Oscilloscope for fft functions

[freebil]

Hello. I would like to buy an oscilloscope which will have a decent fft viewer. I read that ds1054z has poor fft functionality. So, siglent sds1000x is the next move?

[scanlime]

Not sure what your other scope requirements are, but I recently had a similar problem with my Rigol being useful for most things but somewhat lacking for FFTs and deeper captures. I recently bought the Analog Discovery 2, which had a surprisingly nice FFT. IMHO it isn't a replacement for a full bench scope, but if you mostly just need a better FFT occasionally, that may be worth looking into.

[nugglix]

Hi!

Have a look at that one:


Compares a few scopes in terms of FFT.

Cheers

Edit: And now I really need to go to work...   

[rf-loop]

Hello. I would like to buy an oscilloscope which will have a decent fft viewer. I read that ds1054z has poor fft functionality. So, siglent sds1000x is the next move?

If you want Good FFT analyzer with oscilloscope functions also, it is not at all Rigol DS1000Z or Siglent SDS1000X.
DS1000Z is entry level multipurpose toy what is not good in anything. But with its price it have lot of "things".
SDS1000X is strong entry level oscilloscope (but only 2 channel + ext trig) with some totally unbeated oscilloscope functions in its price class.

But if you want Good FFT, forget both, totally.

All depends of course what specifications you have for FFT what you need. Define first exatly what you need, then look what equipment meet your needs. There is so many variables. Some one want good audio FFT with high resolution, fast speed and high dynamic range with very low base noise etc.. some may want good FFT for RF frequencies.  In some audio cases perhaps some PC high quality sound card based solutions are quite good also.

[CustomEngineerer]

If you want Good FFT analyzer with oscilloscope functions also, it is not at all Rigol DS1000Z or Siglent SDS1000X.
But if you want Good FFT, forget both, totally.

Completely agree.

DS1000Z is entry level multipurpose toy what is not good in anything.
SDS1000X is strong entry level oscilloscope (but only 2 channel + ext trig) with some totally unbeated oscilloscope functions in its price class.

Sorry to reorder the quote, and emphasis from me. But, seriously?

Edit: More emphasis on silly statements.

[BravoV]

DS1000Z is entry level multipurpose toy what is not good in anything. But with its price it have lot of "things".

SDS1000X is strong entry level oscilloscope (but only 2 channel + ext trig) with some totally unbeated oscilloscope functions in its price class.

I think you better format the font with this kind of arrangement, much-much better. 

[rf-loop]

DS1000Z is entry level multipurpose toy what is not good in anything.
SDS1000X is strong entry level oscilloscope (but only 2 channel + ext trig) with some totally unbeated oscilloscope functions in its price class.

Sorry to reorder the quote, and emphasis from me. But, seriously?



Edit: More emphasis on silly statements.

Semi seriously.

Rigol problem is high noise level, very  poor sin(x)/x joke (explained here in forum several times and leads example to wrong levels). Trigger and delay time jitter quite high,  possible origin partially from signal noise and partially poor clock aand perhaps also trigger position fine adjust iterpolation is poor. Slow UI. Poor user interface ergonomy including shared vertical control and small display. Poor intensity grading. Lack of true seqmented memory acguisition. Lack of always bacround working waveform history. (just some kind of quite slow  frame recorder). Very unprofessional mask test what give very easy statistical random error in test result. (due to very different speed if mask test pass or fail).  1mV and 2mV/div are not true analog sensitivity.

SDS1000X
max 60kwfm/s average, max >400kwfm/s (quaranteed speed, not average or peak speed)  in sequence (segmented memory)
Full speed hardware based mask test.
Always in backround continuous history recording up to 80000 last captured single waveforms. (stop and look what just previously happend)
Down to 500uV/div full bandwidth true analog sensitivity  low noise front end.
Inputs 1Mohm / 50ohm.
Separate vertical adjustments for channels.
Sin(x)/x and  line modes, both always draw trace via true raw sample points what are not flushed to toilet as Rigol do (starting from beginning with DS1052E, Rigol have this same joke where they flush out true data and supersede it with "fake nice draw using" (imho its betteer name is "smooth").

I have used both. I have also purchased DS1000Z for one my project where need cheap scope standing alone collecting data including fact that if poor luck it can also damage. So I select this because it was cheap and have four channel. But, later I see it was wrong selection. All data nearly like from garbage generator due to noise and trig jitter.

But, for many hobbyist, even if it is toy, it have lot of "things" and for many use - I did not say it is useless. If look how many features get with quite low price....  well,  and if need cheap kitchen all in one multipurpose toy  what can do tens of miracles but nothing well (we know these kitchen machines). This is like same, lot of things, it can do many things but nothing well. And many peoples buy.
Toy is not here as synonym for unusable.

 

[JPortici]

What are you going to look at? what will the fft function be used for?

you have to remember that most oscilloscopes use 8 bit ADC and every averaging/high res mode is actually software vodoo trying to guess what's going on between two LSBs, so the actual accuracy of the thing will be limited on a rigol just as it will be limited in a lecroy, keysight, R&S

This is an  insteresting topic
https://www.eevblog.com/forum/testgear/fft-spectrum-analysis-reviewed/msg920781/

[wraper]

GW instek GDS-1000B series is the cheapest you can get with good FFT. BTW how it looks on the Dave's video is not how it will be with installed FFT app, it becomes separate FFT display, not together with waveform. So his complains about UI are not valid at current moment.

[Wuerstchenhund]

you have to remember that most oscilloscopes use 8 bit ADC and every averaging/high res mode is actually software vodoo trying to guess what's going on between two LSBs, so the actual accuracy of the thing will be limited on a rigol just as it will be limited in a lecroy, keysight, R&S

Not exactly. There's a big difference between the 'HiRes' (boxcar) modes used by most scopes vs ERES as used by LeCroy:

http://teledynelecroy.com/doc/differences-between-eres-and-hires

It's also far from "guessing", you do actually get a real increase in resolution.

[Wuerstchenhund]

Hello. I would like to buy an oscilloscope which will have a decent fft viewer. I read that ds1054z has poor fft functionality. So, siglent sds1000x is the next move?

No, not really. You have to understand that FFT is very processing intensive, probably amongst the most processing intensive applications on any modern scope. Neither Rigol nor Siglent¹ offer scopes that are powerful enough to do that properly.

You didn't say what else you're looking for in a scope, so giving some recommendations is difficult. At the moment, the GW Instek GDS-1000B and GDS-2000E are the cheapest (and only) new entry-level scopes that offer decent FFT (1Mpts). Especially if you need more bandwidth I'd also consider some 2nd hand scopes (for example, my 600MHz Agilent Infiniium 8064A does FFT with up to 32Mpts).

¹ Siglent also has the SDS3000 which is only sold in China and which was developed in co-operation with LeCroy (which sells it in the ROW as 'WaveSurfer 3000'), and this scope offers fast FFT up to 1Mpts.

[Fungus]

Hello. I would like to buy an oscilloscope which will have a decent fft viewer. I read that ds1054z has poor fft functionality. So, siglent sds1000x is the next move?

a) Depending on when you read it, it might not be true any more (firmware update)

b) A much better idea would be to get a spectrum analyzer. No entry level scope has what you'd call a "good' FFT viewer.

[pascal_sweden]

What about the PC software tool that was designed to give the Rigol scope "Spectrum Analyzer"-powers?
http://hackaday.com/2012/10/10/giving-the-rygol-ds1052e-scope-a-spectrum-analyzer/

Has anybody used this PC software tool?

Note that this PC software tool was designed for the Rigol DS1052E.
Is there also a version available for the Rigol DS1054Z?

[nctnico]

GW instek GDS-1000B series is the cheapest you can get with good FFT. BTW how it looks on the Dave's video is not how it will be with installed FFT app, it becomes separate FFT display, not together with waveform. So his complains about UI are not valid at current moment.

Indeed. Currently GW Instek (GDS-1000B and GDS-2000E series) is king when it comes to FFT on an entry level oscilloscope. Peak search also allows you to show up to 10 frequency peaks in a list.

[pascal_sweden]

Can someone with a GW-Instek scope post a screenshot of this separate FFT display?

[freebil]

Thanks for the answers! Can a spectrum analyzer show voltage/time?

[nctnico]

Thanks for the answers! Can a spectrum analyzer show voltage/time?

Not really although some can display the demodulated signal in the time domain. Tektronix has a series of MDO oscilloscopes (mixed domain oscilloscopes) which are both a spectrum analyser and oscilloscope in one device.

[Wuerstchenhund]

Thanks for the answers! Can a spectrum analyzer show voltage/time?

Some SAs can show voltage or power over time (0Hz span), although its no replacement for a proper scope.

[rfbroadband]

The R&S RTO has the best FFT I have seen so far in a scope in terms of usability and performance. It beats the even most modern Keysight scopes in terms of usability. I can post pics if some is interested. As expected, the RTO is not cheap. 

[rfbroadband]

here is pic of the 600MHz RTO 1004 in spectrum analysis mode.

500MHz CW at the input

Important note: in FFT mode you use it like a spectrum analyzer, meaning you will have a GUI that allows you to center freq, span, Res BW etc and the tome domain parameter like sampling rate etc are updated accordingly. Very convenient to use!

Also, you may not care, but then RTO is one of the few scopes that retains the phase information of the FFT calculation. Many scopes throw the phase information away and only offer you the magnitude. As a result you can use the RTO for large dynamic range bode plot measurements plotting mag and phase response over freq.

[rfbroadband]

attached a bode plot measured with the RTO and compared against network analyzer measurements E5061B

[jjoonathan]

You beat me to the punch rfbroadband, I saw this thread yesterday and thought "I should link to info on the RTO to show what a non-toy fft looks like." I wish other vendors took the function half as seriously as R&S.

[rfbroadband]

:-)

[pascal_sweden]

I don't see the point of the R&S RTO scope which costs much more than a Siglent spectrum analyzer that goes up to 3.2 GHz and a very decent digital oscilloscope together.

It's for people who have too much money.

[rfbroadband]

the combination of having a digital 16 bit MSO, a 4 channel scope and a SA in one box for cross domain analysis is a huge advantage.

One can trigger in the frequency domain on masks and signals over a very large dynamic range, and then debug other related signals in the time domain and do decoding of signals on the 16 digital lines.

I have a real high end spectrum analyzer next to the scope but the fact that you have a powerful FFT function and time domain analysis in one box is a huge advantage. I am using the built in SA of the scope more and more just because of the cross domain capability.

And yes it not a scope that a hobbyist would purchase.