randomdreams

My manager's in China, so I can be productive my way rather than his, and I don't feel like lighting myself on fire in front of the building.
We have a complicated test system to test a complicated chip. Among the problems: I wrote the interface software, and that took a while. It took us a while to figure out how to power up the interface, even. (The documentation said you plug in one cable, program it, unplug that, plug in the other cable, and off you go, but it turns out you have to unplug everything then plug in both cables because of an error in the circuit board hardware, that I was apparently the first person to discover. This kind of problem is a perfect prototype for the whole imbroglio.)
The main test system has some hardware problems. Those took a long time to figure out, and a lot of rewiring with fine wire to fix. The instrumentation that interfaces with it is fussy, with lots of hidden settings that have dramatic effects if they're not right, and the instrumentation interface software is full of bugs. The software package we use has funny rules for registering software functions with the overall package, and when we're rewriting broken drivers that still don't run, it's not obvious whether we fixed the driver but didn't register it right, or whether the driver is still broken. It gets worse when my manager tries to fix things in a hurry and mistakenly overwrites working drivers with ones that are broken in new and subtle ways, or when I manage to physically break one of the rework wires so suddenly nothing works. Oh, it's worse than just not working: our chip has what we politely call a poorly defined specification, where if it powers up but doesn't receive communications, it jams all the communication interface lines to ground, and the aforementioned interface board, trying to drive those lines, fries its outputs silently so it appears to still be working, but nothing's talking anymore. Since we have multiple chips in the system, I have to get them all powered up and talking in a short window of time, a fraction of a second, or else one or another of them will ground the communication lines and burn everyone else out. We learned today that rapidly varying the load on the chip will cause it to reboot, and that's right back into the same morass of burning out everything else connected to it. Same with rapidly varying the input voltage. I'm trying to write stuff that cushions the loads, but we're still not sure what the critical loads are, so I'm burning up a lot of stuff. The whole project makes me feel sick to my stomach, and I've been working a fair amount of extra hours, working in the evenings at home, and not sleeping so well as a result.

I drove the Spitfire to the hardware store the other day, to buy caulk for redoing the tub surround, because it was gross. When I came back out and tried to start the car, it went 'click' and nothing more. Generally that means either the battery is low or the starter solenoid is dying. The traditional treatment for the latter is whacking it with something robust, at which point it'll run another ten times or so, and repeating that with the revive interval getting shorter each time. Since the Spitfire's going to be my primary transport next week, when my usual car is in the shop for new head gaskets, I ordered a replacement starter solenoid today. But! one of the lovely things about that weird little car is that I didn't even have to open the hood to get home. I popped it out of gear, pushed it back into the parking lot, which has about a 1% slope, turned the steering wheel to point down, hopped in, and oh so slowly started rolling, then snapped the clutch in third, the car fired up, and I drove home. I went without a functional starter motor on the Subaru for about a week once upon a time, but it was a drag so this time I'm trying to be proactive about fixing the Spitfire BEFORE I need it.

Man today was overwhelming.
Last night just after midnight a big storm came through, with a lot of rain. That segued into a lot of lightning and thunder, including at least two hits to the water tower in the neighbor's back yard. This upset Monty, who started pacing and trying to hide.
She may also have been feeling the aspirin we gave her for her sore hips after yesterday's hike, because after some pacing, she began throwing up, and after a bit of that, I got up and sat with her out in the living room. Her intestines would make horrible noises and then she'd throw up, and then she'd settle back down for fifteen minutes before doing it again.
That took up about two hours in the middle of the night.
I'm really tired.
Today at work we spent the whole time in the lab, desperately working on my project, trying to get it working. We discovered: the clock line from the interface to the part wasn't connected, because the person designing the center of the board gave it a different name than the person designing the outside of the board, and the person laying out the board didn't realize they should be connected; the interface software swaps the high and low byte silently; a problem in the silicon means unpowered chips latch interface pins to ground so nothing else on the communication bus can talk; and the hardware that signals a chip to respond is shared among multiple chips so they all try to talk at once.
That represents about seven hours of labwork.
But about half an hour before the end of the day, we got it working, which is the culmination of about a month worth of debug.
My next several days are going to be very busy, but they should at least be deterministic now.

You know that old thing about how as you get old you walk into a room and can't remember why?
I'm increasingly fighting the opposite problem. I feel so short on time that I do this optimization process: I should go downstairs and move the sprinkler line. But if I'm going downstairs I should take the trash bag down because it needs to go in the can and out to the curb anyway. But if I'm going out to the curb I should get the replacement turn signal bulbs out of my car. But if I'm going to get the bulb I should grab a screwdriver so I can pull off the spitfire bulb cover and replace it. But I don't have enough hands to carry all those things. And then I freeze while I try to reoptimize my path.
Is this better or worse?

My manager came in and tried to help me with my project. About two hours in he was clutching his head and saying "this is a nightmare." However, we did make a bunch of progress, because I started detailing every single step of what I was doing, including stuff that seemed really obvious, and it turned out what seemed really obvious was wrong. There's this board, that I have to reflash firmware onto, which I do by plugging the usb cable into the debug port. Every other system I've used, you continue with the cable plugged into the debug port, and nowhere in the instructions does it say to change that, but when I was detailing what I was doing to some other people via the phone, one said "wait, no, you unplug it after loading the firmware, then plug it back in again, and that sets up the comm port for that side, and then you unplug it and plug in to the target instruction side and interface from there."
I wish I'd known that a week ago. Or two.
ANYWAY. So I did that, and then the board didn't even show up: refused to enumerate as a usb device. I measured and it had no power. So I'm all "you SURE this is how this works?" and he replies "oh, you have to change a jumper setting so it gets power from that usb port." I do that, still nothing, he says "you must have something wrong, because it works for me" and then another person chimes in "oh, wait, I just tried it and it doesn't work for me." There's a silence, and the first guy says "oh, I have a custom firmware package that enables that, and I forgot."
That was one of the points where my manager walked out of my office without saying anything and just walked around the building for a while, as I dealt with talking to them.
So I plugged in both cables and ran the software I'd written almost a month ago and it ran just fine.
sigh

well, not just fine. I ran it, not paying close attention, just seeing what happened, and all sorts of errors showed up about halfway through, and I shrugged because I figured that was not surprising, and ran it again, and now the errors showed up at the second step, and that was surprising, and by then my manager was back and anxious about why that was messed up and why a program wasn't deterministic: it should run the same each time, right?
Except the first pass, it (invisibly, but I happen to know) allocates a block of memory as a buffer for communication, and because the program is crashing, it cannot deallocate that block of memory, so a subsequent initiation blows up because the memory allocation call fails.
I felt really clever to figure that out within about five seconds of it happening.
I'm hoping the experience has convinced my manager that I do know what I'm doing if I have accurate documentation, and that if I have some time, I can work around documentation that is missing things and expert opinions that are flat-out wrong. I think all he saw up until now was that I was spending days of effort and getting nothing, when I was actually getting a lot of experience on what should be happening and how it was likely failing.

I was going to try casting another form Sunday night, but got started too late, and burning out the 3d printed material from the mold took way too long. Last night I fired up the oven and completed the burnout. I've learned the mold should have no trace of melted plastic on it: it should be dusty brown where the material has vaporized.
Tonight I loaded the crucible with aluminum, fired up the foundry, then came inside and had dinner, and by the time I got done, the aluminum was molten and just ready to pour, so it took about ten minutes of actual time to go from start to finished. I'm getting a lot faster at every part of the process except for burnout time.
is what it looked like right out of the plaster mold.
A little cleanup yielded this:

There is a casting flaw on the bottom because the wall thickness was too narrow (and possibly because the mold was cold when I poured: usually I pour when it's just come out of the burnout oven, which means the aluminum doesn't freeze quite as quickly.)

But the important part is the base and hemisphere. I can cut off this section, and machine a new one out of aluminum on the lathe, then weld it to the base.
Here's the part I really like: straight out of the investment, 3 out of 4 bolts go through and into the head, and the fourth doesn't because there was an air bubble in the investment that produced an aluminum lump which prevents the bolt going through the hole. That's okay. When I move to jewelry investment, that cures more slowly, I can vacuum it and not have to worry about air bubbles anymore.

So, what's going on.
threemeninaboat is using the Spitfire as her daily driver for the rest of the week, while her usual car is in the shop. I've been working on getting it as reliable as possible. I drove it over to my brother's house last night to see how it does at highway speeds. As I was backing out of the driveway, the Russian guy from down the street came running out, yelling, "I love your car! I love your car! Take me for a ride!" He still thinks it's a Fiat 124 Spyder. I haven't convinced him that it's like the Vespa ripoff of a Spyder.

My coworker and I, after some long phone discussions, have decided to close Mad Scientist Hut. When the fab company who built the chips I based my projects on ran out of fab space and stopped making chips, and our order stock vanished, orders dried up across the range. It was our halo product.
I will enjoy having a bit more time: that was a sink. But it also paid for a _lot_ of really neat equipment. The downside being: I didn't have time to use some of the equipment that it paid for.

I'm overwhelmed at work, because while we hired someone new, while he's coming up to speed I am training him as well as handling all the things I was doing before, when I was covering for the guy whose death necessitated us hiring a new person. I'm not great at multitasking. However, when I have a largely unsolvable problem, like using language one to call a function in language two, that is returning junk probably because I don't have access to the code in either language one or language two, just the APIs, it's really nice to have a couple of other projects that are deterministic to go spend time on while I try to think of yet another way to fight my way to code victory.

The hip crowd in my office go hiking every Friday morning, getting up at some unbelievable time in the morning and jostling up to the top of a local mountaintop, and then coming into work at our normal start time all full of camaraderie and stuff. They've talked me into coming out with them, not Friday, but Saturday, to go climb a couple of mountains a short distance from Leadville, and because I have a particular fondness for those specific mountains, I didn't take as much talking as they'd expected. So we're all sitting around making plans for how we're going to get up there (120 km of highways, followed by 7 km of serious offroad) and everyone's all talking about food and water and one of my coworkers asks me "what are you bringing?" and I'm all "a really thick hat and some heavy gloves." Conversation stops and coworker says "... how cold do you think it is up there?" I'm all "well, at 5AM, what is it down here, like 65F?" He nods. "So that's 20F up on top." Cue everyone pulling out their smartphones and looking at weather stations and revising their packing plans. I don't like to be the crabby old get off my lawn dude, but I have been stuck, on that mountain, in August, in snow so heavy not only could I not see the path I couldn't see the mountain so I just sat down until I could see the mountain again, and I have vivid memories of how cold I was. Soooo we'll see how that goes. Probably it will be beautiful and sunny and no problem at all.

The other night I disassembled all the dishwasher I can get to, and disassembled all the parts of the stuff I disassembled, and went through everything with a toothbrush and fine screwdrivers, cleaning out every snippet of material I could find. It was a gob about the size of a grapefruit when all collected together. Ever since, the dishwasher actually appears to be washing dishes, for the first time in like a year or so. That's been nice. There is a certain question as to time savings when I have to fussy-wash the machine that washes things.

I saw a recursive tattoo the other day, a tattoo of a woman who had a tattoo.

I'm watching youtube videos on how easy it is to change the sparkplugs on my car. Step three is "remove the engine mounting bolts so we can lift the engine up a bit."
ugggghhhhh I made an appointment to have someone else do this on Monday and was thinking well maybe I should just do it myself and save $200...

Dinner the other night was stirfry baby bok choy and snow peas. I think it almost maxed out my camera's color dynamic range.

We went to a movie at the local Alamo Draft House, that started out as a mildly dystopian science fiction film, started heading towards a Fight Club, and then suddenly turned into a horror movie.
This diagram was on the bathroom.


We went out to dinner with my mom and stared at their saltwater tanks a lot.
Kinda wondering if this is a real thing or a painted rock thing.

These were real starfish, smaller than dimes, because they were moving.


When we got home, my dog had eaten my homework. This is one of my evaluation boards from work, for which I cut a custom heatsink that has a tiny recess in it above each point where there is a switching signal exposed on the bottom layer of the board, to reduce how much the switching signal can couple into the heatsink. Monty had removed the bag from my work duffel bag, taken it into the bedroom, and gnawed experimentally on the corner, before deciding that PCB's don't taste good.

We went to Tiny Town for brunch with friends. There were the same tiny houses that have been there since the 1920's.
As usual, tiny houses aren't really my thing.

I do like the Tiny Town Radio.


Evening, out walking the dog between the colorado national guard armory, the sewage plant, and the 1870's cemetery, because that's where the dogs aren't.
For elusis:

Denver has a shindig called the underground music showcase each year, with dozens of bands showing up and giving free concerts on street corners. We went to see Red Baraat, who are kind of bhangra hiphop, and really amazing. The lead singer went off on a significant tangent about curry westerns and the strong competition for the title of best bollywood action composer of the 1970's, before they launched into covering bollywood westerns for a bit. It was a very high energy show, and near the end, the brass section wandered out into the crowd.




I walked by a building full of brightly colored ceramic penguins the other day.


I also got stared down by a fairly fearless kestrel hawk while out wandering in fields.

Yesterday morning, threemeninaboat's sister came over and dragged us off to the Lavender Festival at a remote design center of the Denver Botanic Gardens.
They had some lavender.


Also a working farm, lots of pollinators, and lots of crafts. They'd opened up the historic farm and outbuildings for us to look around in.
This caught my attention in the blacksmith shed.

It's a wheel shrinker. You heat up the steel wheel rim for a wooden-wheel carriage and smash it together using this, so you can make a wheel rim fit a worn wheel. Typically, in blacksmithing books, if the wheel rim doesn't shrink tight onto the wheel you have to cut it, reweld it smaller, and start over.

I set up the casting equipment yesterday and did another aluminum pour.


This time, I fired the mold more gently, but it still cracked enough that when I removed it from the oven, it came out in three pieces. They were big solid pieces, though, so I put them back together carefully and then set them in a sand-clay mixture, put some steel plates on top to weight them down, (because last time a crack in the mold meant aluminum leaking out floated the mold and it only half filled) and poured it.
Here it is just after pouring.

And once it's frozen.


The stuff that leaked out and ran across the sand made a cool shape.


I pulled the mold out intact.


Here's the result.


In this shot you can see flash where the aluminum leaked into the mold crack.


But the result is a sound casting with plenty of material for finish machining, so this is a successful casting even if it's not a successful mold.


This morning Revel came back over, because we're more fun than his dog sitter.


I tried to convince him that we're not so much fun by taking him out for a 4km walk in 90F weather but he said we were still awesome.