Siglent SDS3000X HD and upgraded SDS1000X HD

[egonotto]

Hello,

in terms of noise at 1 V/div, my SDS3034X HD is superior to my SDS2104X HD. The SDS2104X HD has about 5.5 mV rms at 180 MHz bandwidth and the SDS3034X HD had about 3.9 mV rms at 420 MHz bandwidth.

With a bandwidth of 1 GHz, the noise is about the same as the SDS2104X HD with a bandwidth of 180 MHz.

Best regards
egonotto

[2N3055]

Hello,

in terms of noise at 1 V/div, my SDS3034X HD is superior to my SDS2104X HD. The SDS2104X HD has about 5.5 mV rms at 180 MHz bandwidth and the SDS3034X HD had about 3.9 mV rms at 420 MHz bandwidth.

With a bandwidth of 1 GHz, the noise is about the same as the SDS2104X HD with a bandwidth of 180 MHz.

Best regards
egonotto

Yes, sorry for not explaining better.
I meant for very low level signals with 20MHz BW enabled...

Now you made me curious to go compare in more detail 

[hpw]

Hello,

noise from SDS3104X HD with 100 mV/div.

Best regards
egonotto

Well, I am interested how the SDS3000x HD series performs in low noise as using this instrument as an SA (Spectrum Analyzer).

As using the lowest ranges 500uV ... 10mV scales and various BW & FFT & 16 averages & 1M or 50E inputs load. While noise on FFT may started after a certain freq. as 10...200Hz.


While measure the noise as with samples, there is no description of the internal filtering if applied. Using FFT as looking at the noise floor, it will show spurs or artifacts given from the SA.

My interest as on SA & FFT, to see the idle noise floor given from the today's 12 bit instrument, as I like to know the 14 bit HD3 performance in compartment.

Currently use the SDS2KX Plus in 10bit mode & 100MHz BW as not that bad.

Or I am asking for too much?

[egonotto]

Hello,

from

https://www.eevblog.com/forum/testgear/siglent-sds3000x-hd-and-upgraded-sds1000x-hd/msg5748867/#msg5748867

are a few entries that deal with the FFT and its problems with the SDS3000X HD.

If you want to know more, you can write about it.

Best regards
egonotto
PS: Picture with FFT and ERES 3

[Bad_Driver]

Hi Egonotto,

one question. Is a FFT with 500 MHz span making sense with the reduced bandwidth of ERES3??

[gf]

Hi Egonotto,

one question. Is a FFT with 500 MHz span making sense with the reduced bandwidth of ERES3??

Why not? You can nicely see the frequency response of the digital ERES filter being applied to the noise floor. But I think egonotto complains about the spurs at the marker positions? They originate from imperfections of the ADC.

[2N3055]

Hi Egonotto,

one question. Is a FFT with 500 MHz span making sense with the reduced bandwidth of ERES3??

Why not? You can nicely see the frequency response of the digital ERES filter being applied to the noise floor. But I think egonotto complains about the spurs at the marker positions? They originate from imperfections of the ADC.

Those spurs are at level of -145 dBV or less... so less than 50-60nV RMS..
At 1mV/div top of dynamic range is 8mV P-P, cca 2,8mV RMS. -51dBV.
94dB difference.

Just saying.

[gf]

Hi Egonotto,

one question. Is a FFT with 500 MHz span making sense with the reduced bandwidth of ERES3??

Why not? You can nicely see the frequency response of the digital ERES filter being applied to the noise floor. But I think egonotto complains about the spurs at the marker positions? They originate from imperfections of the ADC.

Those spurs are at level of -145 dBV or less... so less than 50-60nV RMS..
At 1mV/div top of dynamic range is 8mV P-P, cca 2,8mV RMS. -51dBV.
94dB difference.

Just saying.

Sure, you'll have to live with them
Interleaving spurs can be calibrated out to a certain extent, but I suspect this has already been exhausted.

[2N3055]

Hi Egonotto,

one question. Is a FFT with 500 MHz span making sense with the reduced bandwidth of ERES3??

Why not? You can nicely see the frequency response of the digital ERES filter being applied to the noise floor. But I think egonotto complains about the spurs at the marker positions? They originate from imperfections of the ADC.

Those spurs are at level of -145 dBV or less... so less than 50-60nV RMS..
At 1mV/div top of dynamic range is 8mV P-P, cca 2,8mV RMS. -51dBV.
94dB difference.

Just saying.

Sure, you'll have to live with them
Interleaving spurs can be calibrated out to a certain extent, but I suspect this has already been exhausted.

Yes, those are interleaving spurs, and they have been calibrated out to a certain extent.
I can confirm that.

[uc3847]

Hello, I am curious if anone has both an sds3000x and sds6000a and has any thoughts on current benefits of the sds6000a over the sds3000x aside from the really obvious ones (bigger screen, 4*5gs/s, 2ghz bw) ?
My use case is switch mode power supplies up to around 5mhz switching and embedded microcontrollers, fastest signals being spi bus around 200Mhz, I am leaning towards SDS3104x, but would
be very interested in any reasons this might be a bad choice between these two ? Are there any direct noise level comparions floating about already between them ?

[tautech]

Hello, I am curious if anone has both an sds3000x and sds6000a and has any thoughts on current benefits of the sds6000a over the sds3000x aside from the really obvious ones (bigger screen, 4*5gs/s, 2ghz bw) ?
My use case is switch mode power supplies up to around 5mhz switching and embedded microcontrollers, fastest signals being spi bus around 200Mhz, I am leaning towards SDS3104x, but would
be very interested in any reasons this might be a bad choice between these two ? Are there any direct noise level comparions floating about already between them ?

Welcome to the forum.

SDS6000A models are 8bit whereas SDS3000X HD are 12bit.
SDS3000X HD are a very compact 1 GHz platform.
They both support SAPBus active probes.

[Andrew]

I'm looking to upgrade to a scope with triggers on the digital channels and a powered probe interface (and 50ohm, FFT, analysis etc etc). Currently have a KS DSOX2024.

With regards to the memory management of the SDS3000X HD series, does the use of the digital channels and/or math waveforms and/or stored waveforms affect the available memory to the analog channels?

With regards to the probe interface, are there any limitations on the number of powered probes? For example, only 2 powered probes out of the 4 analog channels?

For reference material I'm using datasheet EN01B and manual EN01A, as well as probe datasheet EN01F and SCP5000 instructions EN01A, so hopefully I haven't missed the obvious.

Many thanks

[egonotto]

Hello,

I can't say anything about the probe interface.

If you switch on the digital channels, you still have 400 Mpts for the analog channels. You need at least 10 ms/div for the 100 Mpts to be available for the digital channels. An FFT does not change this. However, you only have 20 Mpts for the addition, for example.

With the SDS3000X HD you don't get much more in segmented mode, sometimes even a little less.

Best regards
egonotto

[Andrew]

@egonotto: Thanks for the quick reply and screenshots.

For the SDS3000X HD FFT, is 4Mpts the total for all Math waveforms using FFT. For example, does using two FFT traces mean a maximum of 2Mpts each?

Many thanks

edit: added "SDS3000X HD"

[2N3055]

@egonotto: Thanks for the quick reply and screenshots.

For the SDS3000X HD FFT, is 4Mpts the total for all Math waveforms using FFT. For example, does using two FFT traces mean a maximum of 2Mpts each?

Many thanks

edit: added "SDS3000X HD"

No. FFT is 4Mpts for each math.

[kladit]

I measured the bandwidth of my new scope via rise time measurement using an Leo Bodnar pulser (BNC-version 30ps).

According to the formular bandwitdh(3db) = 0.35/(rise time) I get 0.35/387.13^-12  = 904.1 MHz.

Is  this ok? I have expectecd more than 1 GHz as all my previous scopes from Siglent have had a
higher bandwitdh than the announced one.

@performa01: you have an SDG7000 contrary to my SDG6000, can you measure bandwitdh with it please.

What results other owners of this scope have measured?

[2N3055]

I measured the bandwidth of my new scope via rise time measurement using an Leo Bodnar pulser (BNC-version 30ps).

According to the formular bandwitdh(3db) = 0.35/(rise time) I get 0.35/387.13^-12  = 904.1 MHz.

Is  this ok? I have expectecd more than 1 GHz as all my previous scopes from Siglent have had a
higher bandwitdh than the announced one.

@performa01: you have an SDG7000 contrary to my SDG6000, can you measure bandwitdh with it please.

What results other owners of this scope have measured?

We keep repeating , you cannot use risetime of pulse  to calculate BW using factor 0.35 because of shape of the filter.
A factor of 0.45 is better approximation.
But that is also rough estimation. Only proper way to measure BW is leveled signal generator and frequency sweep.

[Martin72]

Only proper way to measure BW is leveled signal generator and frequency sweep.

Like this:

https://www.eevblog.com/forum/testgear/siglent-sds3000x-hd-and-upgraded-sds1000x-hd/msg5762537/#msg5762537

[Andrew]

With regards to the probe interface, are there any limitations on the number of powered probes? For example, only 2 powered probes out of the 4 analog channels?

Confirmation from a local supplier that the SDS3000X HD can power four SAPBus probes at the same time.

Hoping to be the owner of one of these scopes soon.

[CTintheUK]

I'm looking to upgrade to a scope with triggers on the digital channels and a powered probe interface (and 50ohm, FFT, analysis etc etc).

I too am looking to upgrade.  I'm wondering if we will see a price promotion on the SDS3000X HD scopes this year, or a bundle offer.  Perhaps when the SDS5000X HD family gets its international launch.  Time will tell.

CT in the UK

[Martin72]

The 5000X HD series is priced in a different league, so that shouldn't affect any promotions regarding the 3000X HD.
But of course, hope dies last.

[2N3055]

The accuracy if one uses the scope to measure or adjust noise levels of generators is influenced by this.
I often use the scope as RF-Millivoltmeter to measure rf-noise levels.

Every question calls for more questions...

Amount of intrinsic noise is not same as scope calibration.
If you apply some external signal what are differences in measured value between channels ?
Noise figures add as RMS, and you will need to subtract instrument noise (whatever it is) from measured value anyways, unless instrument noise is minor compared to signal. In which case even smaller variation on already small signal becomes irrelevant.

In RF measurements, values are usually expressed and specced in dB. What is variation in noise figures for scope channels when expressed as dB on that range?

One note. It is recommended to run self cal occasionally, and when there are large changes in ambient temperature.

[KungFuJosh]

I wonder asbout the great differences  between channels.

Noise of divider resistors, different noise levels at different gains in pre-amp-ic?

Can other owners of  SDS3000XHD please check to what amount noise levels
change in all 4 channel at different sensitivities/division.

Already been done. Scroll back through the thread.

[Martin72]

I wonder asbout the great differences  between channels.

The mean values from the first image:

108µV 110µV 111µV 109µV

The mean values from the second image:

5.4mV 5.5mV 5.2mV 5.4mV

I can´t see great differences between them.

The accuracy if one uses the scope to measure or adjust noise levels of generators is influenced by this.

If I really want to measure/adjust something accurately, I wouldn't do it with an oscilloscope.
Or rather, depending on how accurate it has to be, it's a truism that the accuracy of the measuring instrument must be higher than the accuracy of the test object.

[Martin72]

Here is my scope, after warming up.
There is nothing that would worry me.
After that, I connected my SDG6000X and set it to a 0dBm level.
100kHz and 10khz, then measured with my SDM3065X.
SDM3065X : 223.934mV/10kHz, 223.994mV/100kHz
And who's right now?
You shouldn't worry so much about these things.