An Interview with Ted Yapo

Nov 4, 2019 · 1 hr 25 min · Episode 465 · 75.9 MB

0:00-1:25:22

Streams straight from the publisher. podnod never proxies or re-hosts episode audio.

Ted is a fan of soldering AND stock photography This episode is sponsored by Rohde and Schwarz. Check out AskAnEngineer.us for more info about their value line test equipment.

Ted is a lifelong electronics hobbyist
He studied engineering physics in college.
Full color faxes were a thing in Japan but never took off elsewhere.
After working on scanning equipment, he moved on to doing networking work.
He also did some consulting after Cisco.
Mid-way through his career, he went back to graduate school.
Wrote a paper about finding telephone wires using LIDAR for helicopter pilots.
Transitioned into AR, sort of like Projection Mapping. It was used for architectural daylighting simulation.
Going back to school is tough for what you remember. Ted had to relearn linear algebra. He recommends using Gilbert Strang's lectures, which gave more context.
After school, Ted did consulting with a few partners, mostly New England.
Ted got started posting to Hackaday with 2016 HaD prize.
His first entry was the "Diode clock", which utilizes the fact that diodes act like switches for RF when you pass DC current through them.
"Upping serendipity quotient"
Using diodes as an AND gate
Using pin diodes for a doppler radio project
1n4007 can act as a crude pin diode at certain frequency
Built digital clock that uses diodes for the counters
Assembly instructions - 46 boards!
Power supply is around 6 MHz
One of his other popular projects is the Tritiled, which helps figuring out where things are in the dark
Based around the idea of tritium lights
Modern LEDs are extremely efficient at a certain current
Droop happens at higher currents
The 8 GHz sampling oscilloscope is a more recent project
Research came out of the diode clock
Schottky diodes for sampling
Tek S4 sampling head 14.5 GHz
Diodes can switch in 10-15 pS
Stroboscopic effect
Equivalent time sampling
Tek 11801 from 1989
Nyquist still applies
Verifying on the bench is difficult
Paper from picosecond pulse lab
Jim Williams using an avalanche diode or a step response diode
Looking at RMS error
Another architecture 1975 by SP McCabe
Using a fast comparator to sample waveform, similar to a successive approximation ADC
ADCMP582
That's the time base calibration
What will people be able to build with an 8 GHz scope?
Using it to measure transmission lines on PCB boards
Using Time Domain Reflectometry
Ted has a splitter TDR head
Tracking generator + Spectrum analyzer = Scalar network analyzer
Rigol came out with cheap spectrum analyzer, DSA815, Ted got one of the early units
Tracking generator outputs signals that are the same as the frequency on the spectrum analyzer
Ted gave a talk about this at last year's Hackaday Superconference
There is no phase in an SNA
SNAs are harder to calibrate
Paper was about how to do two measurements to compensate for errors in the generator
Chris has a friend doing Python scripting on test equipment via their ethernet jacks.

You can find Ted on Twitter, Hackaday.io and as a writer for Hackaday