Re: Calibration Of Electronic Equipment In The Home Workshop

doug wrote:

Anthony Fremont wrote:
Chris Jones wrote:

Anthony Fremont wrote:


If someone doesn't need traceable calibration, then why should they
pay for
it? Especially if they have the resources to do it themselves. I'm
thinking of buying a cheap used Rb time base from e-bay so I can cal
my old Protek freq counter and adjust the timebase on my Hitachi
scope, it's certainly cheaper than having it done. Using a PIC
driven by an ordinary can xtal, and a quartz wristwatch of known
accuracy, I was able to tweak the
xtal to within about 1-2ppm over the course of a week or two. Of
course you know that's impossible, don't you?

I don't think the oscillator on a PIC would be good to 2ppm absolute
accuracy even with a very good xtal, unless you FIRST calibrate it
against something that has already been calibrated, therefore it


Like my analog/quartz watch that has a predictable error. I used the CCP
module to time the impulses to the watches stepper motor to the uS. I
already knew the error rate of my watch, so I just played with the
loading caps until I knew that I was sucessfully measuring the error of
my watch
over hours of time. Since my house stayed at a fairly consistent
temperature, I think it's safe to say that it was within 2ppm, but I
guess I could be wrong.


doesn't get you far. It will however be good enough to calibrate
your scope since that would only need 1% or so, and any old crystal
should achieve that, even with a fairly primitive oscillator. For
frequency calibration, your best bet is to receive an off-air
standard, for example GPS or in many countries there are low
frequency standard transmissions (50kHz, 60kHz, 77.5kHz or others,
look up which ones are available in your country). It is quite
feasible to build your own receiver for these. These transmitters
are maintained to a higher accuracy than any piece of hardware that a
hobbyist could afford (e.g. 2 parts in 10^12).
http://www.npl.co.uk/time/msf/ctm001v05.pdf

Chris



This is a dangerous way to calibrate crystals. The temperature
coefficient will give a drift with temperature of up to around
100ppm. I you breathe on the crystal it will go out of spec. If
you want to try this, put a crystal oven on the crystal. There
are temperatures where the frequency versus temperature is flat
and an oven will give you a stability of maybe a part in 10^7 or
so. Then you can do a calibration. Why not look on ebay for an
old hp counter such as the 5328 which often come with a very nice
oxco which is good to up to 10^-8 for maybe a year? I have paid
as little as $50 and then you even have a counter to go with it.
The 5334 are newer, nicer and quieter but may run a bit more.

I think 'dangerous' is not really a sensible word to use. If the crystal is
not inherently stable enough for your application then blaming the
calibration technique is misguided. In any case, the receiver can be set
up as a phase locked loop so that the crystal oscillator is continuously
kept at the correct frequency. This would obviate the need for an OCXO. A
similar technique is used to keep the oscillator of a mobile telephone
continually within 0.1ppm of the basestation reference.

Chris
.

Relevant Pages

  • Re: Time Synchronizing Between Two Servers
    ... So the concept of "good oscillator" is a little different from what you ... temperature variations seen as installed. ... good AT cut crystals are, more or less coincidentally, working at a ... include one Cs standard and three Rb standards. ...
    (freebsd-questions)
  • Re: Calibration Of Electronic Equipment In The Home Workshop
    ... I already knew the error rate of my watch, so I just played with the loading caps until I knew that I was sucessfully measuring the error of my watch over hours of time. ... Since my house stayed at a fairly consistent temperature, I think it's safe to say that it was within 2ppm, but I guess I could be wrong. ... This is a dangerous way to calibrate crystals. ... Then you can do a calibration. ...
    (sci.electronics.design)
  • Re: WTF with my computer clock?
    ... The original Accutron was a steel tuning fork osillator. ... Watch crystals come in a few flavors. ... There was also no temperature compensation or even the use ... Not a single one has a tunable clock oscillator. ...
    (sci.electronics.repair)
  • Re: Watch Crystal dissipation
    ... tuning fork watch crystals make remarkable ... So the issue is not so much that you want to build any old oscillator and then measure the crystal dissipation. ... If this were me I'd implement the loop shaping in a microcontroller and the multiplier in a multiplying DAC; this would not only allow me to play games with the inherently nonlinear effect of regulating the oscillator output by changing it's damping but it would also let me easily communicate the output level directly, since it would be available internally to the processor. ...
    (sci.electronics.design)
  • Re: Defeating the mantis part 3
    ... different video channels and 2 LCDs hanging in front of your eyes, ... The supply voltage has very little effect on the crystal oscillator frequency. ... pretty hard to cope with higher MHz crystals from a wide variety of ... I was first imagining the output to input path as obvious, ...
    (sci.electronics.design)