[time-nuts] My DIY frequency counter and a request for help

Magnus Danielson magnus at rubidium.dyndns.org
Sun Feb 28 23:14:41 UTC 2010


Gerard PG5G wrote:
>    Hello all,
>    First post here, so I'll start with a quick introduction. I trained as
>    an electronic engineer but don't work in that field any more, which has
>    given me the appetite back to do some electronic engineering as a
>    hobby. I have been a licensed ham for over 25 years (more than 60% of
>    my life I realised the other day) and used to be rather active on HF as
>    PA3DQW. At the moment I live in the UK where I am licensed as M0AIU.
>    I recently designed and build a frequency counter and I need some help
>    with verifying its performance. I believe it gives me 11 digits in 1
>    second. I say believe because I have not got the hardware to verify
>    this. At the moment my assumption is based on calculations and limited
>    testing with the equipment available to me.
>    My counter is a "continuous time stamping reciprocal counter". I
>    implemented this as a USB powered device, with the hardware taking the
>    time stamps and sending it over USB to a windows PC. Some software
>    written in C++ takes care of analysing the data.
>    The hardware takes 5000 time stamps per second using a high speed TDC.
>    The hardware is a single PCB measuring about 50 by 80 mm. it requires
>    an external 10MHz reference and apart from using this as the time base
>    it also uses this for self-calibration of the TDC. The unit requires no
>    further calibration.
>    The PC software takes these time stamps and the associated counts and
>    uses regression to calculate the slope. This slope represents the
>    frequency of the input signal. I am sure people on here are familiar
>    with the counters made by Pendulum, and I have to confess that their
>    marketing material was helpful in putting this thing together.
>    Since the hardware is true zero dead time, the final capabilities of
>    this counter are determined by software. At the moment I can
>    simultaneously display the input at multiple gate times (see the
>    attached screen shot). For gate times over 1 second I have the option
>    to use overlapping gates, so that the display gets updated every
>    second.
>    Because there is no dead time I can also calculate Allan Deviation. The
>    two displays at the bottom of the page show both normal and overlapping
>    Allan deviation at tau=10s. I am still working on the software to do
>    this at multiple tau in real time and display it as a graph and a
>    table.
>    So, after this lengthy introduction here is my request for some
>    assistance. Is there somebody on the list who can assist me in
>    verifying the performance of this frequency counter? Ideally somebody
>    with access to two highly stable and known frequency sources. I can
>    send the hardware by mail, but if there happens to be somebody with
>    this kind of gear not too far from where I am (50 north of London) I
>    will travel. In exchange you get to keep the hardware and will be
>    supplied with whatever software I come up with.

Would you consider to disclose your architecture somewhat more?

Could you output time and event values from the time-stamping?
Would allow us to do some off-line processing independently.

Could you try different frequencies/amplitudes (would be good for 
establishing the slew-rate dependency, i.e. internal noise). Measure 
period jitter and plot for different slew-rates (frequency and 
amplitude), use shortest tau possible.

Could you hook up the reference clock with different lengths of coax 
cables. This would assist in measure the background noise and the 
different lengths of cables would allow some indication of interpolator 
non-linearity and input cross-talk.

As has already been discussed, software can do a lot for improving the 
reading, but one needs to be careful in details or else completely 
different measures results and they does not behave correctly. ADEV and 
friends wants the raw time-samples. Frequency or period estimation 
benefits from improved estimators, but then that is not useful for ADEV 
and friends, so it is a dead end for further processing except 
presentation level.

I think I have a fairly good setup including bunches of rock, gas and 
air clocks alongside a fair set of counters, so I could probably do some 
testing, but I am located over in Sweden. However, starting your verify 
exercise with a fellow time-nut excursion yourself should be a nice 
exercise that I recommend regardless. You should have several 
friendly-minded in UK.


More information about the time-nuts mailing list