EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Magnum on May 15, 2019, 08:18:47 pm
-
Hello,
I am looking for a new scope. Right now I use a hacked 1054z which does everything I need except FFT.
I would like to have 4 channel, but IMHO 2 channel should also be sufficient. 70 Mhz is fine for me. Hackable is also nice, but I don't really need serial decoding and only enough memory to have good FFT. The frequencies I need the FFT for are usually below 10kHz (no, it is not for audio, it is for vibration analysis).
The candidates so far:
Rohde & Schwarz RTB2002: At the max range of my budget, the screen looks good and FFT seems to be well implemented. It would make sense to buy the the RTB-PK1 option with it, but then it gets really expensive.
Rigol MSO 5074: Haven't seen much of the FFT functions. It is cheap, hackable and has a touch screen. But I think I won't need much of the functions (e.g. 8GS/s, all the memory etc.)
Siglent SDS1104X-E: No Touchscreen, FFT seems to be good, really cheap
Maybe someone comes up with another one or can comment on these candidates.
Many thanks,
Magnus
-
The frequencies I need the FFT for are usually below 10kHz (no, it is not for audio, it is for vibration analysis).
Would a USB audio interface and a computer be an alternative nevertheless? If the 96 or 192 kHz sampling frequency is enough, the 24 bit dynamic range is hard to beat.
-
I know it works well, I used it for some audio spectrum analysis. There is really good software available. But in this case I want to have a solution without a PC.
Additionally, the sensor gets a supply voltage of 24V DC (IEPE sensor) and I don't know how the audio card will handle that.
Magnus
-
I know you said no PC, but tool for that job is Picoscope 4224 IEPE...
Look at GW Instek MDO 2000E series too..
-
I know you said no PC, but tool for that job is Picoscope 4224 IEPE...
Look at GW Instek MDO 2000E series too..
I suggest to get the more allround GDS-2000E series and use the key generator (found in the GDS1054B hack thread) to enable the spectrum analysis feature.
-
I know you said no PC, but tool for that job is Picoscope 4224 IEPE...
Look at GW Instek MDO 2000E series too..
I would also suggest that picoscope (as it has true 12 bit resolution)
For an audio interface, i would use a DC coupled one, you can find a list here https://www.expert-sleepers.co.uk/siwacompatibility.html (https://www.expert-sleepers.co.uk/siwacompatibility.html)
-
The frequencies I need the FFT for are usually below 10kHz (no, it is not for audio, it is for vibration analysis).
Analog discovery, PC sound card...
-
Thank you for all the replies.
A solution with a PC is out, as I will have to take the unit with me and need everything in one box.
The RTB202 has only 128k of datapoints for the FFT against 1M in the Rigol. Does this make a differnece (esp. at the low frequencies)?
Magnus
-
More FFT points means more resolution in the frequency domain.
-
That is what I thought, the question is, how much is really needed for frequencies below 10kHz.
-
That is what I thought, the question is, how much is really needed for frequencies below 10kHz.
Depends on the sample rate. :popcorn:
-
That is what I thought, the question is, how much is really needed for frequencies below 10kHz.
Well, that's the first thing you need to specify, and it will obviously depend on the problem you are working on: What frequency resolution will you need?
(And, maybe more critical: What dynamic range will you need? The number of sample points, and hence the frequency resolution, is a parameter that can hopefully be varied for most scopes or other spectrum analyzers. But the bit width of the A/D converter, which determines the dynamic range, will be fixed by the device's hardware.)
-
Thank you for all the replies.
A solution with a PC is out, as I will have to take the unit with me and need everything in one box.
The RTB202 has only 128k of datapoints for the FFT against 1M in the Rigol. Does this make a differnece (esp. at the low frequencies)?
Magnus
picoscope + microsoft surface? Already did that, works like a charm.
As eblaster said, the first thing you need to specify is the frequency range and the frequency resolution, that will give you the minimum number of points (Frequency range / points = Frequency resolution)
Then the required dynamic range, which will determine if you can do it with a 8bit ADC or if you need 10,12,16,20,24 bit one.
Ther was a thread one or two years ago on comparing scopes FFT, it was a pretty good analysis with lots of contenders. I can'd find it though :(
Uh, what about the scope rider from R&S? Pricey as hell (much more than pico+surface) but you have an handheld, battery powered instrument with 10 bit ADC (64kpts FFT though)
-
Thank you for all the replies.
A solution with a PC is out, as I will have to take the unit with me and need everything in one box.
The RTB202 has only 128k of datapoints for the FFT against 1M in the Rigol. Does this make a differnece (esp. at the low frequencies)?
Magnus
Let me just say this and won't bother you any more..
Picoscope 4224 IEPE has built in IEPE interface, it's 12 bit, it has 32MPoint mem. Picoscope software has advanced arbitrary math, custom probes. Ability to save complete measurements, setups etc. is very useful and implemented much better on a PC. That is pretty much instrument made for vibration analysis.
And as I said, you can capture whole day long, save it all and take it home to analyse the crap out of it, then create documentation or send data to colleagues, they can open it on a PC without Pico connected and analyse away....
As for portability, if I travel, I have to take my laptop with me anyway. This is one handheld box, that you connect with one USB cable and you are measuring. You get your power from PC, so you can run from battery. I got a small 14" laptop with I7 and high capacity battery, and just throw a Pico in the laptop bag, next to the power supply.
That complete bag is not heavier or bigger than Rigol 1000Z in carrying bag.
If you throw in a Signal Hound SA in other pocket you can carry whole lab with you in single bag.
If you have your reason for insisting on standalone instrument, it's your choice, no arguing that.
But for that particular use, that is most portable, best specced instrument for that price, period.
Sorry for being insistent, I just wanted to lay out all the facts, and from previous posts I wasn't sure you have all the info.
Best regards,
SiniĊĦa
-
Cross-reference:
https://www.eevblog.com/forum/testgear/oscilloscope-for-fft-functions/ (https://www.eevblog.com/forum/testgear/oscilloscope-for-fft-functions/)
-
Thank you for all the replies.
I have to work "in the field" (open air) and therefore don't like to use computers with lots of cables and small usb connections and so on.
But knowing that the picoscope is available with a built in IEPE interface changes my opinion as this will make my work easier. 12bit is also nice and of course having the computer power with a "big" screen (I use 13" laptops).
8 bit dynamic range would even be sufficient. Right now I have to measure in the 200-600 Hz Range, 1 Hz RBW is more than enough. So I won't need many fft points.
-
Thank you for all the replies.
I have to work "in the field" (open air) and therefore don't like to use computers with lots of cables and small usb connections and so on.
But knowing that the picoscope is available with a built in IEPE interface changes my opinion as this will make my work easier. 12bit is also nice and of course having the computer power with a "big" screen (I use 13" laptops).
8 bit dynamic range would even be sufficient. Right now I have to measure in the 200-600 Hz Range, 1 Hz RBW is more than enough. So I won't need many fft points.
From OP
The frequencies I need the FFT for are usually below 10kHz (no, it is not for audio, it is for vibration analysis).
Not FFT but I demo'd a SDS1104X-E recording, capturing and exporting pressure analysis data using the customers pressure logging gear and it worked really well ......once we both got our heads around what we needed to see. ::)
He wanted the data for post analysis and graphing so we used the scopes inbuilt webserver to supply the BIN to CSV file converter utility then converted the BIN file dropped the data into an Excel graph.
I have a scope and AWG out so I'll have a fiddle with some stuff in the sub 1KHz range and grab some FFT screen shots for you. Any particular data type you'd like to have captured too ?
-
Pico uses old school USB type B connectors on the scope, and good quality cables, that are robust enough for everyday work..
-
Swept square wave 200-600Hz.
FFT on STOP waveform.
For the last one, how low do ya wanna go...stopped at 2Hz RBW as mHz offered no useful data.
-
Depending on your willingness to cobble some things together, the Analog Discovery/Waveforms will run on a Linux ARM BeagleBone Black and small touchscreen display, though you could probably pick up a small ARM based tablet for the same price.
Like Yogi says, you could look it up :) https://forum.digilentinc.com/topic/5340-small-computeros-for-the-ad2/
BTW, it will NOT run on a Raspberry Pi :(
-
Thanks for the replies.
I will try a picoscope, as this seems to be the best solution for me.
-
I will try a picoscope, as this seems to be the best solution for me.
So we finally wore you down. ;)
If 12 bit A/D conversion is enough, I agree the Picoscope is a strong candidate. Battery power via the laptop should offset the downside of having to juggle two units, and in contrast to a sound card you get proper wide-range inputs.
Also, you already have a "regular" scope for those scenarios where highly interactive use favors the physical knobs, so the Picoscope will nicely complement what you have.