[time-nuts] HP5061B Modulation Adjustment
Donald E. Pauly
trojancowboy at gmail.com
Tue May 16 22:04:55 EDT 2017
We recently got our second HP5061B to lock after changing the beam
tube. We did no alignment with the exception of changing the gain on
Beam current. Ion current dropped from 14 to 1 with only a few hours
of operation of the instrument. We obtained momentary lock but lost
it after a couple of minutes. The trouble was a blown themal fuse on
the HP10811 oscillator which shut down the oven. We have TEMPORARILY
jumpered it while the correct fuse is on order.
We aligned the C Field on #1 but have not done that alignment on #2 yet.
The 5 mc output of our #1 instrument was compared to the 5mc output of
#2 with an HP10514A balanced mixer. Two input ports L and R are
multiplied by internal Shottky diodes to produce an X output which can
be terminated in 50 Ω. Maximum input current on both the L and R port
is 40 mA or 2 V rms. One instrument was put in Open Loop and the
fine frequency adjusted to approach quadrature. It was then returned
to OPR and steered to final quadrature with the C Field.
We used a digital voltmeter with a resolution of 10 uV dc to measure
the X output of the mixer with the L and R inputs in near quadrature.
The mixer output contains a large 10 mc component of about 2 V peak of
peak which rides on the dc representing the phase difference. The dc
output for in phase and out of phase inputs is about +1 and -1 Volts.
Our calculated sensitivity was 31 uV per picosecond of error when
within a few degrees of quadrature. Lock within a few picoseconds was
obtained by adjusting C field on #2. It had to be continuously
manually steered. We could not look at the X output on the scope
because of the large rf level present. Our digital meter might have
been experiencing rf overload causing inaccuracy at low dc voltages.
Continuous lock was lost a few times on #2 which we attributed to it
needing a complete alignment. Second harmonic was only 22. The
procedure in 5-217 g for modulation adjustment is not very scientific.
It calls for adjusting modulation for maximum 137 cps out of the AC
amplifier board when the beam tube is driven at 5x10^-9 below the
resonance line. Modulation is then to be reduced by 5%. This affects
all other adjustments.
We decided to directly measure this deviation with precision. The
phase noise of the 90 mc output of the A3 multiplier board was
measured as well as the modulation. The spectral width of our
function generators at 5 mc was compared to the HP5061B 5 mc output.
Our adventures with this project will be covered in a separate post.
More information about the time-nuts