gr5

I moved this thread as you asked. It also shows up in the old spot but when you click it, then it jumps to here. Does the bed heat up even when "goal temperature" is 0C? Does your heated bed use a relay? Or a FET? Do you use the mosfet on the ramps board to control the heater directly?

It could be either bad Z screw but if you didn't change the Z screw it is much more likely a temperature issue. Watch the temperature carefully while it prints. See if the temperature varies. It should be consistent within 1C. The pattern you show is what I would expect with 10C to 15C swings in temperature.

Yes. It *could* be implemented. The UMO would have to have an interface to allow you to set 3 things: filament diameter retraction distance retraction amount The changes in gcode for ultigcode are that retraction (and un-retraction) is done with G10,G11 commands instead of moving extruder axis. The other change is that extruder position in regular gcode is mm of filament at the feeder and in ultigcode it is in cubic mm. You need to know the filament diameter to convert between these two units. The code to handle ultigcode is already in a version of Marlin but I don't know if it will (or has already) made it's way into the "standard" ErikZalm version of Marlin. Although Erik works for Ultimaker now (started a few months ago I think).

I used the gas flame from a cook stove. Leave it in long enough to turn the plastic black but not much longer. As you say, don't melt the brass. There is a large temperature difference between burning plastic and melting brass so it's not too difficult. Just don't leave it in there and fall asleep. I moved my nozzle into the flame for no more than 10 seconds at a time. Then moved it out and touched some filament to the outside to gauge the temperature.

STL isn't so good with dual nozzle setup. Currently you have to have one STL file for each color/material. And there are 5 color (5 filament) printers out there.

Unless he ships his computer to Daid

Having said all that they *are* working to improve this. Like I said they just hired more people in support and also they are improving their ticketing system.

For one thing - they are overworked last few days as they changed their ticketing system and/or ordering system and it's a temporary disaster. For another thing even though they just hired a few more people they have been a few weeks behind. So, um, 2 weeks is typical. And YES, the phone works MUCH better. But you probably have to speak english or dutch. Their english is excellent.

This topic moved to this thread: umforum.ultimaker.com/index.php?/topic/8742-how-can-i-test-the-heated-bed-on-um2/

In post #5 above in Coreys photo there is a solder pad under each of the screw connectors - on the "bottom" side of the board in this photo. You could just solder the wire directly to that pad. If you aren't good with soldering, I'm sure you have a neighbor with the right experience and equipment.

Yes. There's another thread about this. But the quick answer is that they switched their "ordering" software a few days ago so even though your order is just fine on the old system you can't see it anymore on the new system. Plus cust support is overwhelmed with people asking what happened to their suddenly missing order. If you just wait your tickets will eventually be replied to but you might have to wait a few weeks. You might get your printer first.

I'm not sure the apm 2.6 is a good choice. Calibrating it has been a ton of work. I'm starting over and buildign a jig to hold the copter but let it spin on one axis at a time so I can calibrate just that axis carefully.

Most c or c++ programs when "killed" do a lot of heap cleanup first. It's silly but that's how we usually write the programs. So all those destructors or free() commands have to get run. When there are a lot of objects in the heap (millions) it can take a long time to delete them all. Especially for linked lists or other linked structures - following all those links and running a delete or free on each one. This can take a long time (e.g. maybe 30 seconds) and also slow the computer down while this happens. There's really no need to do this other then so that you can check for memory leaks when debugging the code. If you do a instant kill on the process (like from task manager process tab versus application tab which only sends a kill message) it skips that step and just frees up the memory allocated to the process on a higher level - not going through all the linked lists. I'm thinking the difference between daid and bot computers for this "kill process" step is that bot had a larger amount of memory tied up in the heap because of a larger model or thinner slice and maybe he was going a bit to page faults on the disk on the cleanup step.

I'm told all the other 3d printer companies have horrible support but this isn't much of a defense. @sod - If you had asked before you ordered (I know - how are you supposed to know to do that) I would have told you to order your printer through the USA based makershed which has a more sophisticated ordering system that knows what is in stock and what is not. @sod - message this guy to get a more accurate delivery estimate - explain that you aren't in a rush so much as that you need to know the date because you are moving soon - so click the link below and then "send me a message" button: http://umforum.ultimaker.com/index.php?/user/423-sandervg/ Make sure to include your order number. It should be a large number that probably starts with the letter E.

As was already said - don't panic - your orders are still in the *old* software not viewable on the web.

The ordering software was switched to new software so even though it doesn't show many orders - they are still there.

I really don't know but it might have to do with how much melted plastics stick to aluminum versus brass. The original reprap nozzles were brass acorn nuts machined on a lathe with the hole drilled out. Many probably still are. You get the hex nut shape and thread built in and they are cheap:

What settings are you using for the fan coming on? I believe the new defaults are 100% at .5mm maybe? meaning if you have .2mm layers it comes on rather quickly (40%,80%,100%). where the old default I think was 100% at 5mm. (4% increase per layer at .2mm, 2% at .1mm). Personally I typically use 100% at 1mm. Of course for smaller parts it's even worse because if it can print a layer in 10 seconds it may be increasing fan speed faster than the heater can catch up.

Just FYI Anders design is moving along. He is having someone make a small batch *and* he has a partner working on making a medium sized batch at the same time.

It seems like you have enough information to make it home anywhere and have z=0 be anywhere (should be when nozzle just touches plate) and have Z=10 be nozzle 10mm above plate and so on. Play with pronterface some more?

Progress. So now your Z direction is backwards maybe?

That's a good sign! Try putting the acceleration back to 5000mm/sec/sec or whatever it had been. You will get better quality with higher acceleration (because you don't want the extruder to ever slow down and if it does you want it to be brief). Um - yeah - a limit switch could also cause this problem. If the LEDs are somehow hitting it or it's getting shorted out - follow the wiring with your eyes and maybe it gets pinched under one of the metal covers (the limit switch for the "bad" axis).

Instead of swapping the pins in the firmware - couldn't he have just swapped the wires to the endstops?

I would like to see a version where I don't have to modify fans or drill hole in fan mount plate.

This sounds like the most unbiased comparison so far? Ultimaker 2 and Original both did well. Makerbot 5th generation did very badly of course: Only 67 printers are mentioned (they need at least 5 reviewers to get mentioned): http://www.3dhubs.com/3d-printers