Breaking News
Comments
Newest First | Oldest First | Threaded View
tom-ii
User Rank
Blogger
Re: Would this have done the job?
tom-ii   1/17/2014 2:19:07 PM
NO RATINGS
@Martin:

How easily can you replace a module with one from another manufacturer? Has the software and drivers reached that state yet? That was the purpose of IVI drivers.

Well, that's the crux of any upgrade, right?  There's a few manufacturers out there that have equipment that will work.  But every piece of equipment (even NI's) has its own way of being talked to - that is, its own commands, its own address space, and so on.

But they are usually pretty well supported, and their tech support can help you with the upgrade.  Caveat Emptor that it really does what you want, but regardless, unless you are replacing with the exact same thing, you will likely need to update the software to run it.  I don't think you can ever get away from that, regardless of the item.

Where it saves dollars, though, is that I don't need to build a whole new box.  Or, in the case of a personality module, I need to build only 1 box, and by updating my software, I get a completely new functionality - rather than building an entirely new test stand for some new test program.  All those development labor hours are saved.

MeasurementBlues
User Rank
Blogger
Re: Would this have done the job?
MeasurementBlues   1/17/2014 1:21:12 PM
NO RATINGS
@tom,

"The nice thing about a PXI rack or FPGA-based system is that you can easily replace a single component,"

How easily can you replace a module with one from another manufacturer? Has the software and drivers reached that state yet? That was the purpose of IVI drivers.

tom-ii
User Rank
Blogger
Re: Would this have done the job?
tom-ii   1/17/2014 1:15:29 PM
NO RATINGS
@Martin,

There was once a time when testing equipment had to be 10x more accurate the the device under test. (Test uncertainty ratio or TUR as it's called in the calibration business.) Then it went doesn to 4:1, then 2:1 and in some cases 1:1 at the high end of measurement.


When you're creating state of the art, sometimes 1:1 or slightly better is all that's achievable.  Think about the "bad old days," when radars were relatively new - each new generation was better than the old test equipment.  I understand the TE companies spent a lot of R&D to keep up with such developments.  Now, there's enough commercial pull for such things that it's a bit easier to get certain types of things.

But still, $500kUS oscilloscopes are out there, and needed att the high end - along with the custom stuff to do the rest ofthe interface.

The nice thing about a PXI rack or FPGA-based system is that you can easily replace a single component, rather than building a whole new rack full of custom equipment.  And that's a hefty cost savings.

 

MeasurementBlues
User Rank
Blogger
Re: Would this have done the job?
MeasurementBlues   1/17/2014 8:38:05 AM
NO RATINGS
There was once a time when testing equipment had to be 10x more accurate the the device under test. (Test uncertainty ratio or TUR as it's called in the calibration business.) Then it went doesn to 4:1, then 2:1 and in some cases 1:1 at the high end of measurement.

Sheetal.Pandey
User Rank
Manager
Re: Would this have done the job?
Sheetal.Pandey   1/17/2014 8:28:26 AM
NO RATINGS
Yes testing devices or setups have to be as good as the devices under test. Testing is extremely important and where there is mass production and variable field conditions, testing is so very crucial. The test cases and scenarios to be designed diligently. Yes Labview is quite oftenly used.

MeasurementBlues
User Rank
Blogger
Re: Would this have done the job?
MeasurementBlues   1/16/2014 8:04:21 AM
NO RATINGS
@Tom,

There was years ago an e-mail list for LW/CVI users. I was on it. There are several resources for LabVIEW users, once of which is called Info-LabVIEW, an e-mail list. It's been in existance for probably 20 years and I'm still on it.

tom-ii
User Rank
Blogger
Re: Would this have done the job?
tom-ii   1/16/2014 7:56:24 AM
NO RATINGS
@Martin:

Tom, I wonder if this would have done the job. You'd have to learn LabVIEW though. It's called CompactUTS from Bloomy Controls.

For the particular job, I don't think that would've done it.  We did a lot of custom work with high speeds & so on, so a very custom front-end was always needed for the interface.  The back-end, though, was where the real saving of the NI PXI rack came in.  The rack & equipment we chose came in at about $45kUS, if I recall correctly.  The legacy stands we were replacing were on the order of $250k to build.

 

I might add that I really like LabWindows.  've toyed with LabView, and while I can see the advantages in certain situations, I find C code much easier to understand and more customizeable.  That said, though, even NI tech support has problems when you call them about LabWindows, as they're extensively into the LabView interface.  I found myself talking to a few of the actual hardware/driver designers to understand how the cards really worked.  The C interface lets you do things that the visual interface doesn't in some cases.

tom-ii
User Rank
Blogger
Re: LabWindows/CVI
tom-ii   1/16/2014 7:49:53 AM
NO RATINGS
@Martin:

20,000 lines of code. Did you have to write drivers for the instruments or was your code all for the application? For that matter, what were the instruments that you had chosen?

 

It's been a few years, now, but from the best of my recollection, we were using power supplies (which were easy to control), a precision timing unit (not so easy), voltmeters (easy stuff), analog outputs (not so bad), a stepper motor driver (to precisely control a mirror), and digital IO/pattern generater boards.

 

We were interfacing to an imager chip (made in-house), and we had to control the settings on the chip, tell it to take an image, and then absorb the raw data back via serdes.  The chip was controlled via JTAG TAP interface - there were a lot of interesting wrinkles involved with this, as it turns out the NI Pattern generator machinery could only output an even number of bits.  This worked out fine, though, as I was already planning on doubling the stream so I could do a verify on the readback.

Most of the code was in getting things from the user interface and displaying images and status on the UI, which ended up haveing about 5 embedded panes that you could tab through.  Most of the stuff was engineering information, but every setup had to write a log file so the conditions could be recreated.  In fact, I had written it so we could import a log file, and not have to do the settings manually.

 

MeasurementBlues
User Rank
Blogger
Would this have done the job?
MeasurementBlues   1/15/2014 9:21:42 PM
NO RATINGS
Tom, I wonder if this would have done the job. You'd have to learn LabVIEW though. It's called CompactUTS from Bloomy Controls.

Bloomy CompactUTS

MeasurementBlues
User Rank
Blogger
LabWindows/CVI
MeasurementBlues   1/15/2014 9:15:13 PM
NO RATINGS
Tom,

20,000 lines of code. Did you have to write drivers for the instruments or was your code all for the application? For that matter, what were the instruments that you had chosen?



EE Life
Frankenstein's Fix, Teardowns, Sideshows, Design Contests, Reader Content & More
Max Maxfield

How to Fly Without Catching the Dreaded Lurgy?
Max Maxfield
13 comments
I am currently not wearing my happy face. In fact, I am a "poorly soldier," as my mother would say when I was a little lad.

EDN Staff

11 Summer Vacation Spots for Engineers
EDN Staff
19 comments
This collection of places from technology history, museums, and modern marvels is a roadmap for an engineering adventure that will take you around the world. Here are just a few spots ...

Glen Chenier

Engineers Solve Analog/Digital Problem, Invent Creative Expletives
Glen Chenier
15 comments
- An analog engineer and a digital engineer join forces, use their respective skills, and pull a few bunnies out of a hat to troubleshoot a system with which they are completely ...

Larry Desjardin

Engineers Should Study Finance: 5 Reasons Why
Larry Desjardin
45 comments
I'm a big proponent of engineers learning financial basics. Why? Because engineers are making decisions all the time, in multiple ways. Having a good financial understanding guides these ...

Flash Poll
Top Comments of the Week
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)