What's some weird/odd/unusual test gear you've seen?

[Keysight DanielBogdanoff]

I've been working with some new-hires at Keysight and have been kicking around an idea about how to think about & classify test gear. I'm trying to break my theory, and was hoping you could help!

What's some of the strangest test gear you've come across? Super niche, out there, unusual?

[nctnico]

I once built an ankle rotation meter.

[tggzzz]

Were you working for Keysight when it was HP during the Carly Fiorina era?

I was, and her performance when she came to the site was instrumental in making me decide to leave. As was her incomprehensible "Rules of the Garage", her version of a dog marking its territory.

Back then the internal usenet groups ensured extremely quick and accurate "horizontal communications"[1] across the globe. I remember hearing this long before it appeared in print:



[1] entirely in keeping with the HP Way

[tggzzz]

What's some of the strangest test gear you've come across? Super niche, out there, unusual?

Devices which test the durability and lifetime of building materials, by alternately sucking and blowing large panels simulating wind loading.

Devices which use lasers to test the surface textures of roads, as a van drives along the road.

Devices which apply a vacuum to a altimeter, to verify the altimeter's calibration.

[Berni]

One particular one that comes to mind is a "engine cranking simulator" i have built some time ago.

Its basically a 20V 5A power supply with 2 quadrant sink/source capability that can be adjusted by a DAC with the bandwidth in the KHz range. It reads a CSV file from a SD card and on a button press replays it on the output. It was used to simulate a 12V DC supply during various engine cranking conditions to make sure the DUT can continue operating normally during that time.

I don't think i have seen any piece of test equipment that does the same thing. I think my Agilent 66332A (Dynamic DC Source) could be made to do it via GPIB since it can do 2 quadrant and adjust its output very rapidly, but the functionality is not available from the front panel UI (Nor does it have any form of removable storage like flash or floppy).

[Tomorokoshi]

Would an HP 5000A Logic Analyzer count?

[babysitter]

I guess one made by me and a co-worker should be in this list: A device that connects+disconnects a hydraulic connector basically by a quarter-turn (done by a pneumatic cylinder with pleuel and some gear), followed by a air leak test. Somewhere the "n cylcles lifetime" numbers must come from somewhere.

[Keysight DanielBogdanoff]

I'll bite.   Over the years, we have had to design a fair amount of one off, custom test equipment.   I imagine this is still fairly common but maybe not. 

Were you working for Keysight when it was HP during the Carly Fiorina era?   I have some friends who were and heard some interesting discussions about her views on T&M.

We definitely still do one-offs, but I suspect not as frequently as we used to.

I was not around for the Carly days, but I've heard... things...

[nctnico]

I guess one made by me and a co-worker should be in this list: A device that connects+disconnects a hydraulic connector basically by a quarter-turn (done by a pneumatic cylinder with pleuel and some gear), followed by a air leak test. Somewhere the "n cylcles lifetime" numbers must come from somewhere.

These kind of cycle test machine are quite common. I've seen something similar for testing the durability of electric outlets at a large manufacturer. They also have machines to check the force needed to throw a switch (and how this evolves over time) but I have not seen these myself.

[DaJMasta]

Something I was looking at a week or two ago without really realizing they existed beforehand, but more of an accessory: a resistivity measurement chamber.

Take a high voltage source and an electrometer, then a pair of measured, characterized plates in a chamber and you can characterize the resistance of material samples, even gasses and liquids in some chambers.  If you're looking at very high resistances, the box also needs to be totally RF shielded and could need to be temperature controlled, depending on the material.



Probably very normal hardware for specific testing... but pretty exotic for a lot of electronics types - more bordering on chemistry, physics, or material science than just electronics.

[tggzzz]

Were you working for Keysight when it was HP during the Carly Fiorina era?

I was, and her performance when she came to the site was instrumental in making me decide to leave. As was her incomprehensible "Rules of the Garage", her version of a dog marking its territory.

Back then the internal usenet groups ensured extremely quick and accurate "horizontal communications"[1] across the globe. I remember hearing this long before it appeared in print:

[1] entirely in keeping with the HP Way

Could you please provide the the books' title or ISBN number?   It looks like an interesting read.

Sorry, no. I don't have the book and can't remember how those scans made it to me.

ISTR the source was fairly obvious but it wasn't something I would want to buy/read. It wasn't Dave's book.

[capt bullshot]

One particular one that comes to mind is a "engine cranking simulator" i have built some time ago.
...

Something like these: https://www.toellner.de/power-supplies?pf=284&pi=286

[capt bullshot]

Some ten years ago, I developed an instrument called "Netzimpedanzanalysator" - Power line impedance analyzer. Its purpose was to measure the (complex) impedance of your common 50/60Hz household and industrial mains line over a frequency range from 3Hz to 20kHz. Worked by pulsing a resistive load with a pseudo random sequence to the line, and measuring voltage and current. Some FFT analysis, math and accumulation of results provided the user with the results. Quite rare, only maybe 10 or 20 units were built and sold.

[Berni]

One particular one that comes to mind is a "engine cranking simulator" i have built some time ago.
...

Something like these: https://www.toellner.de/power-supplies?pf=284&pi=286

Ah yes that's exactly the piece of test gear for this kind of fast transient supply ripple generation. Nice to see it has some memory to hold on to seqences

Tho building one was pretty easy since i just used one of those huge opamps with many amps of output current to simply amplify up a DAC voltage to the final output power. Since only 1 was built there probably was not much money saved when the man hours to design and build it are counted in. But hey it was a fun little project.

[joeqsmith]

Could you please provide the the books' title or ISBN number?   It looks like an interesting read.

Sorry, no. I don't have the book and can't remember how those scans made it to me.

ISTR the source was fairly obvious but it wasn't something I would want to buy/read. It wasn't Dave's book.

Thanks.  I'll try and hunt it down.

Google comes through:


Bill & Dave: How Hewlett and Packard Built the World's Greatest Company
Michael S. Malone
ISBN: 978-1-59184-152-4

[ralphrmartin]

How about bat detectors, which pick up their ultrasonic calls to identify the species? Some work by just downconverting to an audible signal and leave the id job to the user, but others use various machine learning / signal analysis techniques to get an id.

[bsfeechannel]

When I was a kid I thought this was a joke.



When I grew up, I learned it was real.

[Kleinstein]

I came a cross a system with a high power (2 kW AFAIK) pulsed RF generator (up to some 10 MHz) combined with a simple receiver to detect some weak echos.  It was used to excite and measure ultrasonic resonances (some 100 kHz 2 MHz) in small samples to calculate (separate PC) elastic constants.  Kind of scary getting that much power over normal BNC.
The unit may be derived from a simple NMR unit, though I would expect better quality for NMR systems.

At the university they had an odd strain measuring unit: kind of a high resolution laser based bar code scanner to measure a similar stripe pattern with high resolution (e.g. µm range).