 Computers keep changing the world, but their power and safety is limited by their rigid design. The T2TILE project works for bigger and safer computing using living systems principles. Follow our progress here on T Tuesday Updates. This is the 26th T Tuesday Update. Let's get into it. 26 weeks, that's half a year. I don't know that I really believed I was still going to be doing this after half a year. I'm wondering am I still going to be doing this after one year? We'll find out. So last week it was a big update across all the fronts and introducing this idea of the Worms movable linked list kind of structures. That was quite a bit of fun. A lot of people looked at it, kind of engaged in it. Andrew Walpole has already updated his MFM.rocks simulator. So it now has a version of the Worms here. We can use N and make a Worm. His Wormers are actually quite a bit more elusive and hard to catch than the ones I was showing last week. You can build walls around them and they jump right out. But they're quite a bit of fun. I'm still thinking that we're going to get quite a bit of mileage out of these Worms in all sorts of different contexts. I want to figure out ways to make the software underneath them easier to reuse so that we'll be able to get more mileage out of the whole idea. But I'm still kind of excited about that. Didn't push on the whole lot this week. Hope to come back to it soon. Where we left off was we were struggling with, are we going to be able to do this Rojas reduction of hazardous substances when we actually make the boards using lead-free solders? Kind of what it boils down to. And ETS, doing that is not necessarily that easy for them given the particular equipment they have. So I was looking for other places in Albuquerque that might be able to do it. I did get quotes. They were not so compelling that it was a slam dunk let's switch. Especially because ETS stepped up. So even though it's more work because basically they're going to need to hand solder the through-hole parts to do lead-free. They're going to do it at the price that we had already talked about several months ago. So I think that's pretty much a lock as far as I'm concerned. ETS for the win. I've gotten with UNM to get this whole purchase order thing set up so that we can actually get them paid. There's any number of things that can go wrong as we've already seen all the way along. But it's possible we'll actually start to get these tiles assembled as soon as next week. Perhaps not before I talk to you again, but in the week coming after the next episode we shall see. That has been a very long road. What else? Oh yeah, okay. So the 3D printing. I'm sure you're as tired of this as I am. But here's the news on that front. I was getting these ERM intent bed errors that was causing me to lose prints and it was very frustrating. And I talked to the warranty people at Prusa Research and they were telling me I had to take the board apart so that I could look at this thing. And I was trying to look at it and it really looked fine. And I was much more concerned with this little place where the cable containing the power supply to heat up the print bed so that the prints will stick better and stay flatter. Because it seemed to me that it was something wrong there. And I had opened it up. I had fiddled around with it. I had closed it back up. I closed it back up. And then it seemed it was actually working again. I was getting successful prints of three avoiding that back corner because that seemed to cause more trouble. And I got several good prints out of it. Oh yeah. And actually as I was recording last week's T Tuesday update a triple three cases were printing in the background which I'm going to try to avoid doing. I'm not printing anything right now for example because when I listened to last week's episode with the headphones it was kind of aggravating because I could hear them in the background. But in fact this was what was printing. And in fact it finished successfully with like you know a foot of filament left. So that was pretty good because I don't trust the filament outsensor on this. I didn't need to. However, and I got another one eight hours later and another one eight hours after that. It's great. Until it wasn't. The air tempered mint tempered air came back. This one was not very terrible. It was you know barely started the print. So that was great. But then shortly thereafter I had just changed the reel and was working on it. And this one this one lost about six hours of printing time plus a lot of plastic to get it air mint tempered. So at this point you say well come on. I cannot keep doing this if this thing cannot get fixed. Either I have to send it in. I have to give it to somebody or at this point in the Ackley family there's this thing that once we declare an appliance officially dead. So that like you know first do no harm. Well now I can't do any harm because officially dead. Then I can go in and I can screw with it. And sometimes it actually comes back to life. This is the reanimator. Well so the Prusa I3 Mark III was getting to the point of being declared dead. So well first I took some of the cap on tape and I wrapped it all around the thing trying to keep the thing from flexing as much because that really seemed to be where the problem was. And in fact this was just the reverse. So taping it up like this meant I got air mint tempered airs almost immediately. And even before I tried to print anything and so forth. So on the one hand that's like oh terrible. But on the other hand that was basically all but proving that I was right that it was that something about the cable bed interface that was causing the trouble. So I went back I took a look at it. I opened it up again. This time I did not get deterred by the fact that the fabric cable wrapping was sort of sealed down at the end. I just cleared it out. Opened it up and sure enough right in there. This double black cable is the thermistor sensor wires. The big heavy red and black cable is the power cable for heating the bed. And it went right through it and right there it was really squeezing the thermistor cable. I opened it up and right here it's like basically where you've broken your arm and you can feel it's floppy. You know the stranded wire but you could feel like oh man this thing has really been under stress. Well so since it's dead I opened them up. I cut one and I also took it out from being in between these two things. It's like that didn't seem to be doing anybody any favors. I stripped the ends. I got some stranded wire of my own cut a few pieces. Fired up the soldering iron. Twisted the new piece of wire on, soldered it down, crimped the things in a little bit. Used the hot air gun with some heat shrink tubing and made a replacement little jumper. Did the same thing for the other wire. Soldered it up. Heat shrunk it down. Closed the whole thing up. At this point I left the textile cable wrap. I didn't try to stuff it through the hole. I couldn't believe that it's better if I did because it would kind of keep the wires from flexing quite as much. But I just assumed to see it if they're actually flexing so much that they're going bad. And here it is back mounted up again where now you can see the cables as you couldn't before. And I got a beautiful print no problem. Eight hours later I got another one. After that I said hey let's try the four up again. In ten hours, well ten and a half hours, I had four up. I'm using the entire print bed again. This thing is fixed as far as the Irmintent bed error goes. It's still got the problem with the print fan thing being mistakenly detected as not running. I'll bet you a buck that it's the same kind of problem up in the extruder head. But since there's a software switch, just don't bother checking the print fan. At the moment I'm just saying don't bother checking the print fan. Let's go. And then another four more. So now they're coming out. And so that's pretty good. I think, you know, I didn't want to go to this next level of relationship with my printer. But I have now done so. So I think we're going to be able to finish, except for the fact that there's a ton of wasted plastic. I've put it in order to get another couple of rolls of the Prusament Galaxy Blacks, the same stuff that we're doing. That hasn't shipped yet, but it should be shipping reasonably soon. Hopefully before we run out of the last reel that we got here now. So that's good. On the rest of the parts, in addition to the cases, the main thing was these glow rods. This was a picture I showed back in February when I had just cut a bunch of acrylic rods into these little pieces. So that one piece goes over each LED, which is way down at the bottom of the circuit board. And the light comes up to the top of going through the acrylic rod using internal reflection. So each case needs two of these things. I had a bag full of these things in February. By the end of March, I had run dry, ordered another set of them from Amazon. They showed up in this past week. I cut a bunch more and we're back in business on the glow rods as well. So really, you know, things going pretty good as far as finishing up the physical hardware that stuff. The cases that we can do aside from the circuit board assembly. The one other main thing that we do have is the intertile connectors. We need the connectors between them. I also showed this earlier in the year back when we were dealing with the analog signal noise that we hadn't diagnosed correctly. So I was trying to make an intertile connector that had this special little mouth to put down an array of serial resistors, which series resistors, which in the end I thought were actually not necessary. So last week I sent off a cleaned up version that got rid of the series resistor network in the middle and just generally simplified things and cleaned it up. And those things arrived from Osh Park as well in another one of their perfect little perfect purple packages. And here they are got nine of them. And they're pretty much the same things that we saw before except they have actual less on them since there's no component. Each one of these things takes two of these keyed headers. This is what it looks like when the headers are just sitting through the thing before they've been soldered up. After they've soldered up and flipped it over, T212 PD, power and data. These are the full connectors that we use within the Lotus. The word we're sharing power and data. We're going to have the DOs, the data only that we use between the Lotuses separate from these also have to deal with them. There's a lot more PDs than DOs. So deal with the PDs first. So I started up three of them. Here they are. And here's one of them in plugged into the position in a tile. And you can see the thing is way below the tile surface. So if you had another tile coming in here, there's like no way you're going to be able to get these things out. That's why we have the 3D printed holders. Here are three of them. These were stuff that had been printed up ages ago that had just been sitting in a box. Each one of these things consists of three pieces. It's got the finger tab that you grab onto. It's got the loop that goes around the circuit board. And then it's got the little cap that goes on top. And so here it is. Here's the loop going around the circuit board. It's actually keyed. The loop has a little notch in it there to line up with the notch that the keyed header has. Now this actually was showing that there was a bit of a problem that when I tried to get the U to go through the slot in the circuit board, it was really tight. And I'm not exactly sure why it was so tight just on one side. So that may need to be revised. I could either make the 3D printing width a little bit less, but I don't want to do that because I need the strength. But if I make the gap in the circuit board wider, then I lose copper, which I wouldn't really like either. So I'm not sure exactly what I'm going to do there. But eventually I managed to get it seated all the way down. Here's the U thing pulled all the way through. Here's the finger tab with the holes that those things are going to go up through. It's hard to tell, but if you look here, right here, there's actually a notch in one of the finger holes. One of the U tabs that comes up, but not in the other one that actually keys into a slot on the finger tab. So in fact, this thing actually only orients one way, at least when the 3D printing is being accurate enough. So it goes in and the two sides of the U come up through the thing and they're sticking in there. And then finally the cap goes in and it pushes the U's out and locks them into the tab so that you can't pull the thing off. It works pretty well. And here's the cap going on that pushes the U arms apart and there it is completely seated. And so here's a final, completely assembled one with the U strap going down around the piece. The finger tab up on top and the cap holding it all together. Another angle. And here you can kind of see how it jumps around the key and the header. And here's what it looks like from the side. I feel all right about these things, although I mean they are kind of tall. So it's one of these sort of function versus vision sort of things, right? I mean, as it stands now these things stick up a fair bit above the surface of the tile, which in a way is an elegant and if you have a low angle vision on it, it kind of gets in the way. But on the other hand, the whole point is, is that you need to be able to grab it so that you can pull it out. It needs to go up above the level of the tile. So I think it's all right. And yeah, and here's another one where it was really kind of jammed coming through. I had to push more than I wanted to. And in fact, I busted one of the U's right across the part where it's the thinnest, where it's cut away to go around the asymmetric key. So that's a problem. So either I have to, if I make the cut wider, then I'm going to lose copper. And this copper is actually carrying power or ground back and forth. So I don't want to lose more than I need to. We'll see what I'm going to do about that. All right. And okay, so we're six months in the next six months. Well, I would like to say, I think at this point, we have maybe a month, a month and a half, two months tops. We should have the boards assembled, the cases, power supplies with intertile connectors so that it's going to be much more a matter of software. There's a lot to do on infrastructure software, let alone the higher level stuff like the worms and the cells and all that stuff, which is what I really want to get to. By the end of the year, I want to be up at worms and cells on the grid. I would like to, we were talking about this some in the past week in the chat room about ways of maybe being able to live stream the grid and being able to get people to maybe sign up to run their stuff on the grid and then see it live stream back, something like that. I want to get as many people involved as possible. I would like, in fact, to figure out a way if it would help, if there would be people who would be interested and people who might be willing to invest some time and effort to sort of do a programming class online. Sort of, you know, here's how to program in ULAM or here's what we know about how to program in ULAM and SPLAT. If you're willing to spring for a raspberry pine or run the simulator if you haven't got a UNIX box, a Linux box of your own, you know, okay, you're going to have to invest 40 bucks or something like that to get the hardware to run it on. But then maybe we could meet weekly and try to get exercises and so on and so forth, kind of make our own little class to see if we couldn't learn how to program this stuff. I would like to explore that if there's any interest. I don't know. But we've come an awful long way in a half a year. I feel good about where we've gotten to and it's not an anniversary, it's a half an anniversary but it's sort of like a happy T Tuesday this is from last year but I don't think I ever showed the picture before. So I think we're doing great if we can build this grid and get it running in a month, month and a half, two months tops. That'll be good. And then we can move back towards software. And maybe, you know, I would be happy if we can get to the point where we can declare victory on T Tuesday updates and say, you know, let's have a series finale and then move on to something else, you know. We'll see how it goes. Don't know yet. Thanks everybody who's been contributing, supporting, watching the videos in any way it really helps. The next video, the next update will be out in a week. Hope to see you there.