential probe for the SDS5000X? (specs could/should be similar to the LeCroy AP033 which is 500 MHz, 5V CMVR, 50 ohms terminated, input capacitance is low).
Sorry I don't as Siglent's offerings of differential probes are abismall for use with a scope like the 5kX.  :(

I expect they are working on something but I don't know for sure however this is a very good question and I've endeavour to find out something later today when the factory is on deck.
Meanwhile I'll check if the SAP1000 active probe offers differential operation or if the ref connection is BNC ground referenced.
Quick reply from the factory just minutes after they opened today.  :)

We are developing the probe adapters to adapt LeCroy Probus probes and Tek probes (with TekProbe interface). The one for LeCroy will be available by the end of  this year, and the one for Tek will be available in Q1, 2020.

High BW Differential probes are coming but sometime later so these adapters are the interim solution.
Avid Rabid Hobbyist
 
The following users thanked this post: 2N3055

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 2087
  • Country: hr
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #332 on: September 16, 2019, 06:21:49 am »
BTW, do you also happen to know a good compatible high-speed differential probe for the SDS5000X? (specs could/should be similar to the LeCroy AP033 which is 500 MHz, 5V CMVR, 50 ohms terminated, input capacitance is low).
Sorry I don't as Siglent's offerings of differential probes are abismall for use with a scope like the 5kX.  :(

I expect they are working on something but I don't know for sure however this is a very good question and I've endeavour to find out something later today when the factory is on deck.
Meanwhile I'll check if the SAP1000 active probe offers differential operation or if the ref connection is BNC ground referenced.
Quick reply from the factory just minutes after they opened today.  :)

We are developing the probe adapters to adapt LeCroy Probus probes and Tek probes (with TekProbe interface). The one for LeCroy will be available by the end of  this year, and the one for Tek will be available in Q1, 2020.

High BW Differential probes are coming but sometime later so these adapters are the interim solution.

I think those adapters are great solution (if they don't get crazy with the price). Used Lecroy probes sometimes can be had for a good price, and Tektronix has some very fine and quite special probes.
 

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 817
  • Country: at
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #333 on: October 05, 2019, 03:55:57 am »
I took an interesting (but slightly off-topic) discussion about trigger jitter here from another thread:
https://www.eevblog.com/forum/testgear/500-mhz-4-channel-oscilloscope-recommendation-mso8000-any-good/

Here you go.

100 KHz 5ns pulse with 2ns risetime, fastest I have from a SDG6022X. Siglent 1 GHz rated BNC cable.
NO clock syncing, just 2 standalone instruments.

Dots and Infinite persistence.

It is quite obvious that there is no visible jitter for a digital trigger system with properly implemented trigger point interpolator and the screenshot with infinite persistence shows just that.

I don't think the 100ps jitter specification is really meaningful, since I've yet to come across a case where jitter becomes actually visible, even at 200ps/div. This is clearly some worst case specification "by design" and it's a peak value and not RMS on top of that.
I have given this some thought and I think there is something odd going on here. Look at how the signal converges into a single line near the trigger threshold an then smears out. The image would suggest that the trigger jitter is zero but this can't be true because above and below the trigger point the signal starts to smear horizontally. I have not seen a similar effect on other oscilloscopes. On those I see a signal which keeps about the same width from 20% to 80% of the rise time.
For example:


I think Siglent's trigger interpolator takes a very small part of the signal and uses that to place it at the trigger point which creates a somewhat distorted image. It is even possible the trace smears out more horizontally than when using a larger part of the signal.

The spotted problem (if you can call this minor effect a problem at all) is just noise. We should distinguish between trigger jitter introduced by noise in the signal path and the signal itself (which is just inevitable) and jitter caused by the trigger signal processing (the trigger detector and interpolator).

Yes, it seems like the trigger interpolator on the Siglent strictly eliminates any noise at the trigger point, so that noise becomes all the more noticeable with increasing distance. We see a similar effect in the RTM3000 screenshot, just not as pronounced and at the expense of an overall noisy transition and a less precisely defined trigger point.

In any case, the SDS5000X can yield much nicer results as soon as we get rid of the noise, as can easily be demonstrated with a proper setup.

The following screenshot demonstrates two facts at the same time:
1.   Minimal jitter at the trigger point and very little noise
2.   Minimal jitter between channels, even when they do not share the same ADC

250MHz square wave, fed into channels 2 and 4 in parallel through wideband (12.4GHz) power splitter. Since the signal path lengths between splitter and DSO inputs cannot be exactly the same – they appear to differ by about half a millimeter – there is some inevitable skew between the two channels (which is quite welcome here).

•   Dots display mode to make absolutely sure we get the true signal shape (even though it doesn’t make any difference in this case)
•   16x averaging in order to get rid of the noise
•   Infinite persistence recording for 3 minutes to catch any jitter effects.


SDS5104X Pulse 250MHz Avg16 Skew Pers3m

With a picture like that, at 200ps/div, I think we can safely state that there is no visible jitter.

Now let’s have a look at the automatic measurements. The T@M measurement tells us the distance of the nearest rising edge from the trigger point. According to this, we have an error of -10ps for the trigger channel 4 and a pk-pk jitter of 23ps for the measurement data. The standard deviation of 7.25ps is equivalent to the RMS jitter.

Surprisingly, the non-triggered channel 2 yields a very similar pk-pk jitter of 22.7ps, but a lower standard deviation (= RMS jitter) of just 5.86ps.

As can be seen, the channel to channel skew has a peak to peak variation of 40ps with a standard deviation of 8ps which is only slightly worse than the trigger jitter itself. The mean value differs by 8ps (22 instead of 30) which is probably because the channel skew has been measured in the main (not zoom) window, where the timebase is 10 times longer and time resolution for interpolated sub sample periods gets worse.

In general it appears a bit difficult to deal with single digit picosecond values in a scope that has a 200ps sample interval.

« Last Edit: October 05, 2019, 08:55:36 am by Performa01 »
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 2087
  • Country: hr
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #334 on: October 05, 2019, 10:55:48 am »
I just wanted to make a note that Tautech's image shows that effect because of him using SDG6000X for signal source in pulse mode. What you see is SDG6000X undulation problem.
I see absolutely same image on my Kesight, if using pulse mode on SDG6000X.
 

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 817
  • Country: at
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #335 on: October 05, 2019, 12:06:35 pm »
I’m not sure if this really is the “undulation” of the SDG6000X, because this effect used to be much stronger than what we see here in Tautech’s screenshot.

I can show a similar measurement as before, but this time with the SDG6052X, hence only 150MHz and much slower edges (1ns).

150MHz square wave, fed into channels 2 and 4 in parallel out of the two channels of the SDG6052X in tracking mode. Of course we have again some skew between channels, if only because of slightly different cable lengths.

•   Dots display mode
•   16x averaging in order to get rid of the noise
•   Infinite persistence recording for 10 minutes to catch any jitter and noise effects.


SDS5104X Pulse 150MHz Avg16 Skew Pers10m

At 200ps/div, the visible jitter is still pretty much negligible.

Let’s have a look at the automatic measurements again. We now have an error of 8.8ps for the trigger channel 4 and a pk-pk jitter of only 10.4ps for the measurement data! This proves that the SDG6000X is actually a lot better in terms of noise and jitter than the pulse source used for the previous test. I’m really glad that 2N3055 motivated me to repeat this test with the SDG6052X, because the standard deviation (aka RMS jitter) of <1.43ps is fantastic and with this we should be able to compete with the big boys ;)

The non-triggered channel 2 yields a slightly higher pk-pk jitter of 14.5ps, and the standard deviation is 2.26ps.

The channel to channel skew measurement has a high peak to peak variation of 275ps with a standard deviation of 35.3ps which is nothing to write home about, but can certainly be attributed to the slow edges of the SDG6052X. The mean value differs by 17.5ps (-131ps instead of 148.5ps) which once again might have to do with the measurement in the main (not zoom) window.

This is a beta firmware. Skew measurement is not possible within the zoom window and other than e.g. period measurements the channel delay measurements are not optimized for high resolution – but I will suggest this of course.

Anyway, a stable trigger point and channel skew (if measured with the T@M function) can be achieved with the SDG6052X despite its slow transition times. The measured rms trigger jitter of <1.5ps is certainly nothing to worry about. I think the true positive surprise is the fact that the SDS5000X trigger system is so good even when transitions are slow compared to the timebase.

« Last Edit: October 05, 2019, 12:08:06 pm by Performa01 »
 
The following users thanked this post: 2N3055

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17663
  • Country: nl
    • NCT Developments
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #336 on: October 05, 2019, 02:08:17 pm »
Averaging won't do you much good either because you'll be throwing away both the trigger jitter and the signal jitter. If the oscilloscope's trigger jitter is shown uniform (thus without narrowing) it is easier to determine the reference signal's jitter versus a second signal. Another option would be to delay the trigger enough to show the next period but that may add extra uncertainties from the oscillosope's oscillator (think about looking at 1PPS signals which have a 1 second interval).

A clear indication something is wrong with the measurement above is that the SDG6052X seemingly behaves much better than specified. A DDS based function generator with such low jitter is way more expensive -I'm not aware that such a device even exists-. So there must be something wrong with the test setup. I did a quick test with an SDG1010. I fed a 10MHz reference signal from a GPSDO into the SDG1010 and used this reference signal to trigger on. The SDG1010 produces a 10MHz square wave. Channel 2=reference signal, channel 3 is output from SDG1010.
Without averaging:


With 100x averaging:


With averaging the result suddenly looks better but that is only because information has been thrown away.
« Last Edit: October 05, 2019, 02:56:07 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: egonotto

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 817
  • Country: at
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #337 on: October 05, 2019, 03:44:38 pm »
Ah well, I saw this coming...

To really appreciate what’s going on, I’d like to recapitulate how a modern fully digital trigger system works. It is a completely separate processing chain, fed with the sample data stream of the trigger channel. This trigger system doesn’t care about the other channels and whether they are in use or not. Consequently, on the SDS5000X this data stream is always 2.5GSa/s with a 400ps sample interval. This makes it all the more remarkable when accuracy and stability down to single digit picoseconds can be obtained. Furthermore, the channel interleaving as well as the various decimation algorithms (normal, peak detect, average, Eres) are applied to the data channel exclusively at a later stage and do not affect the trigger path in any way.

Of course the objection that averaging also irons out some of the trigger jitter is valid, so I decided to setup one final test, using a highly stable sine wave instead of a pulse. This way we get rid of all the uncertainties with regard to the signal source and quality – and we get faster transitions too (at high frequencies).

Here is the same measurement as for the first test, but in normal acquisition mode, without averaging. Since I’ve already put away the original setup, skew is not exactly the same as in the first test.

1GHz sine wave, fed into channels 2 and 4 in parallel through 12.4GHz wideband power splitter. Some minor skew between channels because of subtle signal path length differences.

•   Dots display mode
•   Infinite persistence recording for >10 minutes to catch any jitter and noise effects.


SDS5104X Sine 1GHz Norm Skew Pers10m NR

At 200ps/div, the visible jitter is still pretty much insignificant.

Let’s have a look at the automatic measurements again. We now have an error of 600fs for the trigger channel 4 and just 7ps pk-pk jitter for the measurement data. The standard deviation (RMS jitter) is <1.23ps.

The non-triggered channel 2 has a significantly higher pk-pk jitter of 15ps now, RMS jitter is <2.12ps.

As can be seen, the channel to channel skew is finally accurate now at some 15ps. The peak to peak variation of 18ps with a standard deviation of 2.4ps meets the expectation.

A very pleasing result in my book.

Yes, the trigger interpolator does a rigorous job and reduces noise and jitter to zero at the trigger point, but we can take the peak to peak jitter of channel 2 as a measure for the maximum trace width at 45° slope, and that’s only 15ps. If I interpret nctnico's screenshot correctly, then there the trace width is uniformly more than 100ps – but that might just be because of a poor quality signal source.

For comparison, here’s the 150MHz pulse from the SDG6052X again, no averaging, infinite persistence for more than 5 minutes.


SDS5104X Pulse 150MHz Norm Skew Pers5m NR

Automatic measurements indicate the following:
-21ps Trigger time error, 18.5ps pk-pk and 2.6ps RMS jitter for the trigger channel 4.
-2.84ps Trigger time, 37.4ps pk-pk and 4.6ps RMS jitter for the independent channel 2.
Channel skew 18.2ps (almost spot on), 39ps pk-pk and 5.25ps RMS jitter between channels.

Even though the jitter values are much higher now, trace width at 45° slope is <38ps, which is more than twice as much than with a high quality sine source, but still a far cry from 100ps.
« Last Edit: October 05, 2019, 03:54:21 pm by Performa01 »
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17663
  • Country: nl
    • NCT Developments
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #338 on: October 05, 2019, 05:47:43 pm »
Even though the jitter values are much higher now, trace width at 45° slope is <38ps, which is more than twice as much than with a high quality sine source, but still a far cry from 100ps.
Still Siglent specifies the trigger jitter is less than 100ps. It would be interesting to know how they came up with this number. Perhaps it is specified at the lowest amplitude it can trigger on. A signal with a faster rise time allows for less hysteresis in the comparator so the results can vary. Either way you have to be careful with assuming a piece of equipment working better than specified.
« Last Edit: October 05, 2019, 05:49:32 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 2087
  • Country: hr
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #339 on: October 05, 2019, 10:05:23 pm »
I don't have super clean low phase noise source here, but some time ago  I discovered that Rigol DG1000Z 10MHz out is quite clean with nice clean edges.
So I fed that to SDG6000X ref in and CH1 of scope and I'm triggering from that. On CH3 SDG6000X output 10MHz, pulse, 50% DutyCycle, 1,3 ns edges. Signals were set a bit apart to be able too see cleanly both edges.

MSOX3104T fastest timebase is 500 ps/div. It doesn't have trigger jitter specification in datasheet. If anybody knows it, I would like to know.

Measuring time between channels, there is 20 ps RMS (stdev) of time variation between signals (jitter). That is combined AWG clock to synthesised signal jitter  + scope trigger jitter...
Persistence is set to infinite.

Second image, 10MHz source directly to ch 1 and 3. Scope alone measures 16,5 ps jitter RMS...

So yes, SDG6000X is, at these settings , definitely better than 100 ps RMS jitter. And SDS5000X has less trigger jitter than MSOX3000T in this case..
« Last Edit: October 05, 2019, 10:06:55 pm by 2N3055 »
 
The following users thanked this post: Performa01

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17663
  • Country: nl
    • NCT Developments
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #340 on: October 05, 2019, 10:25:54 pm »
The standard deviation is not RMS! These are calculated differently. Also you should look at the minimum and maximum values to get the worst case values. The standard deviation for noise is typically low because there isn't really any distribution.
« Last Edit: October 05, 2019, 10:36:39 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 817
  • Country: at
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #341 on: October 06, 2019, 07:56:24 am »
Even though the jitter values are much higher now, trace width at 45° slope is <38ps, which is more than twice as much than with a high quality sine source, but still a far cry from 100ps.
Still Siglent specifies the trigger jitter is less than 100ps. It would be interesting to know how they came up with this number. Perhaps it is specified at the lowest amplitude it can trigger on. A signal with a faster rise time allows for less hysteresis in the comparator so the results can vary. Either way you have to be careful with assuming a piece of equipment working better than specified.
As already stated, I believe this number comes from some theoretical worst case analysis, which might have no practical relevance or maybe even incorrect. Then there’s the fact that other manufacturers specify the trigger jitter as a typical RMS value, which is very different to just a worst case number. I’ll have to discuss this topic with Siglent R&D.

Other than that, I haven’t picked a sweet spot scenario. After several tests I’m not afraid to state that my published results will typically be valid for at least the entire vertical gain range from 10mV/div up to 10V/div.

For the higher sensitivities, noise might introduce additional jitter, so I’ll publish an additional test later in this posting.

Of course, we can always get arbitrarily bad results for jitter if we have a signal with slow transitions and/or signal amplitudes that are less than 50% of the screen height. In the first case it would be unjustified expectations, whereas the second one would fall under the category “user error”. No manufacturer can guarantee any specification if the product is used the wrong way. There is a reason why some datasheets specify a minimum frequency or maximum transition time as condition for certain specifications to be valid.

So at the end of the day, it is completely irrelevant if some theoretical setup might actually give a peak jitter of up to 100ps, because it’s no practical use case and you can certainly find such scenarios in every DSO, no matter what brand or price tag. For all practical use cases, the SDS5000X has proven its trigger system needs not hide behind the A-brand competition.

Now let’s have a look at the real worst case scenario. That’s 2.5mV/div – the highest sensitivity where the full 1GHz bandwidth is available. The increased noise will certainly yield more trigger jitter.

1GHz sine wave, fed into channels 2 and 4 in parallel through 12.4GHz wideband power splitter. Some minor skew between channels because of subtle signal path length differences.

•   Dots display mode
•   Infinite persistence recording for >15 hours(!) to catch any jitter and noise effects.


SDS5104X Sine 1GHz 2.5mV Norm Skew Pers15h NR

Automatic measurements indicate the following:
5.1ps Trigger time error, 20.9ps peak-peak and 2ps RMS jitter for the trigger channel 4.
24.6ps Trigger time, 45.5ps peak-peak and 4.7ps RMS jitter for the independent channel 2.
Channel skew 18.3ps, 46ps peak-peak and 4.8ps RMS jitter between channels.

The persistent picture on the screen confirms the measurements.

When looking at the peak to peak deviations, we should keep the long observation period of 15 hours in mind.
2ps RMS trigger jitter under these conditions is certainly nothing to complain about and the same is certainly true for just 21ps peak-peak deviation.

 

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 817
  • Country: at
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #342 on: October 06, 2019, 08:00:42 am »
The standard deviation is not RMS! These are calculated differently. Also you should look at the minimum and maximum values to get the worst case values. The standard deviation for noise is typically low because there isn't really any distribution.

Hopefully we need not start discussing the basics here again…

From that other thread, that I linked in reply #333, we could learn that not many DSOs actually specify the trigger jitter. If they do, it’s usually as “RMS (typical)”.

The difference between standard deviation and root mean square is that the latter contains a constant (time invariable) term.
•   In case of noise measurements, it’s the DC offset, which certainly is noise at 0Hz, but we are not interested in it because it can be easily compensated or corrected by simple math. So we use standard deviation for noise measurements.
•   In case of jitter measurements, it’s the constant time shift (error), which again can easily be corrected by simple math since it’s just a constant.

There is a reason why RMS (often without the time invariable constant term, as in our cases here) is used instead of peak to peak deviations in statistics. Peak to peak deviation heavily depends on the sample size (number of samples) which in turn is proportional to the observation time. There is a close relation of this topic to the (in)famous 1/f noise.

The peak deviation might matter in some practical cases, but for most applications RMS or standard deviation are way more useful. At the very least, they are the only valid ones for any sorts of comparisons. For the peak to peak deviation, the final value is proportional to the amount of data (which in turn depends on many factors, including observation time), whereas the RMS value is independent, hence fairly constant, but gets more and more accurate with increasing amount of data.
 
The following users thanked this post: rf-loop, 2N3055

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17663
  • Country: nl
    • NCT Developments
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #343 on: October 06, 2019, 11:14:13 am »
The standard deviation is not RMS! These are calculated differently. Also you should look at the minimum and maximum values to get the worst case values. The standard deviation for noise is typically low because there isn't really any distribution.

Hopefully we need not start discussing the basics here again…

From that other thread, that I linked in reply #333, we could learn that not many DSOs actually specify the trigger jitter. If they do, it’s usually as “RMS (typical)”.

The difference between standard deviation and root mean square is that the latter contains a constant (time invariable) term.
•   In case of noise measurements, it’s the DC offset, which certainly is noise at 0Hz, but we are not interested in it because it can be easily compensated or corrected by simple math. So we use standard deviation for noise measurements.
•   In case of jitter measurements, it’s the constant time shift (error), which again can easily be corrected by simple math since it’s just a constant.
Yes you are right. My mistake. But still using RMS as a number to specify something which is noise starts to smell like using a number which is much nicer than the reality. The peak value of noise seems to be 6 to 8 times of the RMS value so that leaves a rather larg margin of uncertainty.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #344 on: October 06, 2019, 11:42:57 am »
The standard deviation is not RMS! These are calculated differently. Also you should look at the minimum and maximum values to get the worst case values. The standard deviation for noise is typically low because there isn't really any distribution.

Hopefully we need not start discussing the basics here again…

From that other thread, that I linked in reply #333, we could learn that not many DSOs actually specify the trigger jitter. If they do, it’s usually as “RMS (typical)”.



The difference between standard deviation and root mean square is that the latter contains a constant (time invariable) term.
•   In case of noise measurements, it’s the DC offset, which certainly is noise at 0Hz, but we are not interested in it because it can be easily compensated or corrected by simple math. So we use standard deviation for noise measurements.
•   In case of jitter measurements, it’s the constant time shift (error), which again can easily be corrected by simple math since it’s just a constant.
Yes you are right. My mistake. But still using RMS as a number to specify something which is noise starts to smell like using a number which is much nicer than the reality. The peak value of noise seems to be 6 to 8 times of the RMS value so that leaves a rather larg margin of uncertainty.

This everyone's known basic fundamentals are quite well and shortly stated here https://www.analog.com/media/en/training-seminars/tutorials/MT-048.pdf  in page 5/6
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 16013
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #345 on: October 11, 2019, 07:05:16 am »
New firmware for SDS5000X models.
New functionality added plus enhancements and fixes etc.

Version 0.8.7R1B1
42 MB
https://int.siglent.com/upload_file/zip/firmware/Oscilloscope/SDS5000X_0.8.7R1B1_EN.zip

Release notes
1. Supported Power Analysis (optional) (Analysis | Power Analysis)
2. Supported Bode Plot (Analysis | Bode Plot)
3. Supported Totalizer (Analysis | Counter). Frequency and period parameters are moved from DVM to Counter
4. Supported 2 math traces and formula editor
5. Optimized FFT
 a) Optimized menu structure
 b) Supported peak and marker (Math | FFT | Tools)
 c) Supported setting max points (Math | FFT | Config)
6. Measurement enhanced
 a) Optimized the UI. In the “Basic” tab the items can be customized (long pressing an item to add to or delete from Basic tab)
 b) Added items: Median, Cycle median, -Bwidth, Time@max, Time@min, 20-80%Rise, 80-20%Fall, +Area, -Area, Area, AbsArea, Cycles, Rising Edges, Falling Edges, Edges, PPulses, Npulses
 c) Supported statistics on maximum 12 parameters at the same time (M2);
 d) Optimized measurement accuracy of Rise/fall
 e) Supported Trend Plot of measurement items
7. Optimized UX of knobs
8. Set the default function of the universal knob as adjusting the trace intensity
9. Optimized SPO display
10. Supported moving the location of the decode buses vertically
11. Supported single step back or forward in Navigator
12. Added bandwidth limit indicator below 2.45mV/div (1GHz, 500MHz)/1mV/div(350MHz
13. Supported Zone trigger in Sequence mode
14. Added entry for Zone trigger in the right side trigger menu
15. Mask Test: Supported failed history (Mask Test | Failure to History)
16. Increased frequency setting digits of the AWG from 3 to 7
17. After gesture control of the vertical gain, the V/div knob still is in the mode that has been used before with the gesture controls.
18. UART/LIN decode/trigger: supported baud rate > 5Mb/s
19. Reference position: Added user defined delay
20. Optimized UI in Zoom mode
21. Deleted the SCPI command which can start Telnet
22. Supported tapping on zone/histogram region to open the corresponding menu
23. Fixed several bugs
« Last Edit: October 11, 2019, 08:43:49 am by tautech »
Avid Rabid Hobbyist
 

Offline tv84

  • Frequent Contributor
  • **
  • Posts: 862
  • Country: pt
Re: At last, Siglent's SDS5054X touchscreen DSO...
« Reply #346 on: October 11, 2019, 08:52:23 am »
New options?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf