Siglent SDS2000X HD, 12bit, 2GSa/s high definition oscilloscope serie.

[Martin72]

Somewhere, I think in the now big SDS2000X Plus thread, a member made a mod to his so to not need to remove the sticker to fully dismantle the scope....darned if I can find it just now.

https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg4167997/#msg4167997

https://www.eevblog.com/forum/testgear/new-electricalcomputer-engineer-student-test-bench-equipment-recommendations/msg4124305/#msg4124305

Yepp, drilled holes..

[TERRA Operative]

Looks like he drilled holes in the plastic frame, but left the sticker intact.

I've already reassembled my scope, so maybe I'll just buy a couple overlays if I can, so I have spares for the next time I cook the front end of my scope....

[temperance]

Something totally different.

I'm using the bode plot function of the scope to plot gain and phase of a 700 W SMPS. Some said the bode plot function is slow but I think I understand why.

The SMPS generates some common mode noise which is turned into differential noise. The bode plot function doesn't seem to care about the extra noise at all and is able to measure signals down to over -40 dB buried in noise without any problem. Pretty nice.

[mawyatt]

Yes this Bode Function is quite good wrt ignoring unwanted signals, and employs a Frequency Selective capability where an effective narrow band filter is impressed upon the signals of interest.

Agree, it's somewhat slow, but acceptable. It's been mentioned before that Siglent might consider allowing the user to select the trade-off between capture Time and Rejection.

If you follow some of our posts, we've utilized this Bode Function for a number of applications from Closed Loop, Oscillators, Filters, Impedance Measurements, even investigating Injection Locking, very versatile and useful tool IMO.

Best

[temperance]

Unfortunately the capabilities of the bode plot function can't be deduced from the data sheet nor is there any video demonstrating such capabilities.

Now I just need a better coupling transformer. I made one from a high permeability ring core salvaged from a CM choke. Maybe I will build an active injector. (A someone stated here some time ago "the PFC people.")

If I find time I will demonstrate the bode plot capabilities to measure the closed loop response of an SMPS. (As an experiment I will remove the Y capacitors from the SMPS under test.)

[tautech]

Unfortunately the capabilities of the bode plot function can't be deduced from the data sheet nor is there any video demonstrating such capabilities.

Coughs....

P13:
Bode Plot
Source C1~C4
Supported signal sources:
Built-in waveform generator
SAG1021I (Connection: USB),
SDG series waveform generators (Connection: USB, LAN)

Sweep type Simple, Vari-level
Frequency Mode: Linear, Logarithmic
Range: 10 Hz ~ 120 MHz
Measure Upper cutoff frequency, Lower cutoff frequency, Bandwidth, Gain margin, Phase margin

Manual
https://int.siglent.com/u_file/download/22_06_08/SDS2000X_HD_UserManual_EN01A.pdf
Usage instructions P248

Much about Siglent Bode plot can be deduced from this older thread:
https://www.eevblog.com/forum/testgear/siglent-sds1x04x-e-bodeplot-ii-(sfra)-features-and-testing-(coming)/

[Martin72]

Much about Siglent Bode plot can be deduced from this older thread:
https://www.eevblog.com/forum/testgear/siglent-sds1x04x-e-bodeplot-ii-(sfra)-features-and-testing-(coming)/

And I've already bookmarked it, amazing.

[temperance]

Coughs....

Creating a bode plot for an inductor or a filter when fed by a clean source in a clean environment is one thing. Obtaining a bode plot for an SMPS where the signal in question is buried in noise is a different thing. That the bode plot function is able to do this is amazing for the price and this can't be deduced from the manual.

Edit: here is what I posted before my comment about the "bad manual":
https://www.eevblog.com/forum/testgear/siglent-sds2000x-hd-12bit-(published-for-chinese-domestic-market-only)/msg5911174/#msg5911174

Here is an example for a linear supply:
https://siglentna.com/wp-content/uploads/dlm_uploads/2019/05/Measuring-Power-Supply-Control-Loop-Bode-Plot-with-Bode-Plot-%E2%85%A1.pdf

That doesn't require to much processing as the signals are clean.

[tautech]

Creating a bode plot for an inductor or a filter when fed by a clean source in a clean environment is one thing. Obtaining a bode plot for an SMPS where the signal in question is buried in noise is a different thing. That the bode plot function is able to do this is amazing for the price and this can't be deduced from the manual.
..............

That doesn't require to much processing as the signals are clean.

100%

Then we need you clever ppl to show us your methods for dealing with SMPS as some guide for the Siglent team to develop an advanced App note for this sorta work.
TIA

[KungFuJosh]

I was testing my bandwidth with the other SML03 (and a cheapo cable), and I noticed a goofy insect.

I wanted to place the Cursors info box close to the -3dB cross point.
Every time I touch the info box, the cursor menu pops up.
Then I align the info box with the cross point, and it looks good.
When the Cursors menu disappears, everything on the screen shifts over except the Cursors Info box.

The attached screenshot is what it looks like after the menu disappears, but was previously aligned. I know, it's a minor bug, but it's still annoying. 😉

[Grandchuck]

Martin72, egonotto, and KungFuJosh were discussing pink noise and the Siglent SDG6000X AWG in another thread.  I found it interesting and decided to try it with the AWG in the SDS2000 but had no luck sending the file to the scope.  But it works fine with the SDG2122X. Josh's file is shown below and it looks about right ... seems to roll off around 10dB/decade.

After the pink noise, I tried white noise with a BW of 120 MHz (the SDG2122X limit).  The last screen shows the AWG turned off and so FM stations are seen.

Thanks guys for the interesting discussion and the files for EasyWaveX.

[egonotto]

Hello,

the files are much too large for the 16 kS of the internal generator. You could try a smaller file that is approximately 0.35 s at 44100 Hz.
The internal generator has no True Arb. This could cause problems.

Best regards
egonotto

[KungFuJosh]

Well, looks like Martin's cable wins! ...for now. 😉

I'm gonna make a shorter one and see if it works better. 

[KungFuJosh]

Here's the bandwidth test again after swapping sig gen hardware around. That really is a great cable, Martin. Thanks! 😉

[KungFuJosh]

Here's some more wacky waves.

[KungFuJosh]

I did another BW test with the shorter hyperflex cable I made today. 😉

[KungFuJosh]

I made another cable and did a better job with the shielding, and that squeezed out a little more bandwidth. 🤣

[Performa01]

I did another BW test with the shorter hyperflex cable I made today. 😉

Watch your reference point.
This should usually be 1 MHz. In practice, it also must be at least twice the step size of the FFT and should ideally be higher than the RBW (so it should be possible here).
If 1 MHz is not feasible for some reason, then 10 MHz should be acceptable as well, but certainly not 100 MHz...

Siece flatness isn't perfect up to 100 MHz, frequencies above 10 MHz cannot be used as reference. Even if the deviations are tinly, such as 0.1 dB, the difference in 3 dB bandwidth could be significant, especially when hunting for single digit MHz differences.

[mawyatt]

Josh,

Cable quality is better served with a VNA than a DSO. Even the cheap nano/tiny miniature handheld VNA types should give good repeatable results for cables in this frequency range.

Of course if you want be absolutely sure of your measurements then consider Gore cables 

Best

[KungFuJosh]

Watch your reference point.

Oops, I wasn't paying any attention to that marker. 😉

Here it is with 1MHz ref:

[KungFuJosh]

Of course if you want be absolutely sure of your measurements then consider Gore cables 

I think these M&P cables will be good enough for me (even the cheap BS cables from Amazon were good enough for me if I'm being honest with myself). 😉

Of course, I'm a dumbass, and while ordering some more connectors from M&P, I also ordered a custom cable using their Stealthflex 7 coax. Curiosity is expensive. 🤦

[kmoonwalker]

lab camo mode 

[KungFuJosh]

lab camo mode 

Now I just need to paint my lab to look like the forest!

The double shield braid looked interesting to me. I want to see how that (and them making it) affects the results.

[KungFuJosh]

Is there a way to convince the markers to go to a specific dBm? No matter what frequency I put in, it just rounds the frequency, and chooses between -3.005dBm or -3.032dBm. Is it possible to make it show -3.018dBm?

Thanks,
Josh

[H.O]

I would guess it's due to "limits" in the resolution.
I mean, isn't a difference of 0.027dBm pretty damn small to begin with?