Author Topic: EEVblog #845 - Oscilloscope FFT Comparison  (Read 34311 times)

0 Members and 1 Guest are viewing this topic.

Offline markone

  • Frequent Contributor
  • **
  • Posts: 361
  • Country: it
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #75 on: February 06, 2016, 06:00:39 am »
Sorry for the OT,

but take a look at this beauty, Dave's FFT reference signal (1Mhz, FM 5Khz, Dev 500Hz) under italian ELAD FDM-S2 16bit-122.88MSa/s sampler analysis, until 60Mhz and 6Mhz span blows out the windows everything showed here in terms of sensitivity, resolution, dynamic and update rate (almost real time) at mere 430 Euro plus taxes.

Not a DSO, but if you need high resolution FFT within its spec, coupled with a DC block and a fixed attenuator and / or active probe for High-Z it does a great job.


 

Offline AlessandroAU

  • Regular Contributor
  • *
  • Posts: 154
  • Country: au
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #76 on: February 06, 2016, 11:00:11 am »
He is the FFT from an AD7760 eval board.

2.5MSPS, 24bit
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2062
  • Country: au
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #77 on: February 06, 2016, 11:12:46 am »
He is the FFT from an AD7760 eval board.

2.5MSPS, 24bit

Now try it with the OP's 1 MHz carrier  :P

Also, did you do the FFT on your PC? How many points? How long did it take? Doing an FFT is trivial, after all, if you have GB of memory and no particular time limit...
 

Offline hpw

  • Newbie
  • Posts: 4
  • Country: ch
    • FFT Spectrum Analyzer (Level, Power, Noise Density) & Wave Generator
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #78 on: February 07, 2016, 11:20:54 pm »

If some is interested about the dynamic range on 12...40 bit

Hp

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2062
  • Country: au
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #79 on: February 07, 2016, 11:47:28 pm »

If some is interested about the dynamic range on 12...40 bit

Hp

That's just a simulation. Try putting in a vaguely reasonable AWGN representing an analog frontend and you'll get very different result. Processing gain from long FFTs is a much more practical way of getting a low noise floor than an absurd 40-bit DAC.
 

Offline hpw

  • Newbie
  • Posts: 4
  • Country: ch
    • FFT Spectrum Analyzer (Level, Power, Noise Density) & Wave Generator
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #80 on: February 08, 2016, 08:50:39 pm »

>> That's just a simulation.

Yes it's a digital & dithered signal to test digital gear! Keep in mind that HW digital mixer often calculates using > 32 bits.

>> Try putting in a vaguely reasonable AWGN representing an analog frontend and you'll get very different result.

Well, I used here triangle dither. Gauss dither will not alter that much. it's also dither level dependent.
 
>> Processing gain from long FFTs is a much more practical way of getting a low noise floor than an absurd 40-bit DAC.

Why is this absurd? I do not play with marketing bits...  ::)

IMHO it was a demonstration of dynamic range f(bits) not more using internal IEEE 64/80 bits..

Keep in mind that PCM <> DSP conversion, even uses higher bit count.

Hp



Offline tautech

  • Super Contributor
  • ***
  • Posts: 13016
  • Country: nz
  • NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #81 on: March 06, 2016, 11:53:52 pm »
An SDS1000X  FFT screenshot from member rf-loop, snatched from another thread:
http://www.eevblog.com/forum/testgear/siglent-sds1000x-series-oscilloscopes/msg832023/#msg832023

Avid Rabid Hobbyist
 

Offline TheSteve

  • Supporter
  • ****
  • Posts: 2591
  • Country: ca
  • GHz or bust
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #82 on: March 07, 2016, 03:42:27 pm »
Wow, Dave used 10 Vpp which is crazy high if he terminated an input with 50 ohms. Seems like a bit of a rookie mistake and a bad example to use. 
VE7FM
 

Offline max666

  • Frequent Contributor
  • **
  • Posts: 367
  • Country: at
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #83 on: March 07, 2016, 09:38:36 pm »
Wow, Dave used 10 Vpp which is crazy high if he terminated an input with 50 ohms. Seems like a bit of a rookie mistake and a bad example to use.

Hmm ... is Dave terminating in at the Rigol DSA815 input? Because then I it would be too much, wouldn't it?
Quote from: Rigol UserGuide
CAUTION
To avoid damage to the instrument, for the signal input from the RF input terminal, the DC voltage component and the maximum continuous power of the AC (RF) signal component can not exceed 50 V and +20 dBm respectively.

+20 dBm = 0.1 W and 10 Vpp AC terminated at 50 Ohm would be 1 W, or am I missing something here?
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4788
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #84 on: December 12, 2016, 11:50:15 am »
Just for reference to this thread the relative prices:

Rohde & Schwarz HMO1202 Series;1232 $US2300
Tektronix MDO3000;  MDO3014. LIST PRICE. (US$4,110)
Keysight 3000X Touch, Keysight MSO 3054T ($13,000)
Lecroy WaveJet 354 Touch $US5000
GW Instek GDS-1104B GWinstek 1104 $700 ($830) 1102B $396, 1054B $366
Rigol 1054z ($US400)
Rigol DS2000 DS2202 ($US1790)

($) are list prices other price is discounted by one popular vendor here, before eevblog coupon code.  FWIW the other Instek is a 2 CH version and 50 MHZ version and have the same processing power, just reduced analog channels and bandwidth respectively.



When Dave said that FFT sucks in the cheap scopes, for some reason didn't mention that instek's 1000B series are even a bit cheaper than Rigol 1000Z (for 50 MHz model if comparing with 1054Z). Also from actual FFT performance it's certainly looks to work better than R&S and Agilent (except fiddly UI) and is second after Tek MDO.

Yes I completely forgot the price of the GW-Instek, and stand corrected.

Best Wishes,

 Saturation
 

Offline Arjan

  • Contributor
  • Posts: 7
  • Country: nl
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #85 on: July 11, 2017, 08:52:40 pm »
Hi,

New member here. I'm considering getting a Rigol DS2074 after basically dimissing the Owon XDS series having seen the 14 bit teardown as I don't want a scope that's likely to break down in a few years due to bad soldering, bad caps or whatever. The Rigol stuff appears to have much better internal design and build quality but then I found this video on its mediocre FFT performance.

However, I'm wondering wich firmware was used in that video for the Rigols because I came accross this:

With the latest firmware (00.04.03 SP2) for the Rigol MSO/DS1000Z Series Oscilloscopes the  FFT functionality has been improved.

The FFT (Fast Fourier Transform) now can be calculated not only from the screen points but also on 16k points from the acquisition memory. This new enhancement enables the customer to perform FFT analyis in the frequency domain with increased frequency resolution.

Calculating FFT only from the screen points? What were they thinking? Anyway, cccording to the page info for that URL this appears to have been released in July 2015 and the 2000 series was released before that so I guess this enhancement was included in the 2000 series from the start but I don't like to guess or make assumptions.

So, is the FFT performance of the 2000 series still as mediocre with the latest firmware? Not sure if it's going to be a deal-breaker for me, I won't need FFT a lot and I suppose I could export the data and have it analyzed on my PC or something, right?

The 2074 seems rather attractive and it's currently on sale at Batronix for € 802.18 incl. VAT (and I don't have to pay VAT as a company) but are there any other scopes I should be looking at for that price or lower? I should say that the 2074 already offers much more than I would actually need, I could probably get by with a €300 scope but I just love the larger 800x480 display, graded display and better update rates and if I understand it can also function as a Wave Generator and replay back captured waveforms which for my application would be very convenient.

Thanks,
Arjan
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 13016
  • Country: nz
  • NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #86 on: July 11, 2017, 08:59:03 pm »
Not long released SDS1202X-E:
1 Mpts FFT

Avid Rabid Hobbyist
 

Offline Arjan

  • Contributor
  • Posts: 7
  • Country: nl
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #87 on: July 11, 2017, 10:56:16 pm »
Wow, thank you. That does look like an attractive alternative. I guess these are the main differences (the ones that matter to me):


Rigol DS2074           Siglent SDS1202X-E
2GSa/s                 1GSa/s[/font]
Waveform Gen           No Waveform Gen
Crap FFT               Decent FFT
€ 802                  € 427     at batronix.com (special 10% discount on the Rigol until 31 July 2017, regular price is €819)


I left out the bandwith because in essence these are both 200Mhz if you're ok hacking the Rigol. Not sure that I am but on the other hand both probably offer plenty bandwidth (even at 70Mhz) and Sa/s for my purposes so I guess I might have to accept the lack of WG and the lower Sa/s on the Siglent.

I have to say that in principle it's a lot easier for me to justify € 427 as I will be using this thing rather sporadically but I've wanted a scope for years and I don't like having to borrow one if I need it. On the other hand I want a quality product that's not going to break down on me. I was a little worried about the Lelon caps but after some Googling it seems they may not be that bad actually (provided they're genuine).
 

Offline Arjan

  • Contributor
  • Posts: 7
  • Country: nl
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #88 on: July 12, 2017, 12:54:53 am »
Bit the bullet and ordered the Siglent SDS1202X-E, it's just an amazing price given the feature set and it does seem like the build quality is quite acceptable.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 15229
  • Country: nl
    • NCT Developments
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #89 on: July 13, 2017, 06:16:54 am »
You could have looked further and bought the GW Instek 1054B. It has 4 channels, 1MPts FFT as well, (limited) lifetime warranty, mature firmware and a whole bunch of extra features (like input filtering) at the same price level.
« Last Edit: July 13, 2017, 06:25:39 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 13016
  • Country: nz
  • NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #90 on: July 13, 2017, 06:26:48 am »
You could have looked further and bought the GW Instek 1054B. It has 4 channels, 1MPts FFT as well, mature firmware and a whole bunch of extra features (like input filtering) at the same price level.
Maybe the 200 MHz BW and 500uV/div was more important to Arjan.
Avid Rabid Hobbyist
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 15229
  • Country: nl
    • NCT Developments
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #91 on: July 13, 2017, 07:14:36 am »
You could have looked further and bought the GW Instek 1054B. It has 4 channels, 1MPts FFT as well, mature firmware and a whole bunch of extra features (like input filtering) at the same price level.
Maybe the 200 MHz BW and 500uV/div was more important to Arjan.
He wrote he would be content with 70MHz so I guess not. Also input filtering makes a lot of difference when looking at small signals. 500uV/div doesn't do any good if the signal is drowned in noise. It is not the big numbers that make all the difference but the small things which make life easier and the (recent) GW Instek scopes have those in abundance.
« Last Edit: July 13, 2017, 07:17:10 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8662
  • Country: 00
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #92 on: July 13, 2017, 05:05:40 pm »
Awww, is the GW Instek not getting enough love?

It seems to be a recurring theme with that 'scope. Nobody wants one. Maybe it's the horrible color and oversized screen bezel.

 

Offline borjam

  • Supporter
  • ****
  • Posts: 621
  • Country: es
  • EA2EKH
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #93 on: August 04, 2017, 07:51:40 am »
Awww, is the GW Instek not getting enough love?

It seems to be a recurring theme with that 'scope. Nobody wants one. Maybe it's the horrible color and oversized screen bezel.
Interesting. Seems that RS is selling the GW Instek rebranded as "RS Pro".

http://uk.rs-online.com/web/p/digital-oscilloscopes/1233543/

 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 13016
  • Country: nz
  • NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #94 on: August 04, 2017, 08:06:18 am »
Awww, is the GW Instek not getting enough love?

It seems to be a recurring theme with that 'scope. Nobody wants one. Maybe it's the horrible color and oversized screen bezel.
Interesting. Seems that RS is selling the GW Instek rebranded as "RS Pro".

http://uk.rs-online.com/web/p/digital-oscilloscopes/1233543/
Yep. They even have rebranded Siglent bench DMM's.
http://nz.rs-online.com/web/p/digital-multimeters/1236466/
Avid Rabid Hobbyist
 

Offline bctouw

  • Newbie
  • Posts: 4
  • Country: nl
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #95 on: October 10, 2018, 06:28:53 am »
I have a recent Rigol 1054Z with all options enabled, exept the 100Mhz,
I have been testing the FFT indept, and i dissagree with you findings 

I have a perfectly usable FFT and even with th cursosr i can measure 60-70 dB of dynamic range.

If you want i can prove it !!,

Granted, it is fiddely to setup, but if yoy take a little bit of time you will get great results,   for a 400 dollar instrument.

Bas Touw, Holland
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 8662
  • Country: 00
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #96 on: October 10, 2018, 07:55:01 pm »
I have a recent Rigol 1054Z with all options enabled, exept the 100Mhz,
I have been testing the FFT indept, and i dissagree with you findings 

I have a perfectly usable FFT and even with th cursosr i can measure 60-70 dB of dynamic range.

If you want i can prove it !!,

Rigol update the firmware since Dave's video, they improved the FFT.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 15229
  • Country: nl
    • NCT Developments
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #97 on: October 10, 2018, 08:36:12 pm »
I have a recent Rigol 1054Z with all options enabled, exept the 100Mhz,
I have been testing the FFT indept, and i dissagree with you findings 

I have a perfectly usable FFT and even with th cursosr i can measure 60-70 dB of dynamic range.
You have to define useable. For example: try to look at a signal in a meaningfull way AND the FFT at the same time. With only a handful of kpts of FFT this is nearly impossible. The GW Instek scopes really set a new benchmark with their 1MPts FFT depths.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline bctouw

  • Newbie
  • Posts: 4
  • Country: nl
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #98 on: October 11, 2018, 05:18:17 am »
That is true
 

Offline bctouw

  • Newbie
  • Posts: 4
  • Country: nl
Re: EEVblog #845 - Oscilloscope FFT Comparison
« Reply #99 on: October 11, 2018, 05:20:04 am »
I can only say about the Rigol
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf