Siglent SDS1104X-E In-Depth Review

[TheNewLab]

OK, I have no idea how to capture another's comment and insert into my own post, so here's a cut and past from the previous entry

If you don't really know what you need in a scope, it is going to be impossible for anyone to tell you what is the best bang for the buck scope for you
I know people mean well, I have seen this so many times. From my own experience as more lame than a Newbie, At some point you just want an oscilloscope, you have seen so much YT's Argh Dave and Martin Lortin....My thing has to keep this thing a hobby, so only toys and a decent multimeter..well, a decade later and thousands of YTs I have really lost it.

As I was saying at some point you get to where you just want a scope,(or AWG, or whatever) so what can I afford. What can I justify so without ending up in  a divorce? Then comes what is the "best bang for buck" and various ways we need to rationalize, try to be logical, whatever. Of course I did not know what to use a real oscilloscope for. I still am not clear and i have just bought three entry level units..instead of my trusty ol' $100 Velleman HPS40, or the $125 plus DO Quad 203..

We just watch and learn all the amazing and cool things that such a tool can do. So, best bang for buck..If is can feel I get that, then I do have something to discover, and learn and what I am going to use it for?? no idea..However, I will find something.

Frankly, I would love to find a series of YT videos, on just using a 100MHz 4 channel (plus/minus) being used in the process of doing actual work, not the how to stuff. I want to discover the features and find some device and scrap electronics and follow along and say "HEY, a scope can also do THIS, too??" "cool!"

for me I am at: searching for this type of YT videos:


and I said COOL! I have a recording engineer friend, I wonder if I could nag him to so some mic audio profiles of different mics!

or,

12 minutes and he is pulling out some of his old projects and scoping them! "I can do that! a can find something and scope out that thing to find that issue!"
plus he has promised to do more videos using other scope features on some other gadgets he has.! Whoo Hoo!

I found a satellite set top box on the street. Now I am saying "how do I inject a signal..where do I inject a signal? ..how will I inject a signal? somehow 'cause I want to scope something in there.."

The detailed review of the Siglent SDS 1000x I saw a photo of him using the MSO and he connected a grabber to every pin on an IC. I said, " whoa...you can use logic probes all on one IC?? I thought it was for tracking signals in various parts of a circuit, not I single IC??" "Cool !! where do i have some digital IC and do that...hum..I need some circuit so it is active..hummm"

to all of this in my long winded post. Sometimes it is just the sheer excitement of getting a tool that can do so many things you just do not yet understand. I have no idea what I am going to use it for, I have no idea what options I should look for. I just want something I
1.) can afford without ending up sleeping on the couch.
2.) can make some explanation that sounds really good that I need that particular scope
3.) If has a lot of features so I can have lots of discovery time and hopefully do some amazing things.

That's all.

Sorry, I have run this topic into the ground.
I do mean to dis' the previous post at all. He really means well and in a good way...

only just sometimes us neo-newbies have gotten to a point, that we feel it is time to get something a little more expensive, and all we really want is as many features as we can get. and they are mostly useful features..

I hope I am not offending anyone here. We all mean well, and try to be as helpful as we can.

[BillB]

OK, I have no idea how to capture another's comment and insert into my own post, so here's a cut and past from the previous entry

Frankly, I would love to find a series of YT videos, on just using a 100MHz 4 channel (plus/minus) being used in the process of doing actual work, not the how to stuff. I want to discover the features and find some device and scrap electronics and follow along and say "HEY, a scope can also do THIS, too??" "cool!"

To capture someone else's post, you click on the "quote" link that is to the far right of their post title.  Or once you start composing your message, you can scroll down and click on the "Insert Quote" link to the far right of the post title.

As far as cool scope videos doing actual work, there are tons of them available.  However, for scope usage and many of the topics he covers in general, w2aew https://www.youtube.com/channel/UCiqd3GLTluk2s_IBt7p_LjA creates some great content.  It is concise, clear, and very approachable for all types.  He sets up real samples and does the work.   

[Old Printer]

OK, I have no idea how to capture another's comment and insert into my own post, so here's a cut and past from the previous entry

If you don't really know what you need in a scope, it is going to be impossible for anyone to tell you what is the best bang for the buck scope for you
I know people mean well, I have seen this so many times. From my own experience as more lame than a Newbie, At some point you just want an oscilloscope, you have seen so much YT's Argh Dave and Martin Lortin....My thing has to keep this thing a hobby, so only toys and a decent multimeter..well, a decade later and thousands of YTs I have really lost it.

I did not intentionally leave you an empty answer to your question, but this section of the hobbyist digital scope market has been under a microscope for the last six months or so. Without knowing what, if any, particular tasks a new scope buyer wants to accomplish there is really not much more to be said. I now see you have very recently picked up a 1054Z at better than a third less than the best price I have seen. If it is new, from a legit dealer, then that is in itself a major break through, though I still see the older 1052 at $300 so I am really curious.

[TheNewLab]

apture someone else's post, you click on the "quote" link that is to the far right of their post title.  Or once you start composing your message, you can scroll down and click on the "Insert Quote" link to the far right of the post title.

Thank you so much BillB

[TheNewLab]

-old Printer,
 I sent you a message in more detail.

For everyone else,
The price I quoted was what I ended up paying after Amazon made a screw up, and then they gave me a huge credit.. my net cost after the partial refund was $226. the original purchase price was $349. I have this kind of Karma where suppliers and shippers make mistakes they really shouldn't have. After running a large mail-order company..I guess I can kind of get under people's skin.

I was so hyped with coffee and two days without sleep when I made that post. And so delighted that I had the courage to try a hack and succeed the first time.

[lordvader88]

So can these be hacked to 200MHz ?

I should really get 1 of these anyway. The SDS1204 is too expensive for now, and the 300MHz versions only have 2ch and are even more money.

[tautech]

So can these be hacked to 200MHz ?

Apparently.
https://www.eevblog.com/forum/testgear/sds1104x-e-hack-to-200mhz-and-full-options/

I should really get 1 of these anyway. The SDS1204 is too expensive for now, and the 300MHz versions only have 2ch and are even more money.

There are no 300 MHz versions of SDS1000X or X-E DSO's but the SDS2000X series offers 2 and 4 ch models to 300 MHz. Until the new SDS5000X models are released soon, the current top model is SDS2304X.

[lordvader88]

Ok thanks

Well there's the siglent sds1302CFL that's 300MHz and with 2GS/s, but would cost +1000USD. Hmmm, they only have 24k memory tho, vs 2M for 1104x-e

Now I have my mind set on 4ch but faster than the ds1054z

[lordvader88]

Whats the fastest sec/div this can be set to ? I don't see it mentioned outright

[Performa01]

Well there's the siglent sds1302CFL that's 300MHz and with 2GS/s, but would cost +1000USD. Hmmm, they only have 24k memory tho, vs 2M for 1104x-e

All non X-series scopes from Siglent are outdated by now. The top models of these may have 300MHz bandwidth and a high sample rate, but slow waveform update rate, hence no intensity grading and lack many of the advanced features the newer scope generation offers. I guess the main reason why these old designs are still around is because there's a demand for such (high bandwidth/buck) instruments in some regions of the (mainly non-western) world.

By the way, the SDS1202X-E, SDS1104X-E, SDS1204X-E have a lot more memory than 2Mpts. A total of 28 Mpts (2 x 14 or 4 x 7) in normal use and up to >100Mpts total in History and Sequence recording.

As can be found in the data sheet as well as my review documents, the fastest timebase is 1ns/div.

[lordvader88]

Ok thanks

In terms of NAND type memoroy or modern DDR(x) ram, what do these modern scopes use and how much ?

[Performa01]

I got a request for another noise demonstration, so here it is.

Ch.1: 20MHz BW-Limit, DC, open input
Ch.2: 20MHz BW-Limit, AC, open input
Ch.3: 20MHz BW-Limit, DC, ext. 50Ohm input termination
Ch.4: 20MHz BW-Limit, AC, ext. 50Ohm input termination


SDS1104X-E Noise TD 20MHz

There are no significant differences and RMS noise is about 29µV on average.

If we really need clean traces, we can use Average acquisition mode. This attenuates any modulation, but doesn’t affect the bandwidth. Since it’s HW-accelerated, the waveform update rate is also nearly the same as in normal acquisition mode.

Here’s an example how just 4 times averaging cleans up the traces very nicely:


SDS1104X-E Noise TD 20MHz AVG4


Finally I can demonstrate the low frequency noise floor (which is always the most critical one) by showing an FFT from DC to 500Hz with ~1Hz frequency step, hence about 3-4Hz RBW with the flat top window:


SDS1104X-E Noise FD 500Hz

FFT trace averaging 16 times to get a clear picture as is common for noise measurements.

EDIT: sorry, I've attached the wrong screenshots originally.

[Performa01]

In interleaved mode (full 1GSa/s per channel) the noise is even a tad lower.
Here's another screenshot showing the peak-peak measurement also.


SDS1104X-E Noise TD CH4 20MHz

[Performa01]

I got another request for a 150µV low frequency challenge.

Well, this is certainly going to be a problem for any 8bit scope, even if it has a low noise 500µV/div setting.

This is about a 150µVrms sine at 10Hz – the frequency where noise is strongest due to the low frequency path in the input buffer, which consists of a high value resistor network at the input and some FET OpAmp with considerable 1/f noise. Furthermore, this OpAmp has to make up for the attenuation of the resistive input divider and then feeds its output to the summing node of the split path input buffer via another high value resistor.

Here’s the test as requested, with 3.0 bits Eres, which acts as a lowpass filter and won’t help with the 1/f noise. Yet I’ve turned the memory down to 14kpts in order to achieve some 800Hz cutoff frequency and exclude at least the high frequency noise by this.


SDS1104X-E_Sine_150µV_10ms_Eres3_14kpts

Well, the result is certainly anything but pretty.

We get much better results by using Average instead of Eres – here’s an example for 16 times Averaging and deep 1.4Mpts memory:


SDS1104X-E_Sine_150µV_10ms_Avg16_1.4Mpts

Now we get the full bandwidth (well, limited to 5MHz by the 10MSa/s sample rate) and the signal looks reasonably nice.


Just as a comparison, a true 16bit Picoscope 4262 with HiRes to 20bits (not even necessary) handles this much better, even though its highest sensitivity is just 2mV/div (I’ve used zoom in this screenshot):


Pico_4262_Sine_150µV_20ms_HiRes20_2Mpts

[kahuna0k]

The first implementation of the MSO option has become available with firmware 7.6.1.20 and I’ve received the SLA1016 digital probe about a week ago. Unfortunately the Sbus cable was missing and even though the connectors are identical, an HDMI cable cannot replace it. So I have to wait until I get the original one.

Until then, I’ll show some details of the hardware. First the contents of the box (minus the Sbus cable):

Did you finally got the SBUS cable and did the review of the SLA1016? I've been unable to find it in the thread. I'm thinking about getting it to be able to trigger on analog. I have a Digital Discovery (https://store.digilentinc.com/digital-discovery-portable-usb-logic-analyzer-and-digital-pattern-generator/) from Digilent that goes up to 800MS/s but is digital only. What's your opinon on the SLA1016?

Thanks!

[tmbinc]

Also I'm still really, really curious if there's a Zynq in the SLA1016. Any teardown pics?

[tv84]

Also I'm still really, really curious if there's a Zynq in the SLA1016. Any teardown pics?

I think there is. Because the SLA1016 has a specific .ADS (with specific Product_ID) with code, FPGA, etc.

[Performa01]

The first implementation of the MSO option has become available with firmware 7.6.1.20 and I’ve received the SLA1016 digital probe about a week ago. Unfortunately the Sbus cable was missing and even though the connectors are identical, an HDMI cable cannot replace it. So I have to wait until I get the original one.

Until then, I’ll show some details of the hardware. First the contents of the box (minus the Sbus cable):

Did you finally got the SBUS cable and did the review of the SLA1016? I've been unable to find it in the thread. I'm thinking about getting it to be able to trigger on analog. I have a Digital Discovery (https://store.digilentinc.com/digital-discovery-portable-usb-logic-analyzer-and-digital-pattern-generator/) from Digilent that goes up to 800MS/s but is digital only. What's your opinon on the SLA1016?

Thanks!

Well, I got the Sbus cable, but then I quickly found a couple issues and wanted to wait with the review until everything was fixed.
Time passed by and some of my complaints just couldn't be addressed because of the given architecture (external device with its own intelligence, hence integration into the DSO isn't great). I got busy with other topics and the SLA1016 review got never completed.

However I have posted some of the old material together with some comments here (Reply #1070):

https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg2007983/#msg2007983
 

[kahuna0k]

Thank you! According to the other post, those tests were done 9 months ago, do you know if:

Mixed channel Pattern Trigger is not supported, so it has to be either an analog or digital pattern.

is still true? If that is the case I'll pass, as that's probably the only advantage of the MSO I was looking for. For the rest of the things I prefer a USB LA.

[Performa01]

Thank you! According to the other post, those tests were done 9 months ago, do you know if:

Mixed channel Pattern Trigger is not supported, so it has to be either an analog or digital pattern.

is still true? If that is the case I'll pass, as that's probably the only advantage of the MSO I was looking for. For the rest of the things I prefer a USB LA.

It is still true, because the SLA1016 is an external device that has its own independant acquisition and trigger circuits.
There is a means to synchronize the records so that the time relation between analog and digital is maintained, but with this architecture, the individual trigger circuits cannot have a high speed link with single digit nanosecond accuracy to implement a combined pattern trigger.

While the combined analog/digital pattern trigger isn't available, the other advantages of an MSO over the LA are still there:

- Real time screen display, high maximum waveform update rate
- Time correlation between analog and digital

However, together with the other limitations, I am glad that the concept of having an external device with its own intelligence is limited to the low cost entry level DSOs (X-E series).

[tinhead]

Here comes an in-depth review of the optional SAG1021 AWG.

exzellent review! I got SAG1021 as well, playing now a bit. The XO in my have 16Hz deviation (AWG 10MHz sine compared to rubidium source), bit more as the one you tested, but (again) better than SDS120X-E counter (21Hz deviation).

The XO seems to be mems from TXC, http://txccrystal.com/images/pdf/ta.pdf

SDS seems to using same mems, i wish i could replace them with decent crystek, but 25MHz is unobtainium.
Overclocking to 50MHz is too much, last stable was 37.5MHz (which make no sense to use, x2 scaling would be ok)

[Rerouter]

Out of curiousity. Are you saying there is no interrupt or trigger style status pin from or to the sla module? As the scope must be syncronising its triggers somehow.

So why can siglent not abuse that to get working pattern trigger? Have the scope trigger on the analog pattern match and have the SLA device use that with its digital pattern match. If they both overlap. Then display?

[tinhead]

btw, thanks for the hint with SCPI commands, even if the SAG1021 does not support sweep from interface, one can send over SCPI sweep command, and that works perfect, even with max freq 40MHz: C1:SWWV STATE,ON,TIME,10S,STOP,40MHZ,START,1MHZ

That makes this baby very useful!

[Rerouter]

You can think of the sag1021 as a fully featured function generator without a displaym you also have bust and arbitary modes. And can load custom waveforms for any of the modes.

[MartyMacGyver]

So I've gathered from this thread that the Siglent PP510 is close to the same performance as their PP215 probe through 200MHz.

That said, of the four PP510 probes that came with my SDS1104X-E, all but one seem pretty solid when using the hook - whereas that fourth PP510 has mad noise at the slightest movement of that hook. In fact, the hook-end spins almost like a top versus the others, suggesting very poor frictional contact.

A look inside the hook assembly reveals the metal hook ends in a kind of fork that friction-presses against the inner probe tip when fitted in its housing. A spring and a plastic retainer round out the interior of the hook end - not much going on in there but obviously prone to problems

So I ask: while it has mostly similar performance, is the PP215 constructed any more durably than the PP510? The '215 is certainly more costly (I'm surprised at just how cheap the PP510 is versus even the Owon T5100's I had on my previous scope... those seemed pretty solid to me), but is basically the same (particularly the hook end)?

I do plan to contact Siglent about the problem as it may be extremely cheap to resolve (I would think the hook end is the cheapest part of this $10USD assembly) but it'd be good to know my other options too.