ddurant

I think Martijn told me that it gets difficult to scale up the design, though I don't think he said what the limit was or if we were anywhere near it. IIRC, one big issue is getting longer X/Y rods that are very straight. As they get longer, it gets harder to get ones that are as straight (and stay as straight) as we need. You can get around this by using thicker rods but then you're adding weight and potentially lots of other changes needed to support bigger rods.. Changing Z is a bit easier - there are a lot fewer bits that need replacing.

:( That's called 'stripping' as in "you've stripped the filament.." The temperature is controlled by the firmware on the machine. RepG (or whatever is feeding it gcode) tells it what the temp should be and that's it - there's nothing else needed on the PC side to keep it at that temperature. You could turn off your PC and the machine probably stay at whatever value you set it at last. I don't see any reason why it'd stop maintaining temperature while start.txt is happening.. Could be but I really don't think so. Personally, I think it's a better idea to use repg as just a something that feeds gcode to the machine. You're better off (IMO) using skeinforge by itself to do the slicing instead of using the version integrated into repg. That's too hot!!! I've been printing at 225 and that's been working pretty well. One thing to watch out for is leaving the machine hot while it's idle. The plastic will work its way out of the nozzle and that can mess up the start of your print and (possibly) encourage heat to travel back up the hot end and form the dreaded PLA-plug that people have been talking a lot about lately. This is especially true if you're really putting the spurs to the heater (which you are). Are you sure you're not starting too low? If the nozzle is too close to the platform, the filament drive won't be able to push more filament into the system and that'll cause it to strip.. Nope - the firmware just sees gcode lines coming in. It has no concept of start.txt. M104 is the "set temperature" command and there's nothing before it to do some other temperature.. Do you print with a raft? That might be a good thing to try as rafts are made of big, slow lines that should be really easy to print..

I think at least some of the SF speed (or lack of it) depends on your machine. More free memory and disabling bits in SF you don't need will help a lot. It's never going to be fast but you can do things to make it less slow... Not exactly.. Comb tries to reroute non-printing moves so that the nozzle doesn't go over any more open space than it has to. This is HUGELY nice feature as it can eliminate many of the strings you'd normally get on complex parts from the print head oozing a bit as it crosses open areas. If you imagine printing a C shaped object, Comb will make it so that the head won't go straight from one endpoint to the other - it will reroute the moves so that non-printing movements travel along the rest of the shape instead. This usually adds to the printing time but results in a much nicer end-product. It does indeed work great! Another HUGE feature. Basically, "volumetric 5D" calculates the extruder speed for you based on other settings (layer height, feed rate, etc) and your filament diameter. Without this, any change to one of the big print settings (again: layer height, feed rate, etc) will require careful and tedious test & tweak calibration runs to figure out the correct flow rate. With it, you tell it the settings you want and it just works. Very nice. Volumetric-style profiles are coming to netfabb. Possibly very soon..

i've seen him posting before and wondered about why he had such a different experience and was so negative.. It makes a lot more sense now. mr_seeker, you should disclose info like that. Not doing that was really.. er.. I'm having a hard time coming up with words that allowed here.

I missed that one.. I'll have to print it! One of my very, very few complaints about the UM is that the stock fan setup there aims the air a couple inches to the left of the nozzle..

Something we don't have here!!

This line scares me. It's not actually hitting that, right Joel?

Looking good! I'd switch the colors around, though. Red makes me think it's something I must pay attention to.

Never ask to be a mod!! The spam definitely seems to be better now - thanks, UM!!

Same here.. Same guess as to why he's seeing wear, too..

I'm not sure I agree with this.. If the point is to quickly iterrate through a bunch of designs before sending a part off to a big & expensive printer or injection molding machine, it seems like a pretty good deal. And I think UM can print fine to show a prototype/mock-up to a customer. Again, I don't really agree.. Eventually, the software we use will be able to analyze the object to be printed and automatically rotate so it sits in the best position and adjusts the settings to deal with bridges and overhangs and areas that need support and all that. We don't have that now so we have to learn some stuff about the software and the physical processes and adjust the settings ourselves. There is no print profile that will work perfeclty for all prints - I look at every object I want to print and poke the settings for every print. This isn't overly hard - it takes minutes, not hours, once you learn how things work. Yeah, well, zcorp machines cost a little bit more. I'm sure the material isn't the $10-15/pound like we get, too.. No, it's not a box you can just plug in and start printing amazing-quality stuff right away with. You need to think about how the physical stuff works and understand the what the software settings do. If you go into it know that, it's not hard. If you go into it thinking it's like a microwave or something you can start using right away without knowing anything new, you're going to be frustrated.

I think there may be settings in the firmware that need tweaking for it to behave just right - like changing source code type stuff. I don't know what those changes are, though. To be honest, the talk of acceleration math and jerk settings and such sorta just goes right over my head. It's awesome that people are really engaged and tackling this complicated (to me) stuff but a version with things preset to what will work for most people without tweaking would be great for those that don't really want to know all the details.

Got a few pictures of stuff you might be printing? The biggest concern, IMO, is that you'll want to print things that FDM-type machines (aka: plastic extruders) have trouble with. These machines are just AWESOME for lots of stuff but some things take a bit more work to print than others. For the hardware, I think you'd it'd be very difficult to find better made machine in this price range. When I built mine, I was quite impressed by how well it all went together and how there was just nothing that seemed like it was there because a certain part was cheaper or what they happened to have on hand. It's really a nice box..

I saw that it was out but haven't messed with it yet. Try it and let us know!! Lots in his blog about things being renamed and moved around - looks like SF44 profiles should work but I'd go through the blog and check that the various settings made it to the right places before printing..

Great progress! edit: and, just a idea, consider having min and max thread widths instead of a single, predefined value. Or maybe allow a specific value or a range. This isn't as interesting on the perimeter but when you do infill, this will allow it to chose fatter/faster lines in the interior and thinner/slower lines on the exterior. That and it will help solve (or reduce) the thin-wall problem where the a specific thread width & shell count prevents areas from being filled..

Yup: http://davedurant.wordpress.com/2011/10 ... of-prints/ .

I think we're already close to it! Or Paul is anyway..

Yup. A number of the batch-1 machines came with a hot pink, rubbery thing that fit over the nozzle to keep it warm. Others use kapton tape, sorta shiny and goldish, and others use ceramic tape, white and sorta fluffy looking. Insulating the nozzle helps it keep a steady temperature while it's printing, especially if the fan's on. When your machine is all about heating stuff up to the right temperature, having it be steady is a good thing..

It's interesting tech. No doubt that it (and powder printers) can produce prints with stunning resolution! I don't see myself ever getting one, though. It's really less of a rapid-prototyper and more of a monochrome printer. Can't see how you'd change layer height or any of the other parameters that enable you to, say, do a fast & low-res print instead of a slow & high-res one. There doesn't seem to be many options for colors, too. Nice? Absolutely. For me? Not unless it's comes out pretty cheap..

LOL!

Z is just part of the G1 gcode line. When Z changes, the firmware just moves it along with the other axes. I think. Dunno. Whatever max-z-feedrate value is in the firmware, maybe? Setting the acceleration parameters isn't the same as doing acceleration! All I meant was that the fiwmare itself is what decides to tweak the speeds at various points, based on what those parameters are..

Z positioning just at the start of the print or throughout it? Several people in the google group reported that prints were ending up too tall. I think that turned out to be fixed by setting ExportGcode Small in SF, though I don't know why that'd fix it. My best guess was floating point drift adding up to an occasional extra step but I also didn't think that was a very good guess.. 1. In SF, you can set the Max Z Feed Rate. It's in Speed in SF44 and Limit in earlier versions. Acceleration is something usually done in FW, not in the slicer. 2. I doubt it - mendels and makerbots are fine with that sorta deal, I think. 3. No idea.

I always use (the windows version of) ~/.skeinforge/alterations/start.gcode. Check the Bookend module to make sure it's active and that the file names are correct. Does that help?? Never seen the .g thing before.. Er.. It's probably something in Export. I've got Activate Export, Analyze Gcode, Gcode Small set and File Extension set to gcode. Is that what yours looks like?

Ah! I recently changed the thickness from 0.3 to 0.285, and I've noted an occasional bumping of the nozzle here or there. What's the best way to reduce the flow rate in SF 35? Tell it that the diameter of the incoming plastic is larger than it really is? That sounds like you're using the print-o-matic stuff built into repg. I know it's there but I've never used it. My preference would be to scrap SF35 and all the stuff that's integrated into repg - use repg just as something that preheats & sends gcode to the bot and download/use SF by itself. Any version past 40 (44 is the latest, dated 2011-10-08) will behave a lot better than SF35 does.

Have you disabled the Fillet module and set ExportGCode Small? What's SpeedTravel Feed Rate set to? Do you hear any sounds when this happens? Probably not interference if all your wires are still twisted and I wouldn't guess loose belts would cause this unless they were really loose.. Another thing that can cause skips is the nozzle smacking a blob at high speed - in general, you don't want the nozzle touching the print at all. There's sortofan exception on overhangs that curl up but those usually aren't too much of a problem. Also, very sudden & rapid changes in direction can cause skips, though I think that's become a lot less common with sprinter/marlin.