Open-source digital oscilloscope "ScopeFun"

[Free_WiFi]

ScopeFun - Open Source Instrumentation
Several instruments in one device
Oscilloscope, Arbitrary Waveform Generator, Spectrum Analyzer, Logic Analyzer, Digital Pattern Generator.
Open source
Software, firmware and hardware sources are available for free. All under open source license.
Cross Platform
Software runs on Windows, Linux and Mac.

Specifications
Oscilloscope
No. of Channels: 2
Analog bandwidth (-3 dB): 100 Mhz
Sampling rate, max.: 250 MSps dual ch. (500 MSps single ch.) / 2,0 Gsps with Equivalent-Time Sampling (ETS)
Sampling rate, min.: 50 Sps
Resolution: 10 bit
Memory depth: 128.000.000 samples per channel
Voltage ranges (with 1× probe): 10 mV, 20 mV, 50 mV, 100 mV, 200 mV, 500 mV, 1 V, 2 V per division
Voltage ranges (with 10× probe): 100 mV, 200 mV, 500 mV, 1 V, 2 V, 5 V, 10 V, 20 V per division
Input Offset: adjustable
Input coupling: DC, AC, GND
Input impedance: 1 MΩ || 18 pF
Max. input voltage: +/-20 V
Overvoltage protection: +/- 50 V ( permanent )
LED trigger indicator

Waveform Generator
No. of Channels: 2
Sampling rate: 200 Msps
Resolution: 12 bit
Output impedance: 50 Ω
Output amplifier bandwidth: 30 Mhz
Waveform shapes: Sin, Cos, Triangle, Saw, Ramp up/down, Delta, DC, Noise, Custom
Custom waveform memory: 32.768 Samples per Channel
Max out. Voltage: +/- 2,0 V
Offset and Level adjustable
Overvoltage protection: +/- 25 V ( permanent )
Short-circuit protection

Logic Analyzer / Pattern Generator
No. of Channels: 12 (logic analyzer / pattern generator: 6-input / 6-output; 12-input; 12-output)
Sampling rate: max. 250 Msps
Interface voltage: Adjustable 1,25 V - 3,3 V in 256 steps
LA Memory depth: 128.000.000 samples per channel
Input Impedace (Logic Analyzer): 200 kΩ
Output Impedance (Pattern Generator): 1 kΩ
Overvoltage protection: +/-5 V (Permanent); +/- 12V (Short term ~ 10 s)
Pattern Generator memory: 32.768 Samples per Channel
Pattern Generator internal clock divider: Adjustable 32-bit (250 Mhz - 0,058 Hz)

Trigger
Source: Analog Ch. 1, Analog Ch. 2, Digital Ch. (external), Generator Ch. 1, Generator Ch. 2
Mode: Auto, Normal, Single (with Re-Arm)
Pre-Trigger: Adjustable 0 - 99%
Trigger Level: 0 - 100 %
Trigger Level Hysteresis: Adjustable
Trigger Holdoff: Adjustable 0 - 17 s (4 ns step size)
Digital trigger: 4 stages (with delay counter for each stage)
Digital trigger: selective channel masking (logic levels: '0', '1', 'Rising', 'Falling')


The price for the full kit is rated more then : $700

Related links
https://www.scopefun.com
https://www.scopefun.com/https://www.scopefun.com/gallery/ScopeFun_v1/1-16-67
https://www.youtube.com/channel/UCd2G5c2wmahLIKfNnVobr7Q
https://medium.com/@ScopeFun/scopefun-oscilloscope-development-update-1-b64f8a2fd085

Arrived at this point,i wish to ask the people of EEVBlog if this scope did really worth the effort to be made in home conditions.
I haven't too much cash in my own pocket for an "high end-mid range scope" but this project looks to be very delicious and so on.....
However tell me please what do you think about,thanks to all.

[NiHaoMike]

I wonder if, by changing the firmware and adding some RF front end circuits, it could double as a SDR platform?

[capt bullshot]

I wonder if, by changing the firmware and adding some RF front end circuits, it could double as a SDR platform?

I'm just an beginner,however your statements sounds like sarcasm?
Do you mean that this dso is bullshit?right?
lol sorry for my ignorance

No, IMO this statement doesn't mean your DSO is bullshit. Indeed, it looks quite nice, way better than many other USB scopes.

To double as an SDR platform means, NiHaoMike thinks this hardware is powerful enough to sustain high data transfer rates, is able to do pre- and postprocessing (like up- and down- mixing) within the FPGA, or maybe the complete signal processing for SDR can be done within the FPGA. So basically by adding some RF inputs and pre-amps, your DSO could be also used as a powerful platform for SDR development.

[Free_WiFi]

So basically,someone of you would be ready to reapeat it by posting the final result on this thread ?
It would be very,very nice 

[james_s]

$700 for 2 channels, 100MHz bandwidth, 500MS/s when only using 1 channel? Logic analyzer is only 12 channels, that's not even enough to debug an 8 bit microprocessor system, I must be missing something because this doesn't sound like any bargain to me. Doesn't sound any better than the low cost Chinese DSOs and those are standalone not tying up a PC. For $700 you could get standalone instruments that exceed the specs of this thing.

[AndreZheng]

 
I do also really love this amazing project. But the target price $700 is really too high for me. You may know $700 can almost buy a really low end oscilloscope, e.g. Keysight(Agilent) DSOX1000 series.

If I were this project's designer, I will make the product's cost down to around $200. And I'll do following modifications to make it happen.
(1) Replace the 10bit@250Msps ADC with 8bit@250Msps ADC.
(2) Separate the single board into two boards. One board for USB3.0 chip CYUSB3014 + FPGA + DDR3 + Expansion IOs; The other boards are standalone boards like Pure IO board, or ADC board or DAC board, etc. In this case, different boards can meet different people's need.

 

[james_s]

Yeah I think at $200 it might be a fair deal for people who are on a tight budget, but personally I'd still rather have a standalone ~$350 Rigol or Siglent DSO. Standalone scopes with real buttons and knobs are so much more convenient than PC based instruments. I think PC based never really delivered on its promises, the cost of the display and processing dropped to the point that the savings of utilizing a shared PC is minimal vs having a dedicated computer and display built into the instrument.

[NiHaoMike]

In my experience, even rather expensive DSOs like the MDO 4000 I have used in an automated test system for work leave much to be desired in many ways. One I have yet to see done right is high speed streaming of data to a PC. I could quickly download screenshots at 1024x768, barely enough for 3 nonoverlapping analog channels at 8 bit, let alone 10 bit channels and/or digital channels, not to mention the annotation that also takes up space. (And if you want a white background, doing it in the scope itself leaves ugly dark fringes around each trace - I ended up implementing a bit of Python code to take the image with a black background, invert it, and remap the colors.) Or I could download the capture data at way too slow a rate to be useful for high speed testing.

And when it comes to bus debugging, the MDO is so slow that I would classify it as barely better than nothing. A $6 FX2 board along with Sigrok gives a much better user experience. (I would like to see the FX3 get official Sigrok support - that would make it the ultimate budget logic analyzer for most hobbyists and even many small companies. I used to work at a company that used Beaglebone Black boards for I2C and SPI analysis because those were cheap enough to provide several for every developer, plus the Ethernet port made it trivial to install in a remote test rack.)

If I were to suggest a path, start by getting a FX3 based logic analyzer supported under Sigrok. Then add a FPGA to allow capture of even faster signals and interfacing to ADCs. I wouldn't invest much effort into streaming to a high speed DAC when the $5 fl2k will do 3 channels at 8 bit/150MSPS.

[tkamiya]

If the board alone is $700, completed project, case, power supply, and everything else will probably run more than $1000.  Then, I assume those functions that this board can do will be one at a time.  Can you have scope running and function generator running at the same time?

If I look at scope spec alone, you could do a lot better by going with used Tektronix.  Mine happens to be refurbished by Tek itself and was $300.  My preference is to go with older but well characterized individual test equipment.  For $1000, you can buy a lot if you know where to look. 

[prasimix]

If everything is reduced to the cost then ScopeFun is not a winner for sure. But, if we put into picture open source design that is not just promised but already delivered, and put into perspective that project team probably will continue to improve software/firmware part of the ScopeFun than it is something completely different. That include real possibility that you and I could ask or influence project team to modify or add new features in software/firmware. A rare luxury in case of commercial brands.
Open source doesn't mean a lot until device failed. I had such situation with Rigol a year ago: bought as brand new that last a little bit over two years and then mysteriously failed. Service cost with shipment was comparable to new unit (500+ EUR) and I decided to buy a new unit. The whole story could probably end differently that Rigol was an open source.

The main obstacle for me to get one ScopeFun is its format: it's unfortunately a PC scope. I found that quite inconvenient in practice (I also have Picoscope 4444): I cannot put PC on my desk top where is DUT, not to mention that most of control is reachable only by mouse that require more time to set desired values then using front panel knobs. With special arrangement of the desk top I probably could make its usage more efficient without constantly moving my head left and right as on the tennis court. Yes, a speech recognition would improve usability a lot when comes to control, but require some integration work since Picoscope doesn't offer such thing out of box.

[OwO]

Open source by itself is not a good enough selling point, your product has got to be competitive on its own without considering that it's open source. I do have a open source hardware startup and my main competitive advantage is still bang for buck, not just that it's open source.

[prasimix]

Thanks for reminder. Luckily for the open source community your worldview even as continue to be a prevalent is not the one and only.

My compliments to David and Dejan for doing a great job. This project is a great contribution to people who is willing to learn something about T&M solutions and push boundaries of DIY/makers solutions to the next level. I do believe that forthcoming crowdfunding campaign will be successful. For other, possibly some company from your neighborhood will start to clone it massively and push the price down. We'll see, that could be a great sign that Scopefun is recognized and attractive and can give Scopefun team a good reason to move it forward.

[2N3055]

Thanks for reminder. Luckily for the open source community your worldview even as continue to be a prevalent is not the one and only.

My compliments to David and Dejan for doing a great job. This project is a great contribution to people who is willing to learn something about T&M solutions and push boundaries of DIY/makers solutions to the next level. I do believe that forthcoming crowdfunding campaign will be successful. For other, possibly some company from your neighborhood will start to clone it massively and push the price down. We'll see, that could be a great sign that Scopefun is recognized and attractive and can give Scopefun team a good reason to move it forward.

Yeah guys that designed this made a nice proof of concept that shows their knowledge. Compliments.
If I were hiring, all they need to do is to show this project and they would get the job.

Is it pushing boundaries of DIY T&M solutions ?  Nah, not really.

Specifications are pedestrian and price is so high you can equip whole lab with pro made equipment for the price. For an average hobby lab repairs are going to be hard, most people cannot replace BGA on board without destroying it all... No tools for that.

It's not a worldview, it is reality. Recommending people to buy half cooked project prototypes for huge amounts of money because it is for a good cause ("Open something") is at best misguided, at worst misleading. And that is just because I genuinely believe you have good intentions. Otherwise it would be simply dishonest and fraudulent.

Open source means something only if a poor kid from a third world country can afford something that would be impossible at the prices that are being charged by likes of Keysight and Tektronix etc...
Sorry to burst your bubble, but low cost Chinese T&M equipment does that job much better than any Open source project now in existence.

Is this a project that designer should be proud of? Yes definitely

Would I recommend it to anybody? No, not really, for the money.
For maybe 150-200 USD, yes definitely, but with a disclaimer it is a developer kit, not finished, ready to use instrument. Cause it is not, and it is not given it will ever be. That still remains to be seen.

[james_s]

I agree, I'm a big fan of open source, but unless I had money to burn and wanted to make a statement by investing in a project like this I would look elsewhere. Open source makes sense when it allows cheap commodity or self built hardware to to the job of much more expensive closed source gear, or when it can do something truly unique or the fact that it can be customized to the extreme makes it more versatile. In cases like this though it seems like it does less than the competition while costing a whole lot more. I've seen too many open source projects get halfway there and then be abandoned by the creator when they get bored and move onto the next shiny thing without a support & development community really taking off.

Open source is good, but in the end it comes down to price. In many cases I'm willing to accept something that is not quite finished in exchange for a substantial savings of money but when I can buy a polished commercial product for less then there's no reason not to. If it does everything I need it to do I don't care if it's open source.

[NorthGuy]

I don't know why you would expect open source to be cheap.

A well funded manufacturer may produce things in millions, and cost per piece will be a fraction of the production of a single item. The price of the item on the market is determined by supply and demand. Most of what you pay for commercial things is to cover sales and management. Production is a smaller portion, R&D is even less. The company's profit may be a tiny fraction of the cost, may even be negative.

With open source you get free R&D, no sales, no management, no profit to anyone. All you need to pay for is production (but in small quantities).

How's that even possible to compare the two?

[james_s]

I don't know why you would expect open source to be cheap.

Umm, because it almost always is?

Linux is free, KiCad is free, Inkscape is free, GIMP is free, virtually all open source software is free. It may not always be quite as good as the high dollar heavy duty commercial stuff but because it's free it gets widespread use and improves. The Arduino and countless clones are dirt cheap, the market is flooded with crazy cheap open source products and derivatives of these products, because they're open source you can build your own copy of them as I have done in numerous cases. ARM boards, FPGA boards, modules and widgets, all sorts of low volume hobbyist projects and kits, it's virtually all either much cheaper than a commercial solution or it does something that is not available in a commercial solution at any price.

Why would I NOT expect open source to be cheap? If I can't save money then why on earth would I go with a half baked open source project vs a polished, completed, supported commercial product?

[2N3055]

I don't know why you would expect open source to be cheap.

A well funded manufacturer may produce things in millions, and cost per piece will be a fraction of the production of a single item. The price of the item on the market is determined by supply and demand. Most of what you pay for commercial things is to cover sales and management. Production is a smaller portion, R&D is even less. The company's profit may be a tiny fraction of the cost, may even be negative.

With open source you get free R&D, no sales, no management, no profit to anyone. All you need to pay for is production (but in small quantities).

How's that even possible to compare the two?

Open source anything should either be very inexpensive or free.
Otherwise there is no reason for it it to exist.

Otherwise it just that some first world hipster wants me to pay him to have a hobby, that nobody will benefit from except him...
And companies that are manufacturing stuff at prices nobody else is stupid enough to pay for.

Simple as that.

That is why Open Source software is possible and can be beneficial to humanity. Many nice people are using their free time to make software that is free to legally install and use to accomplish something they wouldn't be able to afford otherwise. All this human time (engineer hours) is donated by kind people, or even companies (when it helps them or makes them look good or also sometimes purely as goodwill). End users get a usable product (sometimes as good as paid one, or if not as good, good enough for many purposes, aka better than nothing.) and it all works because price of manufacturing copies (replication and distribution) in software is very low. So R&D donated, manufacturing almost free. And it works because it is hell of a deal for the money.
Software people don't like it when someone say it, but software is easy.. No, wait, not fair. It is very hard to make and make right and it not easy. But it is more straightforward to test and easier to fix.
You just apply patch and problem is gone. If after a year, developers realize architecture is wrong, the can (and they do) make a new major release. You download and upgrade. Life goes on.

And that is why Open something in hardware doesn't work.  Hardware mistakes cannot be patched from online update. PCB's need to be respined. Components on bad one cannot be reused.
And once it is made to work nicely, you cannot add 2 more channels just like that. Pretty much you have to start from beginning. Than EMC. Even if you don't care if you radiate like crazy, you have to make it resilient to outside EMC. Otherwise it will be unreliable and won't work properly... Measurement instrument will inject crap into signal you're measuring, all kinds of spurs... That won't be properly characterized and documented so you can expect them in results you you know it's not DUT.....
On Open something hardware, only firmware is open, and that has it's limitations, because embedded processors have limited resources... So you can fix some problems, even redesign few little things but major new release means new PCB and parts for 1000 USD. Old one is electronic waste.

All of the Open Something hardware that exist and is somewhat useful has fixed hardware that is much cheaper than commercial counterparts, and only Open part is software (firmware and PC part).
And many time hardware is cheaper because it has hobby grade specifications, that are great for hobby and education. Great example are many SDR modules out there.
And they are great, because there are tens of thousands people in the world that are given opportunity to learn about SDR on something they can afford.

Like James said, if it is not much cheaper that real thing,it shouldn't exist.  Otherwise we are all simply financing some hipster to have a hobby. No, thanks.

[NorthGuy]

If I can't save money then why on earth would I go with a half baked open source project vs a polished, completed, supported commercial product?

Because you can tweak it. You can get the files, change them as you like, build your own. That's the only difference between open and closed. If someone sells you an assembled scope board for end use, then you're only interested in its characteristics and the price. Why would you care if it's open or not? Would you tell the seller that he must drop his price because the hardware files are open?

If you buy an Arduino board - it's a commercial product. Someone invested money to buy parts, build boards in huge quantity, and now he's selling them to public in small quantities for profit. He could have taken an unmodified open hardware project, or he could have modified it, or he could design entirely his own. Either way, the design expense is a fraction of the overall cost, and there's no difference to the end user.

[james_s]

Who wants to tweak their oscilloscope? Being able to modify it is a nice bonus, but I cannot think of a single successful open source product that costs more than a comparable commercial product. Open source succeeds primarily because it is cheaper than another option, and the ability to modify it yourself partially makes up for the fact that it is nearly always not quite as polished as a commercial off the shelf solution. The open source device you can tweak to reduce the deficiencies (if you are the small percentage who have the necessary skills) vs the commercial product you pay more but you get a finished, tested, supported product that does not have those deficiencies.

Can you think of even one open source project that is popular despite costing more than closed source products that do the same thing? I can't, but if there is one I'd be curious to know about it.

[NiHaoMike]

Can you think of even one open source project that is popular despite costing more than closed source products that do the same thing? I can't, but if there is one I'd be curious to know about it.

Prusa 3D printers have been and still are popular despite there being lots of cheaper printers available.

[james_s]

Is there a comparably priced commercial off the shelf 3D printer that works as well or better? I haven't been following hobby grade 3D printing recently but what makes the Prusa printer popular? Is it the fact that it's open source, or does it do something that a commercial solution can't do or do as well?

From what I have seen, 3D printers are generally speaking another example of a low cost hobby product, they don't work as well as megabuck proprietary industrial 3D printers but their open source nature has enabled them to evolve into a product that is good enough while being affordable and accessible. Once again their success comes down to the fact that they are (relatively) cheap.

[Mechatrommer]

Can you think of even one open source project that is popular despite costing more than closed source products that do the same thing? I can't, but if there is one I'd be curious to know about it.

Prusa 3D printers have been and still are popular despite there being lots of cheaper printers available.

i guess people who bought original Prusa were comparing it with much more expensive unit, but dont want the hassle to tweak/modify nor risk of non working printer, they want some assurance/guarantee/warranty from the seller, this is a portion in the market segment. but for some people with "some" knowledge or willingness to tweak they will buy much cheaper knock off prusa/china/open source unit. and from my limited historical background reading, Prusa maker is among the early mover on this OSHW product (based on Adrian Bowyer RepRap project) which is cheaper than the now bought MakerBot etc where no other knock off brands in competition, so he gathered some "resources" already, so he can continue on.

[NorthGuy]

Who wants to tweak their oscilloscope? Being able to modify it is a nice bonus, but I cannot think of a single successful open source product that costs more than a comparable commercial product. Open source succeeds primarily because it is cheaper than another option, and the ability to modify it yourself partially makes up for the fact that it is nearly always not quite as polished as a commercial off the shelf solution.

I don't understand the juxtaposition. Lots of commercial off the shelf solutions use open source hardware.

... the commercial product you pay more but you get a finished, tested, supported product that does not have those deficiencies.

Looks at the "Test Equipment" forum section. Commercial solutions are anything but polished. There are software bugs, firmware bugs, hardware problems, reliability problems, usability concerns, limitations etc. etc.

Can you think of even one open source project that is popular despite costing more than closed source products that do the same thing? I can't, but if there is one I'd be curious to know about it.

"Popular" products are mass produced and thus cost less. This has nothing to do with the open source-ness. Say, Arduino is popular and cheap. If you decide to produce and sell Arduino in small quantities, your product will be terribly expensive, or you will go bankrupt.

[nAyPDJ]

I've built the v1 of this scope, which has 16 digital channels but just 100MSa/s. I bought the parts for 2, and split the cost with a friend, so it came out to $239.38/ea.

It took me about 7 hours to solder & assemble the scope. I was inexperienced with soldering, so your results may be better.

The software is kinda clunky though, so while I've tested it out and it's worked fine, I've mostly ended up using an old digital scope that I bought off of ebay for about $200. However, I do still keep it around in case I need a logic analyzer.

[HB9EVI]

for 200 +/-50 bucks I'd give it a try since it's open source

good work I'd say if you're not so experienced with assembly