Siglent SDS3000X HD and upgraded SDS1000X HD

[KungFuJosh]

I'm not a somnambulist

Are you sure you haven't sleep walked into a manual or two? 🤣

[2N3055]

I'm not a somnambulist

Are you sure you haven't sleep walked into a manual or two? 🤣

A somnascriptor ??

[Martin72]

It would be great if performa01 would contribute to the manuals.
The documentation is something that, in my opinion, is not yet running smoothly at Siglent.
Be it that the manuals are not updated after various feature upgrades.
Or that certain special functions are not sufficiently explained.
If there are application notes for this, it would be nice if there was at least a reference to this.

[temperance]

I just tried the new software released by Siglent sigscopelab.

But this software seem to require some work.

-Setting the time base cycles the time base on the remote device between 1 s, 2 s, 10 ns instead of following the required setting.
-The application frequently freezes, a command window appears and disappears.

I didn't do any further tests.

Also, this is a free version. I hope they are not planning to release sub standard node locked pay ware like other brands do.

[KungFuJosh]

I just tried the new software released by Siglent sigscopelab.

But this software seem to require some work.

It needs a lot of work, but it's free (at least for now). I tried it with my SDS2504X+, and it worked pretty well, but putting the scope in 10 bit mode made it crazy.

[DaneLaw]

[KungFuJosh]

Nice! Biggest issues I saw:
5:05 didn't know the difference between the internal LA hw or external LA hw connections...
11:32 still didn't know LA hw is external, assumes the zynq chip is handling the LA stuff.
If you don't want to RTFM, look at the product page quickly and you can see the external logic hardware listed.

[ebastler]

As expected, the core architecture seems very similar to the 800X HD series. The front ends differ, of course, due to the 50 Ohm support in the 1000 series.

I am still puzzled by the dual ADCs. Why did Siglent splash out on those, given that the total sampling rate across all active channels is always 2 GSa/s maximum in the SDS1000X HD? Does this imply that the scope could technically run at 2*2 GSa/s (or 4*1 GSa/s) and is just throttled to keep the higher-end models differentiated?

It's the same for the 800X HD platform, of course. But I don't think we ever reached a conclusion there? 

[Martin72]

I would rather enjoy the fact that there has never been so much “scope” for the money as there is now, instead of brooding about it.
Not so long ago, the 12-bit ADCs used were cited as one of the reasons why the 2000X HD was twice as expensive as the 2000X plus.
Now you can get them in a 500€ model.

[ebastler]

I would rather enjoy the fact that there has never been so much “scope” for the money as there is now, instead of brooding about it.
Not so long ago, the 12-bit ADCs used were cited as one of the reasons why the 2000X HD was twice as expensive as the 2000X plus.
Now you can get them in a 500€ model.

Who's brooding? I am just curious why certain design decisions were made and what they imply. Maybe there is something to be learned?

Let's face it, most of our threads here are about "scopology" rather than about using the scopes in practice. (Which is fine by me, there are Project setions etc. which discuss the latter.) No doubt that we get excellent affordable scopes today, which most of the time we won't even use at their full potential. But what's wrong with also being curious about how they work, how they were designed and why? The Test Equipment section would be pretty quiet otherwise.

[core]

I'm waiting for a teardown showing which FPGA and ADC are used in the SDS800X HD. The ADC converter seems to have the smaller capsule vs the SDS1000X HS, but the FPGA might be the same.

Indeed, it's a shame to waste resources, in the sense of not making full use of the ADC converter. 2 x 2GS/s would have made more sense.

If there were limits in the FPGA for example, it would have been logical and cheaper to use an ADC with 4 x 1GS/s channels in the same capsule.

However, we can still enjoy these devices considering the price/performance ratio.

[skander36]

As expected, the core architecture seems very similar to the 800X HD series. The front ends differ, of course, due to the 50 Ohm support in the 1000 series.

I am still puzzled by the dual ADCs. Why did Siglent splash out on those, given that the total sampling rate across all active channels is always 2 GSa/s maximum in the SDS1000X HD? Does this imply that the scope could technically run at 2*2 GSa/s (or 4*1 GSa/s) and is just throttled to keep the higher-end models differentiated?

It's the same for the 800X HD platform, of course. But I don't think we ever reached a conclusion there?

I see two reasons. One or both can be true:

One as you said, the configuration is chosen to create a gap between models.
Or the ADC is configured so, because the FPGA does not have enough resources to process data from 2X2 GSPS.

The double ADC indeed should be capable of a max. 2x2 GSPS in dual ADC config or 4 GSPS in interleaved mode.

[BreakingOhmsLaw]

I would rather enjoy the fact that there has never been so much “scope” for the money as there is now, instead of brooding about it.
Not so long ago, the 12-bit ADCs used were cited as one of the reasons why the 2000X HD was twice as expensive as the 2000X plus.
Now you can get them in a 500€ model.

This, Twice.
A 200MHz 12 bit scope is more than enough for 99% of hobbyist. 25 years ago, digital scopes were in the "small house" price range.
My first privately owned scope in 1995 was a 15MHz analog boat anchor, barely enough bandwidth to service a television.

I could take the 1204X-HD to work and still happily do my job with that. Hell, the R&S I use at work has 10 bits only.

[tv84]

I'm waiting for a teardown showing which FPGA and ADC are used in the SDS800X HD.

The .bit inside the 800/1000X HD FW package is for a Zynq 7z020clg484.

[tv84]

SDS 3000X HD has

Product_ID: 18601

and filesystem contains this password:

root:$5$4se6aot/c2tWs2Lm$UvRirGQV4dp9ytuaWjrbhyqy.glwqqVT8wtMRnc8a57:::::::

Uses Licenses v2.0. 

The app contains references to a new Siglent device's internal name:

HULUWA_2CH
HULUWA_4CH

(don't know if it's one of the recently released as I haven't looked at Siglent's for quite awhile)

[tautech]

The app contains references to a new Siglent device's internal name:

HULUWA_2CH
HULUWA_4CH

The new to come 6&8 GHz models ?
https://www.siglent.com/products-overview/sds7000a/

[tv84]

The new to come 6&8 GHz models ?

First time I see BW references inside Siglent's FW going up to 16GHz !!!

[Martin72]

Bodnarpulse on the SDS3034X HD, risetime about 830ps.

[hpw]

The new to come 6&8 GHz models ?

First time I see BW references inside Siglent's FW going up to 16GHz !!!

This means, any 20..40Gs gear to come. Than we go with Bodnarpulse to the real 30ps risetime 

Requires any better based oscillator & synthesizers as in fs... just look at LeCroy HD series. To get any trusted Jitter figures.

[tautech]

The new to come 6&8 GHz models ?

First time I see BW references inside Siglent's FW going up to 16GHz !!!

This means, any 20..40Gs gear to come. Than we go with Bodnarpulse to the real 30ps risetime 

Requires any better based oscillator & synthesizers as in fs... just look at LeCroy HD series. To get any trusted Jitter figures.

Already available. 
https://siglentna.com/product/precision-frequency-reference/

[hpw]

The new to come 6&8 GHz models ?

First time I see BW references inside Siglent's FW going up to 16GHz !!!

This means, any 20..40Gs gear to come. Than we go with Bodnarpulse to the real 30ps risetime 

Requires any better based oscillator & synthesizers as in fs... just look at LeCroy HD series. To get any trusted Jitter figures.

Already available. 
https://siglentna.com/product/precision-frequency-reference/

Any trusted or valid company would provide any PN figures of the OXCO. In addition any ventilation as vibrations from the gear will reduced the performance of the OXCO to death.
This means also, double oven would reduce the any environment temperature changes.

IMHO, it looks like cheap AliExpress thinking's as no valued gear. While even no performance/improvements & data's provided. 

Hp

[tautech]

The new to come 6&8 GHz models ?

First time I see BW references inside Siglent's FW going up to 16GHz !!!

This means, any 20..40Gs gear to come. Than we go with Bodnarpulse to the real 30ps risetime 

Requires any better based oscillator & synthesizers as in fs... just look at LeCroy HD series. To get any trusted Jitter figures.

Already available. 
https://siglentna.com/product/precision-frequency-reference/

Any trusted or valid company would provide any PN figures of the OXCO. In addition any ventilation as vibrations from the gear will reduced the performance of the OXCO to death.
This means also, double oven would reduce the any environment temperature changes.

IMHO, it looks like cheap AliExpress thinking's as no valued gear. While even no performance/improvements & data's provided.

Tis in the datasheets for the instrument it fits:

OCXO option
Frequency 10M Hz
Initial accuracy -100 +100 ppb 25 ℃
Temperature stability -1 +1 ppb 0 ~ 50 ℃
1st-year aging -50 +50 ppb

[KungFuJosh]

Tis in the datasheets for the instrument it fits:

RTFM / RTFDS?? We have to read to get info now?? 

[ebastler]

OCXO option
Frequency 10M Hz
Initial accuracy -100 +100 ppb 25 ℃
Temperature stability -1 +1 ppb 0 ~ 50 ℃
1st-year aging -50 +50 ppb

I think you are talking about two different requirements here. The optional OXCO seems designed to provide precise and long-term stable frequency, if you want to measure absolute frequencies with the scope. hpw was asking about low jitter and low phase noise, for use as a timebase for very fast acquisition.

[ebastler]

Tis in the datasheets for the instrument it fits:

RTFM / RTFDS?? We have to read to get info now?? 

I would curse more quietly in your place, since you apparently did not read hpw's question.