TLDR: Will you review prototypes and schematics made by the community? Also can we have an R&D channel form for full research projects that we can keep track of online? Also what is your opinion of the extruder? Would you use it, and is it in your opinion an improvement over older multicolored models (disregard my circle making incompetence.)?.
DAY 7 SINCE LAST INBOX. 82 DAYS SINCE THIS QUESTION WAS FIRST HIDDEN FROM VIEW
I spend my nights pondering and asking questions. To which the answer is a swift slap for not letting normal people sleep. Many of my projects that I do out of sheer boredom often go into a filing cabinet full of ideas that I put out, but some I feel there should be more life to them. None get the attention of the 3-D printer that I am (mostly) building from scratch. I often go to many places for advice, but there aren't near enough for proper critiquing by knowledgeable people. The question is would you guys at Teksyndicate accept designs and concept mappings so I or someone else can have another perspective? Such things as sending in concepts, simulations, and prototypes for another person to review.
Also the extruder I was talking about... if you can ignore my incompetence in inkscape in making circles here is a very very rough sketch of the extruder operation and approximate build. The servos adjust the actuating arm which in turn forces a gnarled gear with the filament into one of the main drive gears. This is all controlled by one high torque stepper motor. I am almost done with the printer, and after that I will begin on the new class of extruder. Since green is the cheapest color I will use that as body color, and the other colors will only be a surface attachment. I need a powder printer.
Until I have a solution for black I will only have RGBW
Under current consideration for the design flaws of this printer. Since most flaws are design limitations I am after this construction going to be in persuit a powder based printer. That way I can effectively launch research into a home brew metal printer. This won't be my main priority because I am working on other things that pertain to symbolism. I want the opinion on this so that way I can submit it to a local hospital. What do you think about the usage of braille? Instead of just putting them on/next to doors we should also put it up on direction signs saying room numbers of 1-x. This way a blind person doesn't have to go door to door for room numbers. Maybe this is implemented in where you are, but the hospitals down where I live in Florida don't have this feature. The blind have to scan room to room looking for a specific room number without knowing if it's in Radiology, Labs, or some other area in the building to get them off on the right foot. Instead of having to scour an entire floor moving from place to place we should let them have a rough idea on where they are going like we do.
One concept that I have been kicking around in my mind for a while is a PCI-e RAID controller integrated into the bus. You have PCI-e RAID cards, but the don't integrate them in like they do with SATA. It isn't my favorite idea, but it is one that I don't see discussed. I appreciate the support of my thoughts though.
I'll start by saying that this isn't really in my field, but i'll offer my thoughts anyway.
I guess it is just that there isnt really demand. PCI-e is a pretty neat solution to start with, giving you the ability to swap it to other systems pretty easily. Also, it is better for data security if you dont have it linked to your motherbaord. If your motherboard dies and the RAID controller is integrated into the bus, then you have a bit of a problem.
The speeds of PCI-e aren't a problem, so i doubt there would be any performance gains from having it integrated. The biggest problem with the current PCI-e RAID system i guess is that to have a large array, you have to take up a slot. It's not really a massive issue, and most applications have the RAID array tied to a server, so you don't really need the extra lanes anyway.
Well this isn't my greatest idea, but something that popped into my head were 2 solid state drives on x4 lanes. It wasn't a smart idea, but one that I come up with on the spot.
it'd be good. Only thing would be keeping the costs down. At the moment, all the PCI-based solutions are really expensive enterprise grade cards. If they somehow managed to get the cost down to around the cost of a SATA based SSD then i'd get one for sure.
Yep, and this is kinda like how most things work. The technology could be cheaper if more people purchased it, but more people would only purchase something after the prices have gone down. Someone has to bite the bullet at some point.
I don't have a picture right now, but one of the things that I am designing is a four way extruder that uses only one high torque motor. It uses two drive gears in parallel to run 4 filaments that are attached to their adjacent gears. when the individual gears retract it moves the filaments back with them with a bar. This not only solves the retraction problems that have plagued other implementations of of multi extruder designs, but it also reduces the mass, and space factors that reduce print sizes. I am stuck in the design right now because each retracting gear that holds tension on the individual filaments are powered by small servos. So that is currently limiting me to 4 extruders. I need a minimum of 5 to make it possible to do a full color spectrum of color extrusion. Unlike paper with already has white covered on the page you need white in the print mix in order for it to be represented. So instead of you standard CMYK, you have to modify it to Cyan Magenta, Yellow, White, and Black (CMYWK). This will be extremely taxing on any motor that you include in the design. Even the High torque motors that I am looking at bog down after 4 filaments. I may have solved the problem with space, mass, ability to retract filaments, and made a color scheme for the extruder, but I haven't solved the torque problem. It is possible that the solution will require two stepper motors facing each other. This means that you would have to run the motors in reverse direction of each other in step. You then could get away with a smaller motor of about 2/3 the torque of the original, sacrifice a little space, give it a little more mass, and resolve the problem of bogging down at 5 filaments. The problem arises is if the cost is truely worth it. Should one of the colors be neglected, or do you guys think that I should just make the sacrifice and have all five colors?
One problem I haven't gotten to yet is... Will the entire model be colored, or just the surface?
Cool, I was talking consumer grade though. But I also want to know the motherboard models where the raid plays out as you plug a drive directly into the pci slot.
I'm not at all as technical as anyone else, but I would say try one idea out and see where it leads ya. If it didn't work, then go back and go the other way. Everything has not been made to perfectly fit what was thought to work, most are random stuff that appeared during the making and then became the perfect thing. If you think one thing is better than the other and you got the money for it, go for it. If you want one thing but don't have the money for it, go the other way and upgrade it in the future maybe. How much does it cost to add the 5th colour?
It's not a matter of cost, but it's a matter of practicality. Is it practical or worth any bodies while to add a fifth color. I think it might actually be more practical to use more of the other three colors at first to make black that way I can pursue better plastic technologies. I am almost done with this prototype hardware wise, but the software is coming along slowly. As a 4 way color scheme it works, and it works well in sim. But, as 5 colors I am better off going after a inkjet sintering machine. With this I can modify it eventually to take metal. This is a long shot off, and I have a thousand more practical projects that need completed, things to be fixed, and blogs to be blogged.
