Siglent SDS2000X Plus

[rf-loop]

When the scope is off, but not unplugged from the mains, the power button would still glow green faintly. It could come in very handy in a dimly lit environment. As Dave Jones would famously say: “I like it!”

In this video it appears to pulse rather than glow steadily?

https://youtu.be/TcrSmweH5_w?t=200

Looking with eyes it do not blink. Steady and very dim. I have SDS2kXPlus just front of me and in normal room light can not even detect there is light.
I believe this blinking effect in this old video is just stroboscopic effect with video camera. I do not know if older version have more bright led than today versions.

[tautech]

When the scope is off, but not unplugged from the mains, the power button would still glow green faintly. It could come in very handy in a dimly lit environment. As Dave Jones would famously say: “I like it!”

In this video it appears to pulse rather than glow steadily?

https://youtu.be/TcrSmweH5_w?t=200

Looking with eyes it do not blink. Steady and very dim. I have SDS2kXPlus just front of me and in normal room light can not even detect there is light.
I believe this blinking effect in this old video is just stroboscopic effect with video camera. I do not know if older version have more bright led than today versions.

100%
No Siglent model has ever had a power button pulse like that.

[Michael YYZ]

It’s not flickering (“pulsing”) on mine. The artifact that you’re seeing in the video may be due to the camera frame rate that the video blogger used.

In this video it appears to pulse rather than glow steadily?

https://youtu.be/TcrSmweH5_w?t=200

[Johnny B Good]

When the scope is off, but not unplugged from the mains, the power button would still glow green faintly. It could come in very handy in a dimly lit environment. As Dave Jones would famously say: “I like it!”

In this video it appears to pulse rather than glow steadily?

https://youtu.be/TcrSmweH5_w?t=200

Looking with eyes it do not blink. Steady and very dim. I have SDS2kXPlus just front of me and in normal room light can not even detect there is light.
I believe this blinking effect in this old video is just stroboscopic effect with video camera. I do not know if older version have more bright led than today versions.

100%
No Siglent model has ever had a power button pulse like that.

 That's only true if you're ignoring the SDS1202X-E model (likely most, if not all of the SDS1xxx models too). Mine slowly winks at 420mHz (2.375 seconds cycle time). 

 The SDS2104X Plus and the SSA3021X Plus otoh, both simply show a steady dim glow when connected to mains power and shut down.

[highpower]

Looking with eyes it do not blink. Steady and very dim. I have SDS2kXPlus just front of me and in normal room light can not even detect there is light.
I believe this blinking effect in this old video is just stroboscopic effect with video camera. I do not know if older version have more bright led than today versions.

Thank you for that. I just bought a SDS2204Xplus and didn't notice any light at all in a normally lit room. I thought they may have changed something recently. I haven't looked at it myself in the dark as of yet. 

Thanks to all of you for the clarification.

[Martin72]

Thank god, I don´t care about a lightening button or a more or less noisy fan...

[Michael YYZ]

Two fewer worries for a happy bloke! 

Thank god, I don´t care about a lightening button or a more or less noisy fan...

[Michael YYZ]

Does anyone know what parameter D in the User Manual (UM0102XP-E01B) means? This is on page 214/335, in section 19.4 Frequency Analysis. Thanks.

[DL2XY]

D stands for "Decimation factor" (divisor).
In the example D=4.

[Michael YYZ]

That is logical. But how is this factor calculated? On what basis? Why does it have a value of 4 in the example from the User Manual? What other values can it have?

D stands for "Decimation factor" (divisor).
In the example D=4.

[Michael YYZ]

I don’t know what I’ve done here. I wanted to check out the effect of ERES on random noise, as follows:

• C1 = Random noise from the AWG
• C3 = Square wave from probe calibration output
• F1 = C1 + C3
• F2 = ERES (F1)

And now the 0.5-bit option for ERES is greyed out and cannot be selected. Any idea why?
Perhaps it’s time to better go to bed...

[tautech]

And now the 0.5-bit option for ERES is greyed out and cannot be selected. Any idea why?

Because the DSO is in 10 bit mode.
Look at the timebase tab. 

[Michael YYZ]

Not really. ERES 0.5-bit is still not working with the ADC set to 8 bits. See attached.
However, it works if I do ERES(C1), but not for ERES(C1+C3). This drives me nuts! Why is only the 0.5-bit option affected? Could it be a bug, perhaps?

Because the DSO is in 10 bit mode.
Look at the timebase tab. 

[Performa01]

The SDS2104X Plus and the SSA3021X Plus otoh, both simply show a steady dim glow when connected to mains power and shut down.

My SDS2354X Plus still does the slow blinking. It's a very early specimen though.
The fact that this behavior never changed with FW updates proves that this is in the PSU and cannot be controlled by the FW.

The reason why Siglent has changed this at some point should be obvious: as always, it's been users bitching about irritating blinking lights in their labs.

And I have to agree with them.

[Performa01]

Finally I’ve found some time to measure the frequency response of the PP510 probe on a SDS2354X plus with 500 MHz option.

I ran two measurements in parallel, one with direct coax connection on channel 4 and the PP510 (in x10 mode of course) on channel 2. See attachment.

The signal level was -14 dBV (rms).

At 1 MHz, the probe measurement was -13,78 dBV, so there is an error of +0.22 dB.
At 500 MHz, the probe measurement was -16.72 dBV, the corresponding error is -2.72 dB.

The graph clearly shows that the system bandwidth is 0 to 500 MHz with +0.22 / -2.72 dB deviation.
The probe only frequency response is even better, as the difference between probe and direct connection always stays below 2 dB.

[Emo]

Hi Performa,

These "oscillating" kind of graphs makes me wonder if in your setup the impedances have been off. Is your source properly terminated?

Eric

[rf-loop]

Finally I’ve found some time to measure the frequency response of the PP510 probe on a SDS2354X plus with 500 MHz option.

I ran two measurements in parallel, one with direct coax connection on channel 4 and the PP510 (in x10 mode of course) on channel 2. See attachment.

The signal level was -14 dBV (rms).

At 1 MHz, the probe measurement was -13,78 dBV, so there is an error of +0.22 dB.
At 500 MHz, the probe measurement was -16.72 dBV, the corresponding error is -2.72 dB.

The graph clearly shows that the system bandwidth is 0 to 500 MHz with +0.22 / -2.72 dB deviation.
The probe only frequency response is even better, as the difference between probe and direct connection always stays below 2 dB.

Here is my previous measurement. Yes individual probes etc... all make small differences.

But now if look this coaxial... something is now wrong, really wrong.


My measurement (can find in this thread) Coaxial transmission looks like no very bad mismatch.
ETA:Note, this is PP215 not PP510. But this have compared to coaxial cable. Previous msg have also PP215 compared to PP510 and they are very similar if look only freq response.


Something is now out of order. Perhaps accidentally some mistake or failed coaxial, terminator, connector etc (1) 

Also there may be some small impedance difference between voltage bands but this is so small that it can not explain this.
ETA: added image about it.

(1) BNC connectors are terrible when they wear and also these have many times quality issues.


Here in this animation is perfectly same sweep using V/div 100mV  and then V/div 102mV. Just this step change most low voltage range to middle range. Next change is between 1.00V|1.02V/div



Note. This scope is 100MHz model SDS2104X Plus and just factory new "out from box" condition. As can see -3dB is around 180MHz
So its f.response is what it is but if there is mismatch in signal transmission it can see as "waves". 

As can see there is some "waves" after 250MHz. Most low Voltage band bit more than 102mV/div - 1.00V band.  (note 2dB/div when other images have 1dB/div) But far less than @Performa01 image what
@Emo  noted

These "oscillating" kind of graphs makes me wonder if in your setup the impedances have been off. Is your source properly terminated?.

 

[Michael YYZ]

Oh, boy! This scope gives me a lot of headache! In addition to the previous questions posted here, which are still not clear to me:

• https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3551527/#msg3551527
• https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3551491/#msg3551491

what am I doing wrong this time?

I am trying to set up the user-adjustable rise-time and fall-time measurements. It works okay with percentage values. But I cannot make them work with absolute values for C1, C3, etc.: that is, for any values that I select for Horizontal parameters I get an "Invalid threshold" error message. But they seem to be working perfectly okay for the Math channels, e.g F1 (which is the sum of C1+C3). See screenshots attached.

I even calculated the absolute values to match the percentage values, and the absolute measurement setting still does not work for the C channels. What am I doing wrong here? 

Thanks a lot!

[Performa01]

Something is now out of order. Perhaps accidentally some mistake or failed coaxial, terminator, connector etc (1) 

Thanks (also Emo!) for pointing this out.
During the probe tests, I was so used to wobbly frequency response graphs that the 1 dB peak to peak wobble with the direct coax connection didn't bother me ...

I have no idea. Im not in my usual lab and only have access to a bunch of professional BNC cables (unknown, from eBay, looks similar to RG316) with high quality connectors, which all give very similar results. Since there is no guarantee what they really are, one might suspect they might be 75 ohms. But in this case, I'd expect a much higher error and I also happen to have one crap-quality RG58 cable at hand (this has to be 50 ohms for sure) and if anything, the result looks even worse with this one.

Attached is a screenshot that compares the ?RG316? (F1) with the RG58 (F2).

BTW, quite a while ago I've checked the return loss of the SDS2000X Plus 50 ohms inputs and found them to be perfectly adequate, i.e. >14 dB up to 1 GHz. So this cannot be it, especially not since the first wobble appears at and below 100 MHz already, where the RL exceeds 26 dB.

So if it's not the scope inputs and not the cables, what else could it be?
The generator is different from my usual setup, because I only have the SDG6052X available here. But it's very unlikely the generator can have such an impact, especially as long as the termination at the scope input is reasonably correct.

Whatever; a 1 dB pk-pk wobble is not nice, yet it doesn't change the core statement of this test, which was to demonstrate how the PP510 probe interacts with the 500 MHz SDS2354X Plus.

[TK]

Oh, boy! This scope gives me a lot of headache! In addition to the previous questions posted here, which are still not clear to me:

• https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3551527/#msg3551527
• https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3551491/#msg3551491

what am I doing wrong this time?

I am trying to set up the user-adjustable rise-time and fall-time measurements. It works okay with percentage values. But I cannot make them work with absolute values for C1, C3, etc.: that is, for any values that I select for Horizontal parameters I get an "Invalid threshold" error message. But they seem to be working perfectly okay for the Math channels, e.g F1 (which is the sum of C1+C3). See screenshots attached.

I even calculated the absolute values to match the percentage values, and the absolute measurement setting still does not work for the C channels. What am I doing wrong here? 

Thanks a lot!

I am getting the same errors

[rf-loop]

Something is now out of order. Perhaps accidentally some mistake or failed coaxial, terminator, connector etc (1) 

Thanks (also Emo!) for pointing this out.
During the probe tests, I was so used to wobbly frequency response graphs that the 1 dB peak to peak wobble with the direct coax connection didn't bother me ...

I have no idea. Im not in my usual lab and only have access to a bunch of professional BNC cables (unknown, from eBay, looks similar to RG316) with high quality connectors, which all give very similar results. Since there is no guarantee what they really are, one might suspect they might be 75 ohms. But in this case, I'd expect a much higher error and I also happen to have one crap-quality RG58 cable at hand (this has to be 50 ohms for sure) and if anything, the result looks even worse with this one.

Attached is a screenshot that compares the ?RG316? (F1) with the RG58 (F2).

BTW, quite a while ago I've checked the return loss of the SDS2000X Plus 50 ohms inputs and found them to be perfectly adequate, i.e. >14 dB up to 1 GHz. So this cannot be it, especially not since the first wobble appears at and below 100 MHz already, where the RL exceeds 26 dB.

So if it's not the scope inputs and not the cables, what else could it be?
The generator is different from my usual setup, because I only have the SDG6052X available here. But it's very unlikely the generator can have such an impact, especially as long as the termination at the scope input is reasonably correct.
Whatever; a 1 dB pk-pk wobble is not nice, yet it doesn't change the core statement of this test, which was to demonstrate how the PP510 probe interacts with the 500 MHz SDS2354X Plus.

"yet it doesn't change the core statement of this test, which was to demonstrate how the PP510 probe interacts with the 500 MHz SDS2354X Plus."  Yes also I think it do not invalidate this core statement.
Also it was not my intention when I comment this.
My comment was mainly inspired by @Emo and also myself when I see this image where signal is connected using coaxial and also I have some imagination that usually you do very perfect measurements.

It is now very clear something is somehow wrong. But now after your comment, more mysterious. And then even more after these two coaxials. I assume they both come also from separate generator outputs so that just gen out - cable - scope input.


When I look this image and previous, what FFT window you have used. (nothing to do with this wobbling)
For this kind of things I use always Flattop if not serious mandatory reason to use some other.

Just only for thinking what is big and what is small.
For just example imagined situation where all is pure resistive.
INPUT DATA
Set Source Impedance:50
Set Load Impedance R:45.25
Set Load Impedance J:0

RESULT
Absolute Load Impedance:45.25
Load Reflection Coefficient:0.0499
Load VSWR:1.105
Load Return Loss:26.04dB
Load Mismatch Attenuation:0.011dB

If pure resistive load 33.5 ohm
Absolute Load Impedance:33.5
Load Reflection Coefficient:0.1976
Load VSWR:1.493
Load Return Loss: 14.08dB
Load Mismatch Attenuation:0.173dB

[Performa01]

It is now very clear something is somehow wrong. But now after your comment, more mysterious. And then even more after these two coaxials. I assume they both come also from separate generator outputs so that just gen out - cable - scope input.

Your assumption is correct.

When I look this image and previous, what FFT window you have used. (nothing to do with this wobbling)
For this kind of things I use always Flattop if not serious mandatory reason to use some other.

I've used Flattop window.

Just only for thinking what is big and what is small.
For just example imagined situation where all is pure resistive.
INPUT DATA
Set Source Impedance:50
Set Load Impedance R:45.25
Set Load Impedance J:0

RESULT
Absolute Load Impedance:45.25
Load Reflection Coefficient:0.0499
Load VSWR:1.105
Load Return Loss:26.04dB
Load Mismatch Attenuation:0.011dB

If pure resistive load 33.5 ohm
Absolute Load Impedance:33.5
Load Reflection Coefficient:0.1976
Load VSWR:1.493
Load Return Loss: 14.08dB
Load Mismatch Attenuation:0.173dB

Yes, the load matching isn't ideal - and the scope input is not pure resistive for sure either.
When thinking about it, isn't a VSWR >1.1:1 (10%) per definition nearly 1 dB ripple?
On the other hand, a VSWR of up to 1:1.5 is quite typical for proper 50 ohm inputs on DSOs.
So if I'm not wrong, there is rather the question why your coax measurements look so clean and wobble-free?

[Michael YYZ]

Thanks. Perhaps these are bugs. I will report them in the specialized bugs thread, although I’m not sure if anyone from Siglent reads those posts or actually cares.

Oh, boy! This scope gives me a lot of headache! In addition to the previous questions posted here, which are still not clear to me:

• https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3551527/#msg3551527
• https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg3551491/#msg3551491

what am I doing wrong this time?

I am trying to set up the user-adjustable rise-time and fall-time measurements. It works okay with percentage values. But I cannot make them work with absolute values for C1, C3, etc.: that is, for any values that I select for Horizontal parameters I get an "Invalid threshold" error message. But they seem to be working perfectly okay for the Math channels, e.g F1 (which is the sum of C1+C3). See screenshots attached.

I even calculated the absolute values to match the percentage values, and the absolute measurement setting still does not work for the C channels. What am I doing wrong here? 

Thanks a lot!

I am getting the same errors

[rf-loop]

It is now very clear something is somehow wrong. But now after your comment, more mysterious. And then even more after these two coaxials. I assume they both come also from separate generator outputs so that just gen out - cable - scope input.

Your assumption is correct.

When I look this image and previous, what FFT window you have used. (nothing to do with this wobbling)
For this kind of things I use always Flattop if not serious mandatory reason to use some other.

I've used Flattop window.

Just only for thinking what is big and what is small.
For just example imagined situation where all is pure resistive.
INPUT DATA
Set Source Impedance:50
Set Load Impedance R:45.25
Set Load Impedance J:0

RESULT
Absolute Load Impedance:45.25
Load Reflection Coefficient:0.0499
Load VSWR:1.105
Load Return Loss:26.04dB
Load Mismatch Attenuation:0.011dB

If pure resistive load 33.5 ohm
Absolute Load Impedance:33.5
Load Reflection Coefficient:0.1976
Load VSWR:1.493
Load Return Loss: 14.08dB
Load Mismatch Attenuation:0.173dB

Yes, the load matching isn't ideal - and the scope input is not pure resistive for sure either.
When thinking about it, isn't a VSWR >1.1:1 (10%) per definition nearly 1 dB ripple?
On the other hand, a VSWR of up to 1:1.5 is quite typical for proper 50 ohm inputs on DSOs.
So if I'm not wrong, there is rather the question why your coax measurements look so clean and wobble-free?

I am now more sure that you have some problem in some hardware you use. Scope. Cables. Generator.

I made new test. Parallel test. Using 2 RG316 cables with connectors what also mage perfect coaxial GND and center connection.
(and when I did this test I also find one my matched cable pair was bad so in this test now these cables are not perfectly equal length, nonsense here)

Signals come from SDG6000X. Both channels do same sweep 1MHz - 500MHz. Sweep time 300s.
Signal levels.
Gen Ch2 to scope  CH4 roughly 0dBm. CH4 AC 50ohm.
Gen Ch1 to scope CH1 adjusted so that voltage level @1MHz is equal with CH4. CH1 AC 1Mohm - so really bad "noob style" mismatch when he forget termination.  As can see there is just these waves and very roughly around same level as waves in your signal. My scale is 2dB/div.

In this test I use both voltage bands, I and II. As can see this change and it also change impedance as can see also CH4 small waves after 250-300MHz when voltage band I is in use.

If now look and compare very bad termination situation and good termination it is clear that there is really some big problem in some HW what you use. My upper trace (CH4) is as good signal transfer as can be with my available things here  and lower trace CH1 have 50ohm source with 50ohm cable to scope 1M input. Not as bad as short or open but... still "horrible" and as can see least these waves are in same ballpark with you image.

But it also feels very very mysterious after all you have told... only can know something is failed or other ways wrong in your setup.



Your previous image.





CH1 without 50ohm termination, just 1M input.  RG316. Same kind of waves as in your images.
CH4 50ohm internal load resistor. RG316.
1MHz signal voltage levels matched.
Source. SDG6000X separate channels and both sweeps 1 - 500MHz and swp time 300s.
FFT window Flattop and max-hold.


You Wonder, In my pictures, the signal drawings are smooth.
Well...   When I go on a bus as a gray-headed elderly people here in China, the young peoples get up beautifully and offer a seat place even though I often thank and show that I can stand.
Could it be that the test devices are behaving somehow proportionally same and behaving beautifully front of me.

[Johnny B Good]

 That last observation just seems to be a Chinese version of "Sod's Law". 

https://en.wikipedia.org/wiki/Sod%27s_law