Siglent SDS1104X-E vs. Rigol DS1054Z Advice?

[JS]

I'm after a AWG and a LAB PSU now, and of course considering Siglent, if you know a distributor with reasonable international shipments let me know!
JS

Several but it's very bad form to sell into another authorized distributors patch, period.

Good luck should you need warranty or specialized support from the Asian mum and dad backstreet shops, just good luck !

In my experience a local distributor can be unhelpful as well. It is a bit of a hit & miss and you don't know until it is too late. Usually I use google with the company name and the world 'problem' to see if a shop is any good or not before buying. Last year I bought something from Italy instead of the local NL distributor because the NL distributor seemes to have very bad customer services (sell broken / DOA items and then let customers wait weeks or months for repairs).

  There's the answer I'd gave you, except electro-tools failed to answer my mails so I didn't even considered them. I wouldn't trust a seller who doesn't reply to quotes request to have a good support later. I totally forgotten about them.
  Temtec did answered my mail quickly but they weren't importing any new products, just selling whatever they had left. I remember a 3458A for $5k which as you can imagine is a great deal here but I'm not in the market for that right now, but it catch my eye.
  Sorry, but buying here makes no sense, support isn't great and prices either, this isn't just for your brand, don't take it personally, but for me, rigging up the lab for myself to be able to do what I know to do best I can't justify it. A bigger company, which needs someone to call and blame if something doesn't work, and have some benefits from buying locally, a university which needs to buy locally and doesn't have a choice, and a few other examples, will work with them, but for individuals setting up persona labs to be able to work I don't think so.

  About FFT length, as I had said many times, I pick the data from the PC and run the processing there, I can run a FFT with the full memory depth, so for me being able to capture 24M points with 1GSa/s with some advanced trigger options is the most important feature. I wouldn't go with cheaper with low memory depth, but more expensive with more processing power wouldn't make a big difference for me. Math function limitations are in all the scopes and a bit more in some particular case doesn't make or brake the deal for me. The refresh sending the data to the PC won't be as fast but I can live with that, the times for me catching a rare event aren't that often. Once I had a fast, once in a few month event but I wouldn't trust the Siglent to catch it either, troubleshooting that wasn't easy... Also, 200kwf/s using deep memory isn't a thing anyway.

  I wouldn't even bother in FW updates being easier or harder, is something you do once every few moth at most and be done with it. The key building blocks, as front end, ADC, etc. does bother me, the frond end is a win for Siglent as can go to a lower range better, you do have two ADCs but the same specs as the one in the Rigol, so for one channel are the same, where I would want to characterize a signal. I use as many channels as I have while debugging or troubleshooting but there usually I don't use the ADC to it's full spec. One strong point I see in the Rigol is the 24M for a single channel, I understand the Siglent goes up to 14M in a single channel, and in two channels they are closer 12M against 14M. Again, in this regard, to go to the PC, the Rigol wins. There are so many factors to consider, there is no a clear winner considering the price difference. Siglent is a more modern option with a lot of wins, even more as stand alone.
  Before someone start to attack me why didn't I went with a PC based one to start with, I do use a lot of functionality while using as stand alone, but not while pushing it to the limits. The Rigol makes a great fit for me so far, some people won't be so happy going to the PC and want everything in the one device.
 
JS

[tautech]

<snipped>
  About FFT length, as I had said many times, I pick the data from the PC and run the processing there, I can run a FFT with the full memory depth, so for me being able to capture 24M points with 1GSa/s with some advanced trigger options is the most important feature. I wouldn't go with cheaper with low memory depth, but more expensive with more processing power wouldn't make a big difference for me. Math function limitations are in all the scopes and a bit more in some particular case doesn't make or brake the deal for me. The refresh sending the data to the PC won't be as fast but I can live with that, the times for me catching a rare event aren't that often. Once I had a fast, once in a few month event but I wouldn't trust the Siglent to catch it either, troubleshooting that wasn't easy... Also, 200kwf/s using deep memory isn't a thing anyway.

WRT capturing data for PC processing.
I'll leave this for those that aren't aware of it:
https://www.eevblog.com/forum/testgear/open-source-lxi-tools-and-liblxi-v1-0-released-for-gnulinux/msg1394796/#msg1394796

[bitseeker]

Josh,

Since you work on tube amps, for the $140+ difference between the scope you have and some of the others you're contemplating, you could get a high-voltage differential probe. Although most of the time you take measurements in tube amps relative to chassis ground, with a differential probe you can measure across any two points. I don't know how often that might be useful to you, but thought I'd mention it.

Right now, I see one of the vendors on eBay selling MicSig 100MHz 1400V differential probes has them for $120:

https://www.ebay.com/sch/i.html?_from=R40&_nkw=micsig+dp10013&_sacat=0&_sop=15


For audio work, the higher bit depth of the AD2 (14 bits) or dedicated USB-based equipment for audio analysis (16, 24 or more bits per sample) may be better suited. Of course, it also depends to what level of precision and granularity you're doing things like measuring distortion, S/N ratios, etc.

However, even an old Rigol DS1052E and its FFT can enable you to see if your amp has less than a certain percentage of distortion by sending through a dual-tone test signal, one of which is at an amplitude that represents your desired distortion threshold and at a frequency that doesn't collide with the harmonics of the primary signal, as a visual marker.

As with any technical buying decision, knowing your true requirements will help a lot to sift through all the possibilities. Otherwise, analysis paralysis is always lurking behind every spec sheet.

[KungFuJosh]

Josh,

Since you work on tube amps, for the $140+ difference between the scope you have and some of the others you're contemplating, you could get a high-voltage differential probe. Although most of the time you take measurements in tube amps relative to chassis ground, with a differential probe you can measure across any two points. I don't know how often that might be useful to you, but thought I'd mention it.

Right now, I see one of the vendors on eBay selling MicSig 100MHz 1400V differential probes has them for $120:

https://www.ebay.com/sch/i.html?_from=R40&_nkw=micsig+dp10013&_sacat=0&_sop=15


For audio work, the higher bit depth of the AD2 (14 bits) or dedicated USB-based equipment for audio analysis (16, 24 or more bits per sample) may be better suited. Of course, it also depends to what level of precision and granularity you're doing things like measuring distortion, S/N ratios, etc.

However, even an old Rigol DS1052E and its FFT can enable you to see if your amp has less than a certain percentage of distortion by sending through a dual-tone test signal, one of which is at an amplitude that represents your desired distortion threshold and at a frequency that doesn't collide with the harmonics of the primary signal, as a visual marker.

As with any technical buying decision, knowing your true requirements will help a lot to sift through all the possibilities. Otherwise, analysis paralysis is always lurking behind every spec sheet.

That is a great price for those probes. I have the same probes in a wishlist on Amazon for $159.

Tube guitar amps is what I work on, and they always need distortion, even when they're supposed to be "clean." I'm not really interested in getting rid of distortion as much as looking at it to see characteristics and shape. The amps I make sound great, this is more for curiosity and self-education, with the potential to eventually shape that distortion differently. But more curiosity than anything else WRT to the FFT stuff.

Actually, my biggest headache right now is chasing down a parasitic oscillation in a reverb circuit.  Whichever scope can make that easier wins instantly.

lol, analysis paralysis. That's funny, and accurate.

WRT capturing data for PC processing.
I'll leave this for those that aren't aware of it:
https://www.eevblog.com/forum/testgear/open-source-lxi-tools-and-liblxi-v1-0-released-for-gnulinux/msg1394796/#msg1394796

That looks like another (small) plus for the Siglent. It comes back to exactly what I said about the Siglent being the newer computer. So there. But will there ever be an official LXI release for this? Is this 100% reliant on the github project? If that's the case then that's not a long-term, reliable solution.

[KungFuJosh]

Although most of the time you take measurements in tube amps relative to chassis ground, with a differential probe you can measure across any two points. I don't know how often that might be useful to you, but thought I'd mention it.

I do always measure to ground. This is something I don't understand being so new to scopes. Why would I want to measure across two points on a scope besides ground?

[wpwrak]

Is this 100% reliant on the github project? If that's the case then that's not a long-term, reliable solution.

It's often the other way around: Open Source can be extremely durable, since even when the original author(s) disappears or loses interest, you still have the code and can carry on if you want. "You" doesn't necessarily have to be you as a person, who may or may not have the motivation, skills, and resources to take over maintaining some project, but is the whole user community.

Regarding the Siglent, you may also want to consider this thread:
https://www.eevblog.com/forum/projects/open-source-instrument-firmware-project-104048/

[nctnico]

Although most of the time you take measurements in tube amps relative to chassis ground, with a differential probe you can measure across any two points. I don't know how often that might be useful to you, but thought I'd mention it.

I do always measure to ground. This is something I don't understand being so new to scopes. Why would I want to measure across two points on a scope besides ground?

Think about a half-bridge driver where you want to measure the voltage on the base/gate versus the emitter/source.

[KungFuJosh]

Although most of the time you take measurements in tube amps relative to chassis ground, with a differential probe you can measure across any two points. I don't know how often that might be useful to you, but thought I'd mention it.

I do always measure to ground. This is something I don't understand being so new to scopes. Why would I want to measure across two points on a scope besides ground?

Think about a half-bridge driver where you want to measure the voltage on the base/gate versus the emitter/source.

Not in a simple tube guitar amp circuit. I could see that being useful in other circuits, but for me the question is specific to simple vintage tube guitar amps. Where would I benefit in a simple tube guitar amp circuit?

[KungFuJosh]

To be clear, this is the type of guitar amp circuit I build, and yes, I love heat-shrink tubing.

[tautech]

WRT capturing data for PC processing.
I'll leave this for those that aren't aware of it:
https://www.eevblog.com/forum/testgear/open-source-lxi-tools-and-liblxi-v1-0-released-for-gnulinux/msg1394796/#msg1394796

That looks like another (small) plus for the Siglent. It comes back to exactly what I said about the Siglent being the newer computer. So there. But will there ever be an official LXI release for this? Is this 100% reliant on the github project? If that's the case then that's not a long-term, reliable solution.

The point is, the LXI/LAN speed.
With the existing SCPI command set and the inbuilt webserver that includes the command set panel you can yank info from LAN port in close to real-time, or via the USB WiFi for that matter.

[KungFuJosh]

The point is, the LXI/LAN speed.
With the existing SCPI command set and the inbuilt webserver that includes the command set panel you can yank info from LAN port in close to real-time, or via the USB WiFi for that matter.

I get that, that's why it gets the point. The question is the real-world application of that in the near future, and especially for me personally. How useful will that point be?

Rigol Pros:
Physical side buttons
UI simplicity
Firmware updates probably less buggy.
Laziness *(already on my bench)
Price

Siglent Pros:
hackable to 200Mhz
Higher multi-channel sampling
1M FFT
No side buttons = more screen space for waveforms
LXI/LAN speed
400VDC input (vs 300VDC) (yes, I know it doesn't matter, but a plus is a plus and mistakes happen)
Optional extras that aren't worth the money especially thanks to toys like the AD2

There's also software spectrum analyzers like spek.cc or friture.org that would probably save me the need for an AD2 if I ever want to bother with that...though the AD2 seems fun and has an AWG.

[Fungus]

the AD2 seems fun and has an AWG.

An AWG AND a 14-bit oscilloscope in the same device. This combination allows you to generate frequency sweeps and plot the frequency response/harmonic distortion of things like amplifiers.

[KungFuJosh]

An AWG AND a 14-bit oscilloscope in the same device. This combination allows you to generate frequency sweeps and plot the frequency response/harmonic distortion of things like amplifiers.

Oh, I dig it, but for me it's a toy. I may buy it eventually, but I'm well aware of what it is when I do (a very nice toy). It, and the open-source SA software both negate the ridiculously over-priced add-ons for the Siglent.

  My daughter made me click that.

[Fungus]

Oh, I dig it, but for me it's a toy.

A "toy" that properly set up (ie. with suitable probes) can probably help you do your job better then either the Rigol or the Siglent.

It looks like they've been building up a nice catalog of accessories since I last checked it out, eg. take a look at this bundle. It's not cheap, but very tempting. The BNC adapter gives you two-channel oscilloscope and two-channel signal generator.



Price is relative though - that entire bundle costs less than just the logic analyzer upgrade for the Siglent.

Oscilloscope bandwidth aside, it's a lot of capability for the money. I wonder how electrically robust it is.

Edit: They claim "robustness to withstand student use..."

[bitseeker]

To be clear, this is the type of guitar amp circuit I build, and yes, I love heat-shrink tubing.

Thanks for the example. It's helpful to understand what you're working on. It looks like you already have more than enough for your area of interest. I wouldn't look to spend any more. Use what you have, then you'll better know if and when you need more and what will fulfill that need.

Heat shrink is a good thing. I've seen some mighty scary point-to-point constructions — it makes you wonder how the thing doesn't short out just breathing too hard.

[KungFuJosh]

A "toy" that properly set up (ie. with suitable probes) can probably help you do your job better then either the Rigol or the Siglent.

It looks like they've been building up a nice catalog of accessories since I last checked it out, eg. take a look at this bundle. It's not cheap, but very tempting. The BNC adapter gives you two-channel oscilloscope and two-channel signal generator.

Price is relative though - that entire bundle costs less than just the logic analyzer upgrade for the Siglent.

Oscilloscope bandwidth aside, it's a lot of capability for the money. I wonder how electrically robust it is.

Edit: They claim "robustness to withstand student use..."

I want a bench scope, and that isn't, so it's a no-go for general use as I have no desire to be stuck to a computer. I don't think it's expensive at all, and I actually already had that bundle bookmarked in case I do decide to buy that toy. A reliable scope is what I need, and for actual scopin' I don't think I would call that more reliable than either of the other models. It's also not USB isolated, so I would need to add that to the mix too.

Thanks for the example. It's helpful to understand what you're working on. It looks like you already have more than enough for your area of interest. I wouldn't look to spend any more. Use what you have, then you'll better know if and when you need more and what will fulfill that need.

Heat shrink is a good thing. I've seen some mighty scary point-to-point constructions — it makes you wonder how the thing doesn't short out just breathing too hard.

Thanks, as I mentioned, I do believe they're both overkill. The thing that irks me is the "older tech" repeating in my head makes me feel unsatisfied with the Rigol - as silly as that is. The better processing and networking potential on the Siglent is very tempting...but I could find myself wanting to replace either one of these in a year or two, so  ...

I hear you on the heat shrink, I've seen some serious nightmares. Some people's sloppiness can be bad enough, remove the board on a PTP build and it can get scary.

[Fungus]

I want a bench scope, and that isn't, so it's a no-go for general use

Yeah, I know what you mean. I'd want a really fancy probe before I attach one to a tube amp.

[KungFuJosh]

Yeah, I know what you mean. I'd want a really fancy probe before I attach one to a tube amp.

Are you being sarcastic or do you mean because of the high voltage?

[Fungus]

High voltage.

[KungFuJosh]

High voltage.

Most of my circuits peak around 350VDC. With my current probes (Pico TA131), the x10 covers 600V, so I think I'm okay there. Is there any reason I should be concerned in that scenario?

I'm planning on getting some fixed x100 Cal Test CT2707 probes that do 1200V.

................and I also have some differential probes bookmarked. I really don't want to bother with them though if they're not going benefit me in my simple(ish) tube circuits.

[tautech]

High voltage.

Most of my circuits peak around 350VDC. With my current probes (Pico TA131), the x10 covers 600V, so I think I'm okay there. Is there any reason I should be concerned in that scenario?

I'm planning on getting some fixed x100 Cal Test CT2707 probes that do 1200V.

................and I also have some differential probes bookmarked. I really don't want to bother with them though if they're not going benefit me in my simple(ish) tube circuits.

There will be times when you can't place a probe reference (gnd) lead where you want so don't discount needing a differential probe at some stage.

[KungFuJosh]

There will be times when you can't place a probe reference (gnd) lead where you want so don't discount needing a differential probe at some stage.

I definitely hear that, that's why it's bookmarked. But my builds usually have ground points with good access, and 90% of the testing is probably off the output jack anyway.

I gotta say one other point for the Siglent is the front panel design. I hate the appearance of the Rigol; it looks sloppy by comparison. Aesthetic matters, my day job is design.

[JS]

There will be times when you can't place a probe reference (gnd) lead where you want so don't discount needing a differential probe at some stage.

I definitely hear that, that's why it's bookmarked. But my builds usually have ground points with good access, and 90% of the testing is probably off the output jack anyway.

I gotta say one other point for the Siglent is the front panel design. I hate the appearance of the Rigol; it looks sloppy by comparison. Aesthetic matters, my day job is design.

It's not about having access to a ground, is about needing to know the voltage between two points where neither of them is ground, like if you want to measure the voltage across a cathode and the grid of a cathode follower, to look at the bias while the signal is present. If you probe both in reference to ground which are swinging 20V at 100V from ground and you try to measure that signal for the time you subtracted them both in the math (AKA poors man differential probe) you are left with very little signal, you wanted to look for, as it was a small signal on top of a big common mode signal.

JS

[JS]

<snipped>
  About FFT length, as I had said many times, I pick the data from the PC and run the processing there, I can run a FFT with the full memory depth, so for me being able to capture 24M points with 1GSa/s with some advanced trigger options is the most important feature. I wouldn't go with cheaper with low memory depth, but more expensive with more processing power wouldn't make a big difference for me. Math function limitations are in all the scopes and a bit more in some particular case doesn't make or brake the deal for me. The refresh sending the data to the PC won't be as fast but I can live with that, the times for me catching a rare event aren't that often. Once I had a fast, once in a few month event but I wouldn't trust the Siglent to catch it either, troubleshooting that wasn't easy... Also, 200kwf/s using deep memory isn't a thing anyway.

WRT capturing data for PC processing.
I'll leave this for those that aren't aware of it:
https://www.eevblog.com/forum/testgear/open-source-lxi-tools-and-liblxi-v1-0-released-for-gnulinux/msg1394796/#msg1394796

I tried to post a response to that but tapatalk wouldn't let me do it... not in the last few days.

The main thing on the answer was that seems obvious that more people is working on automated tests, as LXI brings a cheap and easy way even for every home lab to have some gear capable of running on that. Some people is leaving it pass by and I don't know why, because they are likely to be able to learn how and the ROI for that time will be pretty fast as a few measurements go by. Thermal profiles are often logged by hand in some experiments in my university, as the effect is slow enough that with a thermometer, a stop watch, paper and pencil you pretty much have your data 'logger' (sitting in a bench looking at the damn thing for way too long)

JS

[Mr. Scram]

Can a Siglent 100 MHz now be hacked to 200 MHz? I can't seem to find much about that.