Author Topic: Siglent SDS1104X-E In-Depth Review  (Read 37458 times)

0 Members and 4 Guests are viewing this topic.

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 1902
  • Country: hr
Re: Siglent SDS1104X-E In-Depth Review
« Reply #175 on: June 16, 2019, 05:17:42 pm »
Here, quick capture, sweep from 1kHz to 200MHz, 200mV P-P, 61 sec sweep, several passes.
Scopes used MSOX3104T and Picoscope 3406D MSO.. Flat top window on FFT.

Picoscope TA131 and Testec TT HF212..
 
The following users thanked this post: Performa01

Offline bobof

  • Contributor
  • Posts: 6
  • Country: gb
Re: Siglent SDS1104X-E In-Depth Review
« Reply #176 on: June 16, 2019, 05:40:10 pm »
Here, quick capture, sweep from 1kHz to 200MHz, 200mV P-P, 61 sec sweep, several passes.
Scopes used MSOX3104T and Picoscope 3406D MSO.. Flat top window on FFT.

Picoscope TA131 and Testec TT HF212..

Am I reading this right - does it tell you not much more than that you can't usefully compare results for different probes on different scopes without detail knowledge of how the input circuitry compares between scopes?

The HF212's look quite nice; I see RS rebadge them in the UK but that they can probably be had cheaper on their original label.
Am I right in thinking they don't come with any sort of identification bands for the connector / probe?  Do the ones from Siglent fit somewhere on the probe / connector?
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3051
  • Country: fi
  • Starting with DLL21
Re: Siglent SDS1104X-E In-Depth Review
« Reply #177 on: June 16, 2019, 05:46:37 pm »
Here, quick capture, sweep from 1kHz to 200MHz, 200mV P-P, 61 sec sweep, several passes.
Scopes used MSOX3104T and Picoscope 3406D MSO.. Flat top window on FFT.

Picoscope TA131 and Testec TT HF212..

No any word about test setup.
Please, can you explain how was probes connected to signal source, including also source impedance.
Test setup is least as important as result data - images.






If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 1902
  • Country: hr
Re: Siglent SDS1104X-E In-Depth Review
« Reply #178 on: June 16, 2019, 06:09:09 pm »
Here, quick capture, sweep from 1kHz to 200MHz, 200mV P-P, 61 sec sweep, several passes.
Scopes used MSOX3104T and Picoscope 3406D MSO.. Flat top window on FFT.

Picoscope TA131 and Testec TT HF212..

No any word about test setup.
Please, can you explain how was probes connected to signal source, including also source impedance.
Test setup is least as important as result data - images.

Like I said before SSDG6000X, probes with BNC tip adapter directly to the output of siggen. Before measurement, I grabbed 1 MHz square wave with direct BNC cable (50 ohm terminated) into reference trace, and compensated both probes to overlay their response over reference one.. and check with 1 kHz square wave for low frequency top flatness..

HF 212 has 3 trimmers, 1 for LF and 2 for HF compensation.. TA131 has only LF compensation on probe body. I didn't try to lift sticker on BNC side box to see if it has HF compensation there but hidden from user.
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 1902
  • Country: hr
Re: Siglent SDS1104X-E In-Depth Review
« Reply #179 on: June 16, 2019, 06:17:53 pm »
Here, quick capture, sweep from 1kHz to 200MHz, 200mV P-P, 61 sec sweep, several passes.
Scopes used MSOX3104T and Picoscope 3406D MSO.. Flat top window on FFT.

Picoscope TA131 and Testec TT HF212..

Am I reading this right - does it tell you not much more than that you can't usefully compare results for different probes on different scopes without detail knowledge of how the input circuitry compares between scopes?

The HF212's look quite nice; I see RS rebadge them in the UK but that they can probably be had cheaper on their original label.
Am I right in thinking they don't come with any sort of identification bands for the connector / probe?  Do the ones from Siglent fit somewhere on the probe / connector?
Testec is available from many sources including Farnell in UK.
I mark them with label printer.

Unfortunately, it was a quick check so Picoscope is not same vertical scale, but if you look closely shape on lower frequencies track quite nicely. It is only as you go higher, Pico starts limiting things being 200MHz scope.
Shape shown on Keysight is pretty much pure probe response, Keysight being 1GHZ ...
 
The following users thanked this post: bobof

Offline bobof

  • Contributor
  • Posts: 6
  • Country: gb
Re: Siglent SDS1104X-E In-Depth Review
« Reply #180 on: June 16, 2019, 07:49:10 pm »
Unfortunately, it was a quick check so Picoscope is not same vertical scale
I hadn't clocked that, thanks for clarification :)
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3051
  • Country: fi
  • Starting with DLL21
Re: Siglent SDS1104X-E In-Depth Review
« Reply #181 on: June 16, 2019, 08:44:53 pm »
Here, quick capture, sweep from 1kHz to 200MHz, 200mV P-P, 61 sec sweep, several passes.
Scopes used MSOX3104T and Picoscope 3406D MSO.. Flat top window on FFT.

Picoscope TA131 and Testec TT HF212..

No any word about test setup.
Please, can you explain how was probes connected to signal source, including also source impedance.
Test setup is least as important as result data - images.

Like I said before SSDG6000X, probes with BNC tip adapter directly to the output of siggen. Before measurement, I grabbed 1 MHz square wave with direct BNC cable (50 ohm terminated) into reference trace, and compensated both probes to overlay their response over reference one.. and check with 1 kHz square wave for low frequency top flatness..

HF 212 has 3 trimmers, 1 for LF and 2 for HF compensation.. TA131 has only LF compensation on probe body. I didn't try to lift sticker on BNC side box to see if it has HF compensation there but hidden from user.

SDG6000X output impedance is 50ohm. So you have tested probes frequency response using 50ohm source impedance.
Normal standard practice is - and have been tens of years - for test probes frequency and step response using 25ohm source impedance!

Also @Performa01 have used right source impedance, as can see example here:

As can be seen, the circuit already contains the standard test setup: Signal generator with 50 ohms source impedance and 50 ohms through termination directly on its output, resulting in a total source impedance of 25 ohms, as seen by the probe.
« Last Edit: June 16, 2019, 08:59:02 pm by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 
The following users thanked this post: 2N3055

Offline graybeard

  • Regular Contributor
  • *
  • Posts: 95
  • Country: us
  • Senior III-V RF/mixed signal/device engineer
    • Chris Grossman
The Siglent PP510 100 MHz probes are a steaming pile of excrement
« Reply #182 on: June 17, 2019, 06:47:29 pm »
I have been using my SDS1104X-E oscilloscope for more than a month, and except for lacking a trigger reference in the data file I have been very happy with the ease of use, features, and performance of the scope. 

However the 100 MHz PP510 probes that come with it are a steaming pile of excrement because of the flaky X1/X10 switch on the probe.  The contacts on the switch are intermittent and easily tweaked while using.  If it was a dedicated X10 probe without the switch it would be fine, but the poor quality switch renders them highly unreliable and useless for me.
 

Offline BillB

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: us
Re: The Siglent PP510 100 MHz probes are a steaming pile of excrement
« Reply #183 on: June 17, 2019, 06:54:21 pm »
However the 100 MHz PP510 probes that come with it are a steaming pile of excrement because of the flaky X1/X10 switch on the probe.  The contacts on the switch are intermittent and easily tweaked while using.  If it was a dedicated X10 probe without the switch it would be fine, but the poor quality switch renders them highly unreliable and useless for me.

That's surprising.  To me, the switch has a decent amount of mechanical resistance to it, and I must make a conscious effort to switch them.  I really haven't noticed any particular flakiness during use.   :-//   
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15358
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent PP510 100 MHz probes are a steaming pile of excrement
« Reply #184 on: June 17, 2019, 07:58:10 pm »
I have been using my SDS1104X-E oscilloscope for more than a month, and except for lacking a trigger reference in the data file I have been very happy with the ease of use, features, and performance of the scope. 

However the 100 MHz PP510 probes that come with it are a steaming pile of excrement because of the flaky X1/X10 switch on the probe.  The contacts on the switch are intermittent and easily tweaked while using.  If it was a dedicated X10 probe without the switch it would be fine, but the poor quality switch renders them highly unreliable and useless for me.
Chris, we do get the very occasional crook one.

I check every one like this:
Connect to Probe Cal output.
Find a V/div and Vertical Pos setting for 10x probe that can also display 1x probe switch setting without any other adjustment. Ensure a Trigger is maintained with each probe setting.
From one or the other probe attenuation switching must be displayed as a snappy clean change in amplitude, no laggy tails or instability at either 1x or 10x. Gently wiggle the switch at 1x, if it's not perfect claim warranty.

Check each probe carefully, they should be perfect.

When you know for sure how many are crook drop Jason a line and get them replaced.

Anyways, luckily the PP510 is a cheap probe and only $10ea and I keep a few in stock just because I do find the odd faulty one.
https://store.siglentamerica.com/product/pp510-1000-mhz-oscilloscope-probe
Avid Rabid Hobbyist
 

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 1902
  • Country: hr
Re: Siglent SDS1104X-E In-Depth Review
« Reply #185 on: June 17, 2019, 10:13:01 pm »
SDG6000X output impedance is 50ohm. So you have tested probes frequency response using 50ohm source impedance.
Normal standard practice is - and have been tens of years - for test probes frequency and step response using 25ohm source impedance!

Also @Performa01 have used right source impedance, as can see example here:

As can be seen, the circuit already contains the standard test setup: Signal generator with 50 ohms source impedance and 50 ohms through termination directly on its output, resulting in a total source impedance of 25 ohms, as seen by the probe.

Thanks for pointing that out. You are right, that IS standard testing procedure.
As I said, it was just a quick scan, and it did impact high end response.
As a matter of a fact it's actually good to see as a comparison.
Passive probes are actually quite a load to a circuit, and response from them is pretty much not very useful when you pass some 100 MHz in a normal circuit...

So now follows with proper source impedance:

Source was SDG6052X, 50 Ohm Output, terminated with passthrough 50 Ohm terminator (combined 25 Ohm source impedance). Probes connected directly to BNC on terminator which is directly on siggen BNC. No cable.

Sinewave 200 mV P-P, sweep from 1kHz to 300MHz, 11 secs per sweep, multiple sweeps to build spectrum.  FFT uses Flattop window, Max-hold. Vertical scale Keysight 1 dB/div, Picoscope approx 1,1 dB/div (don't ask me why, ask Pico)

TA131 is very good for a nominal 250 MHz probe. It's very good match to 3406D with combined response of +0,5/-1,42 dB for a full 200MHz range. HF212 also does well on Pico, -1,5 dB at 200 MHz, with 0,5 dB dip at 100MHz.

On Keysight both probes have upward slope towards 300 MHz (TA131 +2,5dB, HF212 +3dB at 300 MHz.)

Regards,
 
The following users thanked this post: rf-loop, Performa01

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 798
  • Country: at
Re: Siglent SDS1104X-E In-Depth Review
« Reply #186 on: June 22, 2019, 11:08:46 am »
 I’ve learnt that the distortion in the frontend cannot be neglected when the FFT is used for distortion measurements. It turns out that the linearity is is not constant for the various vertical gain settings.

I have tested all vertical gain settings from 500µV/div to 1V/div for both 6 divisions and 8 divisions peak to peak signals. Here are the results:


Vertical Gain            Strongest Harmonic
                6 Divisions pk-pk   8 Divisions pk-pk
500 µV/div        -66.6 dBc           -68.1 dBc
1 mV/div          -62.6 dBc           -70.1 dBc
2 mV/div          -64.5 dBc           -68.1 dBc
5 mV/div          -66.6 dBc           -66.1 dBc
10 mV/div         -61.2 dBc           -66.3 dBc
20 mV/div         -66.0 dBc           -59.3 dBc
50 mV/div         -57.6 dBc           -55.3 dBc
100 mV/div        -53.7 dBc           -50.8 dBc
200 mV/div        -64.4 dBc           -62.7 dBc
500 mV/div        -59.6 dBc           -55.9 dBc
1 V/div           -53.9 dBc           -49.8 dBc

In order to get the best THD performance, you should prefer the underlined setups and avoid the ones that are crossed out.

It might be surprising that the 6 divisions peak to peak signal can have a higher distortion reading than the larger 8 divisions signal, but this is merely because of the noise floor masking the harmonics at the lower signal levels.
 
According to the table above, an 8 divisions full scale signal at 1mV/div is the winner (on my sample of the SDS1104X-E at least), as it allows measurements down to -70dBc (0.032%) with this setup.

Here’s the corresponding screenshot:


SDS1104X-E_FFT_THD_8div_1mV

 
The following users thanked this post: rf-loop, 2N3055

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3051
  • Country: fi
  • Starting with DLL21
Re: Siglent SDS1104X-E In-Depth Review
« Reply #187 on: June 22, 2019, 02:37:11 pm »

According to the table above, an 8 divisions full scale signal at 1mV/div is the winner (on my sample of the SDS1104X-E at least), as it allows measurements down to -70dBc (0.032%) with this setup.


8div and 6div difference is 2.5dB

So it is value what loose in dynamic range if use 6div or 8 dif signal.
With low harmonics 8 bit adc and noise make result unreliable even if linearity is perfect, so how to setup is compromise between scope produced harmonics, DUT harmonics, noise floor and so on.


But other (partially other) question is why low harmonics measured levels are what they are and how these behave.
Individual scopes analog front end linearity can also differ in different V/div bands.

Other thread I show some images. I have also images what have other settings and totally unexpected results like something (error delta) turns like upside down. (Until I know more I have classified these my own not published images as bullshit)

But there comes noise and ADC resolution and perhaps some other things what I can not explain without lot of more detailed knowledge about front end and how system handle data out from ADC. Even errors are possible explanations for some weird things. Other hand, who is using these scope FFT's for scientific work where all, even error mechanisms and sources need evaluate and explain...

It have its limits.

For accurate low THD measurements this is of course wrong tool. Totally. (whole audiophile park is strange to me. I do not have my ear receiving fidelity specifications including its all impurities in signal handling etc. If someone is talking I do not listen his voice fidelity, I listen what he is talking... same with music. But it is other thing.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 798
  • Country: at
Re: Siglent SDS1104X-E In-Depth Review
« Reply #188 on: June 22, 2019, 06:29:46 pm »
Here comes the complete data from my measurements, which might help to understand some unexpected results, such as the huge difference between 6div and 8div signals at 1mV/div.

First the absolute measurement data:


SDS1104X-E_THD_20kHz_dBV

The following table shows the relative harmonic levels for convenience:


SDS1104X-E_THD_20kHz_dBc

Please note that the 2nd harmonic is not always the strongest, particularly not at the high sensitivities.

 
The following users thanked this post: rf-loop, 2N3055

Offline graybeard

  • Regular Contributor
  • *
  • Posts: 95
  • Country: us
  • Senior III-V RF/mixed signal/device engineer
    • Chris Grossman
Re: The Siglent PP510 100 MHz probes are a steaming pile of excrement
« Reply #189 on: June 24, 2019, 08:28:38 am »
I have been using my SDS1104X-E oscilloscope for more than a month, and except for lacking a trigger reference in the data file I have been very happy with the ease of use, features, and performance of the scope. 

However the 100 MHz PP510 probes that come with it are a steaming pile of excrement because of the flaky X1/X10 switch on the probe.  The contacts on the switch are intermittent and easily tweaked while using.  If it was a dedicated X10 probe without the switch it would be fine, but the poor quality switch renders them highly unreliable and useless for me.
Chris, we do get the very occasional crook one.

I check every one like this:
Connect to Probe Cal output.
Find a V/div and Vertical Pos setting for 10x probe that can also display 1x probe switch setting without any other adjustment. Ensure a Trigger is maintained with each probe setting.
From one or the other probe attenuation switching must be displayed as a snappy clean change in amplitude, no laggy tails or instability at either 1x or 10x. Gently wiggle the switch at 1x, if it's not perfect claim warranty.

Check each probe carefully, they should be perfect.

When you know for sure how many are crook drop Jason a line and get them replaced.

Anyways, luckily the PP510 is a cheap probe and only $10ea and I keep a few in stock just because I do find the odd faulty one.
https://store.siglentamerica.com/product/pp510-1000-mhz-oscilloscope-probe

I was in the middle of my impedance measurement project and I did not want to take the time diagnosing probe problems, so I switched to an older HP probe to finish.  Once I switched, I had no more issues.

Today I had some time and played with the probe some more.  It is intermittent, but the problem is not with the switch.  The problem is the spring hook that fits on the end of the probe.  It's contact to the probe tip is intermittent.  The tip itself is straight and the probe works with another spring tip without issue.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 15358
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: The Siglent PP510 100 MHz probes are a steaming pile of excrement
« Reply #190 on: June 24, 2019, 08:33:11 am »
I have been using my SDS1104X-E oscilloscope for more than a month, and except for lacking a trigger reference in the data file I have been very happy with the ease of use, features, and performance of the scope. 

However the 100 MHz PP510 probes that come with it are a steaming pile of excrement because of the flaky X1/X10 switch on the probe.  The contacts on the switch are intermittent and easily tweaked while using.  If it was a dedicated X10 probe without the switch it would be fine, but the poor quality switch renders them highly unreliable and useless for me.
Chris, we do get the very occasional crook one.

I check every one like this:
Connect to Probe Cal output.
Find a V/div and Vertical Pos setting for 10x probe that can also display 1x probe switch setting without any other adjustment. Ensure a Trigger is maintained with each probe setting.
From one or the other probe attenuation switching must be displayed as a snappy clean change in amplitude, no laggy tails or instability at either 1x or 10x. Gently wiggle the switch at 1x, if it's not perfect claim warranty.

Check each probe carefully, they should be perfect.

When you know for sure how many are crook drop Jason a line and get them replaced.

Anyways, luckily the PP510 is a cheap probe and only $10ea and I keep a few in stock just because I do find the odd faulty one.
https://store.siglentamerica.com/product/pp510-1000-mhz-oscilloscope-probe

I was in the middle of my impedance measurement project and I did not want to take the time diagnosing probe problems, so I switched to an older HP probe to finish.  Once I switched, I had no more issues.

Today I had some time and played with the probe some more.  It is intermittent, but the problem is not with the switch.  The problem is the spring hook that fits on the end of the probe.  It's contact to the probe tip is intermittent.  The tip itself is straight and the probe works with another spring tip without issue.
OK, get Jason to send you another grabber hat. They should have some spares from faulty probes.
Avid Rabid Hobbyist
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf