Oscilloscope Dilemma

[Fungus]

With a modern digital scope, you get an accurate frequency counter and time interval meter built in, and on-screen display of time and voltage values without having to count divisions on the screen.

And FFT to see the signal in the frequency domain - very useful for radio work.

[markone]

Not quite true. Tektronix' TDS210 / TDS220 where real game changers that put 1Gs/s realtime sampling scopes on the desk of electronics engineers as a true replacement of analog oscilloscopes.

Game Boy, this is the name that we gave to those scopes in the lab at the time, the awfullest screen ever seen in my whole life

And designed to a low price point as well. The whole board is like the size of 2 credit cards. I have not seen any of the low cost Chinese brands achieve such high levels of integration.

For the simple fact that it was over engineering, two credit cards inside a much bigger case, no need to make use of aerospace tech for a bench scope.

China world is still in learning phase, i.e. Rigol has already developed several Asics, i would expect Siglent to do the same soon or later and i can see lot of chip designer / maker take life in the far east, not to mention the "queue" outside the TSMC door ...

My Lab is almost equipped with instrumentation from Rigol (DS1054Z and soon HDO1074), Siglent (SDG2042X & SSA3021X), GW Instek (GDS-2072E), if you ask me would say thanks God, China came in this sector otherwise things would have been a lot different, with some rusty & smelly anchor boat on the desk.

Agreed. Tektronix still knows how to make good oscilloscopes. And they are not the only ones who have been milking old architectures.

They are still milking, it's enough to see how many active 8bit DSO they do have (shamlessy) in catalog with 20kpts @ over 1K USD/EURO.

[rstofer]

I have an old Tektronix 485 (350 MHz) dual channel scope and it works well.  You can sometimes find them on eBay for a couple of hundred bucks.  I did...

OTOH, even though it has a lot more BW than my unlocked Rigol DS1054Z, it is retired.  There is simply nothing about an old analog that can compete with a DSO.  Measurements, decoding, triggering, FFT (in some cases), Bode' Diagrams (in some cases) and so on.

I bought the Rigol because the Siglent wasn't available.  Today I would buy the Siglent SDS1104X-E, unlock it to 200 MHz and call it a day.  I need 4 channels more than I need BW.

Really, you're into analysis paralysis and it can last forever so I'll give you a hint:

Buy the Siglent SDS1202X-E for $399 and call it even.  You will be happy you did!

https://siglentna.com/digital-oscilloscopes/sds1000x-e-series-super-phosphor-oscilloscopes/

That's a heck of a price for a very nice scope.  Download the User Manual and see what you will get!

Skip the X-U series, they have been designed down to a price point.

That said, I don't know how you plan to see much of a 400 MHz signal on a 200 MHz scope but I suppose something will show up on the screen.  I don't dabble in RF so I have no idea how it will turn out.

[nctnico]

My Lab is almost equipped with instrumentation from Rigol (DS1054Z and soon HDO1074), Siglent (SDG2042X & SSA3021X), GW Instek (GDS-2072E), if you ask me would say thanks God, China came in this sector otherwise things would have been a lot different, with some rusty & smelly anchor boat on the desk.

Not really. There where quite a number of people (including me) and companies working on low cost oscilloscopes before the Chinese came in and undercut those people / companies. Either way we would have had lower cost test equipment without the Chinese. Take the Welec W2000A as an example (that never got finished).

[themadhippy]

Skip the X-U series, they have been designed down to a price point.

for the price its still a very capable scope,for some  of us the budget is very limited.

There is simply nothing about an old analog that can compete with a DSO. 

my old  analogue scope displayed x/y artwork much better than my new digital one

[tautech]

Buy the Siglent SDS1202X-E for $399 and call it even.  You will be happy you did!

https://siglentna.com/digital-oscilloscopes/sds1000x-e-series-super-phosphor-oscilloscopes/

FYI, these are on special until years end:
https://siglentna.com/product/sds1202x-e/

[james_s]

With a modern digital scope, you get an accurate frequency counter and time interval meter built in, and on-screen display of time and voltage values without having to count divisions on the screen.

There are analog scopes that offer that too, although they are less common.

[2N3055]

I don't know if SDS2202X-E is on sale anywhere, but that one goes to 350 Mhz (more in reality) and would show 432 MHz without problem...

[tautech]

I don't know if SDS2202X-E is on sale anywhere, but that one goes to 350 Mhz (more in reality) and would show 432 MHz without problem...

I did some screenshots for a thread here a while back with SDS2202X-E and pushed to well beyond 400 MHz which is getting close to it aliasing but its frequency counter went a few 100 MHz further if the amplitude was sufficient on the display despite it being well past rated BW......missing my SSG3021X, now sold. 

Edit to add.
Thread link where we tested the BW of a few DSO's and SDS2202X-E here.
https://www.eevblog.com/forum/testgear/dso-bandwidth-test-sds1104x-e-dsox1102g-to1104-gds1054b/msg2391324/#msg2391324
Aliasing was not obvious to me however rf-loop put me straight.

[Andy2]

I've only speed-read this thread, so sorry if someone has already made this point. If you are a radio amateur (like me), at some point you will probably use your 'scope to view your outgoing signal. This is a trivial task for an analogue scope, but for DSO's it can be a difficult job and it will start aliasing and you'll not get the familiar 'AM envelope' display. The way around this is to get one with plenty of screen memory (ie it can store lots of 'points'). My Rigol has 14 million points and will show a good AM display on 160m (1.8 MHz) and up to maybe 10 MHz or so, but beyond this it looks worse than my old Tek analogue.  So when you are inspecting the manufacturer's boast-sheets, look for the 'Memory Depth' spec and aim high. Hope this helps, it had me befuddled for ages....

I am not sure what you are doing wrong, but my DSO has no trouble at all displaying a typical AM signal. Here is a photo of a 29 MHz (10 meters) signal modulated by 1 kHz at 50%:

(Attachment Link)

That's interesting, Baldurn. How many points is the trace memory? As I said, things improve with more memory. My Rigol has 14 M points, which was a decent amount  when I bought it, but these days scopes tend to have a lot more.

[BassClef]

Big thanks to all who chimed in. The "analysis paralysis" didn't last long. I was strongly leaning towards the Siglent SDS 1202X-E. You folks gave me the nudge over the cliff... er, edge. I pulled the (pardon the pun) trigger this evening.

432 MHz wasn't really a priority, as all I do there is local FM or DMR stuff. My main thing is HF. Followed by using the scope as a tool to fix a few things as the need arises, and learn more about what's going on in those wires. Back when I was a kid, I was aligning and adjusting radios with nothing more than a VOM and a 100 kHz crystal calibrator.  So this is going to feel like the electronic equivalent of a luxury hotel.

--Peter

[nctnico]

Agreed. Tektronix still knows how to make good oscilloscopes. And they are not the only ones who have been milking old architectures.

They are still milking, it's enough to see how many active 8bit DSO they do have (shamlessy) in catalog with 20kpts @ over 1K USD/EURO.

More likely they keep these for customers replacing / extending their existing equipment in places where it is too expensive to rewrite manuals / modify automated test setups / train people.

[baldurn]

I am not sure what you are doing wrong, but my DSO has no trouble at all displaying a typical AM signal. Here is a photo of a 29 MHz (10 meters) signal modulated by 1 kHz at 50%:

That's interesting, Baldurn. How many points is the trace memory? As I said, things improve with more memory. My Rigol has 14 M points, which was a decent amount  when I bought it, but these days scopes tend to have a lot more.

That was actually on the picture, but unfortunately too blurry to read. The scope is an Siglent SDS 2304X which is an older version of the scope the original poster is considering. The scope has memory for 140 Mpts but is only using 14 Mpts at 2 GS/s in the picture.

However, someone please correct me if I am wrong, I do believe it is actually the sample rate at viewed timebase that is important here. Because we want to see a 1 kHz amplitude overlay waveform, we need the timebase to be in the range of 1 ms per division. I had 0.5 ms per division in the picture. At the same time we need to capture the RF signal at 29 MHz which means at least 58 MS/s. My scope has 14 divisions and so would need 14 ms of data at 58 MS/s. That calculates to a little less than 1 Mpts.

The scope is doing much more than 1 Mpts and 58 MS/s which just means it has data so I can zoom or I could go much higher up in frequency without trouble.

[rstofer]

Skip the X-U series, they have been designed down to a price point.

for the price its still a very capable scope,for some  of us the budget is very limited.

Then make a spreadsheet of features on both series -E and -U.  Then compare what you can't do with the -U now or forever into the future and see if the savings are worthwhile.

When you can get a nice 2 channel 200 MHz scope with an abundance of features for $399, it's pretty hard to complain about the price.  SDS1202X-E  - Remember, we got along with 2 channel scopes for a very long time.

The SDS1104X-U matches the price at $399; you give up 200 MHz for 100 MHz but gain 2 more channels.

It would cost another $100 over the -U to get the SDS1104X-E which is only 100 MHz but can be unlocked to 200 MHz (details elsewhere, check them out, don't rely on my memory).  This is the way I would go but that's just me.  I want 4 channels above everything else.  200 MHz would be nice but I still have that old 350 MHz Tek 485 for the higher BW requirements.

It gets to be a bit confusing...

[tautech]

Skip the X-U series, they have been designed down to a price point.

for the price its still a very capable scope,for some  of us the budget is very limited.

Then make a spreadsheet of features on both series -E and -U.  Then compare what you can't do with the -U now or forever into the future and see if the savings are worthwhile.

When you can get a nice 2 channel 200 MHz scope with an abundance of features for $399, it's pretty hard to complain about the price.  SDS1202X-E  - Remember, we got along with 2 channel scopes for a very long time.

The SDS1104X-U matches the price at $399; you give up 200 MHz for 100 MHz but gain 2 more channels.

It would cost another $100 over the -U to get the SDS1104X-E which is only 100 MHz but can be unlocked to 200 MHz (details elsewhere, check them out, don't rely on my memory).  This is the way I would go but that's just me.  I want 4 channels above everything else.  200 MHz would be nice but I still have that old 350 MHz Tek 485 for the higher BW requirements.

It gets to be a bit confusing...

Until you have this image as attached, prepared by rf-loop.

Still there are some differences in capability and features between SDS1202X-E, SDS11/1204X-E and SDS1104X-U but it's all in their datasheets and if not in user manuals.
4ch X-E's are the premium range in these 3 and to get close the 2ch 2 GSa SDS2000X-E range shares some features with the 4ch X-E's but also has 1M/50 Ohm inputs none of these 1000 series models have.

[themadhippy]

Then make a spreadsheet of features on both series -E and -U. 

no need,think its on siglents own website, theres a comparison table

When you can get a nice 2 channel 200 MHz scope with an abundance of features for $399, it's pretty hard to complain about the price.  SDS1202X-E  - Remember, we got along with 2 channel scopes for a very long time.
The SDS1104X-U matches the price at $399; you give up 200 MHz for 100 MHz but gain 2 more channels.

horses for courses,i dont have  much use for 100Mhz,let alone 200,4 channels is much more use for the sort of stuff i tinker with

It would cost another $100 over the -U to get the SDS1104X--E

About £70 difference in the uk,So are you going to give me the extra £70? or do i put off buying for another 6 months until i can afford one,meanwhile the money has slowly been nibbled at and/or the prices have gone up

[tautech]

Then make a spreadsheet of features on both series -E and -U.

no need,think its on siglents own website, theres a comparison table

If it's this one it's next to useless:
https://siglentna.com/application-note/differences_xe_vs_xu/

The one I posted done by rf-loop is what it instead should be with the additional features added into 4ch X-E's recently.

[Wallace Gasiewicz]

I use an Agilent 54820 scope that has the HD upgraded. 2 Gs/Sec and 500 MHz Bandwith. I think this would be categorized as an "older" DSO.
I bought the Agilent because there was a bit of confusion about the newer Rigol etc... scopes quality a few years back. Reading now about scopes, it seems that the Sigilent Scopes are really good and probably better than my Agilent.
It took me a while to learn to use this digital scope. Bought it used for $ 400. It is physically much larger than the new scopes. I think it is a keeper.
My only real disappointment is that it appears to get confused and requires that I use the autoscope feature to get a trace sometimes. Mostly happens at low KHz audio freq.
It would also be nice to have an input for an external freq standard. I am of the current opinion that scopes in general do not have a really good internal freq standard. Maybe I am wrong. I cannot find a way to "calibrate" the internal 100 MHz freq standard, if someone knows, let me know.
It is a little difficult to learn the features, it takes time. Things like X vs Y take a few steps to set up. But any computer thing with features is a bit more complicated to set up for these features.
I still cannot use it to get a decent trace for a transistor curve tracer, which is a snap with my analog scopes.

The FFT is OK but I use a spectrum analyzer for my measurements, so I really don't know how to use it to it's full potential.
I think that spectrum analyzers in general are better than most scopes with FFT, again I could be wrong about modern scopes. Bdunham's trace looks very nice. Maybe I should poke around a little and see what I can get on the screen.
My backup CRT scope is an HP 1980, 100 MHz, cost $ 18,000 with freq counter in 1985. (bought it used and broken, now works fine)
It has digital storage for waveforms that is very crude compared to modern DSO. also has autoscope function. And measures rise time and other stuff.
By the way I also have a Tek 475, 200 MHz scope that I can part with, if anyone is interested.
If anyone buys a used scope, please make sure you get the proper probes for the particular scope.
Since using the Agilent, I find it hard to use anything without readouts on the screen and the color screen makes a world of difference. Old CRT scope readouts are not as easy to see at a glance.
The big color screens are a blessing for us who have old eyes. Also gotta love the magenta!
The video output is nice for using a larger computer monitor also.
It took a while but now I find it hard to go back to CRT scopes.

[nctnico]

If anyone buys a used scope, please make sure you get the proper probes for the particular scope.

Why? High-Z Oscilloscope probes are pretty much standard and basically a dime-a-dozen. The only thing to watch out for is whether the capacitance adjustment range covers the specified capacitance of the oscilloscope input. Beyond -say- 100MHz feeding a signal directly into an oscilloscope using a 50 Ohm coax is a good idea if you want to have something on screen that resembles the actual signal.

When I buy a used scope, the High-Z probes that come with it go into the bin unless they are new in a sealed package. Nobody gives away perfect probes with a used scope. IMHO used probes are like used underwear. Really special, expensive probes are a different thing though but these are usually well cared for and nobody has driven an office chair over the cable or let the dog chew them up.

[james_s]

When I buy a used scope, the High-Z probes that come with it go into the bin unless they are new in a sealed package. Nobody gives away perfect probes with a used scope. IMHO used probes are like used underwear. Really special, expensive probes are a different thing though but these are usually well cared for and nobody has driven an office chair over the cable or let the dog chew them up.

It's hardly like used underwear, nobody has had them in their private bits, hopefully. I have a few very old probes that still work fine, if they're damaged or defective I'll replace them. Most of the scopes I've owned are Tek though and rely on the coding pin to display the correct order of magnitude, most aftermarket probes lack that.

[nctnico]

I have been 'bitten' by probes that had contact issues and bad cables. YMMV but in general a bunch of known good Testec probes are cheap enough not to take a gamble. Personally I don't care much about the probe ID pin. In some cases I use direct cable connections anyway and have to setup the attenuation manually which is easy to do on the oscilloscopes I have.

[alm]

Beyond -say- 100MHz feeding a signal directly into an oscilloscope using a 50 Ohm coax is a good idea if you want to have something on screen that resembles the actual signal.

That depends on the output impedance of the node you are testing over frequency. Not every circuit will be happy with a 50 or 500 Ohm load at DC. 10 MOhm at DC going down to say 160 Ohm at 100 MHz might be better than a constant 50 Ohm. Of course active probes perform even better, but at a price.

I would say that in addition the physical form factor, and accessories to hook it up to things like grabbers, are a major factor. I don't think the old 5mm probe tip size with an alligator clip for ground works very well on SMT boards unless you have test points for all the nodes you want to measure. I quite like narrow spring-loaded tips like the Lecroy PP005 probes. Are those available on the cheap Ali-special dime-a-dozen probes yet?

IMHO used probes are like used underwear.

I worry about what you do with your probes. I have bought quite a few used probes over the years, including maybe two dozen common 1x and 10x probes, and all but two have worked well. One was a cheapy with a dodgy switch, and the other had a bad cable. Used active probes are a different story, though.

[rstofer]

Then make a spreadsheet of features on both series -E and -U.

no need,think its on siglents own website, theres a comparison table

When you can get a nice 2 channel 200 MHz scope with an abundance of features for $399, it's pretty hard to complain about the price.  SDS1202X-E  - Remember, we got along with 2 channel scopes for a very long time.
The SDS1104X-U matches the price at $399; you give up 200 MHz for 100 MHz but gain 2 more channels.

horses for courses,i dont have  much use for 100Mhz,let alone 200,4 channels is much more use for the sort of stuff i tinker with

It would cost another $100 over the -U to get the SDS1104X--E

About £70 difference in the uk,So are you going to give me the extra £70? or do i put off buying for another 6 months until i can afford one,meanwhile the money has slowly been nibbled at and/or the prices have gone up

Alas, I live on a pitiful fixed income...

Here's the thing, nobody around here is going to recommend a scope they wouldn't want to use.  There is always something cheaper but likely isn't as exhaustively reviewed simply because no serious user would consider buying it.

Any of the 3 Siglents will do the job and they have been reviewed extensively right here on EEVblog.  Just use the search feature.  The Rigol is also a very nice scope (after unlocking) but it's getting old.  The Siglents are the new shiny things.

If you want to do a more thorough reading, try https://www.eevblog.com/forum/beginners/what-an-oscilloscope-recommended-for-a-woman-passionate-about-electronics/ starting around Reply 150 where Charlotte buys the SDS1202X-E and we get serious about working through the User Manual.  Late in the thread it gets very interesting.  The first 149 replies are just a prelude but could be worth reading as they are more like this thread, comparing features and price.

Many of the same concerns are discussed in great detail.

If price is the primary concern, above all else, drive a stake in the ground and state what it is.  Call it an absolute max, no waffling "well, if I can get better for a little bit more...", nope, it's the MAX price.  You can always get more features "for a little bit more...".  You may not get as many replies if the amount precludes worthwhile scopes but that's the way it goes.

Again, we're not likely to recommend something we wouldn't want to use ourselves.

[tautech]

I would say that in addition the physical form factor, and accessories to hook it up to things like grabbers, are a major factor. I don't think the old 5mm probe tip size with an alligator clip for ground works very well on SMT boards unless you have test points for all the nodes you want to measure. I quite like narrow spring-loaded tips like the Lecroy PP005 probes. Are those available on the cheap Ali-special dime-a-dozen probes yet?

Some manufacturers recognise this and offer relatively cheap products to make our lives easier:

http://www.pintek.com.tw/customer/pintek/product/CP-3501R-PRO-b.jpg

At the end of the day it comes down to the Ahole that designs/lays out a PCB to give us the required test points to undertake rework/repair. However this additional effort/cost needs be balanced against providing replacement boards at reasonable cost. Flip the coin.