[time-nuts] Homebrew frequency counter, need help
kb8tq at n1k.org
Fri Nov 28 09:57:37 EST 2014
> On Nov 28, 2014, at 7:04 AM, Li Ang <lllaaa at gmail.com> wrote:
> The TDC-GP22 is the best choise for me at the moment. Since I am not a
> hardware guy, simplicity is important. It's very cheap in China(About 3$
> for each piece ).
Which is a very good reason to play with it …
> TDC-GP22 has 2 modes and 3 trigger inputs (start, stop1, stop2). The min
> time between start and stop is 3.5ns. max time is about 2.4us.(0 to 2.4 us
> between stop channels).
> Time resolution is 90ps, or 45ps in double_res mode(1 stop input only), and
> 22ps in quard_res mode(1 stop only)
All of which is *plenty* good enough to make a decent counter. That assumes that they are talking about accuracy (even 1 sigma) rather than just the resolution of the LSB. Specs are often confusing on parts like this.
> The biggest problem I met:
> If you trigger start first, 100ns later stop1, another 100ns later
> stop2.Get the result of stop1-start and stop2-stop1. You will find
> stop1-start is very unstable(+-2% level) and stop2-stop1 is stable(+-0.1%
> level ). Dont know if it's my problem or theirs.
This may be the resolution / accuracy thing I mentioned above. Don’t give up quite yet though.
> Here is the configuration of it (the document is very bad, took me 3 weeks
> to play with these 6 registers to get it work as I expected) and schematic.
Gee … that sounds like the documentation on a LOT of parts these days…. It’s also what keeps people from trying this sort of thing.
A few hardware questions:
1) What frequency is the crystal at? (can you drive the chip from an OCXO?)
2) Is there more bypassing on the circuit than shown? (If not, add some more).
3) How confident are you of your input signal? (can you check it with a “known good” counter?)
4) Have you tried jumping the 10 ohm resistor on the regulator output? (it may not be helping things …)
A full counter will have a bit more “stuff” than just this chip. I think getting this part working *before* you work on the rest of it is a very good idea. One small piece at a time ….
> Register_0 = 0x00c42700,
> Register_1 = 0x19498000,
> Register_2 = 0xe0000000,
> Register_3 = 0x00000000,
> Register_4 = 0x20000000,
> Register_5 = 0x10000000,
> Register_6 = 0;
> 2014-11-28 0:20 GMT+08:00 Tom Van Baak <tvb at leapsecond.com>:
>> Your project sounds wonderful. The TDC-GP22 has been mentioned only a few
>> times over the years and I keep waiting for someone to post actual results
>> from this chip, or better yet -- schematics, photos, and source code.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the time-nuts