Author Topic: New Analog Devices/Digilent Analog Discovery board  (Read 18884 times)

0 Members and 1 Guest are viewing this topic.

Offline joelby

  • Frequent Contributor
  • **
  • Posts: 624
New Analog Devices/Digilent Analog Discovery board
« on: May 14, 2012, 09:30:08 PM »
Hi all,

I just saw that Analog Devices/Digilent have released a new "Analog Discovery" board for USD 99 (US student price), along with a couple of other interesting things at http://www.digilentinc.com/Products/Catalog.cfm?NavPath=2,842&Cat=17 .

The specs:

Dual 14-bit 105 MSPS ADC
Dual 14-bit 125 MSPS DAC
16 digital I/Os at 100 MSPS
Programmable power supply

It looks like it's designed to be an oscilloscope/AWG/logic analyser/digital pattern generator, so the usual caveats (5 MHz analogue input bandwidth) apply for such a device, but the screenshots of the software look quite nice and OS X and Linux versions are promised.

Like many here, I'm not too interested in this class of oscilloscope, but assuming it's hackable it could be the basis for a cheap software defined radio transceiver. It doesn't look like a schematic diagram is available, but Digilent often provide them.. we'll have to wait and see after it's released.

Offline rbola35618

  • Regular Contributor
  • *
  • Posts: 238
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #1 on: May 15, 2012, 04:32:43 AM »
Digilent is based out of Pullmen Washington. The company was founded by Washington State people. This is a very cool piece of equipment for the student to have.  I know that my Professor at University of Idaho Ken Noren was using it for his electronic classes. He said the student loved it because it allowed them to perform their labs in their dorms.


Online amyk

  • Super Contributor
  • ***
  • Posts: 2430
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #2 on: May 15, 2012, 07:33:00 PM »
Agilent
Siglent
Digilent

Interesting name. :)

Offline krenzo

  • Regular Contributor
  • *
  • Posts: 91
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #3 on: May 16, 2012, 06:12:54 AM »
I use Digilent's FPGA boards and recommend them.

Offline Smokey

  • Frequent Contributor
  • **
  • Posts: 597
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #4 on: December 10, 2012, 01:59:39 PM »
Sorry to drag up an old thread, but I got a bit of the rant bug in me this evening.

/rant on/

http://www.digilentinc.com/Products/Detail.cfm?Prod=ANALOG-DISCOVERY
I've just started researching USB based scopes that I can use as a simple DAQ system and I came across this thing.  Since the main frequencies I'll be working with are under 100khz, this looks pretty neat at first, especially for the price.  The specs they show like 14bit ADC/DACs, sound really good... at first.  Better than the 8-bits of the general purpose usb scopes like the Hanteks and Vellemans,  but... WTH is up with their documentation and connector choices???

First of all there is no comprehensive useful documentation.  This thing has been out since at least May (from this threads first post) and there isn't even a data sheet anywhere with any meaningful specs on it.  No manual, no nothing.  This cut sheet is all I could find, and it's not even on their site.  It's from Farnell.  Am I missing something?  Since their target customer is educational institutions, I hope all those engineering students don't learn that this kind of documentation is OK.  Those two page examples on the main web site are not really good enough.
http://www.farnell.com/datasheets/1639278.pdf

Secondly, WTH is up with their choice of analog signal connections?  Unshielded pin headers right next to the digital pins?
http://www.digilentinc.com/Data/Products/ANALOG-DISCOVERY/Analog%20Discovery_Pin-Out.pdf
Are BNCs really that expensive?  You can get two cheap but functional BNC 100Mhz 1x/10x scope probes from ebay for 15USD shipped.  One of the joys of being an engineer is collecting test equipment.  Having students buy 15USD probes isn't a big issue.  But one of the most frustrating things as an engineer is fighting with your test equipment.  That's not a lesson students should learn early.  I get that the thing is intended to be cheap for educational use but that's a little too cheap.  Why use the expensive ADC/DAC if you aren't going to put the front end all the way to the connectors to support it?  That brings up the next issue....

I don't see how that thing is worth the 199USD they are charging for non-EDU pricing.  Even at the 99USD student price, it's still expensive for what it is. 

Whats the deal here?  Has anyone actually worked with one of these modules?  Does it in fact ship with an encyclopedia of manuals and you just cant download them?  Is the input circuitry in fact awesome and I'm way off base?  I would hope a collaboration between Analog Devices, Xilinx, and Digilent would get it right, but who knows.

(and what is the headphone jack for?)

Here are some shots of the PCB.
http://dangerousprototypes.com/forum/viewtopic.php?f=2&t=4841&p=46801#p46801

/rant off/

Offline Engineer1

  • Contributor
  • Posts: 5
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #5 on: December 13, 2012, 12:29:28 AM »
Hello,

I felt compelled to register, just to respond to your post about the Digilent box of tricks. I bought one, more out of curiosity than anything else. Since then I've bought another 5, and passed them to colleagues. I am blown away by the capabilities of this device, even at the full $199.00 price (I'm in the UK and get them from Farnell for £127 + VAT)

For the record, I'm a professional electronics engineer (have been since 1991), developing medical devices.

I've tried a few USB oscilloscope type devices in the past, and have always been left pretty dissatisfied. The only USB instrument I actually use (and like) is the Intronix Logicport.

Your points about the connector are kind of valid, but for the vast majority of the intended audience (students and hobbyists) connection of a proper (BNC) 'scope probe isn't essential. It comes with a cable loom that is easy to connect to anything as required. If BNC connection is needed then it's not a big deal to splice up a connector. Certainly in my experience the signal integrity is fine. Sure, if you need microvolt resolution then maybe it's not the ideal tool, but it's not intended for that anyway.

I have even decided to include a pin header on suitable future board designs to allow me to just plug the AD straight on to the board for debug and test.

Your point about documentation is valid, but things are improving. When I got hold of the first device (around August, I think) there wasn't much about. However, since then more and more is appearing on the web. Here are some links that might give you more information:

http://www.analog.com/en/university/course-materials/topic.html
http://www.engr.usask.ca/classes/EE/221/lab/2012/Lab3_MOSFET%20IV%20Characteristic.pdf
http://dangerousprototypes.com/2012/12/04/analog-discovery-teardown/
http://dangerousprototypes.com/2012/04/25/digilents-analog-design-kit/
(I am the Stephen Casey who has posted there)

The software (Waveforms) was something that I had reservations about. But, it's absolutely fine for what it is. Certainly not 'Mickey Mouse'. It would be great to have LabVIEW drivers for the board, but sadly I haven't found anything on that yet. There is an interface to Matlab:
http://www.mathworks.co.uk/matlabcentral/fileexchange/38113-digilent-analog-discovery-support-for-data-acquisition-toolbox-r2012ar2012b

But I know nothing more than the detail in the link. I reckon NI are paying Digilent to not release a LabVIEW driver  ;)

Why not download Waveforms and have a 'play'? (it doesn't require functioning AD hardware connected, and will even work with the PC audio interface, within the obvious limitations of that).

I have already asked Dave to do a review. Maybe one day he will...

The fully differential 'scope inputs make it even more useful - Particularly for students I imagine, who may struggle with the concept of every measurement on a 'normal' scope being ground referenced.

I also love the fact that I can throw it into my laptop bag and barely notice I've got it with me.

So, my recommendation is to get hold of one and try it out. I don't think you'll be disappointed. And, if you are, I don't think you'll struggle to sell it!

Hope this helps to convince you (and others) that this is a great little box of tricks! For the record, I have no affiliation whatsoever with any interested party. I'm just a happy customer.

Cheers.

Steve

« Last Edit: December 13, 2012, 01:44:28 AM by Engineer1 »

Offline Smokey

  • Frequent Contributor
  • **
  • Posts: 597
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #6 on: December 14, 2012, 01:17:03 PM »
Awesome.  Great reply.  That was exactly what I was looking for.

That last DangerousPrototypes thread was the real gem with recent feedback from one of the designers. 
I think I'll wait till they release that analog teardown and noise performance report to pull the trigger though.
Its interesting that they said they are also working on a BNC adapter.  Good stuff.  I wonder how hard it would be to get the API info from the matlab interface if they take forever to release it?

Assuming the noise levels are acceptable, it looks like only issues left for me are the low output voltage of the DACs (since I need at least +/-10V but that would be pretty easy to boost up) and some sort of API so I can jack it into some Python scripts and collect some real data.  Might have a winner. 

The next step up similar solution seems to be the picoscope 4226/4227.
http://www.picotech.com/precision-usb-oscilloscopes.html

I wonder how the Analog Discovery board compares to that as far as the analog performance in the ranges where they overlap.  I know it's not really a 1:1 deal, but still would be interesting considering the huge price difference.

Anyone know of any other >8bit USB type DAQ system like this in between the Analog Discovery and the Picoscope?

Online EEVblog

  • Administrator
  • *****
  • Posts: 11504
  • Country: au
    • EEVblog
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #7 on: December 14, 2012, 01:39:29 PM »
I have already asked Dave to do a review. Maybe one day he will...

I believe there is one in my huge pile of unopened mailbag:


Dave.

Offline ftransform

  • Frequent Contributor
  • **
  • Posts: 728
  • Country: 00
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #8 on: December 14, 2012, 10:14:42 PM »
You know you are livin large when you don't even bother to open the mail ;)


But I suppose you have to open every piece on video in order to give props.

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 6401
  • Country: za
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #9 on: December 15, 2012, 02:19:37 AM »
Not enough hours in the day..... Some occupied with sleep, the 30 minute commute to work and having the kids grow up and recognising you as DADDY, not that strange guy who comes in after we are asleep and leaves in the morning.

Offline aep9690

  • Regular Contributor
  • *
  • Posts: 56
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #10 on: December 21, 2012, 02:08:36 PM »
I would be extremely interested in a review of this.  I am an electrical engineering student and I don't have the space or money for proper setup at the moment, I am considering getting this as a temporary substitute.

Online EEVblog

  • Administrator
  • *****
  • Posts: 11504
  • Country: au
    • EEVblog
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #11 on: December 21, 2012, 03:28:07 PM »
But I suppose you have to open every piece on video in order to give props.

Yep, otherwise its not a real Mailbag IMO, and takes away the excitement for all.

Dave.

Offline Engineer1

  • Contributor
  • Posts: 5
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #12 on: January 08, 2013, 11:33:04 AM »
Hello again,

Jack Ganssle has done a mini review in his newsletter here:
http://www.ganssle.com/tem/tem233.html

Hope this is of interest (although I still have my fingers crossed that Dave will surprise us one day).

Cheers.

Steve.


« Last Edit: January 08, 2013, 11:37:35 AM by Engineer1 »

Offline Smokey

  • Frequent Contributor
  • **
  • Posts: 597
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #13 on: January 09, 2013, 08:27:26 AM »
What we really need is that noise/performance report the design engineer on the DP site was talking about.  I'd expect the thing to do what it says it does specs wise, but there aren't any reports of actual numbers for things like noise floor or crosstalk.  On the DP thread there was a link to a guy that did some Bode plotting and said there was some crosstalk between the AWG and the analog inputs, but didn't say how much. 

Maybe if Dave doesn't have enough time, he could send it to the dude form "The Signal Path".  Since he deals with education he might have an interest in testing some of that stuff out and putting some numbers to it.  I think it would make a good instrumentation video on it's own, and a good product review too.  (Plus if this thing gets picked up at universities, think of all the hits the videos would get from students that are told to buy one)

That API would be great too.  I'm not too hot on doing everything in MATLAB.

I think I'm going to start another thread soon dedicated to these small USB DAQ like products since I have a list of a few now that seem to have similar specs.

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 2261
  • Country: us
Re: New Analog Devices/Digilent Analog Discovery board
« Reply #14 on: January 09, 2013, 05:18:43 PM »
If I got in my mail what Dave gets in his, I'd run home screaming every day to check my mail. 
Still waiting to be climate-changed to death...


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf