[time-nuts] Picking a good HP 10811
bruce.griffiths at xtra.co.nz
Tue Dec 2 06:58:55 UTC 2008
Bruce Griffiths wrote:
> Stanley Reynolds wrote:
>> The HP10811 has about +- 10Hz coarse adjustment would it be a good idea to make sure they don't have the same frequency by adjusting them in opposite directions from 10 Mhz before the test, like 10Mhz+4hz, 10Mhz, and 10Mhz-4hz for three. Would assume the common oscillator would be 10 to 20 Khz above or below the DUT.
>> Or does the sound card method work better with small differences ?
>> The manual recommends grounding the ECC line during tests. Maybe the test rig could set the ECC different on each one in the test. This would result in a much smaller difference and maybe no difference depending on each oscillator zero volt ECC frequency. ECC is listed as 1Hz range for -5 to +5 volts.
>> Would attenuation of the DUT signal physically separated from the mixer but after and buffer amp help ?
> Adjusting the oscillators in opposite directions is helpful at least in
> trouble shooting.
> However AFAIK it wont have much effect on the susceptibility to
> injection locking as that depends on the degree of mismatch between the
> resonator natural frequencies.
> There no real substitute for a high reverse isolation buffer.
> The classical sound card technique works best with small frequency
> There are variants that work better with larger frequency offsets,
> however as yet they are not yet fully debugged/tested.
> Grounding the EFC line minimises noise contributions from the EFC
> voltage source.
> Attenuating the DUT will increase the system noise level.
> High level mixers like the Minicircuits ZP3H and the TUF3HSM+ level 17
> mixers which have high LO to Rf isolation at 10MHz are well suited.
> Operating the RF port below saturation level will minimise the mixer
> phase shift tempco.
> Terminating the mixer IF port in a capacitor will reduce the mixer noise.
> A classical buffer amp with transformer input and output that cascades 3
> or more common base stages can have high reverse isolation as well as
> low input and output VSWR.
> At 10MHz 2N3904's are suitable - they also have less tendency to break
> into VHF/UHF oscillation than higher frequency transistors.
> Shielding everything well also helps.
> Build the buffer/isolation amplifiers in individual shielded enclosures.
Offsetting the injected signal frequency will actually decrease the
required isolation by an amount that depends on the operating Q of the
With an operating Q of 1E6 and an offset of 0.5ppm the ratio of the
injected signal to the oscillator signal will need to be near unity for
injection locking to occur.
It is relatively easy to ensure that this doesn't occur.
Thus it may well be worthwhile doing this as one then (in principle)
only needs 3 mixers (plus 3 simultaneously sampled sound card input
channels) and no offset source, however the maximum achievable offset
will probably result in beat frequencies that are a little too low for a
sound card ADC.
More information about the time-nuts