I have the TI 930 and did run it from a computer - I did it under Windows but had quite a lot of trouble. It worked fine if I used the Windows own command line shell (I forget what it is called and I'm on a Linux machine at the moment) but when I tried communicating directly via the USB port (as COM3) using a c program I found it difficult to get responses. But this may well be my lack of knowledge of USB/RS232 under Windows.
It might well be easier under Linux. Basically the instrument responds to command strings and you can set it to just return values continually which is what I did. I've since got the Tek FS3100 (Pendulum CNT91 I think) which is much more sophisticated and I started working on programming that. Unfortunately I've since got a job away from home so I have no time for electronics and won't have until I eventually manage to move house. If I were you I'd use a terminal emulator, at least at first, and get it going interactively and then just save the output to a file. Once you've got used to how it all works (and there isn't much to it) you can set up a more robust software system. James -----Original Message----- From: Florian Teply <use...@teply.info> To: Discussion of precise time and frequency measurement <time-nuts@febo.com> Sent: Tue, 12 May 2015 22:59 Subject: [time-nuts] TTi TF930/960 linux programming Hi guys, I seem to recall that someone on this list mentioned that he's using a Thurlby-Tandar TF930 or 960 Frequency counter. As I'm considering to buy such a unit for some experiments at my workplace, I figured I'd better ask around here for some suggestions. Has someone already used one of these gadgets in a computer-controlled fashion, with some luck using some Linux environment? Judging from the manual, I probably ccould hack some shell script to repeatedly perform frequency readings and write that to a file, but if someone already has done that I'd be much too lazy to reinvent the wheel... The actual setting I'd plan to use it in is to monitor some ring oscillators (frequency drift) and/or delay lines (output pulse length) sort-of-continuously over extended periods of time. I'd be interested in frequency drifts due to device aging and/or radiation effects, and as especially device aging tests can take quite some time (a few months each...), some sort of stability would be needed. This is not strictly a time-nuts application where one might chase the 10th digit, and I figure I probably could tolerate (and wouldn't even notice without cross-checking) an constant offset in frequency readings even of a few percent, but it would bite me quite a bit if the readings wander around too much when the input frequency doesn't... Any suggestions? best regards, Florian _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
