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

[Martin72]

I´ll do..

[Martin72]

Here, see pics.
I must confess, the menu on lecroys of today are slightly "better", I like the horizontal menu style more than vertical.
Pics showing what to print and in which color, standard, printing or black&white...
In the channel menu (pics tomorrow) you can choose between screen color and print color.

[Martin72]

Yep, this is a "pro" request, normally hobbyists don´t have this "problem".
BTW, printing the screen as it is is maximum horrible to view on documents.
I´ll show some examples tomorrow or the next days and then in the feature thread.

Strongly disagree however that's only my opinion. Convince me otherwise.

Here a print-out with the actual three possibilities.
Leaving as it is, inverted and inverted and other trace color.
The "worst" is inverted with default trace color, followed by non-invert.
The best to see on paper is inverted and "new" trace color.
Best to see on screen is yellow, best to see on paper is blue, so different adjustments for screen/printout makes sense to me.

[Martin72]

Hi,

Saw defpoms video about the SDS6000A a few minutes ago.
I saw that you can choose in the memory management menu "fixed samplerate", something I also know from lecroy scopes - But it´s not avaible for the HD, only auto mem and fixed mem.
Will it be implemented in ne of the next firmware updates or is this missed because of the lower samplerate of the scope ?

[tautech]

Hi,

Saw defpoms video about the SDS6000A a few minutes ago.
I saw that you can choose in the memory management menu "fixed samplerate", something I also know from lecroy scopes - But it´s not avaible for the HD, only auto mem and fixed mem.
Will it be implemented in ne of the next firmware updates or is this missed because of the lower samplerate of the scope ?

Maybe you need install this:
https://int.siglent.com/upload_file/zip/firmware/Oscilloscope/SDS2000X_HD_V1.2.1.1_EN.zip

Also see pic below.

[Martin72]

Damn I was blind, forget this thing...
Looking in the manual which I got on my amazon tablet, it´s already mentioned there...
Dunno why I thougt it wasn´t there.

[Martin72]

Hi,

Concerning the "new" filter functions....
Is there/will be a "white paper" or instruction avaible on how to adjust them/how it works basically?
Will it work like rigol described it here:

https://rigol.force.com/support/s/article/Setting-Filter-Frequency-on-MSO5000-MSO7000-and-MSO8000-Oscilloscopes

[baldurn]

Ahh look at what Santa dropped down the chimney!

[kripton2035]

welcome to the HD owners club !

[Martin72]

We're getting more and more and someday we'll rule the world, ahahahahaha ! 

[Martin72]

(Slightly) OT:

I was fed up with the two boards (deskew fixture and demo-board) being naked, now they have a base plate, which looks a bit better and is also more comfortable to handle.
But at these prices, Siglent could also have installed them.

[tautech]

(Slightly) OT:

I was fed up with the two boards (deskew fixture and demo-board) being naked, now they have a base plate, which looks a bit better and is also more comfortable to handle.
But at these prices, Siglent could also have installed them.

Mine came with 25 x M3 bolts and nuts for legs with the threaded shank as the foot. 
Any nicely finished surface would get scratched.
The addition 4x M3 stainless dome nuts fixed that.

An even cheaper solution would be nylon dome nuts but they must be hard to find in Shenzhen. 

[mawyatt]

Or just Nylon M3 Standoffs, this is what we usually use.

Best,

[Martin72]

This you can do at home or when you build a protoype for lab testing - But when selling it commercial and for not less money, it´s looking very unprofessional.

Mine came with 25 x M3 bolts and nuts for legs with the threaded shank as the foot.

Mine too, guess how I was suprised when open the box for the first time...Are they serious ?! was my first thought. 

[tautech]

This you can do at home or when you build a protoype for lab testing - But when selling it commercial and for not less money, it´s looking very unprofessional.

Mine came with 25 x M3 bolts and nuts for legs with the threaded shank as the foot.

Mine too, guess how I was suprised when open the box for the first time...Are they serious ?! was my first thought.

Me too.
These are NOT cheap items, $205 for STB-3 and $265 for the Deskew fixture is almost obscene pricing. 

STB-3 has a LOT of functionality and arguably is worth the price however not the way it is mounted whereas for just a small piece of plexiglass for a foot it could be so so much better.
Hell, M3 dome nuts of any flavor would be an improvement ! 

The Deskew fixture OTOH is nothing special at all and at one time much the same thing by way of a current loop was standard fare on the front of most CRO's however Tek did it properly with a robust and formed rod that also served as the probe compensation connection.
IIRC even the old D83 I once had provided a specified current compensation output if looped directly to GND.
Real handy it was too for getting input attenuations correct for easy maths when using current probes on a CRO.
Today we just set the input attenuation to match the fixture current pulse and engage measurements......and ppl highly value those old POS. 

But for todays requirements deskew @1kHz is not adequate whereas the Siglent deskew fixture operating near 23kHz much nearer to typical SMPS frequencies so deskew accuracy is much improved but still, it's a very basic circuit that has been replicated here:
https://www.eevblog.com/forum/testgear/scope-probe-deskew-fixture-pcb-project/

[finetuned]

I'm considering purchasing the SDS2104X HD. Does the (apparent) limited rate of capture for digital signals (the reported 16 wfms/sec) impact the scope's ability to do serial decoding? Will it keep up with common serial baudrates like 115200 baud for UART or 1M baud for CAN? These would require way more than 16 wfms/sec, but I'm not sure what to make of the reported limitation.

EDIT: After reading up on the waveform capture rate, I think that to reliably capture a telegram of data on any scope, the timebase should be set such that the entire telegram fits inside one horizontal sweep, because any scope will have a certain 'dead time' between sweeps and will therefore miss some of the bits that fall in between the end of the previous sweep and the beginning of the next one. Is this assumption correct, and should I therefore not worry too much about this issue pertaining to digital waveform capture speed?

[2N3055]

I'm considering purchasing the SDS2104X HD. Does the (apparent) limited rate of capture for digital signals (the reported 16 wfms/sec) impact the scope's ability to do serial decoding? Will it keep up with common serial baudrates like 115200 baud for UART or 1M baud for CAN? These would require way more than 16 wfms/sec, but I'm not sure what to make of the reported limitation.

That is 16 separate trigger events.. Not sampling speed.
It means it can capture whole buffer of data on all 20 channels and show it on screen. 16 times a second.

If you have fast burst of 10 packets and then pause, you can set it to capture whole burst in one go.

Also there is something called segmented mode, where scope capture number of trigger events sequentially without updating screen. That one can achieve some 500k triggers in second..

[finetuned]

That is 16 separate trigger events.. Not sampling speed.
It means it can capture whole buffer of data on all 20 channels and show it on screen. 16 times a second.

If you have fast burst of 10 packets and then pause, you can set it to capture whole burst in one go.

I think the edit to my post and your answer crossed each other. It looks like it confirms my updated understanding of how this is supposed to work, thanks!

[markone]

I'm considering purchasing the SDS2104X HD. Does the (apparent) limited rate of capture for digital signals (the reported 16 wfms/sec) impact the scope's ability to do serial decoding? Will it keep up with common serial baudrates like 115200 baud for UART or 1M baud for CAN? These would require way more than 16 wfms/sec, but I'm not sure what to make of the reported limitation.

EDIT: After reading up on the waveform capture rate, I think that to reliably capture a telegram of data on any scope, the timebase should be set such that the entire telegram fits inside one horizontal sweep, because any scope will have a certain 'dead time' between sweeps and will therefore miss some of the bits that fall in between the end of the previous sweep and the beginning of the next one. Is this assumption correct, and should I therefore not worry too much about this issue pertaining to digital waveform capture speed?

If a gapless serial line capture/decoding is the main purpose, a dedicated logic analyzer would provide a much better service with less money.

[finetuned]

If a gapless serial line capture/decoding is the main purpose, a dedicated logic analyzer would provide a much better service with less money.

Serial decoding will only be an occasional thing that I'll use the instrument for, but thanks for the tip!

[Martin72]

6 months it´s mine and still I can´t take my eyes off this beauty...
Seriously, the design is just very well done.

[elimenohpee]

I'm doing some tax harvesting for the end of the year and on the fence of upgrading to this unit.  But after comparing with SDS2000X Plus, I'm just not sure I see the real advantage this has other than the additional 2 bits of vertical resolution.  Most of the specs between the units look to be nearly identical.  Am I missing something?  What's the case for buying the HD and is it worth the extra $2k+ over the plus model? 

[Martin72]

Hi,
I can list the advantages, whether they are enough for you, you must decide for yourself.

- True 12 bit resolution without any "tricks", it´s not "only 2 bits more"
- ERES and Average in Acquisition menu, instead as a math function
- Memory Management: Auto-Mem, Fixed-Mem/Samplerate
- DVM (Digital Voltmeter)
- Digital Filters, HP, LP, BP, etc
- Way less noise because of controlled fan
- More "heavy" building, look and feel, better incremental encoder

That's what I just thought of now, I may have forgotten something else.
The chances are also not bad that future features from the "big ones" (6000A) will also find their way to this model, because the performance might still allow it, but that of the Xplus no longer (see also memory management, filter).
You also have to remember, for the money you get the 350Mhz model SDS5000X.
With 5GSa/s and the "Pro bus" interface (do not know the exact name).
But that has 8 bit only.
I (who owned an Xplus) faced this decision 6 months ago and took the HD, to date without regret.

[tautech]

I'm doing some tax harvesting for the end of the year and on the fence of upgrading to this unit.  But after comparing with SDS2000X Plus, I'm just not sure I see the real advantage this has other than the additional 2 bits of vertical resolution.  Most of the specs between the units look to be nearly identical.  Am I missing something?  What's the case for buying the HD and is it worth the extra $2k+ over the plus model?

These are another level in cost for sure but they are also another level in design polish and capability.
Only had one for a customer through our hands and wanted to keep it.
X HD are real nice but if you don't have the budget you won't be bitterly disappointed with an X Plus.

Certainly with HD on promo for free next model upgrade grab one as the promo is to end as forecast. 

[Performa01]

I'm doing some tax harvesting for the end of the year and on the fence of upgrading to this unit.  But after comparing with SDS2000X Plus, I'm just not sure I see the real advantage this has other than the additional 2 bits of vertical resolution.  Most of the specs between the units look to be nearly identical.  Am I missing something?  What's the case for buying the HD and is it worth the extra $2k+ over the plus model? 

It is not just 2 bits. 12 - 8 = 4, so we get a 4 bits advantage of course. The SDS2000X HD is a 12 bit MSO, capable of 500 MHz bandwidth, whereas the SDS2000X Plus is 8 bit with an optional 10 bit ERES mode, which only gives a single bit of additional ENOB (in theory) while limiting the bandwidth to 100 MHz.

Apart from this, the question has been answered on page 1 of this thread already:

https://www.eevblog.com/forum/testgear/siglent-sds2000x-hd-12bit-(published-for-chinese-domestic-market-only)/msg3935089/#msg3935089

Additional points are better rotary encoders, better build quality in general and a dead silent fan.