500 MHz 8b 2GS/s oscilloscope

[EEVblog]

Be shore that I know what I talking about, please believe me that I am qualified.
I know that sound impossible for one man to know and do all the stuff… I know.

It shouldn't matter whether anyone believes you or not, the only person who is going to do this project is you.
There is no need to get anyone's approval or endorsement, just do it!

Believe me, if you produce such a thing a lot of people will be interested, but until then you are just yet another person who wants to make their own oscilloscope.
But it would be wise to get people's input on things, like possible price, features, interface, form factor etc, and this forum a great place to ask for feedback as the design progresses.

Dave.

[Bored@Work]

I still have not understood what you want from "us" - the forum readers.

As a general remark, to successfully address potential business partners and potential development partners, you might want to work on the way you pitch the project to them. Growing some thicker skin might also help.

[petar]

>>Believe me, if you produce such a thing a lot of people will be interested, but until then you are just yet another person who wants to make their own oscilloscope.
>>But it would be wise to get people's input on things, like possible price, features, interface, form factor etc, and this forum a great place to ask for feedback as the design progresses.

Exactly as you say.
I am going to work (it is hard...).

[petar]

I calculated BOM price for those interested.
These are single unit DigiKey prices excluding taxes

Semiconductors and connectors 907$   All parts excluding resistors and caps exact value
Resistors and capacitors             100$   I didn’t calculate it but there are lot of them…
Box       aluminum                        27$
PCB 8 layers                               100$   That’s my thought it will be very complicated…
Box   CNC  routing                       10$ 
USPS                                             40$  United States Postal service
                            BOM   SUM  1184$ 
                                 TAX  20%  237$
                                  TOTAL   1420$

So 1420$ would be the single unit price for BOM and all excluding assembling.

I expected that can be cut to half for lets say 1000 units…
And there are many DC/DCs so if I made one larger DC/DC with multiple outputs by myself then it could saved a lot, but for the first prototype I don’t want to do such optimization.

Petar, 

[0xdeadbeef]

Just my 2cents: analog bandwidth and high sample rate are not everything for a good scope. The user interface, measurement/analysis options and customer support are also important - at least in an industrial environment.

E.g. at work, we have some scopes from the Agilent 7000A line and some LeCroy WaveRunners. They are approximately the same price range (10k€ to 14k€) and have similar technical specs: 4GSa/s, 500 to 600MHz bandwidth, the Agilents have 8MSa memory and the LeCroys have 12MSa up to 32MSa memory depth.
Now still, there's no way the Agilents can compete with the LeCroys for more sophisticated jobs. If you need high precision period measurements, advanced statistics, multi stage triggers, jitter analysis or anything that is beyond the simplest of measurements, the Agilent just can't do it. You can't even buy it as an option, where you could easily spend another few k€ for helpful addons on the LeCroy. It's hard to estimate how much time and money went into all this advanced functionality.
E.g. did you ever try the WaveScan feature on a LeCroy?

Honestly: in the industry, it doesn't really matter so much if a scope costs 5k€ or 10k€, especially if the 10k€ scope helps you to find issues earlier.

So while I'd kinda trust you that you are able to design the hardware part of a modern scope that is comparable  to a state of the art device from the big names, I have my doubts that you can really deliver the same quality regarding GUI and measurement/analysis features. Especially if you want to do it all from the FPGA. There's a reason why the scopes with complex GUI and features use a full OS (Windows most of the times) with all its drawbacks.

[tom66]

If you are targeting hobbyists then I see no reason you'd need to go above 100 MHz or even 50 MHz. A sample rate of at least 5x per channel, and preferably a way to combine two ADCs into two channels to get 1 GS/s for one channel. Like Rigol do.

Even most average electronics stays below 200 MHz. High speed FPGAs and CPUs - sure they could be at 300 or 500 or more MHz, but if you can afford to play with them, you can probably afford to buy a 500 MHz scope .

Make a nice UI, add a few bells and whistles (like a 16 channel logic port) and if you can produce it at a decent price, you'd have a winner.

[petar]

Yes my targets are amateurs and my thoughts is that more and more people will use FPGAs especially if they could afford scope which is currently beyond of their $ capability.

It is hard to beat Chinese, no way to be cheaper than them in 50-200 MHz range.
I personally think that their scopes are crap, but for that money you just can’t be better.

Each price range has its buyer you just need to have better price performance ratio.

I will certainly add all of fancy things that people think it is hard to do (that includes SPI, I2C and other serial bus decoding, FFT, pp measurements, averaging, frequency, adding, subtracting, glitch finding, signal frequency sudden changing position, rise time sudden change position and all that I could remember of course all in FPGA). That’s because people are under big companies advertising influence… and brain washing.

It is hundred time better that you have real knowledge of how things behave then to have oscilloscope. I don’t say that I wouldn’t use oscilloscope if I can afford it but I say that every electronic job can be done without oscilloscope. How people think that lets say 20 GHz oscilloscope is born? Do engineers that design it have 100 GHz oscilloscope to debug it? Knowledge is power.

[alm]

So while I'd kinda trust you that you are able to design the hardware part of a modern scope that is comparable  to a state of the art device from the big names, I have my doubts that you can really deliver the same quality regarding GUI and measurement/analysis features. Especially if you want to do it all from the FPGA. There's a reason why the scopes with complex GUI and features use a full OS (Windows most of the times) with all its drawbacks.

I'm not so sure if that's the primary reason for choosing a commodity OS. I think it's more about connectivity and familiarity. I don't really see how programming a decent GUI and analysis features is going to be easier on Windows than on a lighter OS that doesn't take forever to boot. I agree with your point about not putting it all in the FPGA, however.

If you are targeting hobbyists then I see no reason you'd need to go above 100 MHz or even 50 MHz. A sample rate of at least 5x per channel, and preferably a way to combine two ADCs into two channels to get 1 GS/s for one channel. Like Rigol do.

I wouldn't try to compete with the companies like Rigol, your production costs are going to be much higher. In the higher (> 200 MHz) segment, the cheap Asian manufacturers have very few products, and the big boys like Agilent and Lecroy have much higher profit margins.

Even most average electronics stays below 200 MHz. High speed FPGAs and CPUs - sure they could be at 300 or 500 or more MHz, but if you can afford to play with them, you can probably afford to buy a 500 MHz scope .

Don't confuse repetition rate with bandwidth. A lowly ATmega328p might be limited to 10 MHz signals, but the expected rise time could be something like 3.5 ns, requiring a scope faster than 100 MHz to accurately characterize them.

[0xdeadbeef]

I'm not so sure if that's the primary reason for choosing a commodity OS. I think it's more about connectivity and familiarity. I don't really see how programming a decent GUI and analysis features is going to be easier on Windows than on a lighter OS that doesn't take forever to boot. I agree with your point about not putting it all in the FPGA, however.

Sure, there are other reasons for choosing Windows, e.g. MatLab connectivity etc. and of course it would be possible to program complex, touchscreen or mouse driven GUIs with your own mini-OS, but fact is that scopes without a big OS underneath tend to have simple GUIs, less features and no touch screen control.

I will certainly add all of fancy things that people think it is hard to do [...] That’s because people are under big companies advertising influence… and brain washing.

Of course everybody who needs a protocol decoder (like SPI, I2C etc.) is well aware that a decoder is not rocket science and that adding statistics, jitter analysis or whatever to simple measurements is more or less trivial as well. So of course the prices for these decoders and options are much too high from a consumer perspective. Then again, from an industry point of view paying 1000€ for an option is not that much if you think about what an engineer costs per day.

It is hundred time better that you have real knowledge of how things behave then to have oscilloscope. I don’t say that I wouldn’t use oscilloscope if I can afford it but I say that every electronic job can be done without oscilloscope. How people think that lets say 20 GHz oscilloscope is born? Do engineers that design it have 100 GHz oscilloscope to debug it? Knowledge is power.

Well, as always, it's usually more a matter of time and costs than lack of knowledge. Without proper debugging tools, finding sporadic issues in a complex project can get very expensive. And developing specific tools isn't cheap either.
Besides, scopes are also used for documentation, testing against specifications, quality assurance etc.

Anyway, I wish you the best for your project, though I honestly think that it might be too much for one single man to develop the hardware, the software and do the marketing.

[petar]

Thank you, 0xdeadbeef.
For equipment repairs or characterization and completely unknown signals I agree, one must to use oscilloscope, for development no.
Don’t underestimate FPGA power, especially for signal processing.
But FPGA is also peace of junk in wrong hands, but miracle in FPGA guru hands.
DSP processor is junk in comparison to FPGA.
Engineers doesn’t know to use FPGA right way and FPGA manufacturers also want that their client to be let say not smart, to use theirs cores… that are far from optimal… of course they don’t admit it even opposite… they love when something “can’t” fit in specific device and when user need more gates… more gates more…
User interface should be simple, but complete and useful.
I will not going to emulate processor and C programming way of implementing mouse control nor I will follow MPEG2 for display and that doesn’t mean that the screen will be crap. That only means that I don’t play movie on the screen, nor the photo of nude woman.
Processor and software driven user interface is inefficient because it support wider range of feature than is needed for the job and that’s why it is so slow, slow to make me nervous.
Petar, 

[EEVblog]

Just my 2cents: analog bandwidth and high sample rate are not everything for a good scope. The user interface, measurement/analysis options and customer support are also important - at least in an industrial environment.

I think you ultimately need both to have a successful product.
If you don't get the bandwith and sample rate good enough, there is far too much competition at the low end in which you'll never be able to compete.
But have spectacular bang-per-buck in the bandwidth/sample rate/memory department is not an automatic guarantee of success either, if your user interface or and/or software sucks.
It's a fickle business!

Dave.

[petar]

Guys,
If anybody of you knows does all that features like wave scan from LeCroy or other from other “big” players is patented.
Of course that I have no access to their internal documents and that I all design by myself and besides it is such a stupid thing…
But I know that companies like that patented air that we breed…
And I have another question.
Long time ago I wanted to my “imaginary” firm have name OmniDAQ (that name you can see at photos of my previous card…) and I owned omnidaq.com but I didn’t pay attention… and after few years somebody else hold that domain….
About few years ago I had been searching for available firm names (that isn’t total crap) and found that name GigaDAQ can’t be find either as a domain or on any document on the google.
I bought that domain almost two years ago. So gigadaq.com is mine, at least for now.
Few months ago I found on google that:
This is a brand page for the GIGADAQ trademark by OPTIM ELECTRONICS CORPORATION in Germantown, MD, 20874.
But at the later text says:
Word Mark: GIGADAQ Status/
Status Date:
ABANDONED - NO STATEMENT OF USE FILED
5/12/1998
Serial Number: 75190150 Filing Date: 10/30/1996 Registration Number: NOT AVAILABLE Registration Date: NOT AVAILABLE
Does it mean that the OPTIM corporation applied somewhere 1996 for the gigadaq mark and they are abandoned for that mark 1998 because the mark they are claimed actually don’t exist and they just wanted to have that mark to let’s say sale to somebody …
What is your opinion.
Thanks, Petar

[joelby]

If you're concerned about registering and defending trademarks, especially in international markets, you should talk to an intellectual property lawyer. In general, you shouldn't ask for nor rely on any legal advice you obtain online.

[philaburns]

My first thought is that maybe Optim applied for the trademark in the US - where it is essential to also file a Statement of Use indicating the product and/or services that the mark is intended to be used on - and then decided to use another name, therefore this trademark wasn't required and thus they let it go abandoned.  It is not uncommon for companies to file trademark applications for a number of different words/marks and then make a final decision later (possibly one example of in-house council being proactive and trying to stay a couple steps ahead of management who make the final calls sometime in the future...?). Not all countries worldwide require this statement of use to be filed with the trademark application (the US is one of only a few that do require it!) therefore, the particular name you are after may be registered in other countries, although my initial thoughts are that at least in the US, it is probably clear (subject to the usual disclaimers (see below)!).  I did also have a quick look at the Aus TM Register and, while "GIGADAQ" doesn't seem to be registered, there are a few marks that use the "DAQ" sufffix eg "NASDAQ" which 'could' cause issues of similarity.

Note that owing a domain name x.com or whatever or even a registered business name does not give you ANY intellectual property rights to the name! Only a registered trademark in the relevant countries AND for relevant classes will do this!

I suggest that if you want to look into this further you should engage a professional registered trade mark attorney to conduct a thorough search of the trade mark registers for the countries and products and/or services you are looking at and to give you some definitive advice.  If you want me to recommend  some good ones in Oz, send me a private msg.

Good Luck
Dr Phil

Please note my comments above are based only on a cursory review of this particular situation and DO NOT CONSTITUTE LEGAL ADVICE.  If you require anything further, please speak to your trade mark attorney.

[EEVblog]

I did also have a quick look at the Aus TM Register and, while "GIGADAQ" doesn't seem to be registered, there are a few marks that use the "DAQ" sufffix eg "NASDAQ" which 'could' cause issues of similarity.

And of course in this industry there are plenty of "DAQ" names, like of course National Instrument NI DAQ
And there are other SI units DAQ companies too:
NanoDAQ http://www.microsense.net/products-position-sensors-microsense-nanodaq.htm
MicroDAQ http://www.microdaq.com
MegaDAQ http://www.megadaq.com/
So you could argue there is precedent there, if that counts? So GigaDAQ would be likely lost in the noise.
I suspect National Instruments would be the only big corporate fish that would possibly care.

Thanks Dr Phil, EEVblog's resident patent and trademark attorney, who's advice only magically becomes good once you pay him! 

Dave.

[petar]

Thank you guys.
I didn’t have a time for earlier answer.
I definitely can’t have a legal advisor…
But I think that I will give up from GigaDAC and made a new name, one that no body on the planet (if that one isn’t crazy) can possible complain that has any connection with.
And I personally think that firm name in my case isn’t so important.
Thanks,
Petar
I apologize to those of you who think that I have just wrote nothing but I had to say thanks at least to guys that answered me.

[bartek]

Wow sounds like an awesome project! Also great to see you having so much passion, I know trying to create a business is not easy.

I'm not sure if its the language barrier (or just me) but I'm a bit confused about your intension for this thread, it doesn't seem like you want advice...

But anyway here is some advice ive been collecting for this area:

• less features than your competitors but do them really well
• decide what specific thing differentiates your scope from every other scope
• partner with someone who knows how to do accounting/sales/business plan
• improve your "sales pitch", overall it felt quite negative and could turn away people who potentially would want to help you
• be ready for criticism!

[bartek]

IMO there is room to innovate here in form factor, like for example making it an oscilloscope "backpack" for a tablet perhaps? With real knobs and buttons on the sides perhaps?

This is an awesome idea, an ipad app could do a better interface than pretty much any current scope

[ThunderSqueak]

IMO there is room to innovate here in form factor, like for example making it an oscilloscope "backpack" for a tablet perhaps? With real knobs and buttons on the sides perhaps?

This is an awesome idea, an ipad app could do a better interface than pretty much any current scope

People are starting to do this, http://www.oscium.com/index.php?q=products/mixed-signal-oscilloscope-imso-104/ is an example, bandwidth isnt there and I have yet to see one with physical knobs on it.   




for the same price as this stupid dongle you could buy this

[tekfan]

IMO there is room to innovate here in form factor, like for example making it an oscilloscope "backpack" for a tablet perhaps? With real knobs and buttons on the sides perhaps?

This is an awesome idea, an ipad app could do a better interface than pretty much any current scope

It always sounds like a good idea but in reality knobs are just so much faster to use than any touchscreen.

[Mechatrommer]

IMO there is room to innovate here in form factor, like for example making it an oscilloscope "backpack" for a tablet perhaps? With real knobs and buttons on the sides perhaps?

This is an awesome idea, an ipad app could do a better interface than pretty much any current scope

It always sounds like a good idea but in reality knobs are just so much faster to use than any touchscreen.

touchscreen is better at clicking button/navigate menu and scrolling perharps, imho. but for changing value/level/volume, nothing beat old stye rotating knob. i always shit on digital push up and down volume controller esp on car or radio player. for the touchscreen, i only concern its durability in professional or industrial environment, one eg we will not want the screen to be messed with black oily fingerprint.

[Mechatrommer]

I calculated BOM price for those interested.
These are single unit DigiKey prices excluding taxes
...
BOM   SUM  1184$ 
TAX  20%  237$
TOTAL   1420$

well, last time i checked GHz sampler ADC in digikey, it cost a fortune. you want to do 20GHz sampling? you sell me the adc alone i will not be able to afford that. i admire your capability though. the big PC screen and mouse is nice idea (i even have thought about the mouse interface a while ago), about the screen, i'm thinking building my own api like interface, except in serial spi etc comm from dedicated "monitor/UI controller" mcu to/from the front end/fpga/mcu, but thats just imagination/empty talk for now (i believe someone has done that, such as 3d engine, graphix etc)

[petar]

Thank you guys.
Sorry for delayed answers but I had no time.
Bartek, I actually don’t need any advice on how-to implement something I know all the aspects of the design, but I do need feedback if monitor screen and mouse is from your (all of you) point of view good solution for the user interface. I would personally like to design traditional bench scope but as I said I can’t find reasonable price for the 10” to 12” TFT screen and larger screens are even cheaper…
Before I posted this topic I was also thinking about user interface and the idea of having the traditional knobs along with standard PC screen was a bit strange to me because the “keyboard – knobs” is separated from the screen and that’s how someone would have a trouble using it… you know multiple function tasters have it representation on the screen, so I then thought to add some small and cheap graphical display close to multi functions tasters… but it look to me as not good idea, at the end all the tasters – knobs etc, graphical display… PBC for all of that … and I ended up with mouse – screen combination.
I personally think that the combination is good enough and if I make really fast response I think it can be for some things even better that knobs, also there is a while that can act as a rotating knob, so one would have to move the mouse pointer to some position and for example to scroll the while for a level or gain control.
All the instruments that I saw have a slow response to user commands I think if I make a real fast user interface it could be nice to user even in that combination of mouse-screen.
Mechatrommer I am sorry if you can’t afford it, I am struggling for my freedom and I want also to have peoples on my side and if I could make a ADC myself I would make a scope for couple of bucks.
I have also many ideas (that of course isn’t all mine) that can make each person life independent of world economy so no body needn’t to be afraid if he loose the job… and I would like that things lasts and to be good and not afraid of loosing future $ when I have certain amount that is less that many of you thought I don’t need any more $  (and of course I don’t want to collect $, I want to spend it on something that lasts).
That’s how I am not afraid if my future business puff out because I made too good thing that lasts and it is cheap (all of that feature would certainly make further business impossible).
Imagine what could be done if all the engineers around the world instead of fighting with each other have its personal goal that isn’t in opposite direction to all other. The efficiency would be better maybe 100 times and I can’t imagine what we could do.

Thanks,
Petar

[amspire]

Petar,

Really interested in your ideas. I am concerned that you seem to be looking at the something that is head-to-head with the mainstream products, and your bill-of-materials is very high. Even though you may be able to offer a bit more bandwidth, it seems you will be head to head with the new Rigol 2000 series with its great display and impressive features. And they did promise an "incredible" pricing for the range.

I was really wondering whether there are some more niche products that you may be better off with at the start, particularly if you can start with a cheap BOM.

1. Isolated inputs.  This is probably not easy since people will want to use them with switching power supplies. If you have a FET switching 400V in 20nS, that is something like 20 billion volts per second slew rate. If the isolated circuit is powered by a transformer with a 30pF interwinding capacitance, you can still get a 0.6A transient current through this capacitance.

2. Modern Version of the BWD powerscopes. 4 x 600V AC differential inputs. Bandwidth 20MHz or more. Even without much bandwidth, a scope with two separate differential input leads per channel is just incredibly useful. Most of the time, it is much better then the high speed differential probes that are difficult to connect to circuitry.

3. Following on from the idea above, an differential probe like the Fluke DP120.  Again, fantastically useful, if only they didn't cost about the same as two Rigol 100MHz scopes. If you could make a $99 probe, you would have a great product. If you could make a $49 probe, it would be a killer product. There is the potential for a very low BOM, so it could be a great way to start.

4. A 12 bit to 16 bit sampling scope. Bandwidth maybe 1GHz or more. 50 ohm 5V input.  Sampling scopes are never a replacement for real time scopes, but for the occasions when people do need the speed, they often can be made to do the job. They are great for single frequency repetitive waveforms, or looking at timing windows on high speed digital waveforms driven by a common clock.  Also there is the potential for the use of very cheap parts, and the A/D can run at a fairly slow speed, so you can use whatever resolution your circuit can handle. It may be possible to sample with a circuit based on two or four cheap 350pS 0.3pF Schottky diodes. you may have to match your own pairs of diodes. I read recently from a sampling scope designer that you can get up to 10GHz bandwidth using standard off-the-shelf parts (if you know what you are doing). At high frequencies, 50 ohm home made divider probes have higher impedance then passive probes that plug into a 1Meg scope input. It is not much use offering lots of bandwidth if the loading of the probe is going to change the waveforms anyway.

One of the huge things extra resolution gives you is much better FFT results. A high resolution sampling scope would be great for people who cannot afford or justify a spectrum analyzer.

Last thing -  LXI looks like being the new LAN instrumental communications standard. LAN is way better then USB as it has up to 1500V AC isolation, and you do not need a PC on the same desk. You could run the oscilloscope from a smartphone if you wanted using WiFi.

I haven't looked to see if there is a cost, but if you can get you product name on the LXI registered instruments, then your name would get out.

http://www.lxistandard.org/Products/ProductList.aspx

Richard.

[EEVblog]

Bartek, I actually don’t need any advice on how-to implement something I know all the aspects of the design, but I do need feedback if monitor screen and mouse is from your (all of you) point of view good solution for the user interface. I would personally like to design traditional bench scope but as I said I can’t find reasonable price for the 10” to 12” TFT screen and larger screens are even cheaper…

And therein lies the big problem with doing your own scope, and why there exist essentially two camps of devices.
Dedicated hardware scope with real knobs and a built in screen, and fully PC based scopes. You hardly ever see devices that require an external monitor and mouse, and there must be good reason for this.
I've said it before, but will reiterate. IMO a scope that is just a box and needs an external screen and mouse to operate is going to be a very disliked user interface. Why? Because whilst those external screens are cheap and readily available, no one wants one on their work bench taking up space. If they already have a notebook on their bench, then they will want to use that, not to mention the mouse as well.

Also IMO, unless you can come up with some novel form factor, you are much better off making your scope USB PC based. You might hate that, and think it's slow or whatever, but it will appeal to a lot more people than a box that needs it's own screen and mouse to drive it.

Dave.