The (lack of) quality to this kit, and its terrible instructions, continue to spike my blood pressure. But structural assembly is now complete, and wiring up the electronics is about 30% done.
|
real 3D modeling resources.
|
|
Congratulations. While I agree the instructions weren't perfect I did not have nearly as much trouble as you did on the assembly.
E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell." 
The photos are atrocious, especially in the "wiring" section. No zooms available, and so many color definition issues picking out the wires that I eventually had to dig up the Sanguinololu schematics in order to figure out WTF the instructions were talking about. And I still managed to completely miss that the heated bed and the main board need to be powered in parallel straight from the brick.Not to mention that there's apparently not a schematic for the heated bed itself anywhere on teh internetz. Writing tech docs is part of what I do for a living, and these are just shameful. And it's a Wiki, fercryinoutloud! Nobody's even introduced an erratta page?
The extruder: W!T!F! You have to adjust the position of the motor in order to make the gears mesh properly... but with the big gear in place, 2/3 of the mounting screws for the motor are utterly inaccessible. And those &*^&(^(*^!!! nylock nuts on the hobbed bolt -- I've never seen something so well-designed to make it impossible to hold the depth of the bolt -- it was impossible to control which nut was getting tightened. I eventually had to resort to an improv jam nut and some Loctite. And whomever designed the heated bed such that the control cable connector interferes squarely with the levelling bolt needs to be beaten with a nerf bat until they learn to CHECK INTERFERENCES ON THEIR BLOODY DESIGNS! Having to bend the PCB pins by 30deg just to get the plug on is ridiculous. Now, to be fair, my screwups: The Y-axis endstop carrier is on the wrong side of the frame (left/right, not back/front). I'm not taking this bloody thing apart again, so I'll have to improv something. And somehow I soldered the main power socket incorrectly (caught that doing circuit checks before I plugged anything in, thank God, so I didn't fry anything). And those little crimp-together PCB connectors for the motors... well, let's just say that I'm ordering a bunch more from Pololu. Advice to anyone building one of these, or anything using these connectors: buy one of these and for your own sake, read the instructions. Also buy extra packs of the housings, and the pins. And make sure you leave enough slack wire to snip and re-make the crimped ends several times, unless you've already had lots of practice. Trust me. Status: Complete except for re-terminating the motor cables, mounting/wiring the endstops, and getting the main power wiring straightened out. I may actually achieve "smoke test" status by the end of today. SkyeFire Wrote:The photos are atrocious, especially in the "wiring" section.Granted, I went to IRC to get help with that. Quote:I still managed to completely miss that the heated bed and the main board need to be powered in parallel straight from the brick.0.o How did you miss that one? They are fairly clear about that one. Quote:but with the big gear in place, 2/3 of the mounting screws for the motor are utterly inaccessible.yes some of the steps are much easier to do out of order. I disassembled and reassembled some pieces to make it work. Quote:And whomever designed the heated bed such that the control cable connector interferes squarely with the levelling bolt needs to be beaten with a nerf bat until they learn to CHECK INTERFERENCES ON THEIR BLOODY DESIGNS! Having to bend the PCB pins by 30deg just to get the plug on is ridiculous. Ramen brother, ramen. I just went with longer screws to hold the bed up since I didn't dare bend the pins. Quote:And make sure you leave enough slack wire to snip and re-make the crimped ends several times, unless you've already had lots of practice. Trust me. Very good advice. I have had some experience with crimping Ethernet cables so I didn't find it that difficult but it's not easy. Also the crimps you link to look very different from the ones that came with my reprap pro. Quote:Status: Complete except for re-terminating the motor cables, mounting/wiring the endstops, and getting the main power wiring straightened out. I may actually achieve "smoke test" status by the end of today. Congratulations. E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Burn test conducted (sans endstops, for which it turns out that the screws called out by the instructions are completely wrong). Results.... mixed.
Smoke: none (aside from puff from the hot end, which appears to have been the PTFE tape cooking in) Extrduer: turns just fine, and pushed about 1in of filament up the Bowden tube... after which the filament jammed somehow and snapped off at the hobbed bolt when I tried to withdraw it. I'm going to have to completely disassemble the extruder to figure out how it jammed -- there's nothing there for it to jam on! I ran filament through it several times offline without a single issue. (side note) whoever designed the extruder so that the retaining nuts fall out when you open the idler, and made the holes for those nuts nearly inaccessible due to the motor being in the way, needs a nerf beating. Ditto for the the way the main gear blocks access to the motor mounting screws. First improvement once parts are printable: modified extruder). Addendum: somehow, the filament appears to have jammed at the brass exit fitting of the extruder. After an inch of filament was already through it. Still no possible explanation, but clearing this is going to be nearly impossible without wrecking a critical part. The X and Y axes will move nicely (albeit only in + direction right now, due to not having any endstops). The Z axis... I dunno. One motor jumps and turns erratically, the other doesn't even hum. Which is nuts, b/c those two motors are wired in series. Hot end: heats up fast. Thermistor says 200C, my IR thermometer says ~50C. My finger says "ow!". Going to have to look into that. Finding a schematic for the heated bed is now a necessity. As far as I can tell chasing signals with my meter, it looks like the instructions got the control/sensor wires for the bed all swapped around. But given how easy it would be to burn out something, I don't dare just start swapping wires until I know exactly what I'm dealing with. Also, there seems to be something wrong with the instructions for the "current limiting" instructions for the stepper drivers. The Melzi instructions are using the same A4988 drivers as the Sanguinololu does, and explicitly calls out the reference voltage as 0.4V. But the Sanguino instructions reference a page from Pololu that doesn't give a reference voltage, but states that the reference resistors are 0.05ohm. Which, for the 1Amp limit of the steppers, would seem to indicate that the ref voltage should be 0.05 to 0.07V, which is close to impossible to set -- the trim pots move more than that if you breathe on them too hard.
Re: the double power cables. Part of the problem may be that I'm wired to think in schematics rather than verbal (all through the wiring section I would have been better served by schematics than the verbal/photo combo). And looking at that power plug, the entire mess they made of how you're supposed to solder it together doesn't help. Not to mention that the cable that came with the kit was enough for one of their power runs, not both. But I have to admit to getting lost in the details once I went from "comprehensive pre-read" to "step by step" phase.
My issue with the extruder is that the order doesn't matter -- some of those steps simply aren't possible without a 5-dimensional screwdriver. We eventually got around the access problem for the motor mount by iteratively dis/re-assembling the extruder until we had the mesh correct. And the issue with the hobbed bolts and the nylocks had my master mechanic ready to commit murder (or at least minor maiming) on whomever was responsible for that setup. Pins: I linked to both the male and female versions, even though the Hux only uses the females. But I can confirm that the female pins are the same as what came with the Hux kit -- I already had some on hand from Pololu for a different project. Ditto for the housings. Progress: Had to remove the entire Bowden tube and repeat the 2mm reaming, despite having done that and successfully tested a couple days ago. It would appear that my filament has thickness variations -- I happened to start with a piece that fed, albeit stiffly, but after about an inch widened to the point that it jammed hard in the part of the Bowden tube that was compressed by the brass fitting. My hot end keeps reporting the same temp I set, up to 300C, but my IR thermometer keeps saying ~30C. I can actually push filament through and it melts and extrudes, but the extrusion is slow, stalls the extruder motor, and comes out of the tip at nearly 90deg to the axis of the hole, as if the hole was drilled off-angle. I'm pretty sure there's something wrong with my hot end. If I change the setting from 250C to 300C, the software reports the actual temp doing that 50C climb in under 10sec. But holding my IR thermometer on the hot end while that happens only shows 1-2C climb. And now, the resistance from the hot end was enough to kink the Bowden tube where two filament ends adjoined. At an ambient temp of around 75F, my heated bed thermistor is reporting a temp of 9C, and the hot end shows about 25C.
That might be the problem, the little EE I did was long enough ago that I don't remember how to read the schematics, but I can follow the verbal description just fine.
Check your wiring for you thermistor, the join might be bad and have too much resistance. Fiddling with the wire a bit fixed it for me when I had a similar issue. E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Well, I finally got a chance to check out the IRC channel, and got a few tips:1. Don't trust an IR thermometer on a hot end. Given the "spread" of most IR thermos, and the extreme heat gradient in the average hot end, I can sort of see why.
2. If you can push the filament through the hot end by hand, your temp is in the ballpark 3. There's an 8x multiplier that needs to be applied to the Vref value (which is not noted in the instructions anywhere, hah), so that the 0.05V I calculated becomes... 0.4V. Which is exactly what the Hux instructions say to use for the Melzi electronics, but tells you to use something different for the Sanguinololu, without telling you what it is. Nerf bat squad, NERF BAT SQUAAAAAAAAAAAAD..... 3a. So my "stuttering" extruder motor is most likely due to #3, rather than a bad hot end temperature. So, after finishing my 14hr work-Saturday and hitting a co-worker's Memorial Day Party 'til 3AM, I now have the rest of the holiday weekend to keep kitbashing this thing into submission. Hopefully my Z-axis issue is also related to #3. That just leaves the heated bed issue; I still cannot find a schematic for that thing.
So I have spent most of today poking at my reprap to get the calibration finally worked out. The X-axis still has too much backlash but the print quality had gone up massively. I think I'm ready to print something for real.
E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Well, I'm ready to try printing... I think. I've actually downloaded a test object and sent it to the printer, and the printer is moving... but for some reason I've yet to fathom, when I feed an STL file to Pronterface, the GCode it spits out has all the extruder commands commented out. So I'm running a full pattern test print... with no extrusion. Still, I suppose it's a good shakedown for the machine.
I'm also having a problem where most of the STL files I've tried seem to treat the center of the bed as 0,0,0, as opposed to the negative limit of each axis, which my machine apparently does. So I keep getting GCode files with negative values in X and Y. I've also had my extruder tube bust loose from the brass fittings twice, on each end. I'm pretty sure the plastic is getting weak in the threads. And I've had a couple of near-ruptures in the tube along its length, where the junction between two lengths of filament tried to bust out. I'm hunting about for a source of replacement PTFE tubing (2mm ID, 3mm OD), but not having a whole lot of luck so far (Grainger has some 2mm/4mm, but only in 25ft(!) lengths, at $75). Part of the problem is knowing what "brand" name to search for, since most catalogs won't be calling it "PTFE tube."
I had a similar problem with Skeinforge (The slicer in pronterface) so I switched to Slic3r, it works much better and it is much (very much) faster, but I'm still getting notable improvements in objects when I tweak the Slic3r settings. Trying to find the ideal settings still, since my objects are a little too crude still for the stuff I really want to build. I'll share mine if you want.
E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Are you using Pronterface for your user interface? B/c I'm having some really weird issues there.In order to just get some extrusion onto the bed, by way of testing, I tried just entering the motion/extrusion commands into the manual entry window. And I'm getting some deucedly odd extruder behavior.
Now, when I hit the manual "extrude" button, the extruder performs perfectly. No jitter, no stalling, nothing. So I'm pretty confident that my extruder wiring and mechanicals are okay. But when I try issuing a command like G1 X100 Y100 E20 F100, the extruder might turn... or not. And when it doesn't turn, it's b/c the firmware isn't even putting power to the motor -- I've put my hand on it when this happens, and there's not even a bit of hum. And it seems completely random! I've tried prefacing the move/extrude commands with M101 (extruder forward), M108 (deprecated extruder speed command), M113 (current extruder speed command), but no dice. It doesn't seem to make any difference -- the firmware just seems to ignore the E command most of the time. Even when I've hit on a chain of commands that seems to succesfully move&extrude, repeating that same set of commands gets me movement without extrusion. I'm baffled. 
Sounds like a baud rate issue. The communications are happening at one speed, but the receiver is expecting it at a different speed.
-- "You know how parents tell you everything's going to fine, but you know they're lying to make you feel better? Everything's going to be fine." - The Doctor SkyeFire Wrote:Are you using Pronterface for your user interface? Yes, though I havn't bothered trying to enter commands in the window, the reprap-pro guys have a slightly customised version with handy buttons. I would show screenshots but it's on my other computer. I have had similar issues when one of my wires was a bit loose and it drifted in and out of connectivity. My guess is that you are using absolute extruder coordinates, not relative ones. Try the customized version from here; http://github.com/reprappro/Software E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Actually, that's the software I am using, as presrcibed in the kit instructions. I'm starting to lean towards the theory that I'm tripping the stepper driver's thermal overload, although I'm not sure how, since I set the current limiter to match the motor (1A).
At any rate, I'm a bit stuck at the moment. Work has gone into panic mode and eating up all my time (I'm typing this at 0530 local time at the end of a my 10th 12hr+ shift in as many days (plus 2hr drive time)), and my last test attempt my Bowden tube stripped itself out of the brass fittings. That's a major structural weak point, there. And I can't find the tubing to buy for love or money -- the only places that seem to stock it at all are in England, and they're all sold out (and will charge me $35 shipping on a $10 piece of 300mm tube). I found something similar at Grainger Stateside, but they only sell it in 50ft/$75 dollar spools. If anyone knows where I can get 3-6ft of PTFE (Teflon) tubing (not the heat-shrink type) with an inner diameter of right on 2mm and an outer diameter of 3-4mm, without spending ridiculous amounts of money, I'd appreciate it.
check the reprap forums they might be able to help, but I don't have the strength issue with the tube. in fact my extruder motor once ground through the entire plastic feedstock when the wires for the heating element came undone, but the tube and brass fittings are solid.
E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Nothing much to report. I did however find someone online finally who sells 2mm ID x 4mm OD Teflon tubing for $2.50/ft, stateside, an outfit called SeeMeCNC. So I've got fresh tube on order. 3mm OD would be better, but 2x3 seems much hard to come by that 2x4, so I'll take what I can get.
Ah, Catty... if your extruder ground through your entire feedstock, but the hot end wasn't heating and the tube didn't rupture anywhere... where did all that filament go?
it ground through the feestock because the filament didn't go anywhere. think of it as a really messy cut through the feestock, not the consumption of the entire feedstock. does that make it clear? Basicly the extruder motor was going and the part of the filament in connection with the axe was going but the rest of the filament stayed stationary.
E: "Did they... did they just endorse the combination of the JSDF and US Army by showing them as two lesbian lolicons moving in together and holding hands and talking about how 'intimate' they were?" B: "Have you forgotten so soon? They're phasing out Don't Ask, Don't Tell."
Ah! Sorry, I misread that completely. Yeah, sawing through the filament isn't hard. And then you need to clean out the teeth of the hobbed bolt in order to regain traction on the next filament. Been there, done that.
Article in The Atlantic about 3D-printing
-- Rob Kelk "Governments have no right to question the loyalty of those who oppose them. Adversaries remain citizens of the same state, common subjects of the same sovereign, servants of the same law." - Michael Ignatieff, addressing Stanford University in 2012 |
|
« Next Oldest | Next Newest »
|
Users browsing this thread: 2 Guest(s)