And now an entry in the "Plan to throw one away" section.
RF Multiplexer, 8 inputs, 1 output, BNC-connectors, TE HF3 relays specified to 3 GHz, an ULN2803A to pull the relay-coil, and an SPI I/O expander to drive the ULN - should be easy - right?
Well no, PCB trace-geometry does strange things beyond VHF. I clearly don't grok UHF very well.
Onward towards version 2! (any thoughts and advice on simulation or trace-geometry optimizers appreciated!)
My open-hardware design for a 1:8 pulse and frequency distribution amplifier is now available from Aivon LTD.
For isolated 1PPS distribution I made this distribution board.
The input is a TLP117 (or similar) optoisolator driving a LT1711 comparator with a 1.0 V trigger level. An output LED-blink is provided by LTC6993. Outputs are driven by IDT5PB1108 buffers.
In jitter measurements with a HPAK 53230A counter the jitter between two 1PPS pulses (from masers) seems to degrade slightly through this amplifier: from RMS 16-19 ps directly on the maser-outputs to between 21 and 26 ps RMS from the outputs of the ISOPDA. Maybe a faster optoisolator would be better?
KiCad sources available on request.
A mere ~3months 🙂 after moving to a new place I setup the lab-table again.
Despite length-matching traces between a distributor-stage and the individual output-stages on my pulse distribution amplifier there remains a 2-300 ps peak-to-peak output skew between the channels.
Here's a test where a 50 pF or 10 pF trimmer-cap is added just before the input of the output-stage. I found that tuning the cap results in a variable delay of 60-80 ps/pF, so if initially the channels are within 300 ps of each other the 500 ps tuning-range of the 10 pF trimmer-cap is sufficient.
As a test I first tuned all channels to within 20 ps peak-to-peak, then verified this the following day and got 52 ps peak-to-peak. BNC-connectors might not be the greatest for picosecond level repeatability.
Assembled another distribution amplifier today. S/N 004. Measurements to follow at some point.
50mV range with shorted input. Around 4 hours in an office with stable temperature. Less than 10uV drift and offset after maybe 20-30 min of warm-up.
Inspired by EEVBlog forum post.
Here's a tunable coil with an inductance between around 1.9 mH at minimum and 2.9 mH at maximum. The outer coil is around 200 turns of AWG16 wire around a 160mm diameter tube. The rotating inner coil is maybe 60 turns of AWG16 wire around a 110mm diameter 90mm length tube. The inner coil rotates on a 20mm solid rod with 10mm diameter holes through the ends for feeding the wires out. A small geared DC-motor rotates the rod.
Time signal reception with a long-wire antenna and a PA0RDT mini-whip antenna. HP89410A analyzer set to 50 Ohm input, 100 Hz span, 1 Hz RBW. The signals are from:
Antenna in a bottle installed on the roof today. Stay tuned for a future web-SDR...
An article, some theory, some notes on grounding, some comparisons,
There's also a modernized circuit by PA0NHC