||[Feb. 28th, 2019|06:46 pm]
sprockets, sockets, grommets & gaskets
I wrote a test program that turns on various parts of my chip alternately, to see if they interact: if one part is affected when another part switches on or off. Then I added to it to start running at very high frequencies, and then I added a bit more that drives a thermal forcer so I can run the chip between -40 and 130C. It turns out that at high power, at high temperature, at high switching frequencies, it blows up. Flames shoot out. It doesn't seem to be associated with the switching-on-and-off, though, but with being hot.
At the same time, my sweary twin (he looks very much like me but curses a lot) was trying to set up a test system that somewhat emulates mine, so I spent a big chunk of time doing code review with him and walking him through hardware and software choices I'd made when designing this thing.
(Side plot involving explaining the SPI protocol, which is really weird to someone new at it: you write a word to the chip, and then have to wait until the next transaction to get a response. The offset-by-one is surprisingly hard for people to work with.)
At the same time, I was helping my prickly PhD coworker get a board working, which involved iteratively swapping out parts to get a slope compensation value working right, and simultaneously documenting all the changes so we can recreate this later.
At the same time I was also helping my other coworker get his system working. He's using the previous revision of hardware and software that I built last year, and it keeps having communication issues, and no matter how many times I explain troubleshooting the system to him, he never seems to understand the overall concept, so he does the first step over and over and then when it doesn't fix the problem comes to ask me for help. (I got him set up on a software repository system last week, because it's been a requirement for us to use for a year, and I am pretty sure he still doesn't understand that the repository is on a separate, remote, managed system, and is not just another directory on his hard drive.)
And all of these are fine, but the only one my manager will notice is the fix-the-board-for-prickly-PhD, and then most likely all he'll notice is if there were any problems with it. All the other projects are things that he doesn't think are important, so he doesn't find my handling of them relevant.
I know I shouldn't complain about my job, but should instead just find another one.
This entry was originally posted at https://randomdreams.dreamwidth.org/59798.html. Please comment there using OpenID.