Open Source Ecology Development Team, 11 Jul 2017

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: OSE Dev Team Meeting - Jul 11, 2017

Description

See meeting notes at http://opensourceecology.org/wiki/D3D_Log

------------

What you see here at Open Source Ecology is an ambitious program based on a volunteer effort. To help us reach the goals - please consider joining as an OSE Developer in 2017-

http://opensourceecology.org/wiki/OSE_Developers

Take a minute to subscribe to our email newsletter (updates, workshops, etc): http://bit.ly/1LtcM44

A

All right everybody welcome to Tuesday. This is July 11 right now, our next ose developer's meeting. Please take a look at the working document, which is let me paste that, since.

A

Okay, that's the current working document. So please take a look at that. The agenda is very basic. Let me share my screen also.

A

Okay on me, zoom the Sun a little bit.

A

Okay, here we go so here's the current progress in terms of developer, developer hours and and contributor, so we're hovering around like a hundred fifty hours per week, which is a good start.

A

Our numbers are hovering about, like eight eight logged contributions per week from the official developer team out of the unofficial, like we've got like 18 people on an official team, so kind of got to get those numbers and sync, but we do have two new developers, Antonio and Alejandro. So two people will be joining the team, so timeline progress, report, critical path in progress reports during this meeting. So this is where we are right now you can click a link for that critical path right now.

A

The one thing that we're going to add to the to the development. These are all the things are actually going on in the background on different things, but we do want to add a tractor team that there will be a big big new new thing. We haven't done work on the tractor this this year so far, but there's the new people as well as some of the people from Saudi Arabia, as well as some old developers, we're going to do that because a September we are planning on building in the next iteration of the tractor.

A

As far as this, this critical path right here only goes up to the brick press workshop, which is August 25th. So that's the last last part right here. Cd power cube build workshop; okay, that's the 25th! We're going to get that announcement out pretty soon so workshops. Announcements are here, 3d printer announcement that we have to get out by this Saturday. So the the workshop will be August 12, that's right here, factory farm, and that means from this Saturday. That's exactly four weeks to that workshop.

A

So that's that's the kind of timing we want to give people for weeks to sign up for the workshop, so I'm developing the production capacity go to slide four. That's that's! The recent recent work from on-site right here. So I've got a total of seven printers built at this point. Two of them have been tested: five more have to get all the testing testing on them and all that, but that's late last night print cluster, so that we can effectively print parts for the next workshop. Like last time, I had two printers going.

A

Those were the loaves about minis and it took about two weeks to print just pretty much going constantly as much as I could took two weeks to print parts for twelve printers. In the last workshop now we're going to try to get more people signed up, I mean the gold ideally I'd like to have 24 people sign up. Then we need a little more production power so that we're not pressed for time in terms of printing. So it's seven 3d printers that should be I mean right now.

A

Seven I've got actually three more frames that I could build up, but parts for the workshop, as well as for the CNC torch table, which get to be much bigger parts. So so we need this, and this is really good and with Christian we are actually going to meet we're planning on meeting after this meeting to discuss the print cluster as far as the software, the Raspberry Pi getting these connected so that one single computer controls it through a Raspberry Pi.

A

So that would be a great thing and- and at this point we can be talking about you know actually selling parts like kits for the 3d printers or or the CNC circuit mills or whatever we can be using these to produce more more parts and do a website. I mean I'd like to see a web front where we developed the whole process, the whole. Here's how you build your printers, here's, how you run the software!

A

Here's how you run a website sales website, where you can actually print parts I mean I, can tell you that for us there's definitely a lot of a lot of uses like savings are saving ourselves trips to the store when we need plumbing fittings like right now, we're working on on a CD go home and we always need some kind of a fitting, but I mean that's the kind of stuff that we can replace pretty much completely like.

A

If you talk about house building so I mean very practical and making parts for for kids for for more printers, so getting the full open source infrastructure for for a print cluster, including things like people can actually log in to our computer and use our printer remotely I mean. Is that possible? Certainly it could be possible. So, for example, one scenario would be where we, where we just keep several printers, how many we want, possibly with a video camera, and you just order print time and we can perhaps ship the parts out to people.

A

So that's that's an interesting proposition. I.

A

Think one of the challenges is, how do you know print parts are going to print well, I'm, thinking that to address that issue is we have a storefront of accepted parts, parts that people select from a menu and that would assure that the prints are perfected, because we know how to print them well and all the settings are correct and so forth or they're just designed properly, so they are very easy to print, but it's very exciting I mean this is this is really nice just to tell you a little bit about this build I mean it took a bit like all of last week, actually spent building the 13 inch version that that one back there is the complete one.

A

You can see here how I'm kind of fitting all the wires some of the last steps into the cable chain, but took a whole week and then just one day pretty much to build five more because once everything like all the dimensions are proper.

A

Like all the little details of how you know how you the cable routing like, for example, having to do the cable routing and redoing it or refitting some parts, etc that took the time but right now the machine is stands at seven point: five or seven byte 8 inch, build platform and seven point five inches tall and that's using the 13 inch frame. So that's very nice, a nice big build volume, heated bed.

A

Everything is working like all the parts here that you see have been printed with the 16 inch 3d printer that I built here. So the d 3d 16 inch version and once all the parts are assembled, it's really nice, like a good production line, would work very, very well. What I did was prepare all the cold printed parts, the the axes put the motors on them, so XYZ axes prepared the whole a extruder assembly, that's kind of like a big job, bunch of little pieces there and then the whole printbed assembly.

A

Then, once once I had that, once once I started putting the things on the frames, it literally took like two hours to put all those pieces onto to make like from from finished frames and from finished assemblies. It took like five like two hours about to assemble five of these printers, so that's I mean it's pretty amazing. I think the workflow could work really well where, instead of each person doing going through the whole, build we build in teams.

A

So each team has a module and people can switch between teams, but that workflow kind of like a production line works really really well because then, at the end you assemble it very very rapidly, and once again the trick is like the orientation of axes like how the belt is attached and I'll go to the next slide. Here you can look at this video if you go to view present mode, but here actually Shane who did we did the CNC circuit mill, which is what you see here with a spindle.

A

He shows how to tension the belt, but the concept is whether you attach the belt in the upper or bottom hole means that the 3d printer, the axis, would be moving in opposite directions, depending on where you attach it. So that kind of that part is the most tricky part about the entire process like when you build the entire machine. You really want to build it once so in the build here. I had the one machine right in front of me and I modeled all the parts of orientations.

A

It's really about orientations, because I can tell you from last workshop people on average built the 3d printer like three times, meaning they did it. It wasn't the correct orientation, so they had to flip things and so forth, and then, of course, you can fix that in software.

A

But then everyone would be running different software and we wanted to say: okay, the software is going to remain the same fix the physical structure, so that everything is is turnkey upon further down the road, so it works having one model and then assembling everything where you're looking at the orientations.

A

Orientations are rather tricky because even my cell phone I was looking at you know: I'm, making five acts, there's five printers I'm making five axes, and you have to pay attention to a few details like which side is the motor attached to which side is the belt attached? To which side are the tops of the bolts, because then it makes it easy to take the bolts out, etc.

A

So yeah just little details, but definitely definitely tricky yet now I understand why everyone may build the 3d printer three times, because it's actually a lot of detail so I think a workflow where you have a whole team working on modules and there's quality control, because when you're making the whole load of modules, you can quality control at the point that they're being made.

A

So you don't have to go to one person, then another person, all the modules are right there and you can look at one and then you can look at all of them to see if they're identical and that way you know you're right, so that kind of quality control would be much much simpler to do because you're going through a whole set of the modules at the same time. So that's just some notes on the build okay moving on here.

A

So that's the the current work on the on a CNC circuit mill is that Shane has is working on refining the circuit mill to get really nice circuit milling. So we can make our own circuits for everything like controllers and everything, but they're, actually considering writing a small paper on the characterizing, the backlash and all the motion characteristics of the of the universal access system. So that's it's really good.

A

This stuff is spreading to other places. I do really like the axis. It's very flexible I do think it's going to be become widely used in various contexts, because it's a construction set. Okay, next we'll go to lineman, filament extruder. So, as you see it, it's coming together, so the thing is about 3d, printing, the most expensive part is filament, so we're going to make our own filament. So this is part of the filament extruder. If you go to lineman filament extruder page on the wiki, you can see see this work.

A

This is the part library, Thank You Roberto, for doing that. Dude he's assembling a lot of different parts. These are larger a lot of 3d printed parts or some metal parts here, I'm, seeing that Will's doing different parts here, I'm seeing Roberto doing that I'm seeing Abe pulling in some parts. So that's really good.

A

It's coming along together, so so quite good and then just the last thing on updates is we're working on a new, so Joseph is whose contact at you we're working on a new developer invitation, video just to update people with the latest progress and make a more compelling video and we're looking at at creating a the application process would be a video cover letter instead of interview. So once once a person, the interview prior to the freakin exam, it would be a video cover letter.

A

So basically answer all the questions that are in the interview through a little video so that that takes there will be much less work on our part. It would put the more burden on the applicant, but if you I'm keeping detailed numbers and the fact right now is it takes me five hours to get one person on the team. One active developer, wow, that's a lot of time. Actually so I'm kind of getting tired of it, I really need help and I.

A

Think part of the part of this can be simplified by a simple cover letter intro like video, eliminating the interview so that we get people on faster. But that's the facts. I mean one in three end up actually doing the test, so I send up spending this time and then and then also like with the numbers of the actual numbers on the active development side like when you look at the numbers say: there's like eight active developers right now, the numbers after all, is said and done about five hours to recruit one active developer.

A

So that's that's quite a bit of time when you take a look at it, so I'm glad I'm, taking them keeping data on it and so forth, and here's our overall progress but I want to see this definitely go up. I mean we're at like when you talk about 150 hours per week, that's about four full-time equivalent. So definitely you see a lot of room for that. The teams can be bigger and we can actually start breaking up into different teams. So we want to start the tractor team.

A

The tractor team would would begin with so I'm seeing that, let's see who we got Anna, we do have Alejandro. We've got yeah welcome as far as any of you I think we can actually start the tractor team pretty soon, so that would be. That would be really good to get that going.

A

So what I need to do is prepare a document where we have basically the starting of a working document presenting all the work from before I mean there's a lot of work. We've done on a tractor. If you go to a tractor genealogy page on the wiki I mean you can go through copious amounts of all the prototypes, so so life track. Genealogy starts with life track, one with articulated, steering four-wheel, drive, etc. Up to the more bulldozer like devices with tracks up to number eight track tractor of 2016. So you can.

A

You can see more of that if you, if you like, on the wiki but yeah, that's that's good. So we present the prior work start with the requirements. A concept, one of the things we're going to do is also solar, solar drive, which means a tiny electric power cube.

A

The good thing about hydraulics is that even a one solar panel, you know, take one solar panel, one single solar panel. You can drive a heavy tractor with it will go very very slowly. You have what you have is a very tiny power unit power cube, but it can go very slowly so for certain devices. That's really good, like, for example, if you want to do weeding the real application that we want to do here is actually to be pulling chicken tractors. So that means cages with chickens.

A

So you can span a whole field while keeping the animal safe, because everything likes to eat chickens. So you keep them in a cage that just moves and therefore, instead of a cage, they have the effective area of say like an acre or a few acres, because that you're pulling this cage with a solar, GPS tractor, that's moving very very slowly, like maybe a you know like a meter a minute or something like that.

A

Something like that. So that's just notes on the tractor team: I'm going to go back just a little bit to the the critical path here so on a CB press, it's actually time. So the workshop is coming up pretty quickly. I mean it's the next month, the 25th. So about a month. Wow I mean that's like a month and seven days from now, which means in practice we got to get get the updated design and also the CNC torch cutting.

A

So so as soon as I got that print cluster up I mean I'm going to be working on it today, we're going to work on it with Kristian, as far as installing the Raspberry Pi and things like I'm, going to start printing parts for the the torch table and work on that, because I mean we're talking about so torched part torch table. Part printing is right here. It's supposed to be happening like right now. The green line here is where we are currently July 11. So.

A

Torch table part printing and then the build I mean worse, that's starting to build and then part cutting. We have to start that within a couple of weeks, so schedule is pretty tight, but it's moving along, so things are good.

A

So last thing is: let's maybe go to the follow up on the work with people who are here are the things that I missed right now that we need to update, because the thing I think we definitely want to get going is if I have the print cluster going, we can actually start prototyping the parts for the filament extruder, and that would be the filament maker, so we can make our own filament. So that's that's one of the big things to to discuss. I. Think a lot of the team is working on that.

A

Let's see, did I miss anything as far as the status of kind of updating we've got a lot of parts being put into place. I haven't looked into that in detail. I think Roberto knows quite a bit he's been doing the final final assembly here, but the next steps would be to yeah I mean get the final assembly and move forward to to ordering the Bill of Materials parts and starting the actual cutting as soon as we we know from the final design that everything is making sense.

A

So, let's see Roberto, do you want to pipe in or Abe anything on where we are with that or did I cover just about where we are.

B

A

B

B

I need assembly for this two first models: yeah.

B

I used on these, in particular five bark.

A

B

Saw that some people keep doing very complex parts, but with many details I curious about that, because I understood that we have to do the only thing. The back part yeah.

A

Yeah yeah, so so, for example, like I've seen a lot of different simplified parts, but definitely you want to do it simplified versions, because, if we're going to put together the entire entire machine and freecad with all the parts in there, of course the files would just be too heavy right. So yes do the simplified versions for the STL files. Of course they remain complex and they remain and then then we print them. So that's what we would do. They can remain the original, complex, STL files.

A

But when you do the cad model, you don't use those full files. You you like to use simplifications, so you can manage the whole cad very easily and in a fast way. So if anyone's like drawing parts, there are too complex, you don't need to like I'm. Looking at, for example, I mean look at this I mean that's pretty much. What we need power supply, just representing all the different yeah I mean this is a very simple, simple concept, so you're representing all the parts, so you can pretty much say, label them and say: okay.

A

This is these. Are the critical parts that we have labeled, one through n, and that way we can like from the CAD, it's very clear like that. We have the entire machine that we're not like buying extra parts or whatever the build materials comes clearly out of the CAD.

A

So, let's see what would you say Roberto? Where are we in terms of the final like how much work do you estimate still needs to happen to complete the extruder parts.

B

B

Easy to assembly at those parts mm-hmm the other issue that any spot haven't didn't get is hidden pretty so.

A

Right right, so, on that point we should mention is that whenever somebody's drawn up the parts you have to upload all the files so make sure that when you generate a part, save a file version with the with the underlying sketches. So you can edit that and then, if you're doing the final CAD like the the overall CAD to minimize the size, you can simply eliminate all the other details and in and so just upload a simple if even simplified version.

A

So you have a simplified version with sketches and then you would have a simplified version. We just eliminate all the sketches and then save that just upload that to the wiki as a new version of the file, so that in the file versioning you can see. Okay, one file is the original. It's got more detail or or more of the underlying sketches, but the ones you actually putting into the assemblies don't have that they don't need it.

A

You know, because, when you're working with editing the assembly or the parts you're always going into the individual parts and then merging the edited part into the more complex assembly, you're, not editing within the complex assembly, you just take the individual part edit that in a separate document. So that's the that's the workflow that everyone needs to understand, but you do need to have a version. History of everything like, for example, file, thermal components.

A

There's one file here well: I can say that right now you probably want to have more files, but you're saying it's without sketches, so who's done the one with the sketches. Where is that? Do we have that, for example, I have I have that five yeah so.

B

A lot of the full assembly four models have a: they don't want to replicate many files, though, because in food and fuel assembly there are two versions, one with sketches and one without yeah.

A

For the full assembly, we should probably have just the without just to save memory, so it really loads fast and is easy to work with. So we should have the multiple versions at the component level at the part level. Ok, yeah, that's that's an important point, because the workflow is do all the editing on the individual files and then you're just merging, in other words, loading that file into the final document. So you update it and then you load it into the complex assembly.

A

That's the preferred workflow so that it's it's just simply more manageable, because then the final assembly is just as simple as possible and it's easier to work work with it. I mean definitely it's harder to to work with in a document where there's so many different parts and you're trying to edit one it's like just navigating through that whole thing: it's if it's a very complex thing would be difficult. So you just eliminate everything you you have just just the single part file work on that it's easy and fast and you can't mess up.

A

You can't mess up it or if you mess up you're, just messing up one part, so it's just a preferred workflow. So it looks like from this page here that we have so this is a lot about the extruder different extruder parts, but looks like we're missing now the filament winding aspects so Roberta. Would you say that the that the extruder part is pretty much got all the parts in it or is there a bunch of parts still missing like there's a whole load of them missing.

A

B

I tell you: I was working on one of the other modules.

A

Okay io: do you want to update us on the line Amy.

C

A

We can yep oh great.

D

Okay, so I popped on the fullmetal them and half of it polluted, virtually all the SEO power Baku walk on, but that that's out there that still have a steel file attached to them and the higher the number of them still might need help with knowing also access fee at the above. For this one also I can create people power for them.

A

Uh-Huh, so are you saying that some of the STL files are not loading.

D

Up, so if you, if you took that the spotlight free under spool mechanism, there STL file missing for almost half of them,.

A

A

Okay, let's take a look at that: can you paste the link for everybody yeah.

D

Also for fun spoon mechanism, again, I really had moved you out little the image that we have for it, because that there's so many parts and I just can't see how to represent all those parts from that new image. Also.

E

D

I saw to us maybe reversal somebody more specific if we have, if you mini that it stood it path, diagram of something like that stir that and I draw our roots for comprehensible, yeah.

A

So, let's see how much inside we have so in the yeah.

A

So the main working document here so yeah we've got I mean this looks pretty good I mean this is, let's see, let's go down to the spooler parts.

A

A

A

A

Filament winder, for example, when you look at so. If you look at my screen on the filament winder I mean, can you make sense from that a lot or that are you saying? A lot of detail is missing beyond what you would have with what you would need like when you look at this it's this just doesn't do it.

D

Yeah, so just looking at that that image on slide number 10, which represents the collective end,.

A

D

Looks to meet all the way I understand. It's me like a lot of the component that we can see in the pad library that that hidden like couples in you can see them from that image. So I I, really I really count how to like label that go into that image to show where the different components book, one or two of them are cleared. I can see. Forgive me and I can tell what they are. The.

A

D

I am assuming they're like very big behind away from the point of view of that shop. What.

A

Happens when you compare that to slide number 15.

D

Like number 15.

A

A

I mean: can we identify all those I mean, there's a lot of different parts.

A

Can we address.

D

I, want yes, is that that this is what I was thinking when I was mentioning an exploded path? Diagram? Yes, so this this probably was needed that movement in the past I.

D

Probably could identify all the components from this I.

E

Think it's important to remember that some of these photos and the documents they're from mr. Lanham as PDFs and some of the components and there are actually deprecated the photos- still show things that he quit using and also some of the most everything is in the PDFs I think one way or another, especially in those diagrams, so go back to the PDF and the diagrams. If you can't see the detail, those diagrams are pretty good.

A

E

If we should have just used the diagrams more instead of the photos, but photos show a little bit more realism to so uh-huh.

A

Photos so you so, which which do you want to use them photos or diagrams that you think are more.

A

More relevant babe.

E

Yeah I think in in the hi-tech to some degree it almost better would broken some of the stuff down and more modular pattern. Earlier kind of, like we started, do.

C

It in the last.

E

Modules well as soon as share all the parts of the budget line up trying to organize some of the stuff in the sheets and stuff to modules, because it seems like Louise what the modules to match how the system might be assembled in real life or in a workshop situation. So I don't know if we've got all that exactly in that matter. But it looks pretty good.

A

E

Done on the modules I see a lot more detail and I see a button switch and assembly work on the assembly of the modules did too surprised by that I did notice. I was was working on a switch and I guess. There's some overlap between the spooler and the extruder. We have different people working on those I. Just noticed, I was working on a switch and I see somebody already finished. The switch with the spooler so probably ended up with some redundant work there, but I need to double check some of that I.

A

B

You have to consider your friends nature.

B

Maybe the low sixes are because I think.

E

That that maybe yeah I guess we want to use the same things I think and it wouldn't really matter other than that the plastic heart design might be a little different, I mean usually that's which it should be the same. Specs I mean. Hopefully we can just buy generic half-inch switches and they should fit, which is what it looks like what I was working on, but because the switch design doesn't matter that much as long as it fits in the plastic.

E

And ideally, we all have to redesign the plastic printout, although we have all the free catch well, that shouldn't be hard to adjust either if we have to change the print for the switch to fit a little bit or sometimes you have to do that anyway, as I understand.

C

With printouts.

E

It's kind of hard for the plastic pieces to fit together just right, sometimes there's a little variance.

B

One time you said that that speech that was uploaded have enough cases or bring the flow most readable of any way. We have to make another issue.

E

Okay understand this I think to simulator switch is in the extruder and probably different. Toggle switches in the spooler I saw recall a yeast bottle ones. I think I won I think that the switches ease were mostly just whatever was the easiest.

A

E

Supply reasons we I mean obviously we're trying to duplicate I guess as close as possible, just to make it easy, but from the supply standpoint, if you can use all the same parts that would be ideal to but new that might have to be changed later.

A

Yeah um aids, since you did the diagram or you labeled, some of the parts in the slide 15 of the linemen filament extruder doc is that are there any deprecated parts in that or all those parts I mean? That's it? That's exactly what we're going to build. Yeah.

E

I think that the diagrams are more up-to-date. What I can tell there's not really anything displayed? He may have changed some external things that aren't shown there like I. Think in some cases he end up using some of the rubber tubing is displayed knows it is abusing things like duct tape. You know.

A

E

Thanks to because that was just simpler, so obviously that's not going to be displayed in the diagram mm-hmm.

F

E

It's not a significant thing either so yeah.

A

E

Everything heart wise is shown pretty clearly there, okay.

A

So, for example, let's see so that's what are these blocks here? The gray blocks that some of the electronics.

A

Like that's power supply, as that.

E

Motorcycles out see the larger ones electrode efficacy, then the power supply I'm talking back at that PDF and then three six.

E

Control modules: it's like delicious to be the ones with the green underneath we'll take the second boards, probably so.

A

Okay, so let's.

E

Go explain those other parts in that diagram, yeah.

A

Yeah so I OH, for example, I mean how many of the parts within this diagram do. We already have I think you can identify all the printed parts right I mean you I, think they're pretty pretty detailed here. So you do you think you can figure all that out. As far as all the 3d printed parts.

D

Some of the possible familia from here so I can try Oh. Would you.

D

So how'd I approach it out did I try to reconcile with it with the numbering hair. Oh I, should okay, I should use the not brain on the.

D

On the photo library, I guess, yeah, so I guess I'll I'll, lift that drawing and try to try to apply the numbering on on the pad library. I guess that one is done. Yeah.

A

Let's see so if we go to the part library part, so parts gallery spooler mechanism here, yeah, okay, so we recognize some of those pieces.

A

So this is the stuff that's missing yeah those would be, and those look like, maybe they're, not the 3d printed parts so with all those three printed parts of it. Let's assume those are all pretty printed and the rest are not um yeah. Yeah I mean it looks like this yeah, like all these screws and everything I mean see. If you can, let's see how do we coordinate that one diagram with the part library?

A

What's the best way to go, because because we initially talked about okay, let's take let's each person make like like to give you an example like this like what what Roberto did here.

A

Basically, just label just take just about everything and label it and make it very detailed, but but at the same time like Roberto, maybe you can feedback on this. So this is a diagram. That's pretty detailed, but but I mean there's so many little other pieces that are missing like, for example, okay, like what about this bolt this wire, any other things like. Where are we finally reconciling that we actually have all the parts because maybe like this goes from 13 parts, maybe up to like 50?

A

You know like because the idea was that we break it so here in this diagram and slide number 2 of Liman filament extruder, we broke down the thing in two parts and then we said: okay, every person, please label a part pretty much completely so that each so we have seven altogether seven diagrams with a lot of our rows to them.

A

Do you guys think that would be doable and maybe even like? Go to here? I mean I like this like this, this we can understand: okay, there's, seven, seven main parts and then each one of them maybe take one or two pages. So you do have absolutely every single number attached to a part and then that would almost arm that would almost supersede the part library like where we should probably do is use these links in our diagrams to make everything clear like that.

A

We're not missing parts like, for example, here that we label absolutely everything. There's a little bolt here, I see like little bolts. There I see bolts there I see, you know all these different components, so we can visually, observe it and say: okay. This actually has everything, and then we can go to the part library page and we can say: okay, that's where we're just storing the files or representing the individual files and also a very clear way like say you wanna, it would almost be redundant like the part.

A

Library is almost the same as this visual index except the part library is more. You know it's just a list. It doesn't really get you to understand how the parts fit together, but they will also like if you want to see like in this sona's diagram in this visual diagram, you don't necessarily see all the parts clearly.

A

So the the part library would serve to make those parts very explicit, but the most important thing in my view, right now, is to make sure that we are not missing parts and the design is complete and integrated that you know we start building this we're printing out 3d printed parts.

A

We can do that. Definitely, and we can see okay, it all works. So that's level number one and then level number two is put in all the other mechanical components, fasteners motors wires and everything else. But what would really be make it effective is when this diagram has just about everything so that you don't you know like say, I'm, building this there's somebody's prototyping it actually right.

A

Then we don't see ups, we don't get into the conditions where whoops we missed parts and then you have delays like you know, every time you got to get a part, it's like you know a few days or maybe even a week, delay and you go through those several of those cycles and you end up spend a lot of time to get all the parts. So that's why this visual diagram should have everything so can I propose that it there it like so I would expect seven of these diagrams.

A

So so we got say one say one here: you know, let's see one here, so let's call it. Let's get really specific here. So this is one.

A

So we're going up to seven now, let's make it bold and.

A

A

So was this: this was a.

A

Hey this is yours, right.

A

So a for example, can I ask you that in this one, okay I see seven seven items nine items, let's get more more more details, bolts, little bolts wires, so maybe like set another set up, another page four bolts and wires or or add more to this.

A

Does that sound a, but does that sound like a good idea so that we have this and then maybe one ABC? This is like all the main parts. Maybe the second one would be.

A

You know second, one would be hardware and the third one would be wires or something like that, because each one of them is pretty extensive able. What do you think of that I mean? Does that make sense or because because the question is, how do you make sure we have everything in our diagrams, and so when we actually go build it, we can. We can, from this diagram, generate a complete bill of materials. That's actually accurate, it's not like, oh this part may or may not be there. It's for real hey. Can you comment.

A

Hey did you disappear.

A

We can't hear you audio issue. Okay, can you type in those is what I said makes sense, or is that not make sense, go ahead.

B

You're saying Batman, you know.

A

Yeah, that's that's correct in a sense that some details may be missing in other places. Maybe he tells you like a couple of different ways without specifying a part or there may be deprecated parts, that's what I learned from parts that are no longer used. That's what I learned from Abe I haven't examined the document and so much detail myself, but reading through it I can see it's not it's not an exact document. It might be a little general like it will not give you an actual part. So that's why we would have to.

B

A

Free cat is useful, yes, because then we can generate everything else from freaking, yeah, yeah and first and the first answer to your question is that he has not produced. We asked him for the cat, but he's not. He hasn't, provided it to us. So that's why we're rebuilding everything and also for a workflow read simplifying parts. So we actually can manage and work with this very effectively within cat, so we can go through all the other different assets that we need from from like, for example, language, agnostic, instructionals.

A

Even you know, fabrication diagram is just technical drawings. Everything exploded, part animations. We can do everything from that. So but that's the idea, like the shirt, the shirt. What we're after here is that we generate a model that is actually complete and if you look at his model like, for example, take a look at his you know. Page 15, like is his model complete.

A

um Possibly he might have all the parts in there, but but some parts, yeah yeah I, mean I mean this kind of design. What you have here is quite good, but I mean it's not in free cat. So we I mean we don't have access to it.

A

We don't have it. So that's why we're redoing it now. Roberto is what I said about sorry now yeah now we can hear you yeah go ahead. Okay,.

E

Yeah some of the issues would be the PDFs, and things like that are they can be a little confusing, but they should be a huge problem. I think that all the parts that got listed so far or things that aren't difficut, especially in the in the sheet list, so sometimes those are a good reference to and I know, there's a lot of different files to reference to some degree. Event number we haven't ordered so far. I think that the think the most accurate list is probably to split those master part indexes to some degree.

E

As far as the part sent you to get all the parts into sections, with the modules that I thought they belonged in. There are some parts that are kind of general, the bolts. Obviously they just go all over the place, they're listed where it seemed like most appropriate or just in a general area but themselves, but I think that the part index sheets are kind of broken down by the modules.

E

Although some of that may have changed, I think the way people started displaying stuff by the photos in the joints and maybe I saw some Assemblies of more parts as well.

E

A

E

Starting to work on the broader assemblies, instead, besides just the individual parts, which is good, honest it if there are certain simply stages, they seem to match the kind of like the photos, which is good, because that's kind of the way is some of it. Those PDF students do a good job of showing how he you're in those certain some close link for the wiring on the electronics, is thinking that once that plastic shell, the top, is a symbol.

F

E

The wiring would be put in you know at that point, so some of those things get them assembled in in certain groups and then we can add some of the detail that way so there's probably certain ways to break it down, maybe more modules or once we have the full can. Of course we can just kind of hide the parts we don't want and show steps to assembly that one as well obviously yeah.

A

E

Most of the plastic parts are done so that has helped a lot would be overall simply yeah.

A

Yeah definitely we're getting closer, we're, definitely getting closer. So what I would suggest here like? Yes, the master index is very useful by all means and I think the way that our workflow can go is we we go to the so. We've got this, which is the modules breakdown.

A

Then we have things like this, which is the visual diagram for each module and I would also call out for okay, let's do a more detail for the bolts and wires and any little details. So we detailed this out more and then once we have that, then we can go back to the master index like you are saying, and then we can say: okay excellent now we know that okay, there's exactly like you know like, for example, for the extruder electronics. You've got 12 parts listed well.

A

I can tell you right there that there are missing parts like screws right and if you don't have that screw on hand, you're not going to be able to build that. So we would then take our visual diagrams and details and then add more parts to basically finish all these master indices completely. So we can go from the visual to the master index and then from the master index, I'm going to go down this purchasing link and buy everything and get it get it shipped or go to the store or whatever.

A

So this so for me, the end product is this column right here? The purchasing link, that is a verified list. That's been passed through the CAD through the visual diagrams and it all makes sense to us. So that's a deal but I also say you know like wire or wire connectors. We have to be a little more specific, like, for example, wire connectors female Spain.

A

Well, what gauge you know like, for example, when I click on this link, which one am I going to choose, so that should be detailed, a little more out. Do we need like three different ones like there's just spades or there's some ring or whatever just just a little more details so that when we do it it? Actually, you know it's actually working it we're not we're not like. Oh, we didn't get the right one. We got to go back to the store and then spend a week.

A

You know lose a week doing that so yeah, but does that make sense to everybody and it's so like our yeah. First of all. Does that all make sense.

E

Yes, I was going to draw in after he would actually draw the Spade connectors because I kind of report to the electronics. You can't really you need all of that stuff listed it shout in there to some degree. It's not really high-resolution exactly.

A

E

Wiring can be done, I think fairly. Easy I did work with path before on some of that dull stuff, so I think I think I, don't know how that works. It should I, don't think it was that hard to do those paths, and you know just putting some colored wires in.

A

E

And I added the detail to the power supplies of some of the ultra components, so that it would be easier to lay a ball and maybe show some connections from ironing a little easier, I think that was important. We might need to add a little more detail, some of the electronics that should be hard to do maybe fasteners from the fastener workbench. That kind of thing can help with that. No no I just chose to cruise on the spades or wired wiring points.

E

A

E

I know that that probably a lot of that probably needs to be left out of the larger simplified diagram. A kazoo I, don't know how much in fact they don't know how much wiring might slow down. There is too much memory in that and the larger view, of course, also that's inside the box. So if you're, just looking at the box, I think that's all included and if I understand it freak at the display in the rendering speed should be based on. What's what's shown on the screen.

A

Yeah yeah, that's right, but that's the case for, like you know, say you have a hundred parts in a document, but each one of them is 5k. While that entire document is 500k, that's not a problem at all. That's I'm saying! As long as we've got the simple individual items, we should not run into any issues at all with respect to being able to render that and work with it effectively the CAD so yeah does that make sense. It looks like yeah yeah, it does okay. So so, let's, let's just discuss this refine process.

A

So this is what we talked about before we're talking about creating the full CAD and see Roberto log for final assembly example. Let's see final assembly example right there yeah, like that's, that's what we're showing today. So we are doing that and now we're we're looking at the module. So basically, let's go.

A

Let's kind of start from the beginning, as far as what we're actually doing according to what we talked about today, because I think we're getting it, but it's just just complex because it's a complex, you know it's quite a quite a complex machine. So let's say okay, so we've got break down into modules, just so. Let's just document this whole process, so everyone is clear, so break down into modules, which is page 2 I'm, going to link to that directly here, so so break down into modules.

A

Okay for each module.

A

Detail the parts are.

A

List the parts with arrows, so a good example of that is, for example,.

A

Well, let's, let's take this one, that was the first one. So, let's, let's just take a look at so that's a great example.

A

So we're pretty well on that, but overall I think we're missing a few like I know: we've got the big, so that's two. This is three we're looking for seven. This is like four.

A

Okay, this definitely needs a lot of work right. Schooler, electronics, like we need some labels on that.

A

Six is the tension mechanism and then seven is the school, make it look great okay, so we've got placeholders at least for those. So now all these remaining ones need a lot of work right so label it. Exhaust label them exhaustively.

A

So then continue working on each module.

G

A

A fasteners slide, fasteners plus small parts slide and then add a wiring. Add a wiring and electronics detail.

A

Yes, so, basically, that the list parts with arrows that was basically for the bulk structure.

A

Which is obvious, but but okay, that's looking great, but, for example like with number six, we need to detail absolutely everything that bolt there that wire there etc. You know what that's a setter, so we go through that. So I think that's like the you know, these two things so probably end up with two or three slides for each module.

A

So I'm going to say here end up with two to three slides for each module. So that's that's a total of it. There are seven, so that's like 14 to 21, slides total and if it's easy, just use one slide, I mean oh now, let's, let's call it about two slides I mean some of them are going to be much easier than others.

G

A

Add links to very specific part ordering sources.

A

A

So I mean I mean before we're there we're doing the CAD I mean do the simplified, CAD think about think placeholders. Basically, so we don't do not miss anything so simply enough that just represented as a placeholder okay, then we add links to the specific part ordering sources. Then we go back to the master index and we we continue filling in that master index to make sure all the parts are in there.

A

Okay. So that's the next big point so then go back to master index.

A

And fill in any missing parts.

A

Great and then what else is left Venice then we also had the part library, okay, so for the part library, since the part library already exists, there just verify that all the parts are in the part library and if there's parts that don't belong erase them.

A

Verify all the parts are there and you can use the part library where you actually link to the CAD file so both of the document, both the visual diagram.

A

And part library have the same links and that's how you know you're part library is correct, because you've seen it like, like I, look at the visual diagrams as like the ultimate sense making documents they help you make it make sure that you don't miss anything. So, let's say the visual diagrams are used as the sense making documents.

A

And then the the finalcad assemblies.

A

Our our source I mean that's. The actual yeah I mean that's from there. We can source Ford for doing any further work on this project. So say we went through the first prototype. We need to change things, we go back to the CAD and once again we go to the individual part files. We work on those and we merge those into the final assembly dot CAD files. So so that's that kind of the basic workflow here.

A

So this is our July 11 refined process. So with that said,.

A

A

Let's go back to the work allocation document here and with that said, I mean: are you guys good to go on that and what would your next steps be?.

A

Aybe so you're clear on the next steps.

A

So probably do your in addition to your your doc, to this, give us more detail.

A

And then verify the final final master index put it into them master index that make sense.

A

Roberto, for you take this, maybe add one or two more, with all the very fine fine elements and then go into add those parts to the master index for this. This was Dixon I. Think this and the.

A

Yeah yeah just making sure we don't miss any any things like, for example, the bolts that attach this like. I assuming those are bolt holes. How what exactly is attached so so make sure you've got everything here, like whatever else there is.

A

Let's see yeah.

F

A

A little that we.

B

Final communication.

B

Like a cat or something that we could use.

D

B

Think these are, we are precise communication.

B

They'll talk I'm.

A

Coordination yeah how we do that.

B

Talking about black.

A

A

You want to start up an instance of slack.

B

A

If that could help, let's do that, so maybe if you can take on do that and posted if you can posted also on a three DP Network group, yeah are people using the network group I haven't been using up yeah.

D

I thought the network group Oh being tested for that sort of thing, because they're not formation gets very fast yeah, but that's going on for a long time almost possible can find or the network you can organize around and thread, and they can go back to whatever specific read-only piece of information.

D

That will be my preference yeah.

A

I mean right, I mean it's like email, people, I would suggest. Yeah I mean that's. Why that's what we're not using the slackers I mean we had all these projects and it's like at some point. You just lose it. You know just lose lose coherence. I mean you're, welcome to try with people who who are going to do it, but it really is a matter of what's gonna. Stick work that people actually stick with I would say continue on the dev group and just email, people more. You know like email to coordinate more.

A

Maybe you can do it. I mean you're still welcome to start the slack, but I mean if everyone responds to you. That would be great, but we kind of have to I would say you don't possibly do it as an experiment and see if people actually do end up using it. But but we do have the the network which is supposed to be used for that purpose and then emails. We can email people on a developer thread. Like I said we have that email thread on them for the developers, but yeah yeah I do agree.

A

I mean there's issues.

A

Email I mean just email, yeah,.

A

Email more like like official projects of like a dev dev list, so it'll be like a email list: UM's there's not too many people, yet I think we can still continue doing that on email, so I would really suggest between email and a dev group to post those things and Joseph see Joseph you still on no, but maybe I can remind Joseph.

A

No I mean everyone has access to that email group we can follow up with that and I can send an email to make sure everybody has the contact. There's a couple of new people but yeah. We I think that email Fred would be most effective. But let's do that? Okay, let's see, but let's continue so Roberto. Do you have enough clarity to go on for now or.

B

Yes, my first love my love affair, sometimes I'm working on a part of who else he's working the same virus or if anybody has this example, monkey computer already gone.

A

You meet for the overlap, for example, when there's an overlap of parts in different modules,.

A

You mean for the case, because the the modular breakdown is supposed to resolve those kinds of conflicts, you're generating the parts that are in those modules, but you're saying that sometimes one module might have the same part as another. Is that what you're saying.

B

And there were more people.

A

Right I mean for that would suppose right. So so we have a list so the way it works. We do have the list of people like the extruder. We've got the slide number eight. As far as the work, what the allocation was so the thing there should be click on their log. If it doesn't exist on their log. That means you should take.

A

Take it if they're not putting things immediately on their log they're, not following procedure, but the procedure is that as soon as you've got anything, you put it on your log, so you can verify from others logs whether that part is taken already or not. That should be the way it works. I mean that's the only way it can work. That's why we emphasize you got to publish on your log as soon as you've got anything yeah so that that should resolve it. I mean basically click on click on their logs.

B

B

B

A

I explained that.

B

I'm not sure about who is really working on that module of me.

B

Only I can see their names but get off on that.

B

When you write something out.

B

Just one or two people are answering work.

A

Yeah, but that mean the assumption, is if nobody else answers like assume, it's not done so one is assume it's not done, and then you go to their log. If it's not there, it doesn't exist, I mean really. The super rule here is: if it's not on a wiki, it does not exist in the youth and you're. Welcome to take it I mean that's, that's the culture we all have to learn, otherwise, we'll never be able to coordinate more than a few people.

A

So right now, we've got the development log page and in it we've got the you know, every week's agenda, which typically has the work breakdown and also has the list of all our developers, so that the only thing you can say is: okay, I click on his his uh log person's log and like I, mean that's what I do like if I click on somebody's log and it's not updated, my assumption is okay, it's not done period, you know, that's that's all. It means I mean that and that's what we need.

A

That's why we need absolutely need to log all our work as soon as we have something. Even before we have something like put a placeholder, that's you're actually working on something you know. So that's the that's like the important cultural point we all need to learn. Maybe we should emphasize the but I try to emphasize that all the time like if it's not on the wiki, it does not exist period.

A

I mean if our whole team could understand that, then we really, you know, could collaborate to no limit like there's a thousand people and they're working together, you'll, say saying, like twelve person teams, but you know they click on everybody's work and if it's remote work and they're not talking to each other in person, the the substitute for that is your log I mean that's, that's why that vlog is so critical yeah but yeah.

A

Thank you very fine, yeah yeah yeah. Absolutely so so. Maybe that's something I'm learning that we really need to absolutely clarify to everybody and we'll get better at this, but I mean the vision is that you can have an unlimited number of people and the project still doesn't collapse. So you know: I've been on this project for a decade now and I've seen a lot of how this stuff works like how you can have people starting to edit, the wiki.

A

It goes and goes, and all of a sudden just collapses, just total collapse and the page becomes a mess. You know that's basically the state of our entire wiki at this point, because we don't really have good management, you need to have people who understand the process and actually people who manage but part of that big part of that is the culture which says that publish early and often- and if it's not in the wiki, it doesn't exist.

A

That kind of thing and, of course, there's a lot to be done as far as formal processes like if we develop a process- and we say okay, this is exactly the protocol that we use with everybody. Only by such standards can you get large numbers of people to work on it. Otherwise people will be doing things differently and it'll be hard for things to be coordinated, but that's that's kind of like the lesson on on how to collaborate here.

A

So right now and it's like you know, the thing that that may get in a way too is like if a person doesn't do their work right, like you're like well, what's going on is where's that work you're guessing is it? Is it done or is it? Is it not done and the default answer just simply has to be? No, it's not done because it's not on a lock. So we have to. We have to do that as a team, okay, but let's see so, who else? So?

A

So, basically, you know for your time allotment for the week, just tick, you know, look at what has been done if you're finished with yours there's the basic process. We have right now as far as page number three.

A

This is what we're agreeing where we're going with and if you're done with your pardon, you said more time: just you know, go and go further and then just to go with. Let's see so, that's Roberto, that's good I! Think I, oh I, think you've you're pretty decent on what you you can do.

A

Kasi in Israel I'm not sure, because they're, not here, we'll I, think we'll continue. I think you've been I've, seen parts coming from you, so that's good and Abe, that's good good, but but yeah as soon as we get to the complete visual diagrams, and then we refine the master index to verify between the master index and a visual diagram, we're ready to print and we're ready to buy parts.

A

So it seems like we may be, like you know, a couple of weeks away from that, but it's I mean it's definitely moving forward, like you know, like slide number seven right here, you know: that's that's a lot and I think just get those little missing. Details like like, for example, those bolt holes like after we do this. Then we do want another file.

A

That's got even more detail so that in free can, for example, when we, when we end up doing the final, for example, exploded, part diagram, so you ever see how like in some product they have. This really really complex diagram with like a hundred arrows. Well, we should be able to do that with our complete models. That's the end goal because that will be like say say: this is a final product and it's a real finished release and it's the official version.

A

Then people can go, for example, to the full explode part diagram and see how every single thing works if they want to buy like a little part or make it that the explode part diagram can refer them to to all the different parts. So for us, like the eventual explode part diagrams would definitely be hyperlinked where we have links to files or to sources and so forth.

A

So those all those assets will be generated down the road and maybe not exactly right now, but but we do need the absolute full complete CAD to make that possible. So that's kind of a long of it. Okay, so I think that's. We've got a lot to work on on a filament extruder, let's see just to finish up with anyone else, who's on it on the meeting today.

A

Any other updates that we need to bring up to the whole meeting.

H

um I had some idea of maybe how to make the wiki page part live three bit more understandable. Would it be okay? This week, I got a replica page or just added in the page, to discount some some different ideas on the wiki bit more pull and um like just simple things like in the free code. Maybe we can color codings, for instance, let's make all shell parts yellow before exporting them all 3d printing, for example, yeah.

A

H

Then the letter stage, when we make the assembly and cost the color codes, it will be really need to read the reading sample files. Yeah.

A

H

Clear instruction in the later stage, um yellow our idea: I have the dialogue- that's possible, for example, also in 3d printed parts. Maybe we can make engravings interact with text where you put that difficult to do. Yeah.

A

Yeah we know that's where.

H

That's rotating a link to when file is found. Also one foot that you can always trace them back.

A

Yeah yeah know what I.

H

A

Those are improvements, um yeah, I think what would be worthwhile is yeah. If you have some ideas of organizing that page now, would you be able to continue working on some of the part generation like you've been doing or.

H

Ya can also like there's still a lot of parts in the spool mechanism and also in that big box enclosure I can just keep parts to free, color girl, yeah.

A

I think yeah. Let's focus on that. Let's prioritize that, but definitely I mean you're, also welcome to edit any page on the wiki as well I mean the wiki does have a version history so that we can revert to older versions. If you want to so things like that, mm-hmm yeah, yeah sounds.

B

Good but alive memory, there's a lot of pieces that maybe.

B

A

Yeah I mean we should feel free to just trash them. I mean just get rid of them, I don't know which those are, but if you see the ones that just definitely aren't used yeah yeah, do it.

C

A

Yeah Oliver: do you have any updates on any of your work or.

C

A

A

Go ahead, yes, go ahead!.

C

Okay, I went on with producing Sanwa heads for the top right controller needed from couple covering, which I am us flat. Yeah.

A

C

My locker photo of it um the most ominous and I have made a little video of it on YouTube, but I made some screenshots so that we have to talk about Ricky I. Did that and connected this with a shooters quad yeah, hey, yes, yeah.

A

Oh wow, yes, yes, ok, come cute.

C

It is with quarks er carbon, so meters, Senhor probe, which is better than the oscilloscope probe I, had for the first test, mounted it on my tester ik and then started doing some term and at first, where I moved the text rig test reeked by a stepper motor are in steps of hundred microns or coins, one millimeters in a time one hundred fifty seconds each over a distance of ten fifty meters, no.

A

C

And then I wrote a program in processing processing this visualization language, with which this yeah famous for being the basic whether Arduino IDE is made from the.

F

C

Stuff and but it's good for visualizing things and I wrote a program which can get the data from the PCB over USB ports and make some some curved diagrams yeah, and you can see they are very straight going up and down according to what happens on the or what movements are happening on the test rig now in this setup. It is that the pastor II is moved by an external controller, which is the old manual THC controller I wrote the firmware which moves it continually forward.

C

A network and the new PCB is only doing so far. The measuring and I think the results. Look good relatively good. They are has some quirks or irregularities where I am wondering about I mean last Nia was exploring the stuff and getting trying to get a feeling of. What's going on. There are small things which irritates me, for example, I was watching out his there's me little drift or something but not really.

C

I made a test where I left it in one position overnight and on the next day it was nearly showing the same value so who drifts there. But on the other hand, the whole system is quite sensitive against any influences of electromagnetic fields. For example, if you get near with your hand, you as a human are full of electric signals and as much influences and so on. However, one discovery I made, which was really.

F

C

A great thing for me was that I could get better signal when I'm doing some grounding, but grounding means in this case not connect everything to the normal ground, but I have ever there's a follow, but instead to the shielding I mean the 87, 7 or 7 process. I have a pin which is related to use. This is connected with the surrounding a shield on the coaxial cable mm-hmm, for example.

C

My test rig was not electrically connected with ours in any way, but it was a great improvement when I tried out was a crocodile clip to make a connection between the shielding.

C

Fabric using wire and and the structure of the author of the text break that made the signal quite quite better, but I also had found that, for example, if I then change the position of where I connect. My my shooting crocodile clamp that can make a huge difference in where the starting level of the author of the measurement range.

C

Where, when starts one time, I have effect that it made a difference in one situation, I could go over a distance of three centimeters very well and then I made some changes and then suddenly I had a distance or distance of five centimeters. This is what I mean as a kind of irregularities which I'm still exploring and I think what what is may be still easily kind of calibration.

C

There are in the datasheet written something how to calibrate it and doing stuff with some internally itself and so on. Oh well, but that sounds quite complicated, but maybe I can do a kind of relative translation.

C

I'm still thinking about it and another point which I was called a irregularity is, um there is a formula which describes how to calculate the capitis capacitance value, which we would expect by independence of some nature concerns like the permittivity, but also on the surface area of the sensor probe and the distance between these things and I've, calculated that for for my sensor, probe and I came to an expected value that the capacitance should be in the range of 0.5 picofarads and in the net during range of the trip is about 8, pico, farads or 0.5 would quite nice, and that work was most value which I have calculated.

C

This is my theoretical expectation. Where uses this, what I expect to measure what I measured instead is more in the range of half later. My my diagrams with me hand to heart, which is another word for a tour for us.

C

It's about four magnitude orders smaller. So, on the other hand, the signal is quite handy, I mean like I. Can I can work with it so which irritates me why I get a very smallest signal than I have calculated.

A

The calculator.

C

Uxr and some Antley, as you can see in the diagrams, there seem to be a reasonable Sigma. I have made three kind of experiments. First was what I said before is a variation along one centimeter with Europe point 1 million meter steps, which was accordingly to what we have seen last week in the in the video of this guy clémence victor. Is it or what the name was you remember he he did his video with a same set up.

C

So that was the reason why I started with this and then I made a second run where I made a variation along five centimeters distance, which also works well and in the diagram, is your broader range where it goes over and then I made a search. A runt experiment run where I moved manually, not with the little stepper motor app I've, switched off the stepper motor and moved carriage by hand along my my my marks, which I have painted on the on route a streak in one centimeter distances.

C

It's always 1 centimeter or more over total distance of 5 centimeters. So on each mark I waited 1 2 3 seconds, so that in the diagram, then there is a little plateau and you can be on the beginning of each platter, a little peak which there.

F

C

I put my hand onto the carriage. This is electrically influence of my body, but what you can see in this diagram is kind of of stair stair up and stare down here.

F

C

This, where I variated in 1 centimeter why's that five centimeters in total forward and backward- and that is yeah- but you can see from this- is about the linearity. The thing is not completely: Linea means the lowest step and the biggest height, and if you go upwards, higher higher steps, meaning more bigger distance and then and the range is getting a little bit other this air case height is getting a little bit smaller. This is what what we discussed already last week and.

H

C

Not a real problem, especially since we are probably mostly interested in the mirror and um now or maybe not. Maybe we know I think we don't want to be as tight as possible on the surface. We want to be as as distance from it because of the metal plate is hot and it's, of course better. If you can move 22 to centimeters or 20 millimeters above the surface and but still have a reasonable signal to check our height yeah and so far these are all brand new experiments.

C

I'm I'm I'm, really working literally until the last second or the meeting began that real while I was a little bit late. However, and some what I'm doing now is my next step.

C

I said here and we've shown experiments, the tax trick was moved externally, but the next logical step is to make a marriage between the cats and Xing part of the new PCB and, of course, the periphery, which was the drop wheel and the PD six zero-zero driver, and my next step will be to put all these modules into the firmware so that the thing can control itself and its own movements. Yeah I mean it's interesting to see if, for example, stepper motor is controlled from the same board.

C

Maybe this may have some some noise influences or whatever I hope not, but we will see that, as in the morning, mate little test run, which seems to suggest that there will be no problem, but we will see I have at the moment mix the old firmware of the of the menu height controller, which is only doing the pitifully with the firmware from the new PCB, which is so far only doing the kept sending stuff I have to mix it together, and that will be then we and residing somewhere and with this then I have everything tested on the board, meaning not only the cat's handling, but also how the cats handling is working together when, when the periphery comes into play, then that so far my state well.

A

You said the calculated value is a picofarad.

C

Yes, the calculated that value should be 0.5 picofarads and the real measured value is one point, zero point, one tenth of our meaning. We are about the front to essential meaning for magnitudes of order. Lower than expected I mean.

F

C

Has made Fault in my calculations, not I'm, not right, it's creepy I mean the other guys were working with that have also seen that the signals in the femtofarad range and this what-what-what-what can also be seen under on the monitor or program where I always have in the right corner, showing the original value. As it came from the kept sending born from the mr inverter, the film are small, small aremy d that can be made better. I think there is not an issue of calibration.

C

It seems to be a kind of systematic error in the film. Well, maybe.

F

C

The it's not such a big problem as it normally would be, because the signal is sort of clear and I feel I can work with it, but I also want to from enlighten the background also seems and of course, as a scientist, you know you have an expectation value and if it's for magnitudes of all are smaller yeah or it's a good thing to compare and use to measure that data with and take the expectation, value, yeah.

A

Absolutely so your range sorry yeah as a capacitance should be larger. If you have a ring like like the typical rings that I see or maybe like like I, don't know, maybe like a half inch, half inch wide and you've got. How wide is your ring like half inch, maybe like a 2 or 3 2 or 3 I mean like say, 3.

C

A

Inch across yes.

C

The ring is at the moment of 42 millimeters outer diameter meter and 32 millimeters in a diameter that gives us the ring. Broadness is 5 millimeters and from this broadness I have calculated the surface area, and this value is created to my Zandro, meaning. If I would take a bigger diameter ring, then I would get another layer, but I have take into account. Surface area of my sandals was what's float into the formula and therefore the irregularities layer, meaning with a bigger diameter.

C

You get probably another value, but maybe also for magnitude of other different from strong from what you measure right.

A

Do you have a plan for how you resolve that difference or or no no idea again?

A

Do you have a way to I mean how do you know that it's just maybe not a simple calibration issue or something because.

C

Of the 4 magnitudes or orders is too big for my tanks. That's why I as humans, I would more assumes that there's, maybe some thing in the formula: where is the somewhere kept incidences calculated?

C

Yes, something yeah.

A

C

It means if this could mean is that the case that would mean that, in fact, the AP trip sandals people value, but somehow we calculated in the film where smaller than it is in real time. What? What argues for that is that the single is sort of we are stable or strong or, however, you will call it right right, I mean you can see in the picture of years.

A

C

I mean the groups I have has made their most nearby electric stepper motor driving thing and stuff. So therefore, we have relatively small noise yeah.

A

Yeah, that's amazing, yeah and.

C

And I said: if it's the case and if there is a little calculation error and the firmware, that's that's what I, what I call the systematical error? And but since we are dealing with relative distances, you.

A

C

May be not necessary to solve this problem, that's what I mean mental in the beginning, when I said, maybe we can do some kind of relative generation miniature like in the beginning, you move it tied to the surface and press a button and check the value, and that's our Bureau value. It's the home position. If you want and then we go and say three centimeters upward and that's our position or whatever like this and set set an example, these values on the beginning of each run, and then he knows what it is.

C

It's range and from this range we calculate the millimeters and the following: all the movements balancing movements of the of the head, yeah.

A

C

A

You say: I was.

C

Europe when, when I was married the summer when, as mixed everything in one firmware and have this complete and yeah, then I I will I will try here from there try to develop some method of kind of relatively calibration or whatever you know, I have to say, of course, with this I'm really getting on the borders of my knowledge, I mean and biologists and know physically this to exist in me. Yes, I do my best and yeah, but sometimes there are practical experiments. Ten replace your critical knowledge, yeah.

A

I'll try to get some people to review that, like Shane, might be able to put provide some feedback on this he's: an electronics guy, okay, the guy who did our circuit yeah, maybe.

C

Also can do is ask Aidan what is really his last from where I think I have this last one, but maybe they have one version made one or two days later, but have not LinkedIn or something or you could check that simply maybe that could be the solution follow or the expected value difference could.

A

Be I mean the firmware you're just using using Aidan's firmware.

A

Yes, okay, I mean that could be one source, so that I mean as long as we've got a stable value that you can track. I mean the relative is pro help me find. So that's good excellent media wow. That's.

C

Pretty good, it haven't, read it as long as it is stable. We can work with this and the only concern is. It is really in that very, very small range I mean we are talking of you to para part, that's really smart. It really should be users range. Then it clear that is quite more sensitive against other influences of noise. Then it would be in a higher range, but if it really operates in the range, the truth and.

F

C

Is some calculation error producing wrong wrong wrong data? Then?

C

A

That's right and when you did your experiment, do you have the grounding to the shielding and then like the.

C

Can you repeat that question when.

A

You do your experiment that you have the proper grounding, or this was done without the proper ground.

C

No, this was with the part of our grounding where I have connected to sing Auto, but, as I said, the grounding are doing. Changes with the grounding has much influence of the basic level where well, distinct, where things start, for example, I've made use these experiments and the diagrams you can see a range from, let's say, zero, let's say 120, 140 million, so for us, and then I exchanged the PCT and took one of the other PCBs and then I suddenly was in the range between zero and three hundred milli central Farhat.

C

But when I made the movement when it was within these three hundred range of hundreds there, it was very 18. So it's a kind of continuity, like I, said I'm still playing around with it and exploring it and trying to getting a feeling for what what what what is having durable enough boards instead of around one arc. If you cleanse the the shields croco diagrams different for this noise, you make it loose and clamp it on the same position that can make a shift in the ground level, and it feels a bit like this.

C

If by this actions server, indeed, chip is maybe public or our influence, or something like this- that that's that's my impression if, on the new leather and I, switch the board off and switch it on again, then it fits again on this level. Then it somehow has learns that never or Wassa insignia, but.

F

C

I'm still discovering English and I think in the end situation, you will have a stable set up. I mean my test break is also quite wobbly or little bit wobbly because I just fitted on my table and can move if I push it a little bit or from the vibrations of the stepper motor, for example, in the first experiment, where I made the one centimeter step forward backward forward, backward I have made a run where I've made I had sampled it over a longer period and I was seeing that the thing was slightly going upwards.

C

Like the drift like it was drifting. But then later, I I found that when, when when the drift was getting clearer or more I had I had a gap between the arm and the head and the surface, and they probably from the vibrations of the stepper motor that the ceiling has moved a little bit. So I'm I have Hume that this is not a really drift address than artifact rewriting from my wobbly setup. Okay,.

F

A

C

Keep an eye on that here.

A

C

Most problem: okay,.

A

When you look at my screen, know the experiment that you did there for that wave there. What was the distance that you were measuring was that from 0 to 5, centimeters or yes,.

C

And the latest set up well, which is also presented in videos. It was from 0 to 5 centimeters. You can get even a few millimeters forever, but in this region the more distance there is, then then it flattens on top yeah yeah, no based.

F

C

Completely linear, it's going a little bit like going in the saturation. It's lit the same. What we have seen in the video from Clemens, this cold yeah.

A

Okay, that's a distance of 0 to 5 centimeters yeah. We shouldn't have a problem, getting reliable measurements with this. That is good mm-hmm. That.

C

Means that, in the close, a distance you get a clearer and better signal. What what is the distance, what you have in mind or what you think by experience, maybe what is appropriate, of which, how much distance should we at least have between sin, go ahead and the surface later in running the torch table, when it's getting hot at you and any value 5.

A

Millimeters, how many thousand five millimeters for.

C

Five millimeters is a global distance. We could get. That is cool because in this range we have of course, the best scale.

A

Ya know sometimes I put.

C

And we made from here I mean I, don't know how much electrical noise the rest of the table table is producing, but here from my set up women we can step up model next to it. I am so feeling that we could even go up to 20 millimeters, so I guess five millimeters up to 20 millimeters. That will be our our.

G

C

But when you say five millimeters, that means also, probably that is what we can expect that place. The iron plate will be deforming if it's get getting caught. Is that right, yeah.

A

We're gonna have a yeah, that's correct. Okay,.

C

A

No, this is great, looks, looks very positive for initial data, we're gonna we're going to work this out. This is good and I'll see if I can get somebody to comment on this. Some of them might be in a know about capacitive, sensors, okay, excellent work, man, good job for the home team, another score against global oppression, okay, very good. So we can I think we.

A

We want to wrap up the meeting it's good two hours, so yeah I think good progress and we'll continue working on the 3d print clusters, so I think I'll call it call it here. Does anyone else have any other comments or questions at this time?.

G

Well, if I, if I, can help you and anything just tell me.

A

Who was that speaking, sorry.

A

Was it Alejandra, hey.

G

I'm, okay, yeah.

A

Water excellent, if.

I

I can walk yep.

A

Yeah, what so, what.

I

I'm performing okay.

A

Okay, we should get you going on like immediately. We should start on the filament, extruder, stuff and I know how much you you understood from that, but maybe we can follow up. So maybe we can let everybody go. The only thing is I want to follow up with Christian as well Christian we're talking right now as well to do the print cluster work, because I need like half hour to relax for a second and them, which are you still going to be out Christian, no problem.

A

Okay, I'll call you back in half an hour and as far as as far as the meeting up I just want to Alejandro that maybe we can talk a little, but otherwise, let's, uh let's call this a day thanks everybody and the next meeting is once again 1:00 p.m. Tuesday of next week. So we'll see how far we get till that time, and otherwise great progress we'll see you at that time: okay, great, okay, Farkle, excellent Alejandro's! Let me just ask you for a little bit.

A

How much of the conversation did you follow regarding the filament extruder work? Did that make sense to you or.

I

Yes, some things are a little advanced to me, but I.

C

I

To understand, I'm recording this conversation and I'm review later and I'm, trying to understand yeah.

A

Right what I, what I would say we want to do see if you can go through the yeah yeah just review the part about like what I thought yep.

I

Connect the connection, the connection so sounds that sounds.

A

I

Don't know my connection, okay,.

A

A

Let's, let's have you do this I mean just what I'm going to do is.

A

B

A

You know what Roberto, if you want to get together with Alejandro, you guys speak the same language. So maybe maybe you can fill them in a little more on the details and maybe have him. Can you maybe have him take on a couple of simple tasks with you like you can guide them through? That they'll be great.

B

A

Yeah, that would be great.

G

A

So I'll introduce you to like I'll email. You guys right now and you can then guide Alejandro on some of the specific tasks. So you have a Roberto. You have a good good handle on what's what's going on there, so if you can have them step right into the simple task, there's a lot of simple tasks to be done there. So you could do that yep!

A

Okay! If you look at, if you see the chat, you see his email.

A

Yeah, okay, so we'll connect you on an email and then otherwise thanks a lot, everybody and we'll meet again next Tuesday 1:00 p.m. CST. Okay, thanks a lot everybody, my bike.