Back in Maryland, I was busy working away on my tasks, when it started to strike me (pun intended?) that something rather odd was going on outside weather-wise. There’s only one window in the lab I worked in. By the time I realized it was raining sideways, it was too late to protect the thousands of dollars worth of keysight equipment sitting on the desk for myself and my colleague to use, not to mention, prototype units of the DUT.
So, of course, as you can imagine, I immediately started unplugging things. There wasn’t even time to safely shut things down, they wen’t unplugged right then and there. You can hear the sound of transformers arcing outside from who knows what.
I guess my supervisor didn’t care if lightning blew up our equipment or DUTS. The test boxes were on loan from Keysight anyway, so why should he care? I think otherwise. If there’s no UPS, and you have no ability to discern the weather that is coming, you’re basically throwing away money. Which is ironic, because typically people try to avoid that sort of thing, not the other way around.
I’m sure the building, or at least the lab, was on some sort of power protection. A brown-out can still damage a power supply. I know from experience that power supplies in test equipment usually have to have some of the tightest tolerances in a power supply you can manufacture. Dirty power can and will result in poor quality of measurement, for a variety of reasons.
This particular storm knocked out power all over the place. This must’ve been a major thing for Maryland, because here in Florida we tend not to have issues like this.
After my DUT had been vibration tested, it was now receiving the “Act-of-God” test, aka HiPOT in the most visceral way possible.
For my own sake, I should’ve recognized this as a red flag. First, you can’t work when the power is constantly on and off again. Even if you could attempt to boot your computer, the power would fail anyway. Second, while I understand prototypes are of limited quantity, if you’re going to take a good DUT and do vibration testing on it, what and how am I supposed to be sure that the measurements I make in my prototype software yield reliable, repeatable, accurate results? Third, the LabVIEW API we were using to talk to the hardware was essentially a SSH terminal hidden behind the scenes, so to speak. SSH is not meant to be used like this. I find it hard to believe that they couldn’t have used an RS232 port, or even a RS232 to USB adapter to talk to the linux console running on the SoC. At least then, you don’t need to pay for an SSH API from a 3rd party company. It’s not a perfect solution either, but then, neither is GPIB, so what else is new? Perhaps, better yet, JTAG could come in handy, but I wasn’t responsible for that so I don’t know if JTAG was used.
Anyway, yeah. That job was ridiculous, and all because it was under the pretentious guise that since they were paying for an expert (at doing FFTs in Excel), I should’ve been done with all the code and ensuring everything was polished up nicely in about 5-6 weeks time. I’ve worked in manufacturing. I know what goes on in these places. You don’t just sit down, cruise through the code, and then expect everything to work right from the start.
Add in to this the random occasional condescending attitude (not just from my supervisor), and you have the makings of the most horrible job I have ever worked. Even putting 50 lb bags of dog food on a shelf at Target was better than this.