Products > Test Equipment

Siglent SDS1104X-E and SDS1204X-E Mixed Signal Oscilloscopes

(1/419) > >>

Lot of information in this thread is or is perhaps more or less obsolete except very basic things due to several FW updates. Latest FW is:  *.   (* can be 7 or 8, no need think, just do not care in this case) Release Date: 26.SEP.18
So take care about possible obsolete information. Of course there is still LOT of valid information.

ETA2 // 28.05.2019:
Lot of information in this thread is or is perhaps more or less obsolete except very basic things due to several FW updates. Latest FW is:  *.*.6.1.33   (* do not care in this case)
So take care about possible obsolete information. Of course there is still LOT of valid information.
Specially my messages/tests related to BodePlot and FFT.

I recommend for all peoples who tell experience or do some test or find some errors or other ways think that some things need develop and then show examples with his scope.

Important: Please tell always your FW version! when you show some tests or experience with this scope

Please also keep this thread for THIS oscilloscope model functions, experience, tests etc.
For other things use other treads. Or just open new.

This is more than just only 2 extra channel add to SDS1202X-E using cheap way by interleaving same ADC for more channels. 

This 4 channel version have 2 x 2channel 1GSa/s system. Around same principle as in SDS2000X series.  When all 4 channels are simultaneously in use, 500MSa/s for every Ch.   Many fetures like SDS1202X-E but also some more. Deatails later but example bode plot with phase included.

All we know that after domestic public release and launch it takes time before export.  My personal guess is, without any silent knowledge, 1Q2018. Of course these need all independent certified laboratory inspections for proof they meet all safety etc rules in different marketing area etc.

More details about features etc later.

But, it is coming. Also we can now see what is coming.



It is now arrived to markets.

Shared vertical adjustments. I do not like shared controls but this lounge need eat or keep starving - for price and compact size.

First time in Siglent oscilloscope it have now also Sweep Frequency Response Analyzer, SFRA aka BodePlot with phase.
This is bit more than previously launched Keysight InfiniiVision 1000X models very extremely limited performance BP.
Who can say "Scrap a toys and get real oscilloscope"  SDS1004X-E series is lot of in its price class. It can be also MSO. It can be FRA analyzer and it have 1M FFT and really fast segmented memory acquisition aka Sequence mode, up to 80000 segments, down to <2.5µs trigger interval in fast Sequence mode. It have also full speed waveform history buffer always working backround. If you see some fast change in signal and stop scope, you are always late. Not with Siglent. Look previous waveforms in buffer it can still perhaps find.

ETA: Lot of information about BodePlot is obsolete. After FW version 6.1.33 Bode Plot is Bode Plot II.
It is very highly improved (or better say: whole new design) after original first generation BodePlot.

New BP II information can find here.

BodePlot (SFRA)  Max 120MHz or limited more low depending what is used signal generator sinewave max.
Optional 1 channel 25MHz AWG (can control only using scope)
Other compatibles are Siglent SDG family. Just connect cables and BP function take full automatic control, no need even touch SDG front panel - except turn power on).

Removed old version BP image.

Sweep step 10Hz (5000 Hz span, 501 data points.)
Removed old version BP image.

Sweep step 1Hz (500 Hz span, 501 data points.)
Removed old version BP image.

Of course fact that also this oscilloscope have 8 bit ADC means quite low dynamic range. This limit can see also in these images.

Bode blot can do low, medium and high horizontal plot resolution up to 501 data points and up to 3 channels simultaneously (1DUT input with 3 DUT outputs).
If compare to Keysight DSOX1102G it is like from other planet if look resolution - 10 points per decade *   |O   - nearly useless toy. Oh well, ok barely for educational playing with   one pole RC filter etc.
*Ref:  Keysight Technologies, 2017
Published in USA, January 8, 2018

SDS1000X-E  4-channel models. New Firmware FW
Now with new FW measured average max wfm/s is 118kwfm/s (peak max 128kwfm/s)
(single channel, dots, 50µs/div)

EDIT: Old compare table removed.
Reason: This was partially obsolete due to some Rigol FW bugfixes and due to Rigol offer for free Options.

ETA: (tested using FW *.*.6.1.26)
Serial decode: SDS1004X-E. Full memory length, but decode limited to 3000 bytes. Two independent decoder, both can have also different protocol. Example for UART it means total 4 data streams and of course full duplex. Decoder 1, Rx, Tx. Decoder 2, Rx, Tx.  Also it can decode memory stored waveforms, example in waveform history buffer or segmented memory. Decode can do without serial trigger.  Full decode result (data) can also save as .CSV file.

SDS1004X-E have web server so it can watch and control using web browser. (no need install anything, so it is also totally independent of computer OS.)
Update speed is ~10 times/sec using  100M LAN between scope and PC. (image updfate speed do not mean wfm/s speed. Oscilloscope itself works using it normal speed)

Just random nonsense example.

Just other random nonsense example. (69x4 messages and if take history buffer, with this settings there can be 36 these 7s records and all can decode and also full decode lists can save to USB flash (CSV)
Messages from Arduino Mega UART 0,1,2 and 3 Tx (every UART  send around 100ms interval but individual time)
Of course whole memory is decoded, not only zoomed bottom window. Siglent decode full memory and also no meaning if run or stop and same for history buffer (including of course also sequence mode).
Decoder 1: Rx, Tx 19200,8,E,1
Decoder 2: Rx, Tx   9600,8,O,1
(tested using FW *.*.6.1.26)

About noise. Because frequently in many threads some peoples rise question about some oscilloscope front end noise I add it here about SDS1104X-E.
But, it need really also understand what is noise and how it can characterize and measure so that different scopes are compared as apple to apple. It is more or less complex but least it is not simple. We can measure peak to peak noise. This is important value - exept if you read Agilent application notes aka "AN-advertisements". (but still I recommend to read these but just know that they find best settings for they scopes when they compare and other scope perhaps just bad settings or least not optimal)
Many times we need live with peak values - worst case values in many places when we do measurements. Even if we talk about time jitter or signal noise. If we are heating our room with electric we need RMS value.

But then, how we can compare different oscilloscopes if we look peak values. We can easy but result is mostly not apple to apple compare.  Noise is more or less random. Other scope may collect and show less amount of data and propability to get highest peaks are different. Fast cope may show more noise just because it show more data and random distribution more rare events come also more visible. It need also note that Siglemn do not decimate data for display. On the screen trace length is 700 pixel. If there is 7000 data points capture length then 10 data point is mapped to one time position on the screen. If data length is 14M then there is 20000 sample for one time position on the screen. (one pixel column on screen). This kind of things affects lot of how much data you see on the screen and more fat trace when there is noise because more highest peaks.

There is borderline good method for compare. We can compare these values but it is not what we see, so if you read only that scope noise level is measured and it is 50µV RMS - do not think you see 50 or 100uV thick trace. You see more or much more fat trace. Perhaps over 400µVpp depending your scope and scope settings. (Memory, sampling speed, wfm/s speed, amount of displayed data, Horizontal speed... )
But, mesasuring RMS make results more comparable than measuring "visual" peak to peak.

RMS is bit tricky. Because if we have very high sensitivity there exists very easy some DC offset. RMS include this DC offset - of course.  How to clean away this mean value from result. Of course IF scope have very accurate DC offset nulling then RMS is ok but often it drift some amount, specially if scope have high sensitivity.
But wait a momement: Do we really want measure RMS. No.
We need measure RMSnoise. Here we are interested only this noise power but not its mean power value what include DC..  If we can run this (offset)DC  including noise via DC block  then we can just use RMS. But we do not have this. So we need use math. Trick is here.
It is quite simple if oscilloscope have this measurement. Instead of using RMS there can use SDEV. Are you surpriced. RMS = SDEV. (No they are not at all same but in this special case when we measure random noise and mean (DC offset) is removed.)

Here is noise with (100MHz BW because model is 100MHz SDS1104X-E)
Most oscilloscope have 20M BW setting, so, with it 500M and 50M scope can somehow compare also.

500µV/div 100M BW (as can see DC offset removed noise RMS is <60µV)
500µV/div 20 BW (as can see DC offset removed noise RMS is <<30µV )
1mV/div 100M BW (as can see DC offset removed noise RMS is ~60µV )
1m/div 20M BW (as can see DC offset removed noise RMS is ~30µV )

Awesome. 200 MHz and 1 G Samples. I wonder how expensive it will be. If it is similar to SDS1202XE then I am personally going to miss waiting for some more time. ??

Sent from my iPhone using Tapatalk

Is the HDMI connector labelled SBus for the logic analyzer?

How bode plot was made? Does it come with a built-in 30MHz signal generator?

Excellent. Would be nice to compare this with the 1054Z 'bang for buck'.
Thank you for the update.



[0] Message Index

[#] Next page

There was an error while thanking
Go to full version