Year 2013 Oscilloscope Choices...

[casper.bang]

Well, considering the Rigol DS1000 series design is now 7-years old, better to compare the UNI-T's current design to Rigol's current design...

Which would be the DS2000 serie? It costs 2-3 times as much here in Europe and is thus *far* beyond the 360-400€ category range the OP started this thread on. I'm sure your skepticism is well founded, but lets compare similar price points since it is the primary parameter above everything else. The Siglent and the UNI-T might well be the best fit in this category, too bad reviews and teardowns of these are so hard to come by.

[marmad]

Which would be the DS2000 serie? It costs 2-3 times as much here in Europe and is thus *far* beyond the 360-400€ category range the OP started this thread on.

Agreed, but I was only referring to the design at that point. Rigol has moved beyond that particular style now, and you have to admit, it's a bit out-of-date (which means, of course, the DS1000 series and the Hantek are both also looking old-in-the-tooth). 

I'm sure your skepticism is well founded, but lets compare similar price points since it is the primary parameter above everything else. The Siglent and the UNI-T might well be the best fit in this category, too bad reviews and teardowns of these are so hard to come by.

Price and 'listed' features are important, to be sure, but when comparing DSOs that are all fairly similar, I'm not sure it's more important than a large community support base or detailed tests of actual capabilities.  Given all that I know or have read about various brands, I think the Siglent is probably a fairly safe bet - but I would definitely make sure I could return the UNI-T if I bought it - and the first thing I would do upon receiving it is to open it up and check the internals. 

[tinhead]

Maybe they did the same with the oscilloscope.  May be they didn't.  The fact is that
we don't actually know.  We can only speculate.  You say that because in a previous
product they gave wrong information then this would be the rule for all the products.

my proposal - buy the UTD 2102 CM R from Reichelt, it is 100eur cheaper than the "white" version,
or ask pinsonne-elektronik.de for UTD2062CM (or better price for UTD2102CM )

I was the first person outside China who purchased Tekway DSO (at that time they sold only ~240 DSOs!)
my first one was even 100% with Chinese only menu, it was this one -> http://www.tekwayins.com/product.asp?ArticleID=7
It was a nice Tektronix TDS1/2 clone, with similar specs, but already 400wfms/s (so 5x "faster" as Tektronix).

I was as well the first person who bought their current model (which is today as well known as he Hantek/Voltcraft etc),
it wasn't even ready at time of purchase (firmware, language files, even hardware got updated between my and retail model).

I did it becasue Tekway was unique (not the very first model i got, but the second - the current one), large screen, 2500 wfm/s
and many usefull things i haven't seen on other Chinese (or as well TEK TDS1/2xxx). Sure, no GPIB, no LAN, crap PC software, no support at all, etc.
But they got a chance from me, and i still remember ppl in other forums (talking only bullshit) talking things like "no, this is not possible, that not as well, this is china crap, etc".
Today there is large community, Tekway/Hantek/Voltcraft sold about 15000 DSOs, today there is support available, etc.

So when you think UNI-T CM is ok for you, or you think to give them a chance, then buy it, make teardown/some pictures and when you able to test, test the wfms/s.

[marmad]

So when you think UNI-T CM is ok for you, or you think to give them a chance, then buy it, make teardown/some pictures and when you able to test, test the wfms/s.

Well sure - I did the same with the Owon SDS here - if he wants to do the work, we would all be happy to see the insides and results of tests done on the UNI-T by someone else, no doubt about it    But, IMO, he should then make sure he can return it.

[hgg]

Marmad, thanks.  I will try to contact him there.

I am not sure if they will accept to send me a unit for test and if I don't like it to send it back...
I am still waiting a reply from Reichelt.

[marmad]

Marmad, thanks.  I will try to contact him there.

I am not sure if they will accept to send me a unit for test and if I don't like it to send it back...
I am still waiting a reply from Reichelt.

In EU - in most countries - there is at least 14-day return policy for anything bought online (Batronix has 30-day return policy); then you just lose shipping costs.

[tinhead]

right, both Reichelt and Pinsonne have to accept return within 14days without any reason. All you have to pay is the shipment - that's all. Reichelt as far i know can even pickup (not from everywhere available), Pinsonne can probably (in case you decide to buy another model) reduce the price (they did it for me once). So whatever shop yuo chose, it should be no big deal/loss when returning.

[hgg]

The Siglent has dual time base!

Looks like that the Siglent is going to be the winner...

[tinhead]

The Siglent has dual time base!

Looks like that the Siglent is going to be the winner...

right, all other DSOs on your potential candidate list (UNI-T, Hantek) have only ALT trigger with single timebase.

[hgg]

So I have finally... made up my mind and I will go for the Siglent oscilloscope.
Either the 70Mhz or the 100Mhz.  I've read that the 70Mhz behaves almost like a 100Mhz one!
Now the only thing that remains is to find a nice price and buy it.  I've already found one seller
in ebay that says its the official representative in Europe, but when I ask him about the firmware
of the oscilloscope, he says that he doesn't know.  Is this possible?

Thank you very much for helping me choosing my first oscilloscope!
I hope I've made the right choice.
Thanks! 

[nctnico]

The Siglent has dual time base!

Looks like that the Siglent is going to be the winner...

Why would you need dual time base on a DSO? Just capture at a high sample rate and then zoom in.

[rf-loop]

I've already found one seller in ebay that says its the official representative in Europe, but when I ask him about the firmware of the oscilloscope, he says that he doesn't know.  Is this possible?

Yes it is possible.

Seller do not know these things if he (or his counterpart) just only receive cartons from other seller or factory and then ifsomeone buy  they ship this carton forward. 

I think some of these sellers are just sellers who sell "nice boxes".  No matter what is inside box, milk,  parfumes, oscilloscopes, shirts or what ever where forward shipping get some money.
(I know who seller you mean and I think he is not this lowest class of sellers who works just as "dropshipper".  Perhaps he also can test and look version if you wait some time?)

But this tell that seller sell just cartons where is equipment inside. He do not check and test units before sell. So it is just  "as is    from factory" what is just littlebit different as surplus "as is" in ebay.

If seller check and test that unit is really ok, he also know this information.




-----------------------------------------------------
Why I test before I sell any unit:

I think it is very sad situation where customer get new fail unit. It is really sad situation.
I do not trust any chinese factory so much that I will sell anything without my own tests for every single equipment.

In my tests I have stopped 99% of "new fail" units. Still there is of course possible that equipment fails in some rare cases.

All these companies have proofed to me that this is important:

example these:
Rigol have teached me that this practice is wise.
Hantek have teached me (lot of) that this practice is wise.
Owon have teached me that this practice is wise.
Siglent have teached me that this practice is wise.
MCH have teached me (lot of and much more) this practice is wise.

With all these exept two, this falure rate what I find before sell is quite low but after then lot of more rare. Yes it need work and also invest for test gears.  There is no free lounges.
But I can walk around and no need run away if I meet buyer afterwards.
 



Btw, FW is one thing but more important is perhaps HW version!

But if know FW, know also HW becouse it is same information screen in oscilloscope what tell all.
It tell how many times scope is booted, FW, HW, Model and serial number.

Then, if not at all open carton  but IF carton is really physically available for check. In Siglent cartons, outrside of carton is serial number. Part of serial number include information about manufacturing time.  If seller do not knoe even this...  perhaps it is wise to think why seller do not know this or do not want know or tell.

 

[rf-loop]

The Siglent has dual time base!

Looks like that the Siglent is going to be the winner...

Why would you need dual time base on a DSO? Just capture at a high sample rate and then zoom in.

Have you never used oscilloscope for other work than just looking one single signal?
Do you know why there is also other channel and for what all it can use.  Do you know that some times need really look same time two signals what are not syncronized with each others. For this is ALT in many scopes. But then, it is somehow very limited if both signals can use only same horizontal speed.
I do not at all understand your zoom comment related to this. Do you know what is this whole case about ALT and dual timebase ALT.  I can understand your zoom comment if you really do not understand what is this.

Tell me now how you look with normal zoom 2 separate signal what are not in synch.
How if these two are so that other is example 50MHz and other is 1kHz and not in synch.

Perhaps next then you ask why need look this kind of signal and who need this. This world is made so that if people do not know what he do not know he may think all is simple and no need know or use anything he do not know.

Time ago I go to restaurant and I ask cup of coffee. Seller ask why I want coffee, we have tea.
Without words I walk out.

[hgg]

rf-loop thank you very much for your reply!

Apart from the dual timebase which is like having two oscilloscopes on one screen, the Siglent has
also a feature that I am not sure that many other scopes on the similar price range have.  On the
X-Y mode it has a variable sampling rate from 15Ksa/s to 250Msa/s

I also like that it has a hardware frequency counter instead of a software one.

The only thing that I am not sure about, its the fact that while on a single channel the sampling
rate is 1GSa/s, in the dual channel mode its 500MSa/s but only when the timebase is faster than
50ns/div.  What if its slower than the 50ns/div?  How low will it go?  250MSa/s or more?

[marmad]

The only thing that I am not sure about, its the fact that while on a single channel the sampling
rate is 1GSa/s, in the dual channel mode its 500MSa/s but only when the timebase is faster than
50ns/div.  What if its slower than the 50ns/div?  How low will it go?  250MSa/s or more?

The sampling speed of DSOs is fixed - so above a certain time base, all DSOs automatically start slowing down their sampling rate: if you double the time for a 'sweep' (e.g. go from 1us/div to 2us/div), either the sampling rate has to be cut in half - or the sample size has to double. Since the latter is not really feasible, the rate is halved with each doubling of the time base.

[marmad]

The Siglent has dual time base!

Looks like that the Siglent is going to be the winner...

Why would you need dual time base on a DSO? Just capture at a high sample rate and then zoom in.

Have you never used oscilloscope for other work than just looking one single signal?

I think he misunderstood and was thinking of a delayed time base (real time zoom) - and not a dual (completely separate) time base on each channel.

[hgg]

It makes sense.  Thank you for the explanation.
So for up to 50ns/div is the 500MSa/s typical among these range of scopes?

[marmad]

It makes sense.  Thank you for the explanation.
So for up to 50ns/div is the 500MSa/s typical among these range of scopes?

It's really a math formula determined by the sampling speed and the sample length (and the waveform display size). I can't really speak about other scopes without looking at their data but I can give you some examples using my Rigol DS2000:
With two channels, I have a maximum rate of 1GSa/s.
With a sample length of 7kB, the sampling speed switches from 1GSa/s to 500MSa/s when I go from 500ns/div to 1us/div (and ~halves for every further increase in time base from then on).
With a sample length of 7MB, the sampling speed switches from 1GSa/s to 500MSa/s when I go from 500us/div to 1ms/div (and ~halves for every further increase in time base from then on).
But the amount of samples visible on the screen remains the same - the only thing that changes is the data that is captured pre- and post-display - which you would see if you stopped the DSO and 'zoomed' out.

[marmad]

In terms of the formula - if your DSO is sampling at 500MSa/s, that means each sample represents 2ns. So if your sample length is, for example, 1kB - and your waveform display area is 10 divs wide - I would think you could sample at 500MSa/s up to 200ns/div - i.e. 200ns / 2ns * 10 divs = 1000 bytes).

[rf-loop]

rf-loop thank you very much for your reply!

Apart from the dual timebase which is like having two oscilloscopes on one screen, the Siglent has
also a feature that I am not sure that many other scopes on the similar price range have.  On the
X-Y mode it has a variable sampling rate from 15Ksa/s to 250Msa/s

I also like that it has a hardware frequency counter instead of a software one.

The only thing that I am not sure about, its the fact that while on a single channel the sampling
rate is 1GSa/s, in the dual channel mode its 500MSa/s but only when the timebase is faster than
50ns/div.  What if its slower than the 50ns/div?  How low will it go?  250MSa/s or more?

These attached images have samplerate tables with different memory and time/div settings.
Siglent SDS1102CML, 1072CML and 1072CNL samplrates.
Rigol DS1052E + DS1102E
Owon SDS7102
Hantek 5000B. (hantek is older HW version and I have heard (but I have not tested) there is some improvement specially with long memory)
-------

In all these scopes X-Y mode is very poor. 
Siglent:  In long memory mode you can turn samplerate up to 250Msa.. but. Its all. Result is "there is something in display...what it is..can not know"  With 25M and much more low it may work better and even with some signals and some samplerates it can use for some things.

It can do something but I do not know if any useful. Perhaps some low freg and low accuracy things just barely but more like just for play as toy. This is weak point in all these. It is nothing..really nothing like in good analog scope where you may run even some kind of slow scan "TV"

Example if need adjust two 10MHz oscillators for exactly same frequency.. this lissajou image with X-Y mode is just "I do not want touch this 
 Perhaps it can use for some other slow freq low accuracy things but.. "not for my eyes".
I remember some people who make some kind of component tester what draw "curve" but picture is very extremely "rough".

My opinion is that with these oscilloscopes better method example  for freq adjustment is: reference to CH1 and DUT to CH2. Trig CH1.  Then adjust vertical and horizontal so that you have reasonable resolution in horizontal scale. 


About displayed and real sampling speed.
ADC's work always full speed" Even if there is 1ksa/s
(now I'm not sure how Siglent is constructed for this. 1Gsa/s it use all ADC's together and samplerate is 1GSa/s . If go to 2 channel, this ADC group is splitted.
But if with single channel go to speed where sampling is 500Msa/s (50ns -> 100ns/div this ADC is splitted to two. (if now there is only one channel in use this other half of ADC group only works for nothing)
I suspect that 100ns/div and slower speeds, independent of 1 or 2 channel, true samplerate for channel is 500MSa/s. This is meaningful only if think peak mode limits.
Example if samplerate is 5MSa/s it just drop out 99 true samples and keep one. (this is meaningful if think how "peak" mode works.  (and how high resolution mode in some oscilloscopes work - in this scope, there is no high res mode)

Samplerate tables.

[hgg]

I see.

rf-loop I've sent you an email.  Did you receive it?
Thanks!

[nctnico]

The Siglent has dual time base!

Looks like that the Siglent is going to be the winner...

Why would you need dual time base on a DSO? Just capture at a high sample rate and then zoom in.

Have you never used oscilloscope for other work than just looking one single signal?
Do you know why there is also other channel and for what all it can use.  Do you know that some times need really look same time two signals what are not syncronized with each others. For this is ALT in many scopes. But then, it is somehow very limited if both signals can use only same horizontal speed.
I do not at all understand your zoom comment related to this. Do you know what is this whole case about ALT and dual timebase ALT.  I can understand your zoom comment if you really do not understand what is this.

Tell me now how you look with normal zoom 2 separate signal what are not in synch.
How if these two are so that other is example 50MHz and other is 1kHz and not in synch.

Watching two signals which have no relation is always difficult with an oscilloscope. Dual timebase is not triggering on two different signals. That makes no sense because where is your reference point? Dual timebase means you can trigger on a signal and view an enlarged portion of it after an adjustable delay (usually its called B sweep). On an analogue scope you can use this feature to zoom in on part of a signal.

With a DSO you just capture a long record and zoom in on the part of the signal you want.

[hgg]

Hi,

Thank you for the voting.
The results are interesting indeed!

Rigol is clearly the winner but maybe this is because more people have worked with it.
It is the safest choice.

Poor Uni-T has only one vote...  but maybe this is because nobody has any experience
with it yet.

Despite the fact that Rigol is the safest choice I think I will still go with the Siglent because
it looks like an upgraded Rigol that is not yet tested enough.  Larger display, better bandwidth,
2.5 times faster wfrm/s, more knob controls and a silent fan.  I think i will try it and I hope I am
not making a mistake...

Anybody that has any real negative experience with the Siglent please tell me so before I make
the mistake... 

One last question.
My first projects will be basically Arduino circuits and measuring photodiodes and photoresistors.
Do you think that the 70Mhz will be enough, or should I go for the 100Mhz?

Thank you very much for your help!
rf-loop thank you for the countless pages of unbiased information!

[eevblogfan]

HOLD YOUR HORSES !

I am with contact with Uni-T distributor please wait till I'll get the Uni-T and I am going to test it . I really think you should wait and consider that

[Bored@Work]

rf-loop thank you for the countless pages of unbiased information!

Unbiased? He is a Siglent/Atten reseller, who in the past even ran a few sockpuppets here to promote his business. That and rather strange claims about Siglent quality.