SparkFun Forums 

Where electronics enthusiasts find answers.

Have questions about a SparkFun product or board? This is the place to be.
By Dorje
#148341
Hello everyone. I'm a first time poster but a long time lurker; I really do enjoy the threads here. The whole sparkfun forum community seems really thorough and helpful and responsive. Anyway, a bit about myself: I'm a mechanical engineer with a startup engineering firm. One of my primary offerings is a "system tune up" for an industry specific machine. Long story short, I need to analyze some fluid mechanics and piston position sensor data with some PC software. The trouble is, the entire piston stroke is only about 0.002 seconds long total.

I'm looking for some kind of plug and play setup that can read a position sensor and a pressure transmitter (both analog, 0-5V) at about 40kSa/s or more, giving about 160 data points per event minimum, and save that data. I am on a tight budget right now for this system. I'd like to keep it under $200, but up to $250 is passable.

I've looked into buying oscilloscopes; correct me if I'm wrong, but they should be able to take that data, even if it's not a constant waveform, right? Ideally, I would like to be able to link the start of the data collection to the start of the cycle (just hit go and it starts collecting for a second or however long), but that's not really necessary. Saving the data as ADC output (0-1024 corresponding to 0-5V like on Arduino) or voltages or something else I can easily scale would be great, and so would saving in csv or excel file format.

I originally liked the DSO Nano V2 for this purpose, but it is only one channel, and people have a lot of complaints about the interface and glitches. The same goes for the DSO Quad 203, though I may be talked into buying that. I would prefer to not have a steep learning curve for the system/scope. The only scopes I've used are the $10-15k ones from University and from the engineering firm I worked at for a few years.

I've heard people suggest getting an older, used scope in this price range that will have higher sampling rates, but I haven't seen any that work with modern data storage protocols (USB, SD cards, etc).

I imagine (as is the case with dozens of electromechanical projects I've done) that there is some kind of system I'm just missing entirely. For example, I have a tendency to use an Arduino UNO to make Up/down counters. I know that there are much better, more cost effective ways of doing this, but the familiar (and prebuilt!) Ardy system saves me tons of time getting up and running.


TLDR:
Need 40+kSa/s on two (plus?) channels (the more the MUCH better, this is bare minimum for limited budget)
Need accurate time correlation
Need to be able to save the data
$250 budget firm, <$200 preferred
Plug and play; please don't ask me to program more microcontrollers - just don't have the time!


Thank you all in advance for your time. You are the first ones I'm reaching out to because I know how knowledgeable and skilled you are at what you do here.
#148406
I'm not sure you are looking at this quite the right way. For a startup engineering firm this is probably a long term investment. Not a one time expense. If you have to spend a little more well shucks.

If you need 2 channels you could buy two of the DSO nano V2's or spring for the quad. As far as learning curves go. There is always going to be one. If you already know the basics of using an O-scope then I think you won't have a whole lot of problems with whatever you buy. I see on amazon the DSO2090, 2 channels less than 200 bucks. That could be your cup of tea.

I know one of my professors was working on an O-scope in the sub $100 range that ran on the PC. He was tied up in legalities with the university, so I don't think it was ever going to see the light of day.
#148427
I normally like to start any search by finding out what actually exists, then filtering by other criteria afterwards.

I'm inclined to suggest that this sounds like quite an important piece of capital equipment, and that there needs to be an adequate budget available to purchase whatever tool is needed to do the job. That could be a PIC, a handful of passive grit and a piece of stripboard for £10, or a high end storage scope for £10000.

I doubt even an old storage scope that's up to the job will be available within budget. A scope like my Tek TDS754D would probably do the job (provided you have patience and a floppy disc drive), but it's still about 10x too costly.
#148428
Wow, as I expected, this is the right group of people to ask. Your responses were all very helpful; I really appreciate it.

Grimfox, that is a very fair point. There are many factors limiting the budget right now for this system, but primarily, the goal of the project is to have more of a "plug and play" testing system, where we can plug the sensors into the machine and effectively hit go. If we pass around one high end oscilloscope, while it is feasible (routing a bunch of BNC connectors and switching them out, for example), a dedicated system is currently the best option. If it comes to it, we may very well just hold off until the budget allows and make that investment.

As for the DSO Nano/Quad, the more I read, the less I like. There are lists of posts here and here and Amazon and a few other posts like mine looking for entry level scopes/data loggers where people advise away from them. Of course, if anyone has intimate knowledge of them, and says that it can do what I need easily, and the 3" screen isn't an issue, I suppose I could be swayed. Most of the features people complain are missing are things I've never heard of, and I don't need to generate signals anyway.

Looking at the picotech site, the only data loggers that fell into my price range went for about 1 kSa/s per channel, 0-2.5V, so those are out. I went through the scopes, and at ~$260 (before tax and whatnot), the PicoScope 2204 could have what I'm looking for. 10 Mhz bandwidth (how much do you think I would need? is this just dependent on resolution of the samples?), 50 MSa/s for two channels (way beyond spec, actually), 8 kSa buffer, advanced triggering, runs/saves data to PC. Honestly, I don't know what most of the other specs mean. The accuracy seems a bit low though - +/- 3%?

The Diligent Inc Analog Discovery is pretty interesting. On first glance it has the 50 MSa/s for two channels, 5 Mhz bandwidth, appropriate voltage ranges and is actually under the conservative budget before tax. It sounds just like what I'd be looking for. Do you guys see anything I'm severely overlooking?

Another question that has come up during this project is what's the difference between using a scope and using a data logger in this case? I'm not generating functions or waveforms, or even re purposing the item any time soon. Should I be looking for a "DAQ" instead of a scope?

AndyC_772, it is a point well made, and you posted while I was typing this, so I know you haven't seen this post, but again, this is to be used more as a dedicated system with a limited budget instead of the primary scope for the whole R&D section.

Again, thank you all for the help. I really do appreciate it.
#148430
I felt much the same about the DSO Quad. It looks like a really nice, cute little piece of kit, and clearly most of the hard work has already been done - but by all accounts the firmware just isn't as complete or reliable as it needs to be. I'd have bought one without a moment's hesitation if it didn't have any major issues, but I hear a lot about triggering problems. Sadly my R&D time has to be spent on paying customers rather than fixing the tools of my trade, so it's not really an option for me.

40k samples/sec is slow for a 'scope (though they'll all do it), but fast for a data logger. If the event of interest is .002 sec long, and you're sampling at 40k, I make that just 80 data points - and that's well within the record length of even the most basic storage scope. Unfortunately even the most basic scope is likely to be over budget, though - the cheapest new one I can find is this one:

http://www.sjelectronics.co.uk/acatalog/GDS1042.html

I'm sure there's plenty of others out there too - Ebay is your friend.

What about something like this?:

http://www.ebay.co.uk/itm/Hantek-DSO-21 ... 19cefcdbd5
#148632
Should I be looking for a "DAQ" instead of a scope?
Definately yes. Pretty much any decent DAQ would be sufficient, and simpler than using an oscilloscope.

As for which one, there are many. The Digilent one troposphere suggested is good. National Instruments is the default source for many, but they're somewhat more expensive, if more fully featured and supported. Two other options are this and this. There are many more inexpensive DAQs available, but not that many with 50ksps/channel capability for <$250.