2

I have a lab instrument from the late 1980's that has a printer as its sole storable data output (it gives you the readout on the screen in real time but does not save it to memory, as it has none, you can just view the last measured data point).The printer is an Epson FX1050 but the instrument also has the option of using an "HP" printer (just putting it out there as I understand the standards are different in both cases). The instrument itself performs measurements and the calculations to get to the results that become the output. The time the instrument takes to perform the task will vary between samples.

I went ahead and purchased a data collection software "Collect 6.1" by labsoft/Perkin ELmer to see if it was the solution to my problems. I understand a bit about data parsing but the tech was a bit overwhelmed at the amount of data a single measurment would generate (over 100 lines) with only a couple of specific places where the same text would ever get repeated and I have not had any luck getting it to parse the data the way I want to, I think it times out even though I am setting the timing out very high, measurments are about 3 minutes on average. I can get it to see what is being sent but it does not do much with it, and I can only view it while in test mode and it gets saved as text.

After many frustrating days I am trying to change my approach. I have an article from 1995 where they connected this same instrument to an external computer with a communications software (PROCOMM which downloaded an ASCII file) and then they converted it to Lotus 123 / Excel for data processing. I would be in "hog heaven" if I could do something similar but I am thinking a printer file might be easier to work with. The instrument has the option for setting up hardware handshakes, CTS output, RTS input, etc... and from my failed attempts with the Collect software I know I am using the correct cables to get data from the instrument to the computer but now I am thinking of switching cables and just trying to get the printer output into the computer if there is an "easy" windows based importer for the printer output data.

The advantage of the "printer output" is that only the critical numbers, with the results, are sent, saving me from sorting through the individual measurements. I have read the other "intercepting printer data" questions and I did not find them 100% applicable.

As a warning, I am not particularly talented with computer softwares, apps or macros, just the basic windows type stuff and some better than average Excel skills. I stumbled across this forum as there were other printer data exporting questions and I became hopeful. Thanks in advance.

Smandoli
  • 6,919
  • 3
  • 49
  • 83
Abipin
  • 29
  • 1
  • Maybe add some whitespace... – rene Dec 05 '13 at 17:04
  • Please provide the details of the instrument, make and exact model if possible - it would make us more able to help. – tutuDajuju Dec 05 '13 at 17:24
  • 1
    Your instrument is apparently putting out plain text to a printer and you need to get that text into a format that can be processed. Your problem begs for a virtual printer. There are many to choose from, the most expensive being Adobe PDF, the least expensive being several free ones. Converting printed text output to binary data isn't a new concept so I recommend you look into those options. – Carey Gregory Dec 07 '13 at 07:41
  • sounds more of a which tool question than programming question – AndersK Dec 07 '13 at 07:47

1 Answers1

0

The mention of CTS and RTS suggests you have a UART between the instrument and the printer. Have you managed to capture any of this data in to a file? If so, could you supply a sample?

One gotcha is that the instrument might be running some sort of protocol against the printer, and won't emit the data without the right responses from the real printer. If that's the case, and you're planning on replacing the printer with a PC, then you might need to get the PC to spoof these responses to keep the instrument happy.

If you are keen on parsing the data yourself, you could write an application in the language of your choice that responds to whatever the UART spits out, then extracts the good bits. If you can do Regular Expressions, that would be a great help in parsing this kind of stuff - you can match on the features surrounding the data, then capture the important data and emit it to a .csv or whatever.

Good luck with this.

Chrisky
  • 567
  • 3
  • 9