Project: Tom Waits

I was watching a bunch of TED Talks as research for uni and came across this talk by David Allen which resonated loudly with this particular project. I’ve had 2 conversations that this reminded me of in the last couple of days, one where I was asked how I’d gone from a panel and a couple of batteries to all that I’ve learned and built since, the other where I was talking about how I keep my work space small, and clear it away regularly before letting it reclutter. Each time the clutter builds again things get better and the creativity flows. This puts it more clearly and succinctly than I did. The specifically relevant part goes from around 10:40 to 11:40, but strongly recommend you watch the whole thing.  YOUTUBE.COM The Art of Stress-Free Productivity: David Allen at TEDxClaremontColleges Productivity guru and coach David Allen talks about "Stress Free Productivity" at…

Project Tom Waits: Why do we fall down, Master Bruce? In science we learn from what we get right. We learn even more from what we fuck up. Sometimes it’s a case of “so THAT’S what happens when you put too much water in the battery” and others it’s “this is why we check the other end was disconnected as well”. Here’s the thing: We don’t all learn well in a classroom. We don’t always learn well from books. I realised recently that the reason I zone out in those situations isn ’t that I’m lazy or inattentive, but because abstract concepts without application literally don’t stick. I need to poke things with a reference I can check, then poke them some more and the knowledge builds cyclically. I need to be able to use what I’m learning. Tom is in a state I couldn’t have fathomed when I started, but I grok every piece of the puzzle in away I couldn’t have if I’d just built a kit. Course, part of that is getting it wrong occasionally. I knew that reverse polarising connectio...

Last weeks experiment was about setting the batteries and panels up for 24v rather than 12v. That wasn’t he hard bit tho. The issue is that my little inverter and main night light are exclusively 12v and will either cut or burn out if you try to drive them too high. You need a power regulator or buck converter to drop the voltage efficiently, and I don’t have one of those. Sure I could buy one, but that’s hardly the point of all this now, i s it? Building from scratch requires some fairly specialised parts I just can’t readily get from salvage. So Tom got recabled, made MUCH easier with the arrival of some crimpable alligator clips so I can make connections without resorting to cable ties that come loose and short out right when you’re shoving 16A through them. Now, I don’t have a buck converter, but I DO have a device that will take a 18-50v and regulate it down to 12v: Brains. Brains is a solar charge controller. It sets its target voltage by seeing what the batteries ...

“This is where it gets good. And by good, I mean: Fucked.” - Ambassador To Chad AliExpress had a sale on just as I was speccing and pricing parts for Project Harry Hart, so I tacked in a couple of choice upgrades for Tom. There’s only so much you can do with salvage and my skills aren’t up to the task of designing and building an MPPT regulator. The Tracer is a more flexible and efficient version of Brains (blue and black unit on the right). I sw apped them over when the shipment arrived, having conveniently just rewired the batteries to handle more current. This also gave me a chance to play with a concept I’d seen mentioned online: using two controllers to manage separate panel strings on the same battery bank. Quick version: it works. Brains pulls juice from my little 160w panel and pushes and extra 4-6A (when the sun is shining) to the batteries. I position the panel to catch more of the morning sun, so by the time the 2 x 250w panels get light the batteries have already go...

The "ghetto fix" side of the project is still going strong. Tom's Brains was overheating a couple of days ago, which finally gave me a use for the old NVIDIA graphics card passive cooler I've been trying to find a use for. 10 min with pliers and a hammer later and I've got effective passive cooling. Testing out my balance charging theory was an exercise in "how the fuck can I get this connected without tearing it all apart. Mismatched cabl e diameter and lengths everywhere. Resistance balance? We don't need no stinking resistance balance! Not in Test anyway. The Prod rewire is a little kinder and a lot neater. This weekend's AliExpress sale resulted in a massive hole in my credit card and the promises of many packages on my doorstep over the next month or two. I have some better tools coming, data logging kit, plus a lot of upgrades. I'll lose some ghetto chic, but I can't justify spending half an hour making dodgy connectors when rea...

Now uni is finished for semester and everything's quieting down I finally have enough brainspace to do an update post. Tinkering continues. I managed to be in the right place to catch when a couple of 250w panels fell off the back of a truck, so there's now ~320w of power available (I'll spare you the essay, but PWM voltage modulation is ridiculously lossy if you don't get the right input voltage). A drop in at Battery Wizards Welshpool yielded 3 more ~90  Ah batteries, 2 of which (Hardy and Strong) have passed testing and cycled into use. Bombadil is showing massive voltage drop under load and may have been cooked, but is getting a test to see if he still has life or if it was a wiring problem. On the topic of wiring, I ran an experiment in balance charging that looked positive, and some reading showed I was on the right track so Tom has now been completely rewired to give me better battery management and even loading across the cells. At the moment I'm ab...