[time-nuts] Broken Ovenaire OSC 85-50

Bruce Griffiths bruce.griffiths at xtra.co.nz
Sun Jul 5 10:07:58 UTC 2009


Pascal Nguyen wrote:
> Hi Joe
>
> I am just a radio amateur and I build a lot of oscillators for my need,
> including a GPSDO "tunable" with CW25-TIM GPS receiver chip set. Please
> allow me to share my usual  trouble shooting when oscillator is not working
> properly.
> I see an oscillator is just only a high gain amplifier with  positive feed
> back via "narrow band pass" filtering to sustain an oscillation in a
> controllable manner.
> I had a look at your circuit with all Vcbe of all transistors transcribed,
> it seems to me all transistors are within normal limit (bias # 0v7).
> Please look at the "narrow filter", my suggestion is to check this circuit.
> 1) 22pF in parallel with a serial (10uH - 30pF) is a  diplexer, high
> impedance to ground for 10 MHz and low impedance to ground for unwanted
> spectrum.
>   

A poor description that misses the function of this inductor entirely,
such a network acts as a mode suppressor by making the impedance to
ground inductive for the crystal mode that is being suppressed. If one
of the  2 feedback capacitors in the Colpitts oscillator is replaced by
an C || (L+C) network oscillation won't occur in the frequency range for
which that network is inductive.

> 2) 80 pF feed back capacitor.
> 3) 10 MHz Xtal
>
> Now I can perform a differential diagnosis:
>
> All transistors are correctly biased, can eliminated fault from power supply
> and transistors.
>
> A quick and dirty approach is to do a loose coupling (capacitor or inductor)
> with the 10 uH inductor, connect to an external 10 MHz variable output
> source, check gain of all transistor stages to the last output
>
> Check high impedance to ground for 10 MHz, disconnect the "diplexer", check
> magnitude of output signal, signal is increasing, hurray we see the problem.
>
>   

Bad idea, the (overtone??) crystal will then probably oscillate at the
frequency of the mode that this network is intended to suppress.
Better (if possible) to actually measure the impedance of this network
as a function of frequency to check that its has the required
characteristic.

As far as I can deduce this circuit limits the amplitude by having the
oscillator transistor saturate for a part of the oscillation cycle.
Its better to use transistor cutoff as the limiting mechanism as its
quieter.
However this requires a larger oscillator transistor Vcb than is used in
this circuit.

Excessive loss in the frequency determining network or a lossy varicap
or circuit board can significantly reduce the crystal current to well
below the design value.
> Same signal level, put the diplexer back to the circuit, check the 80 pF
> cap, all else failed, the last component of the "narrow filter" is the Xtal.
>
>
> Good luck
>
> Pascal Nguyen
> vk2ihl / xv2pn
> _______________________________________________
> 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.
>
>   

Bruce




More information about the time-nuts mailing list