Siglent SDG1032X Harmonic Distorsion

[mawyatt]

Since everything is based upon the ADC captured data and time reference to the ADC clock, any randomness in the ADC clock or input signal will appear when compared to a uncorrelated signal like we are doing. At least we "assume" these are completely uncorrelated and hopefully no subtle non-linear effects like injection locking are taking place.

Anyway, whatever reference and up-conversion PLL the SDS2102X Plus is using is certainly working well as shown. Resolving 1ns after 1,000,000,000ns is impressive!!

Wish there was a means to adjust the DSO internal reference frequency like the SDG2042X has, or allow an external reference input, but that's a very minor point. 

Best,

[DL2XY]

...
I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.
...

Sorry for disagreement, but holdoff delay is totaly different to timebase delay.

Holdoff: delay>arm>trigger>sample ...
Timebase delay: arm>trigger>delay>sample ...

So with holdoff you will get the samples immediately after trigger, but we want the samples at defined time behind trigger.
The use of zoom is just a workaround to get the delay as high as one second at 1ns timebase.

 

[mawyatt]

...
I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.
...

Sorry for disagreement, but holdoff delay is totaly different to timebase delay.

Holdoff: delay>arm>trigger>sample ...
Timebase delay: arm>trigger>delay>sample ...

So with holdoff you will get the samples immediately after trigger, but we want the samples at defined time behind trigger.
The use of zoom is just a workaround to get the delay as high as one second at 1ns timebase.

  (Attachment Link)

Don't think we have a disagreement, as mentioned both show about the same result.

Zoom to 5ns/div from 1s sweep


Holdoff for 1s of 5ns/div sweep


Best,

[TurboTom]

...just to show what the "jitter cross correlation" of two really good oscillators looks like on a moderately fast scope, the MSO4000. In this screenshot, the internal HP 10811 OCXO of a 53310A is providing the external reference to the scope while my Efratom LPRO101-based Rb standard is connected to channel 1 of the scope. I arranged it that way because the square output of the Rb standard is really fast and provides good steep slopes for superb observability of the trigger crossing width. Prior to taking the measurement, the oscillators had been running for only about 20 minutes, so they didn't even reach thermal equilibrium.

As arranged by the previous contributors, the display window is delayed by one second vs. the trigger point and the scope is running with infinite persitance for approx. five minutes, recording a trace every second. Unforunately, the scope hasn't got those nice histogram functions or colour grading features but the cursor measurement shows a 260ps peak-to-peak jitter width. 

[mawyatt]

That looks very good indeed for the Rb standard and the HP external reference

How does this compare with the scope using it's own internal reference?

Best,

[TurboTom]

How does this compare with the scope using it's own internal reference?

This is something I specifically left out... When the engineers at Rigol designed that scope, they must have had something in mind like this: "If we provide an input for an external reference, why should we install a good internal one..."

Actually, hardly seen a scope with worse internal reference jitter. In the view as attached to my last post, the slope would be all over the place if run on the internal reference. 

[mawyatt]

Sorry, I don't know anything about Rigol, maybe a replacement oscillator with a better, lower jitter one would help?

Wish Siglent had a means to adjust the reference frequency like they have with the SDG2042X, and/or an external reference input.

Best,

[tautech]

Sorry, I don't know anything about Rigol, maybe a replacement oscillator with a better, lower jitter one would help?

Wish Siglent had a means to adjust the reference frequency like they have with the SDG2042X, and/or an external reference input.

Best,

All/any SDS2000 series DSO's have never had Ext clock references however the SDS5000X does.

[rf-loop]

...just to show what the "jitter cross correlation" of two really good oscillators looks like on a moderately fast scope, the MSO4000. In this screenshot, the internal HP 10811 OCXO of a 53310A is providing the external reference to the scope while my Efratom LPRO101-based Rb standard is connected to channel 1 of the scope. I arranged it that way because the square output of the Rb standard is really fast and provides good steep slopes for superb observability of the trigger crossing width. Prior to taking the measurement, the oscillators had been running for only about 20 minutes, so they didn't even reach thermal equilibrium.

As arranged by the previous contributors, the display window is delayed by one second vs. the trigger point and the scope is running with infinite persitance for approx. five minutes, recording a trace every second. Unforunately, the scope hasn't got those nice histogram functions or colour grading features but the cursor measurement shows a 260ps peak-to-peak jitter width. 

"Efratom LPRO101-based Rb standard"
So, do it mean LPRO is main reference inside some Rb standard and  is disciplining something what is short time better than LPRO itself. If (when) this is case, do you know what OCXO is before output and how LPRO discipline it.

Three decade count of LPRO101 (second hand) what I have seen long time ago, some even repaired and most of them calibrated, have never been in this "jitter" class. Most of them are really crap when we talk short time stability (depending highly also individual LPRO), even when longer time average is semi good or good in 5*10 exp -11 class (factory initial). (yes in previous life I have been time nutter) 
Inside LPRO there is only crap cheap "OC"XO and cotrol loop is "hopeless" if think short time jitter. They are mainly made/used for time keeping and there they meet standards. But using LPRO is possible to make better freq standards...
Of course also 260ps P-P jitter is not good but is it random symmetrical jitter or how much it include LPRO and 10811 drift due to very very short time after power up. But yes, some 10811 versions are good and best versions some individuals are very good, nice reference for many purposes. But of course if they are long time power off they need long long time before they come back near or fully to individual nominal ageing curve (xtal ageing retract)



"moderately fast scope"

what is this meaning in this test setup where you get under 300 acquisitions in 5 minute?

[rf-loop]

I think these delayed or "zoom" type jitter measurements are asking a lot from the scope trigger and sweep performance. Triggering and waiting 1ms, 50ms even 1s and looking for ns randomness seems difficult for any DSO much less a modest priced DSO. Actually impressed they can do this well.

If you think about what's happening, the trigger circuit is trying to trigger on exactly the same sampled voltage, but noise, quantization and other things place some randomness on this, then the delay circuit must wait the prescribed delay period which is DSO clock dependent and subject to some jitter, then the sweep starts for the display which is also subject to some uncertainty. Sure all these are created from the sampled data from the ADC, but still subjected to the ADC clock jitter which is based upon the DSO clock.

I see very little evidence of jitter from any source; SDG2042X, SDS2102X Plus DSO AWG or SSA3021X 10MHz reference regardless of the delay, so the DSO must not have much jitter on it's clock either.

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)

The SDS2102X Plus AWG with 1s delay
(Attachment Link)

Best,

Can you now carefully explain what are these images showing.
What have 1s delay and where and after then explain how it trig after holdoff. 
My eyes are far over best before date so perhaps I can not see something important...
I believe I see only rising edge trig just in trigger time position and as we know Siglent this scope trigger jitter is inside 10ps rms class.
Even if you run random time hopping signal you must get this when you look just this edge what is triggering and edge shape stay enough constant and amount of noise is enough low.

P.S. I know exactly what is trigger Holdoff (what can not generate trigger jitter at all and if holdoff time itself have jitter or inaccuracies also it do not generate time jitter what you can  see in this kind of test what are in images. It do not work so at all.
After delay it wait NEXT trigger event in signal and  trig perfectly to this point including fine interpolation between raw samples and fine positioned then in display buffer)

I think this is now good to handle just for basics how oscilloscopes works for avoit next thinking trap in future.. With scope is easy to do mistakes in some tests. It is very important to understand how oscilloscope works.

With holdoff. After last trig it start holdoff time and during this time it do not listen trigger engine for acquisition trig.  After time is full it start listen again trigger engine.  If trigger Holdoff timer have jitter or not it do not produce visible jitter at all. If holdoff time have big jitter or signal have jitter still in trigger time position they are positioned in image to same position. Independent of if your signal is random hopping or holdoff time have enormous time jitter and drift etc. Of course if we go to some special rare situations all can happen depending how it is programmed there deeply inside trigger engine.

But time delay after trigger trigger position inside this one acquisition... it is very different and it can tell truth added with some errors. Example if you look pulse to pulse jitter you set trigger to rising edge and you look next edge position what is delayed one  period after trigger. You see pulse to pulse jitter. You can zoom to this next rising edge and your zoom window center is delayed one pulse period from trigger position.  If you look position 1s after trigger position you see jitter over this time time period.
But if you stay looking in trigger time position and look it, zoomed or not, and you set trigger Holdoff time to 1s or what ever you do not see signal or holdoff time jitter. You see just as your images here show, even if your example square wave signal is random time/freq hopping.

Holdoff is, with other words, delay time after last trigged event before enable next possibility for trig.
Delay after trig done is very different and it can have negative or positive delay. If this time have jitter it can see signal trigger position jitter just same as if signal itself have jitter what can see in this delayed position.
If this Holdoff timer itself have jitter or not it is not normally visible in this kind of situation.

[TurboTom]

"Efratom LPRO101-based Rb standard"
So, do it mean LPRO is main reference inside some Rb standard and  is disciplining something what is short time better than LPRO itself. If (when) this is case, do you know what OCXO is before output and how LPRO discipline it.

Three decade count of LPRO101 (second hand) what I have seen long time ago, some even repaired and most of them calibrated, have never been in this "jitter" class. Most of them are really crap when we talk short time stability (depending highly also individual LPRO), even when longer time average is semi good or good in 5*10 exp -11 class (factory initial). (yes in previous life I have been time nutter) 
Inside LPRO there is only crap cheap "OC"XO and cotrol loop is "hopeless" if think short time jitter. They are mainly made/used for time keeping and there they meet standards. But using LPRO is possible to make better freq standards...
Of course also 260ps P-P jitter is not good but is it random symmetrical jitter or how much it include LPRO and 10811 drift due to very very short time after power up. But yes, some 10811 versions are good and best versions some individuals are very good, nice reference for many purposes. But of course if they are long time power off they need long long time before they come back near or fully to individual nominal ageing curve (xtal ageing retract)



"moderately fast scope"

what is this meaning in this test setup where you get under 300 acquisitions in 5 minute?

In my LPRO100 based reference oscillator is no post-processing circuitry contained that would affect short term phase noise / allan deviation. I read about the sortcomings of using an Rb oscillator module directly and probably all your assumptions regarding the situation of such units in private ownership fit my Efratom module as well -- I repaired it myself  . I've got other Rb modules (Temex LPFRS) that perform considerably worse. Do you actually have some hands-on experience with an LPRO100?

Nevertheless, its performance (also short-term phase noise wise) is better than any other reference oscillator that floats around in my basement lab, inside some gear or externally. And this includes the HP 10811, after "moderate" warm-up periods of a few hours, though they are close. As yet, since I received the 53310A with the ovenized timebase option only shortly, I didn't bother to give it a really long time heat-up period. Since I haven't got much experience with this HP oscillator yet, it may also be a rather poorly performing one. For whatever reason, the LPRO101 that I've got appears to perform (much) better than average. Or all the other references that I've got are really crappy... 

The "moderately fast" oscilloscope helps a lot visualizing the fairly steep slope of the input signal without much smear and as such makes the cursor measurement possible at all.

I may try another test by supplying both the reference and the input signal from the same source. This will show how much the internal processing (sampling clock PLL) of the scope contributes to the jitter.

[Johnny B Good]

Get yourself an MDA (HP 53310A) and be happy! 

There's probably no better tool (toy??) for a (soon-to-be) time-nut. It won't take long until you need to own a GPSDO, Rb or even Cs time base...    I can tell from my own experience. There's no way of getting out of this -- join TEA

I try to stay away as much as I can... Don't tempt me... 

 Yeah, me too.

John

[mawyatt]

I think these delayed or "zoom" type jitter measurements are asking a lot from the scope trigger and sweep performance. Triggering and waiting 1ms, 50ms even 1s and looking for ns randomness seems difficult for any DSO much less a modest priced DSO. Actually impressed they can do this well.

If you think about what's happening, the trigger circuit is trying to trigger on exactly the same sampled voltage, but noise, quantization and other things place some randomness on this, then the delay circuit must wait the prescribed delay period which is DSO clock dependent and subject to some jitter, then the sweep starts for the display which is also subject to some uncertainty. Sure all these are created from the sampled data from the ADC, but still subjected to the ADC clock jitter which is based upon the DSO clock.

I see very little evidence of jitter from any source; SDG2042X, SDS2102X Plus DSO AWG or SSA3021X 10MHz reference regardless of the delay, so the DSO must not have much jitter on it's clock either.

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)

The SDS2102X Plus AWG with 1s delay
(Attachment Link)

Best,

Can you now carefully explain what are these images showing.
What have 1s delay and where and after then explain how it trig after holdoff. 
My eyes are far over best before date so perhaps I can not see something important...
I believe I see only rising edge trig just in trigger time position and as we know Siglent this scope trigger jitter is inside 10ps rms class.
Even if you run random time hopping signal you must get this when you look just this edge what is triggering and edge shape stay enough constant and amount of noise is enough low.

P.S. I know exactly what is trigger Holdoff (what can not generate trigger jitter at all and if holdoff time itself have jitter or inaccuracies also it do not generate time jitter what you can  see in this kind of test what are in images. It do not work so at all.
After delay it wait NEXT trigger event in signal and  trig perfectly to this point including fine interpolation between raw samples and fine positioned then in display buffer)

I think this is now good to handle just for basics how oscilloscopes works for avoit next thinking trap in future.. With scope is easy to do mistakes in some tests. It is very important to understand how oscilloscope works.

With holdoff. After last trig it start holdoff time and during this time it do not listen trigger engine for acquisition trig.  After time is full it start listen again trigger engine.  If trigger Holdoff timer have jitter or not it do not produce visible jitter at all. If holdoff time have big jitter or signal have jitter still in trigger time position they are positioned in image to same position. Independent of if your signal is random hopping or holdoff time have enormous time jitter and drift etc. Of course if we go to some special rare situations all can happen depending how it is programmed there deeply inside trigger engine.

But time delay after trigger trigger position inside this one acquisition... it is very different and it can tell truth added with some errors. Example if you look pulse to pulse jitter you set trigger to rising edge and you look next edge position what is delayed one  period after trigger. You see pulse to pulse jitter. You can zoom to this next rising edge and your zoom window center is delayed one pulse period from trigger position.  If you look position 1s after trigger position you see jitter over this time time period.
But if you stay looking in trigger time position and look it, zoomed or not, and you set trigger Holdoff time to 1s or what ever you do not see signal or holdoff time jitter. You see just as your images here show, even if your example square wave signal is random time/freq hopping.

Holdoff is, with other words, delay time after last trigged event before enable next possibility for trig.
Delay after trig done is very different and it can have negative or positive delay. If this time have jitter it can see signal trigger position jitter just same as if signal itself have jitter what can see in this delayed position.
If this Holdoff timer itself have jitter or not it is not normally visible in this kind of situation.

These are intended to show the effects of scope introduced jitter, and I know exactly how Holdoff works at least in the old Tek analog scopes, assumed the DSO works similar and why I indicated as you highlighted above!!

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

I've been using analog Tektronix scopes since 1966, and about 40 years ago actually studied how the triggering was achieved in the Tek scopes. They had an almost amazing ability to capture signals, and found out they were using the fastest ECL logic chips available. These were used as analog limiters and comparators, not the intended logic use, although they did use some as logic functions.  In these old analog scopes the Holdoff function could not achieve a 1 second delay with a 5ns/div scale factor (not enough range), and neither could the B Delayed function since the update rate would be 1 second and you can't "see" the result because of the finite CRT phosphor persistence. I believe the old Tek analog scope B delayed function is similar to the way the Zoom function works on the DSO for the scope time base. On some of the old Tek scopes (like the 2465) you can select to "Retrigger" on the B Delayed function which allows a Holdoff like behavior but with more range, although you are still limited by the trace persistence issue.

Anyway, all this tells me that the oscillator used for timing and ADC capture in the DSO is very good from a jitter standpoint, and why I stated such. If the delayed retriggered (Holdoff) and triggered then delayed but not retriggered (Zoom) trace look the same then the scope is doing it's job quite well thank you

I just did a quick test on my Tek 2465 using the B delayed function with the AWG waveform at 10MHz. With the sweep set to 1ms/div and B Delayed sweep set to 10ns/div the trace has considerable jitter due to the scope clock and circuitry, using the retrigger feature (like Holdoff) locks the trace in with no apparent jitter. This is the best I could do because the trace is very faint (had to darken the room), and my eyes aren't good either


Here's couple photos on the Tek 2465, 1st is B Delayed (~7.1ms), note the significant jitter introduced by the scope.


2nd is delayed  (~7.1ms) with retriggering enabled, note the lack of jitter.



Best,

[mawyatt]

Sorry, I don't know anything about Rigol, maybe a replacement oscillator with a better, lower jitter one would help?

Wish Siglent had a means to adjust the reference frequency like they have with the SDG2042X, and/or an external reference input.

Best,

All/any SDS2000 series DSO's have never had Ext clock references however the SDS5000X does.

Can you loan me the price difference between these 2000 and 5000 series scopes, so I can get one with an ext ref input

Or maybe a big discount

Best,

[Johnny B Good]

Ok John, I think I understand where you are coming from regarding these time/phase/shift calculations. Need to see if I can, as you say "get my head" around this.

Anyway, I don't see any of these "jitter" type effects on the SDS2102X Plus internal AWG, nor the SDG2042X AWG.

Best,

 All of which leads me to conclude that I'm simply seeing the effect of a faulty cheap DDS type of smd xo rather than some feature of a particularly cheap 'n' nasty dds type smd xo. After looking at the datasheet for the Silicon Labs one dollar 50 cents in 1000 off quantities dds smd xos (just how cheap were those earlier dds type xos, famed for this very effect? ,

 I can't really see Siglent skimping on such a vital part of their low jitter strategy. I can well imagine a company like Feeltech opting for the ancient dds versions at 10 cents each but not Siglent.

 Torturing a poor defenceless DSO into revealing this particular flavour of jitter by zooming into a 100ns window some 45 to 50ms to even as much as 1000ms after the trigger event when you happen to have at least one low jitter reference signal (a 10MHz gpsdo, a DOCXO or, better yet, an Efratom LPRO-101 or even better an LPRO-101 and a gpsdo) to hand to do this the "Easy Way"(tm) just seems so masochistic an act of folly to my mind.

 I do agree however, that when dealing with such an unusually jittery signal, you need to be doubly sure that what you're seeing isn't the result of a very flawed test set up. In my case, I had two independent and uncorrelated 10MHz references, the RFS and the GPSDO to trigger from with the other acting as a referee to validate the results.

 I even went so far as to run the 100ns zoomed window delayed from the trigger event by 50ms test, aka "Doing it the hard way" (I wasn't sure I could use as long as a 1s delay for this test  but even so, the framerate was going to be rather glacial to say the least and the infinite setting on persistence didn't seem to be working).

 The 50ms delay was sufficient to show enough of this jitter and I did question the timebase jitter by triggering from one then the other of my references to see how much jitter, if any, was present in the timebase itself (and, for that matter, between the two references) with the indication being that what small (tens of ps worth of?) jitter appeared to be all down to the DSO's timebase with none observable between my two uncorrelated references. From these 'sanity checks', I think I can safely conclude that the DSO's contribution to the jitter displayed on the AWG's signal was less than 1% of the total.

 As I've already mentioned, this method is the resort of the man who has absolutely no access to an independent low jitter source of test signal to use the "Easy Test" method. Such an independent test source doesn't even need to be a highly stable GPSDO or RFS or DOCXO type just as long as it can remain frequency stable to within +/-1Hz for more than ten seconds at a time and has negligible jitter of its own. A salvaged DIP14 XO in the range 5 to 15MHz could suffice for this, even if plugged into a solderless breadboard as long as it's protected from random air draughts (a bit of rag or a cardboard box or WHY).

 Anyway, all that aside, I'm returning my new toy as "Unfit for purpose" for a full refund. Incidentally, it is possible to apply a 1mHz offset to a 10MHz frequency setting (and a 100uHz offset to a 9MHz setting) with the SDG1000 and 2000 X series of AWGs despite the displayed frequency being rounded to 10.000000MHz when the "MHz" suffix stamps all over the last three digits after pressing the MHz button. You need to use the external 10MHz clock option to eliminate the jitter and lock it to your reference source before the effect of these tiny frequency changes can be observed.

 I first tested this by entering 10.000000000 until it wouldn't accept any more zeroes followed by a single backspace to replace the final zero with a "5" before hitting the MHz button. I saw a slow but quite discernible drift to the left despite the display now showing 10.000000MHz.

 When I tried this trick using 9.9999999999 and moving back one digit to enter a "5" expecting to see a similar drift to the right, I'd overlooked the fact that I'd gained an extra digit's worth of resolution and had entered a 500uHz reduction rather than the 5000uHz increase I'd applied to the 10MHz setting which resulted in only a tenth of the drift rate to the right that I'd been expecting. However, after a minute or so, I did see a rightward drift and only then did the penny drop.

 The result of this test strongly suggests that, given a re-write of the frequency entry algorithm in the GUI, it could be made to emulate the Feeltech's 14 digits of resolution frequency entry. complete with the selected digit being remembered between each adjustment session. The frequency setting algorithm as it stands is an insult to the AWG's capabilities and whoever was responsible for that execrable effort in data entry programming should be hanging their head in unmitigated shame.

 It is only with that thought of a possible re-write of the frequency input algorithm in these two series of AWG that I'm even entertaining spending yet another 200 plus quid on an SDG2042X to get hold of a more upmarket AWG that does offer a little more improvement versus the SDG1032 which, surprisingly offers so much less  Even if my unit hadn't exhibited this nasty and insidious jitter on its internal clock reference, I think I'd have been returning it as "Unfit for purpose" anyway.

John

[TurboTom]

Couldn't agree more with you on the U/I shortcomings of Siglent's AWGs. Even their top end model that I've got (SDG6000X) shares the same troubled user interface. It all gets down to usability of the instrument without having to look at the display all the time. The plain and easy rule when changing parameters incrementally is "stay on the digit" whatever decimal's carry is going on. Moreover, when switching between parameters to adjust alternatingly for example amplitude and frequency (or whatever other parameter may be of interest), remember the digit that the cursor had been on when leaving the parameter. There are a few more things (like the choice to set limits for the adjustable range of a parameter) that would be really nice to have. I wonder if Siglent will actually address this - if they do, these instruments would surely gain even more popularity. If not, well, other manufacturers aren't sleeping...

Cheers,
Thomas

[mawyatt]

All of which leads me to conclude that I'm simply seeing the effect of a faulty cheap DDS type of smd xo rather than some feature of a particularly cheap 'n' nasty dds type smd xo.

As I've mentioned a few times, this I believe is the source of your observed excessive jitter.

I even went so far as to run the 100ns zoomed window delayed from the trigger event by 50ms test, aka "Doing it the hard way" (I wasn't sure I could use as long as a 1s delay for this test  but even so, the framerate was going to be rather glacial to say the least and the infinite setting on persistence didn't seem to be working).

Even extending this to 1 second and using the DSO "Zoom" feature shows no excessive jitter with the SDG2042X

Anyway, all that aside, I'm returning my new toy as "Unfit for purpose" for a full refund. Incidentally, it is possible to apply a 1mHz offset to a 10MHz frequency setting (and a 100uHz offset to a 9MHz setting) with the SDG1000 and 2000 X series of AWGs despite the displayed frequency being rounded to 10.000000MHz when the "MHz" suffix stamps all over the last three digits after pressing the MHz button. You need to use the external 10MHz clock option to eliminate the jitter and lock it to your reference source before the effect of these tiny frequency changes can be observed.

 I first tested this by entering 10.000000000 until it wouldn't accept any more zeroes followed by a single backspace to replace the final zero with a "5" before hitting the MHz button. I saw a slow but quite discernible drift to the left despite the display now showing 10.000000MHz.

 When I tried this trick using 9.9999999999 and moving back one digit to enter a "5" expecting to see a similar drift to the right, I'd overlooked the fact that I'd gained an extra digit's worth of resolution and had entered a 500uHz reduction rather than the 5000uHz increase I'd applied to the 10MHz setting which resulted in only a tenth of the drift rate to the right that I'd been expecting. However, after a minute or so, I did see a rightward drift and only then did the penny drop.

Yep, that's the way to get more digits with the SDG2042X as well, they don't show on the display tho.

 It is only with that thought of a possible re-write of the frequency input algorithm in these two series of AWG that I'm even entertaining spending yet another 200 plus quid on an SDG2042X to get hold of a more upmarket AWG that does offer a little more improvement versus the SDG1032 which, surprisingly offers so much less  Even if my unit hadn't exhibited this nasty and insidious jitter on its internal clock reference, I think I'd have been returning it as "Unfit for purpose" anyway.

BTW I used the "sliding waveform" method mentioned in a separate post to quickly align the SDF2042X Internal Reference (has internal control DAC), this achieved a 0.002ppm deviation from my stable SSA3021X Plus 10MHz reference (previously indicated as a good reference) for > hour quick test. I'm sure it's not that good over a longer period, but plenty good enough for my immediate needs

Best,

[mawyatt]

Couldn't agree more with you on the U/I shortcomings of Siglent's AWGs. Even their top end model that I've got (SDG6000X) shares the same troubled user interface. It all gets down to usability of the instrument without having to look at the display all the time. The plain and easy rule when changing parameters incrementally is "stay on the digit" whatever decimal's carry is going on. Moreover, when switching between parameters to adjust alternatingly for example amplitude and frequency (or whatever other parameter may be of interest), remember the digit that the cursor had been on when leaving the parameter. There are a few more things (like the choice to set limits for the adjustable range of a parameter) that would be really nice to have. I wonder if Siglent will actually address this - if they do, these instruments would surely gain even more popularity. If not, well, other manufacturers aren't sleeping...

Cheers,
Thomas

Agree, this isn't the best implementation of a UI, and some parts are quite annoying, not limited to just the rounding of the frequency display. Since this is evidently a common interface with all the AWGs, maybe Siglent will address some of these shortcomings in updated firmware

Anyway, with the performance I've seen and the chips used (shown in teardown) in the SDG2042X I honestly have few complaints, and outside a larger screen, all with the UI 

Best,

[tautech]

Can you loan me the price difference between these 2000 and 5000 series scopes, so I can get one with an ext ref input

Of course the SDS5000X was a totally new design and a new venture into higher BW's for Siglent and the SDS2kX Plus was a follow on using a very similar form factor but with shared controls to keep size and costs down.
In a recent chat with the product manager he disclosed the the 2kX Plus rear housing is exactly the same as used for earlier 2k products which I'd never noticed before. 
So this means 2kX Plus is the same physical size as Siglents earlier 2000 products.

[rf-loop]

I think these delayed or "zoom" type jitter measurements are asking a lot from the scope trigger and sweep performance. Triggering and waiting 1ms, 50ms even 1s and looking for ns randomness seems difficult for any DSO much less a modest priced DSO. Actually impressed they can do this well.

If you think about what's happening, the trigger circuit is trying to trigger on exactly the same sampled voltage, but noise, quantization and other things place some randomness on this, then the delay circuit must wait the prescribed delay period which is DSO clock dependent and subject to some jitter, then the sweep starts for the display which is also subject to some uncertainty. Sure all these are created from the sampled data from the ADC, but still subjected to the ADC clock jitter which is based upon the DSO clock.

I see very little evidence of jitter from any source; SDG2042X, SDS2102X Plus DSO AWG or SSA3021X 10MHz reference regardless of the delay, so the DSO must not have much jitter on it's clock either.

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)

The SDS2102X Plus AWG with 1s delay
(Attachment Link)

Best,

Can you now carefully explain what are these images showing.
What have 1s delay and where and after then explain how it trig after holdoff. 
My eyes are far over best before date so perhaps I can not see something important...
I believe I see only rising edge trig just in trigger time position and as we know Siglent this scope trigger jitter is inside 10ps rms class.
Even if you run random time hopping signal you must get this when you look just this edge what is triggering and edge shape stay enough constant and amount of noise is enough low.

P.S. I know exactly what is trigger Holdoff (what can not generate trigger jitter at all and if holdoff time itself have jitter or inaccuracies also it do not generate time jitter what you can  see in this kind of test what are in images. It do not work so at all.
After delay it wait NEXT trigger event in signal and  trig perfectly to this point including fine interpolation between raw samples and fine positioned then in display buffer)

I think this is now good to handle just for basics how oscilloscopes works for avoit next thinking trap in future.. With scope is easy to do mistakes in some tests. It is very important to understand how oscilloscope works.

With holdoff. After last trig it start holdoff time and during this time it do not listen trigger engine for acquisition trig.  After time is full it start listen again trigger engine.  If trigger Holdoff timer have jitter or not it do not produce visible jitter at all. If holdoff time have big jitter or signal have jitter still in trigger time position they are positioned in image to same position. Independent of if your signal is random hopping or holdoff time have enormous time jitter and drift etc. Of course if we go to some special rare situations all can happen depending how it is programmed there deeply inside trigger engine.

But time delay after trigger trigger position inside this one acquisition... it is very different and it can tell truth added with some errors. Example if you look pulse to pulse jitter you set trigger to rising edge and you look next edge position what is delayed one  period after trigger. You see pulse to pulse jitter. You can zoom to this next rising edge and your zoom window center is delayed one pulse period from trigger position.  If you look position 1s after trigger position you see jitter over this time time period.
But if you stay looking in trigger time position and look it, zoomed or not, and you set trigger Holdoff time to 1s or what ever you do not see signal or holdoff time jitter. You see just as your images here show, even if your example square wave signal is random time/freq hopping.

Holdoff is, with other words, delay time after last trigged event before enable next possibility for trig.
Delay after trig done is very different and it can have negative or positive delay. If this time have jitter it can see signal trigger position jitter just same as if signal itself have jitter what can see in this delayed position.
If this Holdoff timer itself have jitter or not it is not normally visible in this kind of situation.

These are intended to show the effects of scope introduced jitter, and I know exactly how Holdoff works at least in the old Tek analog scopes, assumed the DSO works similar and why I indicated as you highlighted above!!

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

I've been using analog Tektronix scopes since 1966, and about 40 years ago actually studied how the triggering was achieved in the Tek scopes. They had an almost amazing ability to capture signals, and found out they were using the fastest ECL logic chips available. These were used as analog limiters and comparators, not the intended logic use, although they did use some as logic functions.  In these old analog scopes the Holdoff function could not achieve a 1 second delay with a 5ns/div scale factor (not enough range), and neither could the B Delayed function since the update rate would be 1 second and you can't "see" the result because of the finite CRT phosphor persistence. I believe the old Tek analog scope B delayed function is similar to the way the Zoom function works on the DSO for the scope time base. On some of the old Tek scopes (like the 2465) you can select to "Retrigger" on the B Delayed function which allows a Holdoff like behavior but with more range, although you are still limited by the trace persistence issue.

Anyway, all this tells me that the oscillator used for timing and ADC capture in the DSO is very good from a jitter standpoint, and why I stated such. If the delayed retriggered (Holdoff) and triggered then delayed but not retriggered (Zoom) trace look the same then the scope is doing it's job quite well thank you

I just did a quick test on my Tek 2465 using the B delayed function with the AWG waveform at 10MHz. With the sweep set to 1ms/div and B Delayed sweep set to 10ns/div the trace has considerable jitter due to the scope clock and circuitry, using the retrigger feature (like Holdoff) locks the trace in with no apparent jitter. This is the best I could do because the trace is very faint (had to darken the room), and my eyes aren't good either


Here's couple photos on the Tek 2465, 1st is B Delayed (~7.1ms), note the significant jitter introduced by the scope.
(Attachment Link)

2nd is delayed  (~7.1ms) with retriggering enabled, note the lack of jitter.
(Attachment Link)


Best,

I have also much over 50year experience with many kind of oscilloscopes etc. In history including these 2465 and if go more far in history also other models. Also couple of 7000 series "RollsRoyces" and also couple of Hewlett-Packard ones. And of course Solartron ones etc. And then some total hobby ones like example most crap shit Heathkit what I get when I was young after my grand father frustrated I always mess his professional instruments.
But I have never ever messed and mixed basic fundamentals between digital scopes and analog scopes. This is also one reason I still sometimes use analog scope for some purposes due to fact they are fast never ever fool me by aliasing  and always things can fast set withoit searching any menu system finding this and that setting.

Now I can see here that something is wrong in thinking scope itself timing jitter and some DUT timing jitter and how to detect if there is and roughly how much.

I do not agree your explanations and tests here based to your words and 5 images.
As far as I understand your explanation and SDS2000XPlus  images  you did not detect any jitter. Even when you say there is 1s delay. (but I can not see any delay there)
In every image your timebase delay is Zero! Also when you set trigger Holdoff to 1 second what is just nonsense in this meaning. Even if your scope have enormous reference clock jitter with timebase delay zero and what ever trigger holdoff time you can not detect this jitter at all.
Also I can see  @DL2XY  note this to you but it looks like you know about oscilloscopes so much that of course you are right...  and your answer is... as can see images have no difference even if Holdoff is 1 sec or not Holdoff. But you forget that of course they are same because Holdfoff do not affect anything but slower wfm/s rate in this case.



Here is just two of these images just for example and because SSA3000X ref clock is most less jittering signal thing in your displayed set of things and this is what is best for analyze - if you want - your scope timebase jitter. (But not as you have done it with Trigger Holdoff 1s what do not accumulate time base or DUT time jitter at all to display).


@myawatt:
The SSA3021X Plus 10MHZ reference with 1s delay
[ Attachment Invalid Or Does Not Exist ]



@myawatt: Here's using the Zoom with delay of 1s on SDG2042X AWG
[ Attachment Invalid Or Does Not Exist ]

Now it is possible that I do not understand your scope screen image information related to your words because I do not have SDS2000XPlus here my position now.

Scope screen display time delay zero when I look it with my old eyes. Your image looks like it is just in trigger time position. Not delayed. So if this is case it do not accumulate also any possible time jitter in DUT signal or in scope timebase. It can not as long as you look trigged position.
If you take even 200ms long capture with this 10MHz signal and zoom(!) to example 100ms delayed position AFTER trigger position then there is jittering DUT and Scope timebase producing jitter what is accumulated over this 100ms time. And if not see jitter take more long capture and go to example 1s position after trigger and zoom to this position, zoom in to fastest time scale it can. If still can not see any jitter then need very very hard explanation why not and what is fooling us.
 
Please draw me free hand image and/or some other method.  Please, teach me so that also I understand where is this said 1 second delay in both of these images. Between what and where and how it produce displayes time jitter if jitter exist.

[rf-loop]

I think these delayed or "zoom" type jitter measurements are asking a lot from the scope trigger and sweep performance. Triggering and waiting 1ms, 50ms even 1s and looking for ns randomness seems difficult for any DSO much less a modest priced DSO. Actually impressed they can do this well.

If you think about what's happening, the trigger circuit is trying to trigger on exactly the same sampled voltage, but noise, quantization and other things place some randomness on this, then the delay circuit must wait the prescribed delay period which is DSO clock dependent and subject to some jitter, then the sweep starts for the display which is also subject to some uncertainty. Sure all these are created from the sampled data from the ADC, but still subjected to the ADC clock jitter which is based upon the DSO clock.

I see very little evidence of jitter from any source; SDG2042X, SDS2102X Plus DSO AWG or SSA3021X 10MHz reference regardless of the delay, so the DSO must not have much jitter on it's clock either.

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)

The SDS2102X Plus AWG with 1s delay
(Attachment Link)

Best,

Can you now carefully explain what are these images showing.
What have 1s delay and where and after then explain how it trig after holdoff. 
My eyes are far over best before date so perhaps I can not see something important...
I believe I see only rising edge trig just in trigger time position and as we know Siglent this scope trigger jitter is inside 10ps rms class.
Even if you run random time hopping signal you must get this when you look just this edge what is triggering and edge shape stay enough constant and amount of noise is enough low.

P.S. I know exactly what is trigger Holdoff (what can not generate trigger jitter at all and if holdoff time itself have jitter or inaccuracies also it do not generate time jitter what you can  see in this kind of test what are in images. It do not work so at all.
After delay it wait NEXT trigger event in signal and  trig perfectly to this point including fine interpolation between raw samples and fine positioned then in display buffer)

I think this is now good to handle just for basics how oscilloscopes works for avoit next thinking trap in future.. With scope is easy to do mistakes in some tests. It is very important to understand how oscilloscope works.

With holdoff. After last trig it start holdoff time and during this time it do not listen trigger engine for acquisition trig.  After time is full it start listen again trigger engine.  If trigger Holdoff timer have jitter or not it do not produce visible jitter at all. If holdoff time have big jitter or signal have jitter still in trigger time position they are positioned in image to same position. Independent of if your signal is random hopping or holdoff time have enormous time jitter and drift etc. Of course if we go to some special rare situations all can happen depending how it is programmed there deeply inside trigger engine.

But time delay after trigger trigger position inside this one acquisition... it is very different and it can tell truth added with some errors. Example if you look pulse to pulse jitter you set trigger to rising edge and you look next edge position what is delayed one  period after trigger. You see pulse to pulse jitter. You can zoom to this next rising edge and your zoom window center is delayed one pulse period from trigger position.  If you look position 1s after trigger position you see jitter over this time time period.
But if you stay looking in trigger time position and look it, zoomed or not, and you set trigger Holdoff time to 1s or what ever you do not see signal or holdoff time jitter. You see just as your images here show, even if your example square wave signal is random time/freq hopping.

Holdoff is, with other words, delay time after last trigged event before enable next possibility for trig.
Delay after trig done is very different and it can have negative or positive delay. If this time have jitter it can see signal trigger position jitter just same as if signal itself have jitter what can see in this delayed position.
If this Holdoff timer itself have jitter or not it is not normally visible in this kind of situation.

These are intended to show the effects of scope introduced jitter, and I know exactly how Holdoff works at least in the old Tek analog scopes, assumed the DSO works similar and why I indicated as you highlighted above!!

I don't see any difference between using the delay "Holdoff" trigger or "Zoom" feature.

I've been using analog Tektronix scopes since 1966, and about 40 years ago actually studied how the triggering was achieved in the Tek scopes. They had an almost amazing ability to capture signals, and found out they were using the fastest ECL logic chips available. These were used as analog limiters and comparators, not the intended logic use, although they did use some as logic functions.  In these old analog scopes the Holdoff function could not achieve a 1 second delay with a 5ns/div scale factor (not enough range), and neither could the B Delayed function since the update rate would be 1 second and you can't "see" the result because of the finite CRT phosphor persistence. I believe the old Tek analog scope B delayed function is similar to the way the Zoom function works on the DSO for the scope time base. On some of the old Tek scopes (like the 2465) you can select to "Retrigger" on the B Delayed function which allows a Holdoff like behavior but with more range, although you are still limited by the trace persistence issue.

Anyway, all this tells me that the oscillator used for timing and ADC capture in the DSO is very good from a jitter standpoint, and why I stated such. If the delayed retriggered (Holdoff) and triggered then delayed but not retriggered (Zoom) trace look the same then the scope is doing it's job quite well thank you

I just did a quick test on my Tek 2465 using the B delayed function with the AWG waveform at 10MHz. With the sweep set to 1ms/div and B Delayed sweep set to 10ns/div the trace has considerable jitter due to the scope clock and circuitry, using the retrigger feature (like Holdoff) locks the trace in with no apparent jitter. This is the best I could do because the trace is very faint (had to darken the room), and my eyes aren't good either


Here's couple photos on the Tek 2465, 1st is B Delayed (~7.1ms), note the significant jitter introduced by the scope.
(Attachment Link)

2nd is delayed  (~7.1ms) with retriggering enabled, note the lack of jitter.
(Attachment Link)


Best,

I have also much over 50year experience with many kind of oscilloscopes etc. In history including these 2465 and if go more far in history also other models. Also couple of 7000 series "RollsRoyces" and also couple of Hewlett-Packard ones. And of course Solartron ones etc. And then some total hobby ones like example most crap shit Heathkit what I get when I was young after my grand father frustrated I always mess his professional instruments.
But I have never ever messed and mixed basic fundamentals between digital scopes and analog scopes. This is also one reason I still sometimes use analog scope for some purposes due to fact they are fast never ever fool me by aliasing  and always things can fast set withoit searching any menu system finding this and that setting.

Now I can see here that something is wrong in thinking scope itself timing jitter and some DUT timing jitter and how to detect if there is and roughly how much.

I do not agree your explanations and tests here based to your words and 5 images.
As far as I understand your explanation and SDS2000XPlus  images  you did not detect any jitter. Even when you say there is 1s delay. (but I can not see any delay there)
In every image your timebase delay is Zero! Also when you set trigger Holdoff to 1 second what is just nonsense in this meaning. Even if your scope have enormous reference clock jitter with timebase delay zero and what ever trigger holdoff time you can not detect this jitter at all.
Also I can see  @DL2XY  note this to you but it looks like you know about oscilloscopes so much that of course you are right...  and your answer is... as can see images have no difference even if Holdoff is 1 sec or not Holdoff. But you forget that of course they are same because Holdfoff do not affect anything but slower wfm/s rate in this case.



Here is just two of these images just for example and because SSA3000X ref clock is most less jittering signal thing in your displayed set of things and this is what is best for analyze - if you want - your scope timebase jitter. (But not as you have done it with Trigger Holdoff 1s what do not accumulate time base or DUT time jitter at all to display).


@myawatt:
The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)



@myawatt: Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

Now it is possible that I do not understand your scope screen image information related to your words because I do not have SDS2000XPlus here my position now.

Scope screen display Timebase delay 0.0 when I look it with my old eyes. Your image looks like it is just in trigger time position (trig position symbol top middle of screen). Not delayed. So if this is case it do not accumulate also any possible time jitter in DUT signal or in scope timebase. It can not as long as you look trigged position.
If you take even 200ms long capture with this 10MHz signal and zoom(!) to example 100ms delayed position AFTER trigger position then there is jittering DUT and Scope timebase producing jitter what is accumulated over this 100ms time. And if not see jitter take more long capture and go to example 1s position after trigger and zoom to this position, zoom in to fastest time scale it can. If still can not see any jitter then need very very hard explanation why not and what is fooling us.
 
Please draw me free hand image and/or some other method.  Please, teach me so that also I understand where is this said 1 second delay in both of these images. Between what and where and how it produce displayes time jitter if jitter exist.

[TurboTom]

I actually wanted to stay out with further posts regarding the signal / TB jitter issue since it's somewhat off-topic but @rf-loop made me reconsider, and he is right. I didn't understand where the jitter/delay is in your scope screen shots as well. The functions of the "good old time"  scopes basically make it very clear: timebase B delayed by A or "zoom" mode without retrigger is the actualy "delayed timebase mode", where a single trigger event causes the whole trace display sequence to happen and thus will accumulate the TB A (or signal, whichever is predominant) jitter.

Anything that just delays the trigger arm function for the "display timebase", will not accumulate any jitter since the triger event for the display timebase will be on-screen. These are holdoff or anything that incorporates a re-trigger function (as it's called on the analog flagships).

Since in my previous post, I mentioned to try to measure the 1s delayed performance of my MSO4000, I won't retain it from you. Here I decided to disable the vector display but rather have dots displayed since this reveals some more information. This first screenshot shows again the measurement of the HP 10811 (reference) vs. my Efratom LPRO101:



And now, I used the LPRO101 to supply both the scope's reference signal as well as the input, so this should depict the scope's own one-second delay jitter:



Both measurements were taken over approx. the same time of a few minutes and with the instruments warmed up for maybe an hour. It's quite curious to see the individual "islands" of the sample locations not completely bleeding into each other in the latter measurement. But it also shows that this kind of measurement (at least with the gear used) is not accurate enough to quantitatively characterize the jitter characteristics of two fairly good oscillators. I only wish the MSO4000 had density colour grading, in these kinds of measurements this feature would be really useful!

[rf-loop]

I actually wanted to stay out with further posts regarding the signal / TB jitter issue since it's somewhat off-topic but @rf-loop made me reconsider, and he is right. I didn't understand where the jitter/delay is in your scope screen shots as well. The functions of the "good old time"  scopes basically make it very clear: timebase B delayed by A or "zoom" mode without retrigger is the actualy "delayed timebase mode", where a single trigger event causes the whole trace display sequence to happen and thus will accumulate the TB A (or signal, whichever is predominant) jitter.

Anything that just delays the trigger arm function for the "display timebase", will not accumulate any jitter since the triger event for the display timebase will be on-screen. These are holdoff or anything that incorporates a re-trigger function (as it's called on the analog flagships).

Since in my previous post, I mentioned to try to measure the 1s delayed performance of my MSO4000, I won't retain it from you. Here I decided to disable the vector display but rather have dots displayed since this reveals some more information. This first screenshot shows again the measurement of the HP 10811 (reference) vs. my Efratom LPRO101:



And now, I used the LPRO101 to supply both the scope's reference signal as well as the input, so this should depict the scope's own one-second delay jitter:



Both measurements were taken over approx. the same time of a few minutes and with the instruments warmed up for maybe an hour. It's quite curious to see the individual "islands" of the sample locations not completely bleeding into each other in the latter measurement. But it also shows that this kind of measurement (at least with the gear used) is not accurate enough to quantitatively characterize the jitter characteristics of two fairly good oscillators. I only wish the MSO4000 had density colour grading, in these kinds of measurements this feature would be really useful!

Excellent. Thanks.
But... is it also so that this 1 second delay is exactly based to its reference clock. Now your oscilloscope reference clock is based (with some sidenotes) to same signal what you use for input.  If you use LPRO or good 10811 as input and give scope run with its own reference you see this reference jitter what is totally much much more. I believe can not stay in this display width at all.

What is this tiny jitter we see here, roughly perhaps 100-200ps class depending what amount trigger jitter is included.
I believe it include many things.Freq locking quality to ext ref signal, some jitter/PN generated in PLL etc system what clock ADCs and acquisition, some part come from trigger itself jitter (I do not know how it fine interpolate trigger  between sample points (inside 250ps interval) and position signal inside this sample interval. In image looks like it do not make very fine fine interpolation and positioning. Last bright dots related to trigger (time) position.
So we do not even know sure if it display some trigger or is it  mainly trigger jitter due to some lack or lousy fine positioning.

If it is mainly trigger jitter it is not very good but if it is mainly system other kind of "jitter", not bad when external reference is in use. Without external reference situation is sure totally different.

But, it have Ext ref input.

Scopes what do not have... there is good and important to know its time jitter when it run based to its own reference because there is not  any choice. (except if user do HW modification for better internal or ext ref)

[mawyatt]

@myawatt:
The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)



@myawatt: Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

Now it is possible that I do not understand your scope screen image information related to your words because I do not have SDS2000XPlus here my position now.

Scope screen display Timebase delay 0.0 when I look it with my old eyes. Your image looks like it is just in trigger time position (trig position symbol top middle of screen). Not delayed. So if this is case it do not accumulate also any possible time jitter in DUT signal or in scope timebase. It can not as long as you look trigged position.
If you take even 200ms long capture with this 10MHz signal and zoom(!) to example 100ms delayed position AFTER trigger position then there is jittering DUT and Scope timebase producing jitter what is accumulated over this 100ms time. And if not see jitter take more long capture and go to example 1s position after trigger and zoom to this position, zoom in to fastest time scale it can. If still can not see any jitter then need very very hard explanation why not and what is fooling us.
 
Please draw me free hand image and/or some other method.  Please, teach me so that also I understand where is this said 1 second delay in both of these images. Between what and where and how it produce displayes time jitter if jitter exist.

Obviously the second image is incorrect, as there is no "Zoom" feature enabled in the display, so I must have used the incorrect image file

I've also discovered the DSO "Zoom" feature does not work exactly like the Tek B Delayed function since the trace starts with 0 delay and begins at the center of the display, thus invoking Zoom does not introduce any delay and this is subtly displayed above the lower Zoomed trace window. I'm not sure if another means is available to introduce a long delay, but the <> Knob is hopelessly inadequate if a large between Zoom timebases is required. The only way I could get this to work was use the finger "gesture" method, which was trying indeed

Eventually I was able to get ~1/2 second delay with a normal timebase of 100ms/div and still have a 5ns/div Zoom timebase. As expected the 1 sec delayed trigger shows no jitter, the Zoom trace shows a small amount of jitter with a delay of ~1/2 second.

As mentioned I'm impressed with this DSOs ability/performance in every aspect of it's use and behavior as I learn more about it's capabilities

Hopefully someone can point out a better method to achieve these widespread time bases with the Zoom feature, I suspect there IS a better method

Edit: Just found the Zoom Timebase window that shows up on the waveform has numerical entry for the delay. It allows a 1 second delay with a 5ns/div timebase. So I'll add a third image with 1 second delay with 5ns/div Zoom timebase and 100ms/ normal timebase. So there IS a better method 


10MHz reference with 1 second Delayed Trigger



10MHz reference Zoomed with ~1/2 second delay



Added 1 second Zoom delay.



BTW I saved up and purchased a Heathkit scope when I was 12~13, this was the original tube version

Best,

[rf-loop]

@myawatt:
The SSA3021X Plus 10MHZ reference with 1s delay
(Attachment Link)



@myawatt: Here's using the Zoom with delay of 1s on SDG2042X AWG
(Attachment Link)

Now it is possible that I do not understand your scope screen image information related to your words because I do not have SDS2000XPlus here my position now.

Scope screen display Timebase delay 0.0 when I look it with my old eyes. Your image looks like it is just in trigger time position (trig position symbol top middle of screen). Not delayed. So if this is case it do not accumulate also any possible time jitter in DUT signal or in scope timebase. It can not as long as you look trigged position.
If you take even 200ms long capture with this 10MHz signal and zoom(!) to example 100ms delayed position AFTER trigger position then there is jittering DUT and Scope timebase producing jitter what is accumulated over this 100ms time. And if not see jitter take more long capture and go to example 1s position after trigger and zoom to this position, zoom in to fastest time scale it can. If still can not see any jitter then need very very hard explanation why not and what is fooling us.
 
Please draw me free hand image and/or some other method.  Please, teach me so that also I understand where is this said 1 second delay in both of these images. Between what and where and how it produce displayes time jitter if jitter exist.

Obviously the second image is incorrect, as there is no "Zoom" feature enabled in the display, so I must have used the incorrect image file

I've also discovered the DSO "Zoom" feature does not work exactly like the Tek B Delayed function since the trace starts with 0 delay and begins at the center of the display, thus invoking Zoom does not introduce any delay and this is subtly displayed above the lower Zoomed trace window. I'm not sure if another means is available to introduce a long delay, but the <> Knob is hopelessly inadequate if a large between Zoom timebases is required. The only way I could get this to work was use the finger "gesture" method, which was trying indeed

Eventually I was able to get ~1/2 second delay with a normal timebase of 100ms/div and still have a 5ns/div Zoom timebase. As expected the 1 sec delayed trigger shows no jitter, the Zoom trace shows a small amount of jitter with a delay of ~1/2 second.

As mentioned I'm impressed with this DSOs ability/performance in every aspect of it's use and behavior as I learn more about it's capabilities

Hopefully someone can point out a better method to achieve these widespread time bases with the Zoom feature, I suspect there IS a better method

Edit: Just found the Zoom Timebase window that shows up on the waveform has numerical entry for the delay. It allows a 1 second delay with a 5ns/div timebase. So I'll add a third image with 1 second delay with 5ns/div Zoom timebase and 100ms/ normal timebase. So there IS a better method 


10MHz reference with 1 second Delayed Trigger

(Attachment Link)

10MHz reference Zoomed with ~1/2 second delay

(Attachment Link)

Added 1 second Zoom delay.

(Attachment Link)

BTW I saved up and purchased a Heathkit scope when I was 12~13, this was the original tube version

Best,

Now it looks better.

But as previously told first image can not accumulate any jitter related to waiting time. It can not see in image but I assume you have again used Trigger Holdoff and of course this Holdoff time do not accumulate any jitter due to zero delay when next trigger event after Holdoff time elapsed.  Result looks same if Holdoff or not, exept delay next displayed acquisition.

DPO zoom have nearly nothing to do with analog scope zoom even when result may looks like somehow same kind of image.
This is also small Achilles heel in conventional DSO and more modern DPO (or SPO like Siglent name it) but mostly there is way to walk over this. But dual independent beam dual time base dual delayed trig  -  just forget with these digitals.
This, and most other simple ones,  DPO can only do one whole sweep based to one trigger engine trig and nothing else.
For Zoom window we take only more or less long part from this one acquisition and from time position how user set zoom window position. There can not generate example other delayed trig for zoomed detail. There is not second "Timebase" where scope sweep part of trace more fast or do sequentially slower sweep and then for zoom faster sweep for adjusted position. No, DPO do not at all have other than just one "timebase" so it is even bit wrong to name horizontal axis adjustment as timebase. It is only time scale. With Zoom there is just this same original one "sweep" aka acquisition in memory. When we zoom, we only take part of this acquisition in memory and show it using different time scale on display.  And in this case now here, time jitter come only visible using this zoom so that we look other memory position than trigger position. There we can see if signal detail exist sooner or later. After then we know that oscilloscope timebase have jitter or signal under test have time jitter or both have time jitter. For know more we need know scope timebase jitter. If we know it is far less than signal under test we can name this observed jitter is signal under test jitter + error from scope timebase jitter. If we Know signal jitter is example "zero" after then we can tell that jitter we have seen is oscilloscope own jitter in timing.

In your images (zoomed ones) can see quite small time jitter.
Now small problem there is that slope is very slow for this purpose.
Decimated samplerate is 100MSa/s
It mean that there is decimated sample interval 10ns.  In Zoomed window there is only 5 samples in one acquisition and rest visible things are interpolation. Due to lack of knowledge about just this Siglent model I can not estimate what is perhaps possible trigger/fine interpolation/display positioning jitter amount and what is timebase related jitter and signal jitter. I believe trigger engine just after ADC use full non decimated samplerate but this is only now hope and believe.  But so or so, it can say --  scope timebase jitter is small!  Better than I expect. And it is good to know due to fact it do not have ExtRef input.

Oh well, example Hethkit IO-12 and perhaps even one bit older.. is still deeply "burned" in my memory, I can even remember smell of it and my first old Collins Rx from 40's.