Welcome to the SeeMeCNC Community

This will be the biggest thing I've printed yet! Took forever to get a slicer to actually slice the thing. Also had to scale it down a couple percentage points to get it to fit.
[youtube]http://www.youtube.com/watch?v=Lc2r2RKJebQ[/youtube]
20 hours to go!
https://www.thingiverse.com/thing:1458545

Looks like something I would like to print, perhaps in a bronze or Copper but the thing would probably take 8 rolls.

cool model best of luck with the print

if you do it in lost PLA for casting (Patinated bronze would be nice) and then give it glass light up red eyes and a smoke generator(use a tank type Vape unit) with prox sensors so it lights and smokes from the nose and mouth as it is approached...


should cut down on visits from various missionaries

[youtube]http://www.youtube.com/watch?v=UrCHtCNfITw[/youtube]

Might be a bit too big for our printers... 4 hours-ish in.

Sad to see that. Looks like the major problem was drag from previous layers causing the plate to shift. I'm sure you could probably start over, and that better retention of the plate would help, as might a higher nozzle temp, and a slightly lower extrusion multiplier (To reduce the volume of plastic per layer). Looked good until then. PLA or ABS?

PLA. I'm using the free spool I got MWRRF. I finally got Slic3r to slice it, had to up the threads to 8. Added a 2mm hop on movements, couldn't find that setting in Cura. The funny(haha) part is it's printing now, but says it's going to take 27 hours! I hopped it up to .2mm layers, cura did it in 21 hours with .1mm layers.

Weird issue with the calibration??? Notice the honeycomb isn't as defined at the bottom of the picture as at the top.
[img]http://lh3.googleusercontent.com/-HLILR ... 152355.jpg[/img]

Nope. Not going to try this anymore. After a kink in the filament caused me to have to figure out which layer it stopped extruding, then about 20 more layers in it hung up on a curl, it's just not going to print.

barry99705 wrote:Weird issue with the calibration??? Notice the honeycomb isn't as defined at the bottom of the picture as at the top.
[img]http://lh3.googleusercontent.com/-HLILR ... 152355.jpg[/img]

Looks like the bed isn't level. I think the part that is bad is sloppy like that because the nozzle is higher above the print than it should be.

forrie wrote:

barry99705 wrote:Weird issue with the calibration??? Notice the honeycomb isn't as defined at the bottom of the picture as at the top.
[img]http://lh3.googleusercontent.com/-HLILR ... 152355.jpg[/img]

Looks like the bed isn't level. I think the part that is bad is sloppy like that because the nozzle is higher above the print than it should be.

It's as level as I can get it by eye, but I've never had good luck getting anything to stick out at the edges of the platform.

It really is worth getting a cheap dial indicator off Amazon and running either manual or auto OpenDACT etc. The FSR kit is pretty cheap as well for auto calibration. Longer arms 300mm or 325mm help as well to print over the entire bed.

The dial indicator certainly helps for a manual calibration. Bit it still is very tedious. I tried it, and the more accurate your dial, the more certain your results will disappoint I stopped trying to get a perfect calibration, because that seems virtually impossible. I want fsr desperately, and it is not expensive. But the only firmware that it seems to work perfectly on is the DC42 and for that you need to replace the Rambo with a Duet. That is just a bridge to far for me at the moment...

DeltaCon wrote:The dial indicator certainly helps for a manual calibration. Bit it still is very tedious. I tried it, and the more accurate your dial, the more certain your results will disappoint I stopped trying to get a perfect calibration, because that seems virtually impossible. I want fsr desperately, and it is not expensive. But the only firmware that it seems to work perfectly on is the DC42 and for that you need to replace the Rambo with a Duet. That is just a bridge to far for me at the moment...

That is one of the problems with this equipment, it gets to be very expensive.

Eaglezsoar wrote:That is one of the problems with this equipment, it gets to be very expensive.

It would help if there was a capable firmware that fits the Rambo. There are still no newer "official" (read SeeMe) firmwares available. Does a auto calibration take so much memory that it cannot be fitted into a rambo?

DeltaCon wrote:

Eaglezsoar wrote:That is one of the problems with this equipment, it gets to be very expensive.

It would help if there was a capable firmware that fits the Rambo. There are still no newer "official" (read SeeMe) firmwares available. Does a auto calibration take so much memory that it cannot be fitted into a rambo?

+1

SeeMe doesn't develop the firmware, they provide a version with a preset configuration for the Max.
Delta auto calibration has only very recently reached a point where it's actually useful IMO, judging by the effort SeeMeCNC seems to have been putting into the accelerometer based probes, I'm sure their intent is to have the printers self calibrate. I have no idea what electronics and firmware they are running on the Eris.

You can manually run the one in the DC42 firmware with a dial indicator using this webpage http://escher3d.com/pages/wizards/wizarddelta.php , It takes perhaps 15 minutes you do have to negate all the numbers you read from the indicator.

Adding DC42 like auto cal to repetier is probably not all that hard, but it's open source and unless someone has the motivation and aptitude to do it, it's not going to get done.

DeltaCon wrote:

Eaglezsoar wrote:That is one of the problems with this equipment, it gets to be very expensive.

It would help if there was a capable firmware that fits the Rambo. There are still no newer "official" (read SeeMe) firmwares available. Does a auto calibration take so much memory that it cannot be fitted into a rambo?

Short answer: Yes.

According to 626pilot, who develops autocalibration for smoothie, he is rapidly running out of space on which to implement it. The smoothy has twice the flash, eight times the ram, and a far higher clock speed. The RAMBO's processor can barely handle doing delta kinematics in realtime, much less auto-calibration and kinematic corrections.

Polygonhell wrote:You can manually run the one in the DC42 firmware with a dial indicator using this webpage http://escher3d.com/pages/wizards/wizarddelta.php , It takes perhaps 15 minutes you do have to negate all the numbers you read from the indicator.

Yes I have been trying that, and it works to a certain level. But when you forget to negate one number, you start from zero. Also I have trouble setting the dial to zero, because after touching the dial the measurement is off sometimes by 0.05 because. Not sure if resetting is needed though, probably not. My results are not very repeatable, but that might just be the source of the calibration problem I'm beginning to think... I only have one printer, so no possibility to reference. Anyone in the Netherlands Limburg or Brabant provence around here?

DeltaCon wrote:Also I have trouble setting the dial to zero, because after touching the dial the measurement is off sometimes by 0.05 because.

That's a bit of a worry. You should be able to get near perfect repeatability on a bed that is temp stable. I'd be looking into that before going any further otherwise your results will be frustrating.

Personally, I would zero it only at the start of the operation, and then just remember or write down your reference points. Many dial indicators are sensitive enough and the backlash and flex in the assembly is large enough that re-zeroing will knock you off, at which point you need to start over on that iteration of measurements. I'm just glad we aren't using dial indicators mounted on armatures, because those are a pain in the ass to sweep around and not jostle.

I wish you the best of luck, especially if you're using a cheap dial indicator. We have enough problems with mitutoyo ones that I can't imagine that a cheap solution would be any more user friendly.