Journal98 nov
From Simreal
November 1998
Mon, 2 Nov 98
Halloween has come and gone, and I had lots of fun... not Trick-or-Treat in my apartment complex, but we had a small party and put up a bunch of decorations just for fun.
I did more testing on the Boris Brain Board since the last post -- and it seems my ALE signal is messed up. Most of the time it has only a 3-volt peak, and shows signs of fairly strong capacitance on it... though when I flex the board, it goes to full value. Some of the data lines also look funky... my guess is that my initial crummy soldering job damaged the board traces... so I am going to slowly and carefully stuff my backup board as soon as I can get duplicate components.
When I started with the first board, I had no idea what I was doing -- what to look for in testing, or anything. So I just drove ahead gung-ho, balls-to-the-wall, fire when ready Gridly! Didn't work. But now I know what to look for, and I have test code for the various devices. This time, I drop in the 8515 and test its ports, lines, signals, and addressing to make sure it is solid.
Actually, I'm going to do that BEFORE I stuff the board with a DIP 8515 I have, as a reference. Then I check the stuffed SMD 8515 to see if it is good. Then I add... one component at a time, checking each line of each chip to see if it does what I want. Now that I know I can program the 8515 on board (the ONE thing that works reliably and consistently), it simplifies the rest of the testing. It will be a couple of weeks before this rigorous process is complete... given my part-time status.
If THIS fails... I redesign the board to use serial components instead of a parallel memory-mapped bus. But I'm not beat yet! But I definitely bit off a large chunk this time... education can sure be painful (and expensive) sometimes.
I got an e-letter from Charles Davis yesterday... asking how the project was going along, how Shane was doing with my parts. It's nice to hear from him -- I guess his business is growing quickly, and massive success is looming just over the horizon for him!
Sat, 21 Nov 98
My supply delivery problems are beginning to really torque my twinkies. Two weeks ago I got a promise of parts from this guy (me: "Hey, I want to buy some stuff", him: "heck, I'll just sample it to you", me: "great!")... the stuff didn't show, so I called him up the following Wednesday; oops, order processing had held it up. So he got it pushed through, and then that Friday I got packages! Rip, tear, shred... tubes of thru-hole discretes, and no MCU.
ARRRGGGHHH.
On the phone, "hey, you screwed up! Where's my MCU!" Well, the MCU and memory showed up last Monday so I put the MCU in place and checked it out, and it looks like my soldering is working just fine on this board. Yay! But where's my discretes? Can't do much without address decoding logic. Called again last Thursday... "hey, where's my parts?" Claims they are on the road, straight to me from God who made them fresh. But did I get a package yesterday? Noooooo.... I should have just ordered the damn things.
I need to finish Boris.... I really want full assembly by the new year, and the project wrapped up and done February. I've got things to do! I mean, I've already started side projects...
Anyway, that's why no updates the last couple of weeks. I've been on hold, and the Muzak is beginning to wear thin.
Thu, 26 Nov 98
If it seems that nothing has been happening on the Boris project for the last month or more... well, then you're not alone. Yesterday I finally gave up on my samples (I guess I know one company I shan't be using anymore) and just ordered the diminutive things from DigiKey -- a company with good response time, dedication, and reliability. Not to mention a good UPS tracking interface. If only they carried Atmel AVR parts; I like that series.
Called Shane up too. I guess his class load has him bogged down to a crawl on my project; he sounded frustrated when I called him. Anyway, he has finished the part type, which is good.
I have no idea when I'll get to finish Boris... but my DigiKey parts come in next week, so next weekend at the latest I'll be back to debugging the Brain Boards. I also sent off my non-functioning brain to Randy (Hi Randy!) so he will be fiddling with it, too.
I haven't been inspired to go to my workshop recently... it's just too far away. And though I'm supposed to be moving into my new house this month they haven't even finished the outside yet and might begin the insides early next month... the estimate is January for closing (though they didn't even seem sure of this). If and when I move in... I can get my shop working again, and fabricate all those little metal parts I've been waiting for from Shane.
Anyway, to balance all of my complaints and frustrations, I want to take a moment to notice those things I am thankful for too. After all, it is Thanksgiving today.
I was able to come up with a good design for a walking robot, one that could be built with limited mechanical resources. When I started, I had no idea how I would fund the expensive bits, but that money magically appeared when the time came -- I'm sitting on over $3,000 of pneumatics plus an $800 compressor right now.
All that is left are the electronics, manufactured brackets, and assembly. And programming. And this will only cost about a grand, which I can find easily. I have 80% of my robot here! And that's a heck of a thing. It might even be complete by now if it weren't for the many delays in supplies, my move to Texas, and my over-reaching my electronics ability with the brains (but of course, that is how I learn). I'm still excited about robots, and I want to whip together a monster wheeled robot soon, too. After this legs experience, that should seem downright simple.
Have a great holiday season!
