The Summary Siglent SDS1202X-E Thread

[niekvs]

The one thing I don't see addressed there is that the serial decoding doesn't work properly on history frames. Specifically, the decoded content shows the value of the last frame even if you switch to previous frames. I saw this while using CAN decoding and haven't tried to check it with other formats.

Could this be related to fix #14: "Fixed the CAN source bug. If the source selected was CANH or CANL, decode would not work correctly."?

[nctnico]

I too updated using the USB stick and performed a self-cal as instructed.  It now reports SW 5.1.3.13 but I now have a triggering issue.  Using the test output, I connect to Ch1 and set DC trigger at 464mV and the scope probe set to x1 with matching selection on the Ch1 menu.  As I increase the vertical sensitivity, I can get down to 200 mV/div and all is OK, then I click once more to 100 mV/div and the display goes weird - see pictures.  The horizontal time base doesn't seem to make any difference.  I'm not sure if this was there before 

That is the overdrive recovery you see and every DSO does that to some extend so no problem here.

[Gandalf_Sr]

That is the overdrive recovery you see and every DSO does that to some extend so no problem here.

Interesting, can you please expand on that slightly? I think you're saying that the signal is actually overloading the input and that the input protection circuitry is causing what I see?

[nctnico]

That is the overdrive recovery you see and every DSO does that to some extend so no problem here.

Interesting, can you please expand on that slightly? I think you're saying that the signal is actually overloading the input and that the input protection circuitry is causing what I see?

It is not the input protection circuitry but one of the amplifiers in the input stage. In short: When you overdrive an opamp some of the transistor(s) inside get saturated and will take longer to get back into their linear operating mode. This results in the effects you see.

[Gandalf_Sr]

Thanks, that makes a lot of sense.  So I agree - no bug there.  I'll go back and edit my previous post.

[rf-loop]

That is the overdrive recovery you see and every DSO does that to some extend so no problem here.

Interesting, can you please expand on that slightly? I think you're saying that the signal is actually overloading the input and that the input protection circuitry is causing what I see?

Your full scale with offset zero setting is +/- 500mV  when you are with 100mV/div
Also maximum offset range is in this case is +/- 2V   
+5V signal is too high for this setting = user error.  But good for "know your equipment" purpose and good for many readers for training.

[Loboscope]

Did anyone check, if the XY-Mode will work now fast enough as it should be and will now be useful?
I am curious about it because I want to use the XY-Mode, but currently I am in holidays and far from home, so I can not check this function by myself.
Thank´s for testing and answering!

Greetings from the Netherlands coast!

Jürgen

[rf-loop]

Did anyone check, if the XY-Mode will work now fast enough as it should be and will now be useful?
I am curious about it because I want to use the XY-Mode, but currently I am in holidays and far from home, so I can not check this function by myself.
Thank´s for testing and answering!

Greetings from the Netherlands coast!

Jürgen

X-Y mode is not fast. As can see in image. Persistence is 5s. Just count updates.

I am always interested in everything to learn more. I have used oscilloscopes for hobby and work nearly 50 years. In some cases, old times, I have done some curve-tracker things with analog some analog oscilloscope equipped with storage crt. 

But, what is real use today for cheap scope very limited X-Y mode. I can not imagine nearly anyreal use so I want ask  -  for what?

Yes I know some peoples play and show example clock on the screen or some other things just because it is fun. But for important things, what is need for XY mode. I am interested to know. 

But here one image. Input CH1 500kHz and CH2 roughly 0.03 Hz below 1MHz

[nctnico]

XY mode is usefull for looking at phase changes between signals you can't see when using the scope in YT mode.

[rf-loop]

XY mode is usefull for looking at phase changes between signals you can't see when using the scope in YT mode.

Yes, some times. THis is why I still keep some bit higher quality analog oscilloscope with adjustable persistent crt.

But phase between signals and YT. Why not, in some dedicates cases.

Here two 1MHz signals. With fixed 1 degree phase shift. But as can see after long record time it do some times random 0.1 degree jumps to higher phase shift. When it do not jump phase stay quite well.  This can detect also quite fast phase shifts. Cursors spacing is roughly 1 degree with this signal. Many things are how we want do. But yes, XY mode have some advantages because it is not related to trigger and trig quality.  Analog scope is still hard to beat in some cases without truck load of money. Exept if buy phase measurement equipment or some vector...blablaa.

Typically when I adjust example frequency references I always use YT mode...just for save time and get more accuracy. Also it is nice to get some imagine about short time phase jitter (frequency jitter). Of course, the quick interpretation of the picture requires a little bit of experience as well. (I do not mean this attached image what is very simplified case  )

[nctnico]

Why would you need an analog scope? Many years ago I used XY mode on an old HP DSO (with a CRT screen) to look at sub-ppm frequency drifts to see how a PLL tracks a reference frequency.

[Loboscope]

One of the uses I have for X-Y-Mode is a very simple task for a scope (or at least it should be): In the field of musical instruction I want to show musical students the so called phenomenon "beat frequency", that is, if you detune for example a perfect fifth a bit, you will hear a more or less significant wow and flutter. For example, if you have the two frequencies 400 Hz and 600 Hz and you will detune the 600 Hz to 601 Hz you will hear this beat frequency, it will be a wailing of the sound with the resulting frequency of 1 Hz. (And 0,1 Hz at 600,1 Hz, 0,2 Hz at 600,2 Hz and so on.)
Using the X-Y-Mode of a scope will show this phenomenon simply, because the pattern on the screen will move with the frequency of 1 Hz. And seeing the phenomenon will help to identify and hear it more precise.

But if the scope will be too slow in this mode, the pattern will not move smooth, but will jump jerkily, and this will not help at all to to make the beat frequency visible. So does the 1202X-E (resp. did with the old firmware as I could state).
Unfortunately it will be a shame for a scope with this comparatively modern and fast hardware. Even my first scope, the GW-Instek GDS 1022 manages this simple task with a smooth screen update and movement - and this is certainly one of the most basic scopes ever!
I wanted to replace this simple scope with a more modern one with a much better screen, but i this regard the Siglent seems to fail completely 

By the way, this is not the only task for that I will use a scope. I am musician and I will use it also for maintaining and in case it fails my technical gear. But  the task I described above is a very useful task in my educational practice.

[rf-loop]

Why would you need an analog scope? Many years ago I used XY mode on an old HP DSO (with a CRT screen) to look at sub-ppm frequency drifts to see how a PLL tracks a reference frequency.

Sub ppm... oh well. I need some times adjust freq drift in sub ppb... but this is easy. It can do even with old analog multimeter - if know how. But with enough fast scope it is more easy because also can easy see drift direction when use YT mode. And also get some imagine about possible short time phase  jitter what mostly can not detect at all with XY mode or in YT mode with slow old dso. Also here wfm/s is important. First limit is 5 to <8 bit resolution in X and Y and other limit may be slow update rate, depending scopeXY mode is decades too slow.

[rf-loop]

One of the uses I have for X-Y-Mode is a very simple task for a scope (or at least it should be): In the field of musical instruction I want to show musical students the so called phenomenon "beat frequency", that is, if you detune for example a perfect fifth a bit, you will hear a more or less significant wow and flutter. For example, if you have the two frequencies 400 Hz and 600 Hz and you will detune the 600 Hz to 601 Hz you will hear this beat frequency, it will be a wailing of the sound with the resulting frequency of 1 Hz. (And 0,1 Hz at 600,1 Hz, 0,2 Hz at 600,2 Hz and so on.)
Using the X-Y-Mode of a scope will show this phenomenon simply, because the pattern on the screen will move with the frequency of 1 Hz. And seeing the phenomenon will help to identify and hear it more precise.

But if the scope will be too slow in this mode, the pattern will not move smooth, but will jump jerkily, and this will not help at all to to make the beat frequency visible. So does the 1202X-E (resp. did with the old firmware as I could state).
Unfortunately it will be a shame for a scope with this comparatively modern and fast hardware. Even my first scope, the GW-Instek GDS 1022 manages this simple task with a smooth screen update and movement - and this is certainly one of the most basic scopes ever!
I wanted to replace this simple scope with a more modern one with a much better screen, but i this regard the Siglent seems to fail completely 

By the way, this is not the only task for that I will use a scope. I am musician and I will use it also for maintaining and in case it fails my technical gear. But  the task I described above is a very useful task in my educational practice.

Siglent XY mode looks like it is not at all true XY mode.
 It looks like it is derived from bacround running YT mode (without history buffer).  If it is (and I believe it is).  This may explain why in XY mode there can also do full set of automatic measurements, includind also phase-angle what really is not measured from screen memory, as also example risetime etc. ERES works, wfm average works. (also trigger works )  For build whole XY display it need use lot of YT acquisitions. This is perhaps reason what makes it some slow. Some good, some bad.

Images have same signal. First XY mode and then YT mode.
It can see that it keep same measurement resolution in both modes. With this setting, 7M memory and 500MSa/s XY update speed is quite slow. Roughly in 1 second class.

Of course if this scope do real XY plot from ADC directly to display map memory  it can be very fast. nearly like analog scope without ADC resolution and noise limits.

What I hope is that Siglent add also fast true XY mode. Yes then loose some things but get high speed and perhaps can add even intensity gradation(?).






(my mistake, in this image cursors irrelevant)

[rf-loop]

Bugs Summary

3. There's a menu item for 50 Ohm/1 Meg Ohm and it doesn't do anything

As discussed previously this input internal impedance selection need delete. (no 50ohm HW inside)

Also discused that it need keep in some place so that user can tell to scope that he is using extrernal termination. This is important for example FFT dBm measurements.


They did it it in last FW update. They did more what was asked.

There is now free selection for externally added terminator/load impedance.

RF work mostly is used 50ohm impedance. But also others are some times used.

But then example audio things looks like use many kind of impedances, one common is 600ohm.

For serve all needs, user can now set impedance of termination what he is using in scope input externally.
Adjusting range is from 1 ohm to 1 Meg ohm.

But now, user need understand it and take care about is so that setting is right. Remember set/check value is ok before measurements for avoid wrong measurement results.

When you are in FFT menu, Page 2.  Adjusting menu opens when you select Unit/Load. There you can select Unit: dBVrms, Vrms, dBm.  There you can adjust ExtLoad 1ohm  to 1Meg ohm.  (1 - 99 ohm step 1ohm, 100 to 1000 ohm 10 ohm steps, 1k to 10k 0.1k steps etc)

Well done Siglent!

but, woks need continue....

[rf-loop]

New FW5.1.3.13 changes many things in FW. I am quite sure that published changelog do not include all fixes what they have done in FW. Only some visible and major things.

wfm/s update speeds have changed more or less but still "up to" max peak speed is 100kwfm/s - but motre important continuous average speed bit slower now in some cases but also then some positive things. Somehow dots-lines-sinc on/off is hadled now bit differently. Also waveform fine adjust to interpolated trigger positioninside sample interval is - well, they have least one perfectionist in team. (as perhaps all know inside 1ns sample period there need interpolation, linear or sinc and where this iterpolated curve or a straight line crosses trig level then this position need adjust to trigger time position and there interval os 20ps.)
But, as many times noted, wfm/s speed is not all we need. There is many other things, one is also fact that many times oscilloscopes hide part of captured length outside and because they are there unvisible it may some times even seriously rise visual blind time ratio. This do not happen at all in Siglent, never.

But, here is new table for SDS1202X-E (HW 0c-01)  FW 5.1.3.13 Old table, if it still exist in some place,  is valid only for FW5.1.3.8.  (Yes I leaked it before 1.8.  if final check I find some error - I hope not.  If someone else find  some error or fatal error there, please fast feedback time constant )


Table
All measurements done using oscilloscope watching trig out AND HP53131A connected also to Trig Out. Manual trig mode and gate time 3s for enough reliable results. Slowest times gate time doubled.


Trig out with other scope for look speed peak values ( what are not really important at all).

Sequence speed is low in fastest time bases. But there is really not any single reason for use 1 or 2ns/div when 2 channels are in use or 1 - 5ns/div if 1 channel is in use if need more speed in sequence mode. This is because same max amount of segments still available with better speed and in segments view mode (History viewer) there can use window zoom or whole screen zoom if want.

[eurofox]

I have a silly question, I build up my lab again.   

My last scope was a 1 GHz recent Agilent scope sold to a member of this forum.

Since I start again with some DC applications and microcontrollers I need a decent scope. 

I would like to have some feedback from users of this Siglent 200 MHz scope with serial decoding if it is really a “decent” scope to work with? 

[rf-loop]

I have a silly question, I build up my lab again.   

My last scope was a 1 GHz recent Agilent scope sold to a member of this forum.

Since I start again with some DC applications and microcontrollers I need a decent scope. 

I would like to have some feedback from users of this Siglent 200 MHz scope with serial decoding if it is really a “decent” scope to work with? 

If you live in Finland you can come visit and test and test with your applications and then you see if it pass for your needs. 
If you have known specified personal needs only way is that you personally test with your special nees.  You never know other peoples opinions how and what they have exactly done. Every people and every need is different.

You live in Belgium. Why you do not use Batronix 30 days free trial period if you can not walk to place where you can test it before buy.  Local seller who have enouh knowledge how to use equipment is best way. Because without knowledge it takes some days before find right way to use equipment for some defined needs. If not, next solution is use these sellers who can offer free test period beforew buy. 
Batronix is one well known and trusted dealer  for Siglent equipments.

[eurofox]

I have a silly question, I build up my lab again.   

My last scope was a 1 GHz recent Agilent scope sold to a member of this forum.

Since I start again with some DC applications and microcontrollers I need a decent scope. 

I would like to have some feedback from users of this Siglent 200 MHz scope with serial decoding if it is really a “decent” scope to work with? 

If you live in Finland you can come visit and test and test with your applications and then you see if it pass for your needs. 
If you have known specified personal needs only way is that you personally test with your special nees.  You never know other peoples opinions how and what they have exactly done. Every people and every need is different.

You live in Belgium. Why you do not use Batronix 30 days free trial period if you can not walk to place where you can test it before buy.  Local seller who have enouh knowledge how to use equipment is best way. Because without knowledge it takes some days before find right way to use equipment for some defined needs. If not, next solution is use these sellers who can offer free test period beforew buy. 
Batronix is one well known and trusted dealer  for Siglent equipments.

Thank you for your invitation to your country, I visit it many times on business trips a few times on a vacation trips.

I suppose you are a professional with electronics, are you satisfied with it? Does it fulfill your needs?

The only Chinese made oscilloscope that I use is a Rigol 100 MHz, is not Keysight or R&S but is OK, is not the same price tag.

Thanks in advance for taking the time commenting on this instrument “satisfaction”

[tautech]

Some discussion on pimping your SDS1202X-E with SDS2000X knobs:
https://www.eevblog.com/forum/projects/new-knobs/

[tautech]

New FW5.1.3.13 changes many things in FW. I am quite sure that published changelog do not include all fixes what they have done in FW. Only some visible and major things.

You are correct, not all changes are listed in the changelog.

Auto firmware updates feature has been removed completely.
This feature was to allow online updates of firmware directly to the scope........no unzipping or USB stick required.



To use this auto-update was quite simple, in I/O settings the DHCP need be turned on to auto negotiate an IP address and then in the FW update UI the Online update set to Status ON. (OFF by default)

Within a second or two the X-E contacts the Siglent server and updates the screen with the status of the FW available, see below.



With only an older pre-release version available and not the latest V5.1.3.13 one can only imagine Siglent have for the time being abandoned online FW updates. 



As mentioned previously and as stated in the firmware update instructions the SDS1202X-E must be self calibrated after the firmware is installed to remove the channel offsets.
This is how far from 0V the trace can be after FW installation.



After Self Cal has been run:

[rf-loop]

This is important to know!

If user do Security Erase, it also affect self cal produced cal data!

After Security Erase, it still use ok cal data as long as you do not shut off it. Next time when start scope it is now totally out of cal without any warning! But if user do not know it and just believe what he see on screen it may be more or less false.
Siglent need solve this thing way or other. Least it need information message on the scope screen about this.

After SecurityErase it NEED do self cal.

My recommendation is: If Security Erase have done, please immediately after it shut off and back on, Wait enough components get again thermal balance and do Self Calibration (it takes nearly 8 minutes).
If cold scope started, it need over 30min warming. If just warm scope rebooted ... well least minute or two. If you do not it just after SecErase it is easy to forget.

Also because some times we can see failed or poor calibs..  It need remember that nothing is conneted to inputs when run self cal.

[Loboscope]

Back from holidays I loaded the new firmware (without any problems) and I tested the X/Y-Mode myself.
Unfortunately as mentioned already by @rf-loop I had to realize, that the X/Y-Mode is still too slow and the trace jumps jerkily. It seems to be a bit faster as before, but still only twitchy and not continuous - sad  .
The Rigol DS1104Z does better (but also a bit twitchy), the best remains my old and super-simple GW-Instek GDS-1022!

[tautech]

Anyone out there with suspect "not quite right" probes ?

As I now unbox and install new firmware into each unit before dispatch and now have to run the Self Cal to negate the 0V offset, I use some time waiting for the unit to fully warm up to compensate the probes and fit the coloured rings to indicate the channel #.
I had one probe today that would not compensate correctly, it was impossible to obtain a nice clean square edge along with a nice flat top. 
Probe swap as a sanity check only proved the probe was crook.





Siglent will replace it with a warranty claim but IMHO their probe supplier QC is substandard. 

[rf-loop]

@ tautech, lets hope they do not use same kind  source what  Keyshit for cheap probes. All things must not copy.

Back from holidays I loaded the new firmware (without any problems) and I tested the X/Y-Mode myself.
Unfortunately as mentioned already by @rf-loop I had to realize, that the X/Y-Mode is still too slow and the trace jumps jerkily. It seems to be a bit faster as before, but still only twitchy and not continuous - sad  .
The Rigol DS1104Z does better (but also a bit twitchy), the best remains my old and super-simple GW-Instek GDS-1022!

Trace do not jump jerkily if it is used for speeds what it can do.
As told previously it looks like it is not made as fast XY plotter can do with this HW. If they do it, it can be faster than both you named.

But, this looks like it works like normal YT behind XY screen. XY screen is perhaps then derived from these waveforms in acq memory.

What this can do is thing what your bit named scopes can not.
Sigglent can do full resolution measurements in XY mode.
Signal in image is horizontally CH1 60.000001kHz sine and CH2 60.000000kHz pulses. Pulse width 25ns and rise and fall time around 7ns. In picture phase was without reason around 180 degree slowly walking due to freq difference.

Lets hope Siglent is kind and add some day user selectable fast XY plotter so that peoples can plot images and clocks on the screen.   But perhaps first some other new tool functions. Now there is brute force inside box for do some....
Here on the screen this measurement in XY. Of course due to slow 500MSa/s it can only 2ns resolution.