Seeing agreement between my spreadsheet thermal model and FEM simulations to within 1.5 K. The former is isothermal, much easier to deal with

Been running a 2-week long vakuum test of our electrostatic instrument. Standard deviation after the system has fully evacuated is around 12 mV/m over the entire period, which is pretty darn stable

Reading about the cryogenic radiators on . Deceptively simple, until you get into the fact that thermal conductivity and emissivity is very much temperature dependent

Doing some rough thermal simulations. Things get a bit tricky when ground temperatures range from 100K to 400K ❄🔥

Working on an instrument paper, trying to come up with a new introduction. Drawing a blank so far

Grr, Traco and Murata use different pinouts for their SIP style regulators

Refactoring boards, reducing part counts, freeing up space

Gooping unobtanium (low-outgassing structural epoxy) on some electrical components that have been exposed to unplanned blunt trauma

A quick fix is to star ground the motors via cables, and drill out the copper in some mounting holes so return currents don't flow through the chassi

We have a new interface specification coming in for our moon instrument, so I'm taking the opportunity to go over grounding. Right now everything goes in the same ground net, which is less than optimal since it's shared between motors and high impedance circuitry..

This is fun: someone calculated how many packs of Skittles one would need to open before finding two identical ones. Then they went and did it, and the answer is 468 packs :]

Dammit, Mastodon cuts off links. It's about what happens when you try flying with a Nobel prize

Yikes, those bearings really start to howl as soon as the pressure starts to drop. Current spike too, from normal 200 mA @ 24 V up to 600 mA

The amount of formally verified code in the moon project now clocks in at 690 LOC, of which ~2/3 is annotations.

707 goals are currently proved by qed, alt-ergo, cvc4 and z3, in around four minutes on a laptop..!

Beyond knowing that all loops terminate and all memory accesses are within bounds, I also have numerical bounds on many variables. This helps verify that accumulators are using appropriate integer types and therefore can't overflow

Woop, motors plain bearings arrived from the workshop. Just have to wait until tomorrow for the glue to dry, then we can test them in vacuum over the easter weekend

-C quirk of the day:
int Q[3] = {0,0,0}; //this works
int Q[3] = {0}; //this doesn't

We're going back to pre-1800s tech in our instrument by switching the ball bearings in our motors to plain brass bearings. Hopefully if we make them just the right size they'll have "lagom" contact with the shaft, providing grounding, without being so tight that they just seize

thardin boosted
Spent four hours getting electron microscope and X-ray spectroscopy pictures taken today. Below is a zoomed-out view of a worn coil of silver plated copper wire

Heh, Gallager's original LDPC paper calls bits "binits"

Show more
Scholar Social

Scholar Social is a microblogging platform for researchers, grad students, librarians, archivists, undergrads, academically inclined high schoolers, educators of all levels, journal editors, research assistants, professors, administrators—anyone involved in academia who is willing to engage with others respectfully. Read more ...