Low Cost Rigol or Siglent Oscilloscope and FFT Function

[Shadders]

Hi,
I am looking to purchase a Rigol or Siglent oscilloscope for less then £500, and one aspect I am not too certain about is the FFT function and how it relates to the waveform captured.

The FFT function seems to have a set number of points used, and the frequency width of each bin I would expect to be the 2 * (sample rate/(number of points)), so for a 1Gs/s oscilloscope, and a 1024 FFT, then the bin width will be 1.95MHz.

The resolution of the FFT is limited by the FFT size. If you are interested in only a small spectrum, such as 2MHz bandwidth, can these oscilloscopes reduce the bandwidth calculated ?

If you are examining the 1kHz response of an amplifier, then you will never see the harmonic distortion, or class D switching artefacts of the amplifier, since it will always be within the first 1.95MHz FFT bin.

So, do the oscilloscopes have the ability to take the currently viewed waveform horizontal span as the timebase for the FFT, and the FFT is calculated by decimating the data (filter, then downsample) to obtain a restricted FFT frequency range ?

Thanks and regards,
Shadders.

[Sighound36]

Hi Shadders

I would suggest hanging on and spending a wee bit more, for audio (at sensible price and performance, plus a lot more cost effective than an R&S UPV!) these are a rare nugget of sanity in a high cost test equipment world.

https://www.picotech.com/oscilloscope/4262/picoscope-4262-overview


You will be able to look at most switching amplifiers critical areas of operation (250-750Khz) plus have up to 20 bits of adc resolution and 96dB headroom. For the money it also is more desirable than the DSA boat anchors of old!

We have two of these great little scopes top product.

[tautech]

Most of the info you seek about the Siglent's SDS1104X-E FFT implementation is in this post and the links within it:
https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg2454291/#msg2454291

[Mechatrommer]

For 1khz, increase timescale to reduce capture rate.. as for controlling fft parms in dso, cheap rigol ds1054z cant, you just accept the retarded display as is, thats why i built my own pc sw to make full use of 24Mpts memory. Cant tell you about cheap siglent and others..

[Shadders]

Hi Shadders

I would suggest hanging on and spending a wee bit more, for audio (at sensible price and performance, plus a lot more cost effective than an R&S UPV!) these are a rare nugget of sanity in a high cost test equipment world.

https://www.picotech.com/oscilloscope/4262/picoscope-4262-overview


You will be able to look at most switching amplifiers critical areas of operation (250-750Khz) plus have up to 20 bits of adc resolution and 96dB headroom. For the money it also is more desirable than the DSA boat anchors of old!

We have two of these great little scopes top product.

Hi,
Thanks - i already have a picoscope - one input failed with a permanent DC offset, and now the second has failed too, since it uses a dual FET on the input stage. The picoscope software now is very good, but the problem i have is that it is not very robust (max 20volts input) and PC based. I want something that works anywhere without have to set up a PC. For audio, i will eventually go for the dedicated hardware specific for audio (https://quantasylum.com/collections/frontpage/products/qa401-audio-analyzer)

Most of the info you seek about the Siglent's SDS1104X-E FFT implementation is in this post and the links within it:
https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg2454291/#msg2454291

Thanks for this - i downloaded some of the pages. What i am not sure, is that from the tables, it seems you can set the sample rate of the scope by varying the timebase ?. I can see some graphics with different sample rates, so this is what i would expect to do to perform an analysis for a set bandwidth.

For 1khz, increase timescale to reduce capture rate.. as for controlling fft parms in dso, cheap rigol ds1054z cant, you just accept the retarded display as is, thats why i built my own pc sw to make full use of 24Mpts memory. Cant tell you about cheap siglent and others..

Thanks - you seem to be confirming that changing the timebase will modify the sample rate.

Thanks to all for the answers. I definitely want a standalone scope with a display, to use for other than audio projects also. I did look at the manual for the Siglent lower end scope and it did not have the detail as per the write from Performa01.

Regards,
Shadders.

[nctnico]

Another model to consider is the GW Instek GDS-1054B. Hackable to enable 100MHz and decoding using a key-generator and slightly cheaper compared to the Siglent (depending on where you are in the world). Also has 1Mpts FFT and selectable memory depth.

[khutch004]

Thanks for this - i downloaded some of the pages. What i am not sure, is that from the tables, it seems you can set the sample rate of the scope by varying the timebase ?. I can see some graphics with different sample rates, so this is what i would expect to do to perform an analysis for a set bandwidth.

I have an older Picoscope (3600 series perhaps?) and while the FFT function is quite good I did get tired of having to set up a PC too every time I wanted to look at something quick with a scope. I now have a Siglent SDS1104X-E that I chose because it had a more capable FFT function than most in the price range. I am quite happy with it. Yes it does have its limitations but even as an RF engineer who uses very high grade spectrum analyzers at work I still find it quite useful for the types of things I do at home. I don't recall exactly how you control the sampling rate but it is easy enough to do. I would love it if it were to have a spectrum analyzer app to do all that fiddly stuff for you. In other words a user interface that looks like a spectrum analyzer so that you adjust things like sweep rate, frequency range, resolution BW, etc from front panel controls and the scope firmware does all the math to make that happen for you instead of making you remember how to use traditional scope controls to do what you want to do. I know this is possible, I believe it is Rhode & Schwartz who has a relatively low cost digital scope that does exactly that.

I can't tell you what is the best brand and model for you but the Siglent would certainly be in the running. It's well worth a look.

[nctnico]

The GWI GDS-1054B may be hacked to have a spectrum analysis mode but I don't know whether it actually works. The SA mode hack does work on the higher end models though (2000 series).

[tautech]

Most of the info you seek about the Siglent's SDS1104X-E FFT implementation is in this post and the links within it:
https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg2454291/#msg2454291

Thanks for this - i downloaded some of the pages. What i am not sure, is that from the tables, it seems you can set the sample rate of the scope by varying the timebase ?. I can see some graphics with different sample rates, so this is what i would expect to do to perform an analysis for a set bandwidth.

Typically when using FFT you use a slowish timebase to have lots of data points for the FFT to work with and then within the FFT UI select the amplitude range, units, Span, CF and Hz/div. Some of the setting choices are related/restricted by the choice of timebase setting as in the result wouldn't be satisfactory at some particular settings. It takes a few trials to get the hang of the available choices within the UI to see how these all work together. Look more closely at the screenshots and the FFT graticule and then you can see the vertical and horizontal settings used in the examples I linked.
The FFT result can be displayed with the source channel visible or hidden.
Sorry my SDS1104X-E is out on loan while waiting for new stock so I can't post further examples for your study. FYI, the SDS1202X-E FFT works exactly the same and while it's only a 2ch model might also fit your needs and is a good bit cheaper than the 4ch X-E's.