convert file data to analog input

[atari]

I hope this is the correct place for this....... I need something that can accept the data from an Excel or csv file and can output a 4-20mA signal to represent this data, so it can be recorded on a recorder I have that will accept this signal. Along the lines of a digital-to-analog converter, I think. My device has only 8-bit input resolution, and an accuracy of 0.5% of span. This probably seems a little bass-ackward, but I really do need to get the data from the PC into my device. I would not mind having to "roll my own" if an off the shelf solution isn't available but I'm not real familiar with selecting the necessary components.

[Ken Moore]

I think you will need a DAC board: http://www.accesioproducts.com/go.cgi?p=.....put/output.html and then you will have to do a little programming. If you could give a little more detail on what you are trying to do, there may be a better way. Seems like you're using a Rube Goldberg approach.

[atari]

Ken- temperature data is pulled from a plc-5 every 15 seconds and logged to a server. I can get this data in the form of a csv or excel file, but I need to feed this data into a device that can output a 4-20 mA signal (4mA= 0 degrees, 20mA = 2000 degrees) so my device can record this analog temperature signal. The tricky parts (for me at least) seem to be finding the equipment that feeds in the digital data at the correct rate, and having the analog output hold itself steady until the new number is processed and output. I've never done this before (from pc to analog recording device- usually the pc IS the recording device). I have also never done any VB programming, but it looks like I might have to start. trying not to be ignorant, but what's a Rube Goldberg approach? Edited August 8, 2006 by atari

[Ken Moore]

Rube Goldberg was famous for coming up with complicated machines to do simple task. One simple task might have ten or more steps. Since the PLC-5 already has the data, why not set up an analog output from the PLC to the recorder, would require very little ladder logic, and avoids using the PC and some other device.

[atari]

Ken, have we met? I am famous for making things more complicated than they need to be. Just ask my wife or anyone I work with. I will say in my defense, though, that my plc programs are written to provide for just about any eventuality, with plenty of warnings and indicators of what is going on! A little more detail- I had already set up the plc to to provide the recorders' input. It provides a 4-20mA that is scaled to represent the process temperature. However, the recorder missed capturing some data for a few days that we really need to have printed out. This recorder (Omega RD-821) has been used here for a few years, its sofware is used to set up the graphs, and to keep everything standard and everyone happy, we need to print out the data using the supplied software. The data that the recorder missed is not lost, fortunately, because the plc temperature data is also logged to a server. So, if I can convert the file data to an analog 4-20, I can get the device to record the data it missed.

[panic mode]

how much data we are talking about? it may be possible to create few datatables, dump the data in, download and let plc send the data out through analog out. or don't store data in plc, just send it to plc through what ever connection you have (maybe RS232 for example) and pass it on to analog out. so what cpu you have? looks like you have all hardware, all you need is some software...

[atari]

I never thought of using the plc itself. If I didn't have such a large amount of data (number is collected every 15 seconds. I need a weeks worth ) it could have worked. The plc is a 5/25 with a rather large program (400+ rungs) with lots & lots of PID loops.

[Ken Moore]

1 week=168 hours 168 hours * 60minutes * 4 per minute =40,320 data points. A Plc-5 data file can have 1000 registers, you could do it with two files, move 1000 points to one plc file, start the recording process, while that is going on, download another 1000 to the second file. Switch the recorder to the second file, download another 1000 to the first file, repeat, repeat, repeat. Sounds like a lot of work, but it's something you can do already, the other choices require obtaining hardware, and learning a new program enviornment, testing, debugging, then finally recording the data. You say you need this data for graphs, why not just graph the data directly from Excel?

[panic mode]

since the streaming rate is very slow, you could avoid need bur buffering and datatables in PLC. just write small PC program that sends new value every 15 seconds. PLC should just pass it to analog output and keep it like that until next sample arives. Eventually you could do scaling in PLC but even this could be done in PC.