Show Posts
166
The Lounge / Re: Sheepdog still out there
« on: March 17, 2015, 10:19:05 AM »This looks great. Really wish you would have given it a plug on here when it was a kickstarter.Sorry about that.
What is the likely leadtime on an XL kit at the moment? I have been planning on building a bigger machine for a while but this looks like a quick solution rather than the issues I was looking at scaling a cartesian machine. Do you use a heated bed? How does the Gorrilla glass handle heating? I generally print with PLA so need a heated bed, but am looking at printing with HIPS soon so a heated enclosure might make more sense.
I have always wondered if a Delta bot could be made to flip half way through the print to be able to use the entire height for the print? ie. Raise the print head; then drive the print head outside the normal print envelope right off to one side; and then flip the opposite carriage to below the print head. Then repeat for the other two carriages. Then carry on printing with the carriages below the head right out the top of a shorter fatter printer... I hope this crappy description makes sense.
I would also love a small waterjet. Been wondering if it would be practical to make a sort of two foot square one for the home. Waterjet has so many advantages and if a typical pressure washer could be adapted to provide the pump it might be pretty affordable. The hardened and scruffy edge on plasma and laser cuts is a big issue sometimes.
All the best with it.
George
Lead time... depends on our electronics supplier, we've gotten our ass kicked by them. The Chinese are a royal pain in the ass to work with on anything. We hope to have a bunch start shipping early next month, and to be caught up by mid to end of the month. The problem is, we are about sold out of our first batch, before all our parts even arrive and we don't know how long it will take to resupply (again, China). Batch 2 will also have a price increase, but be a bit better and have a heated bed option on the mini and mid. The XL is impractical to heat as it would take about 750 watts to heat it. Not only can the controller not handle 60+ amps (solved with or external power), but that is a small space heater running for hours or days at a time in a small room. We do have some customers who plan to do it though. Price is set to rise this week ($200), if my partner hasn't done it already.
Just so you know, you can only do about 100-150mm in ABS without problems even with a heated bed, to get more, you really need a heated chamber, or get creative. Which is another reason heating an XL is kind of silly, since you would still be limited to that length before it curls in on itself. Heated beds also create errors in the print dimensions. I've seen many times where the print is hourglass shaped due to the temp differences as it went up. ABS needs a heated bed (3% shrink rate), as does some nylon (1-5% shrink rate), PLA does not (.03% shrink rate). It's the shrinkage that causes it to warp. We also do T-Glase (PETT) without a heated bed, in fact we are using T-Glase instead of ABS for our effectors, it's a bit more expensive but a WHOLE LOT less fickle than ABS (I HATE ABS*). Enclosures create a lot of issues, and were/are under a patent (as were windows until recently!).
There is also the OS (Open Source) version of the printer, which starts at under $500 to build, and can be upgraded to a Pro. I have a full B.O.M. with links to suppliers and an assembly manual in the forum. The Pro grew out of the OS, which was built to be home buiilt, cheaply, while the Pro was meant for production and was built with a dream list of parts. The Pro will also be open sourced for the most part, though a few parts will not (hobb, bed plate..) since they aren't really easy for someone to make without access to a machine shop or laser cutter.
Gorrilla Glass takes heat WELL, it spreads heat about as good as aluminum and doesn't chip like Borosilicate. We've crashed the head a ton and not scratched it, it's probably the best glass you could use. While it can be flexed to remove a print, I've shattered one doing it. It's tough stuff when hit head on, however, if it's hit on the side, it shatters, just like an Iphone. It's also not cheap. Corning wanted $700 for the XL glass prototype, as a result we use normal tempered glass on that model. Even in a batch of 500, they wanted $250 wholesale for the glass. Our cost on it in bulk for the mid and mini is what most other glass retails for. We are still evaluating the practicality of it, if it doesn't sell well, we may abandon it, but for the moment, we are the only ones with it. If it does do well, we'll get the price down some, but also square pieces.
Cartesians don't scale well, neither do most common deltas, the Griffin and Wolfstock (another open source design) were built to be scalable from the start. If you do a cartesian, look into h-bot and Corexy (modified h-bot), they are a better option.
Yes, you could flip the delta as it went up (and that made sense), and I thought of building a reverse one, however, your build volume at the top would be significantly smaller as your arms could hit it as they move. Otherwise, it's just a matter of making sure you have enough range of motion and finding someone to code that into the slicing system (or do it manually). To have it built into slicer, if you had to hire someone to write the plugin, based on our current investments in software, you would probably be looking at about $1000-3000.
The thing to remember on a delta is that it's very efficient in terms of deskspace, the loss is vertical, so it's not a huge deal for most people. on the other hand, a cartesian becomes more and more efficient as it gets larger. CoreXY makes more sense for a short, wide printer, it also runs fast like a delta and handles dual nozzles better.
Not all is well with big printers, our XL already caused some work for slicer creators as our XL was causing them to crash due to size. Going much beyond that brings a whole host of issues. The largest you can really do with standard parts is about 16in unless you know what to watch out for. There's just a lot of little gotchas as things get bigger, and that applies to deltas and cartesians. Not to mention build times, our XL regularly goes 24hours+ and has gone as long as 100 hours. We had one have a slicing issue at hour 38 of 40. Our Ironman helmet did that, and there's $160 worth of plastic in that print. That said, we are working on a 28inx28in (70cmx70cm)build volume delta. It will allow dual nozzles in two modes (dual filament types or 75% speed increase), 3mm filament or pellet feed and eventually a clay system. No heated bed though. It's about 80% complete and a beast, it will probably weigh 120 pounds (50 kilos)when done and start at $3-5k
If I were you, I would do a CoreXY, and a cheated chamber. I would probably use dual "all metal" hot ends, with water cooling, and put the motors and electronics outside the box (something many miss). I would NOT use Arduino, and instead use Smoothie and 400 step motors with 1/16th or 1/32nd stepping and 16t pulleys. Not a small or cheap project, but would give you the best results for space and performance. If you haven't built a printer (or similar) before from scratch expect this to run you up around $2000 to $2500, with experience, it could be done for about $1000-$1200. Just how it is with projects like this, your first one tends to cost double by the time it's working how you expect.
There was a guy on Hackaday recently who built and documented a small waterjet, almost the size you want for about $5k. Scaling it up isn't really a big issue once you understand the mechanics of it all (cnc, 3d printers, lasers, waterjets all move in similar fashion and use similar controls), but there's just not a good way to get the pressures needed cheaply. The pumps and nozzle are what get you. A pressure washer usually runs 1000-2000psi with high end ones reaching 4000. Even a low end water cutter is 10,000, and some are now up to 80,000. You might get through wood with a pressure washer, but probably not metal.
I haven't done much with laser cutters, but the one at the hackerspace doesn't seem to have rough edges. Not sure what the deal is there. The only thing I can see it being is either it's garbage, or you are talking a VERY high wattage laser on metal, which I haven't dealt with.
*Abs can be roughly about 10% stronger than PLA when done right, and handles heat better (won't warp in the sun). It also flows faster, so in theory you could print faster with it. Unfortunately it's so fickle in terms of cooling, that if you get it wrong, you can crumble it in your hands, and it's very easy to get it wrong. Keeping it from delaminating is a hassle. If you spend the time and get it right, it's fine, usually slwo and little or no cooling works, but it's just a pain in the neck and it stinks. T-glase has a temp range between PLA and ABS, and is as easy as PLA to print, but costs almost twice as much. Considering how often ABS screws up though, it can be cheaper in the long run. Also T-glase is made by one company, who knows what they are doing, whereas ABS can have different formulas, or be made by questionable companies, I've had some so out of spec it jammed by feed tube and hot end to where I needed a wire coat hanger and hammer to get it back out after it jammed. I learned on ABS, and sometimes there is no replacement for using ABS, but these days I avoid it like the plague.
Thanks for that huge reply, much appreciated since I know you are busy. I did check out the Griffin site and browse through the boards you have there so I knew your thoughts on big heated beds not really being viable (which I agree with to some extent, though I would think it doesn't need to be that bad if the bed is well insulated underneath and there aren't fans blowing half the energy away). I haven't even tried my HIPS yet, but the word seems to be that a heated enclosure is better anyway, and a Delta would obviously suit this well. But my worry with this is that any printed parts of the printer itself are going to start softening and melting, so ideally everything in the enclosure needs to be heat tolerant (at least up to say 80C).
In terms of inverting a delta, I can see it having limited value due to the arms striking the print, but if the hot end is underneath then some clearance is created and most prints tend to be smaller by the top anyway, it would add print height or let the side corner rails be shorter and therefore cheaper and stiffer. But I can see the code being a pig too.
My original plan was to use my current Prusa as a doner for parts to build a new cartesian but I might try a Delta as a second machine and be able to print parts for each other or print two things at once... not really short on space.
Water jet leaves such a clean cut. I often want to make small shims or plastic washers that are very difficult with most cutting methods. Lasers leave a thick edge which makes the part useless. Waterjet cut parts have a perfect edge and dont need any more preparation before welding which is nice too. With jet washers I was wondering if they could be run in series to boost the pressure.. though admittedly three 3000psi pumps in series is getting a bit much probably.
G.
