Custom softwarez

[ELake]

Heres a little project I've been developing for awhile now..

Mainscreen.jpg is just that - the main screen.

Valvecontrol.jpg shows the different valves that can be tested, when selected, selector switches pop up to allow the operator to energize solenoid valves on the valve to be tested. The neat thing here is that depending on which one you select, the number of switches visible changes. The slider isn't shown very well but this is just an analog output slider. 0-10VDC output. This is the preset voltage going to the PID control card that positions the valve to be tested proportionally. When the slider is at 10VDC, the feedback from the valve should be 20mA. The ramp was a bitch to program and I don't really use it much but it basically allows an operator to ramp the valve to a specific setpoint with an adjustable time base.

xducercal.jpg sounds like exactly what it is, this is what calibrates the different transducers on the test stand. I also added some custom controls for modifying the sample rate and duration of time to sample for. The bottom output is the number of cells the spreadsheet will contain once the operator starts the flow shut off test.

datafile.jpg this popup allows the operator to create a custom file name for the valve to be tested. There is a template that the program reads with header information pertaining to what is to be tested.

I tried to attached a .xls showing the results of a 9800GPM shut off run but thanks to Pete's 500K limit I had to take a screen shot. You can see how we plot flow, valve position (LVDT), upstream, downstream and also calculate CV. There are some other rows that I used in an attempt to do my own flow calculation because the velocity output on the transducer has a moving average that it uses to smooth out the output. However, we had an application in which the moving average was really screwing with us because the valve would trip so quickly. I had to crank the sample rate up to 5000 samples per second to actually see that the valve was doing what it was supposed to. On the X axis you can see that the whole thing only took around 600 milliseconds to achieve close to 10,000GPM and that the valve closed during that short period of time.

[Pete]

big things Mr. Software Developer.  Are you still running that Vista theme on XP?

[ELake]

Yeah its on a big gay IP67 rated POFS with a 2gb SSD and an AMD processor

[Pete]

weak.  CBC Computing need to get in there 

[ELake]

For sure. Next build will have a PC sitting in the panel and some VGA cables running out of the cabinet so I can represent with a C2D

[ELake]

Newest revision is utilizing National Instruments modular DAQ system. The capabilities are freakin insane.
The program was all developed in labview 
 
The advanced options screen is probably my favorite, it allows you to enable/disable channels that plot to excel. We ran a simulation test for a customer and we were polling data from 19 different devices at a sample rate of 5000 per second for a duration of 5 seconds. The excel spreadsheet pops up at the end and then the thing just hammers data into the cells. It's pretty badass.

The manual control screen proportionally controls the valve being tested's position and also allows the operator to control the pressure supplied to the valve. We upgraded on this new setup and have two 75hp motors now. They can be started/stopped from the software as well.

[Pete]

will the disk array in your server keep up with that?   

[ELake]

nope I save all of the data locally (sometimes 15-20 megs after I graph it all!)