I’ve done a lot of experiments with toner transfer and etching my own boards. The highest accuracy with the smallest clearance I can achieve is with old inkjet photo paper in my old HP laser printer. I have tried several different photo papers. They all act a little differently, but they all take up far more toner powder from the drum roller. I’m not sure why. It can be a pain to get the paper backing off of some of them, but when it comes to etching, it approaches photoresist levels using the typical overhead transparencies.

Just an idea to throw out there. This looks great already.

I have a feeling we’ll need to cite this often.

Yeah, I just have to waste a day making a doughnut every time because the memorized workflow is not compatible with my use case. Blender is made for people that use it every day. I’ll use it to fix a mesh once in awhile when that will save me substantial time, but that is about all I can do with it efficiently.

I’m using principles of 'close enough. I almost exclusively use the Part Design WB.

Technically I should probably model this in Blender. I know Blender has a way to wrap and unwrap like I hinted. This post was really a fishing attempt to see if anyone here knew of a way to do the same thing in freecad.

To model this, the first trick is to pick the best origin point. I’m using the screw pivots where the arms fold to hold my X-axis. The inside face of the lens is 146.6mm in diameter and is my primary reference surface for the lens. The shape of the lens is all done by creating sets of 2 sketches that are only used for references. These references are only used to attach a datum plane. The datum planes hold the sketches that pad and pocket the solid shape.

If I really wanted to make this better I could do so by making more complex sweeps while only building the part with reference to the outside face of the lens, because it is not a perfectly circular dimension. After the surface features are complete for this face, I could use the defeaturing add-on workbench to extract just the outside face from the PD body. With this face by itself, I could create a filled offset in the Part workbench. Then I would be able to cut the circular inside diameter from the X/Y axis to rebuild the 146.6 mm inner face.

In my experience that tree would have around a 50/50 chance of failing if I tried it. The task would cone down to how well the offset in the Part workbench is calculated. Even with this entire setup, I would not be able to control things like a varied edge finish on different parts of the lens so that I could model the mechanical properties better. I’m not super worried about the surface in freecad because I will need to sand and polish the print anyways.

Thanks though. Hopefully you can follow the explanation. When I started with freecad years ago, I might have tried something like the draft WB but almost everything belongs in a Part Design workflow. It is simply a matter of knowing how to pick good references, setting your origin well, avoiding referencing off of π as much as possible, and NEVER importing edges or points from 3d geometry.

Slowly trying to learn sh while using mostly bash. Convenience is nice and all, but when I encounter something like OpenWRT or Android, I don’t like the feeling of speaking a foreign language. Maybe if I can get super familiar with sh, then I might explore prettier or more convenient options, but I really want to know how to deal with the most universal shell.

It follows the first to market principal in many of the most active communities, and it is the most federated instance. Many instances that are not federated with each other are federated with .ml. You still won’t see those comments between instances. Like from my main account here on .world, I can’t see hexibear or beehaw stuff, but from my .ml account I can see them. I have accounts on many instances in order to help federate new communities and to check biases/instance behaviors.

I came over a few days before the rexodus and subscribed to the active communities before the influx. That sub list is still centered around the most active communities, and the majority of those are from .ml and before I joined Lemmy.

Instances all have different flavors. I don’t like using my .ml account as a main. I’ve tried it. But I find they are the center of the most interesting and productive conversations for a more broad audience, while Beehaw has the most positive and friendly conversations overall. The main benefit to .world is the speed of connectivity, general audience scope, but with a strong anti asshat policy.

The .ml instance is run by the developers. If it isn’t well configured we would have no reason to be here.

Not sure. There was some controversy with some of the devs making alt front ends and admins complaining about the slowness. I’ve seen mention of one of the two devs learning Rust just to participate. So it is not entirely a walled garden. The front end devs wanted to make an alt from scratch but in something like JavaScript although I don’t recall the details exactly. There were a lot of red flags related to privacy and understanding the community at large in the posts I saw from them. When asked why they weren’t adding pull requests with Rust in order to address their complaints I got no reply.

All that said, I’m no dev. I can read in to around half the code I come across if I really try, and can successfully modify maybe half of that if I spend a few days on it, but I suck at clever code and the DRY cult types. I haven’t tried to look into Lemmy in any depth beyond figuring out the basics.

The Lemmy algorithm:

https://join-lemmy.org/docs/contributors/07-ranking-algo.html

The instances communicate between each other using special bots in the background that transport info between instances. Dot world is too big and makes too many requests to other instances. Most instances are reducing the number of transport bots back to dot world right now. There is supposed to be a fix in the next Lemmy version, but the code base is somewhat slow moving due to only two devs and it is written in Rust. Rust is a hard language like C, and not too many here are able to contribute to it, even though it is like the new gold standard of code.

So you might see a delay between posting and replies or the interaction may come in bursts that correspond with the transport bots carrying content between instances as the host admin have configured their instance.

Maybe look at galnet too:

https://f-droid.org/packages/gal.sli.digalnet/

That one is totally offline and is a simple translation dictionary. I use it and Wiktionary on a browser tab regularly.

They no longer directly maintain a flathub beta. You must use the .apk from the website like it was in the past. I guess no one wanted to maintain the extra effort for flathub.

I found a Python project that does enough for my needs. Jq looks super powerful though. Thanks. I managed to get yq working for PNG’s, but I had trouble with both jq and yq with safetensor files. I couldn’t figure out how to parse a string embedded in an inconsistent starting binary, and with massive files. I could get in and grab the first line with head. I tried some stuff with expansions, but that didn’t work and sent me looking for others that have solved the issue better than myself.

That is freaking awesome! I tried both ABS and ASA with the MK3 and a sealed enclosure, I can do some fairly large prints up to something like large grapefruit scale, but anything like your enclosure will separate on the corners. I think it is simply a bed slinger issue that is unsolvable. Awesome to see the tangible results and justification for building a 2.4.

Awesome project. Reminds me of Hexibase YT type stuff. I’ve got a spares shelf of drivers but would need to build an Arduino project to characterize them and the output empirically, and still haven’t gotten motivated for that level of math project.

I assume PLA? If that is ABS I’ll lose my mind.

So no, the marketing shysters are simply trying to silence the self aware minority. /s

Never talked about the important thing. Are the kernel modules fully open source and is the SoC fully documented publicly. If those two aspects are not a yes, this entire affair is an attempt to steal hardware ownership just like phones.

I did proper assembly from the start where I cleaned and greased them when they were brand new. I’ve never had any issues since. No (cheap) linear bearings come with grease. They only have assembly oil and that is not even a load bearing lubricant; it is a corrosion inhibitor. This is a good thing really, because you need to know exactly what grease your bearings contain. You should never mix greases of any kind. They all have different formulations and will act unpredictably when mixed; often failing in a coagulant that provides no protection from metal on metal contact.

Many cheap printer manufacturers will dab a bit of grease on the rails outside of the bearings when new. This is useless in practice due to the bearing seals. The seals are designed to let a small amount of grease out, but block any old grease from reentering the block itself.

If the blocks were run dry without grease, they are contaminated and need to be cleaned out completely. Likewise if they need service and have unknown grease inside them. If you clean them out to the point they are spotless, and then you manually pack them with a quality grease, you’re unlikely to ever need to service them again for a very long time.

I build my own bicycle wheels and service my bearings and hubs about every 10k miles riding in all weather. I was sloppy with how I serviced bearings for a few years before I really narrowed in on my issues. They must be spotlessly cleaned, without any old grease whatsoever; like clean enough to eat off of them. This is the difference between 2k-4k between problems and 10k+ on a daily ridden bike. Same thing applies here if you want to only do the job once.

Maybe. Smarter Every Day on YT had a couple of uploads where he implied that SpaceX’s lack of commitments and schedules or even proof of critical technology was a major problem for Artemis and all other phases of the program. There probably isn’t time for everything else.

I only need the ram to initially load an AI model.

The best deal is probably going to be looking for a used machine with a 3080Ti. There were several of these made with Intel 12th gen CPU’s. That is probably the cheapest way to get a 16 GB GPU. They can be found for considerably less than $2k. Anything with a “3080Ti” where the “Ti” part is super important, has a 16 GB GPU, (the “3080” is 8GB). That was the only 16 GB laptop GPU until the newer Nvidia 4k stuff.

That can play any game, and can run some large models for AI stuff if you become interested. On the AI front, you want maximum system memory too if possible. My machine can only address 64 GB of sysmem. Some go up to 96 GB. I wish I could get like 256 GB.

Just because a machine comes with Linux does not mean the problems are solved. You will find many times when people buy machines that have peripheral kernel modules that are orphaned and not part of the kernel. Orphaned kernels are not real Linux and are like phones. Indeed this is the exact mechanism used to steal your phone and prevent you from using it for its true hardware lifetime.

The real solution is https://linux-hardware.org/. Use that to see what works where. You also need to understand modern secure boot with the TPM chip and package keys. These exist outside of the Linux kernel. If delving into this system is too much for you to deal with or of no interest, just stick to using either Ubuntu or Fedora. These both have a special system outside of Linux that will handle the keys for you. Presently, these are the only two distro choices that do this; not derivatives either, it must be vanilla Ubuntu or Fedora. You won’t be able to change anything in kernel space when going this route, but if the keys issue is unimportant, that probably won’t be an issue.