I Can't Get A Break

[Zulfe]

If you've checked my recent posts, I've been having trouble with filament recently. I finally fixed the issue and printed off some quadcopter pieces. I noticed the next morning that they had layer separation on them, which was weird being that they were in a small, closed space (better environmental heat) and they were short (better inner heat). When I tried heating up my bed for tonight's prints, it wouldn't heat. The screw-down connectors for my bed are melted. The board has burn marks on it and the connector is done. I can't think of what to do next other than buy a new board or find a way to run the bed off of D09.

Any tips?
Photo Oct 14, 12 08 28 AM.jpg

[Mjolinor]

Take some better pictures.

It doesn't look too bad but it isn't good to see. Certainly I have seen a lot worse.

You should be whinging at whoever built the printer as that is caused by not tightening the wires enough.

It seems that for some reason a lot of these 3d printer makers decided they should solder the ends of the wires that fit into screw connectors. Tinning the wires for screw connection is an absolute no, the wire should be stranded and not tinned with anything if it is to fit into a screw connector. If you do tin them then this is what can happen.

[Zulfe]

I built the printer myself, but I know for a fact that I tightened all the screws down very much. I've broken screws in the same type of connector in the past due an issue of over tightening in the past. Maybe vibrations caused the screws to loosen over time? I could easy be the problem too, as much as I'd like to think otherwise.

[Mjolinor]

Did you put solder on the wires before putting them in the connector?

[printbus]

Like Mjolinor said, the picture doesn't show us much. Another possibility is that the wires to the heat bed were allowed to flex at the screw connector and wire strands slowly started breaking off, leading to remaining strands overheating at some point.

[gmay3]

Sorry to hear it Zulfe, don't be discouraged by these few issues you're having, stuff like this can just happens when you build your own 3d printer! You'll get it working!

Personally, I've tinned my wires that go into each screw terminal because in my experience, connections pull out more easily and are weaker when left stranded. Regardless, whether they are tinned or not, I don't think this alone is the root of your issue. It is more important to make sure that all of your wire slack is anchored to the frame so the wires aren't pivoting at the board connection when the printer moves. Your issue is that that one wire touched another while both were in the connector either by loosening slightly after being pulled on by the printer or by having too much wire exposed between the connection and the insulation during the installation. This also makes me want to go back and check that my screw down terminals are still tight!

Before buying a new board, does it still provide all other functions except the heat bed? If so, could you remove the burned connector and solder the heat bed wires directly to the board? To test this I would power down your printer first and remove the heat bed connector. Once it's cleaned up command the heat bed on with prontorface and measure the voltage between the two pins on the board where the connector used to be, also being careful not to touch the multimeter leads together. If everything's working the best practice would be to order a new screw down connector to replace the burned one. If you need help finding a replacement I'm sure we could help you look.

UPDATE BEGIN
I recently redid all the wiring and wire routing on my printer and after testing the strength of tinned wires and untinned wire terminations in the screw terminal, I 100% agree with printbus and Mjolinor that the wires in a screw terminal should NOT be tinned. Especially if you are trying to fit more than one wire into the terminal. Make sure that you twist the ends of your stranded wire before inserting the untinned end into the screw terminal. Many thanks to printbus and Mjolinor for educating me!
UPDATE END

[printbus]

There's probably a NASA workmanship standard out there somewhere that details it, but Mjolinor is correct in how stranded wire should never be tinned when the wire will be crimped or tightened under a screw head.

Yes, it appears to provide a more solid initial feel for the crimp or the screw terminal. The problem is that solder isn't as solid as it seems. When put under pressure (as in a crimp or screw terminal), the solder will slowly creep and flow out of the way, loosening the connection. It's just a matter of time - maybe months or more, but it will eventually lead to a poor connection.

[Zulfe]

Thanks for all of your suggestions, guys. It really shines some light on this crappy situation. I'm actually quite lucky I didn't burn my house down when this happened, as my printer is located in my room's closet. If the wooden frame caught fire, who knows what could have happened.

I'll take the board off the frame and analyze it some more and take some pictures when I get home. Maybe we can find a solution that won't require me buying a whole new $50-$80 board/shield. Funds are tight right now and I'm really counting on my printer being up and running to create pieces for some of my current projects. I don't really see the usefulness of a screw terminal on a vibrating machine since the potential of the vibrations loosening the screw is always a possibility. I would've preferred a pin connector like the rest of the board uses for the motors and thermistor. I also assume that it was entirely my fault for this happening since there was a very small amount of exposed wiring sticking out of the connector, creating the possibility for this issue to occur. My first solution is to modify the board's code to run the bed off of D9, which is completely unaffected. In the pins.h file, the bed runs off of '5', so I'm not sure how to go about moving it to D9. I'll eventually upgrade to a RAMBo board anyways, but I just need a quick fix in the mean time.

Thank you everyone.

EDIT [10/14/2014 @ 9:30AM]: Colin just replied with:
"I have seen that before if the heat bed is struggling to maintain the target temperature. There are a couple things that can help with this, first make sure you have insulation under the heat bed, next a heat bed relay will be the biggest help, if you get the heat bed relay you can move your heat bed to d9, if you do not get a heat bed relay d9 will quickly get damaged if trying to run the heat bed. The relay will let you reach higher heat bed temps at faster speeds, otherwise 110c may be difficult to reach.

From what I have seen in the past this is just due to very high current, not from shorting. The heat bed technically is shorting the two wires out so it can't be protected.


Here is the relay:http://www.makerfarm.com/index.php/p...bed-relay.html"

[gmay3]

Zulfe,

See this link for the replacement part. Search "J2"

http://reprap.org/wiki/RAMPS_1.4

It seems like it has the exact same connector in black, or a replacement two part pin connector that will also fit in to the board. The second one seems like what you were hoping would have been there in the first place

[printbus]

If you don't have a heat bed relay, I don't think you can move the bed heater to D9. D8 gets 12v power through an 11-amp polyfuse to protect the board (and your house) from a short in the heat bed circuit. This is the larger of the two yellowish postage stamp looking things. D9 gets 12v power through the smaller 1/2-amp polyfuse. You'll open the small polyfuse and shut down your printer quickly if you try to feed the heater directly off D9. Ignoring that problem, there's another. The MOSFET switch for D8 has a heatsink on it. The MOSFET switch for D9 does not. You'd likely overheat the D9 switch by drawing heat bed current through it.

FWIW, when polyfuses are subjected to overcurrent, they go into a high resistance state acting somewhat like a fuse. When the overcurrent goes away, the polyfuse resistance will restore to normal. So, I somewhat disagree with Colin. Since RAMPS has an 11-amp polyfuse, the board layout should be designed to handle at least that without any issue. I don't know if it was or not, but this makes me lean towards thinking the problem was in the wiring to the board.

If you do have a heat bed relay (or do as some have and wire in an automobile 12V relay to use for one), you should be able to swap functionality of the D8 and D9 connectors by swapping the values for the FAN_PIN and HEAT_BED_PIN in pins.h. Double check your pins.h file. For motherboard type 33, HEAT_BED_PIN should normally be defined as 8, hence the heat bed connection to the D8 port.

When a heat bed relay is used, only minimal current related to the bed heater is flowing through the RAMPS board. Also note that when you use a heat bed relay, the on-board polyfuse is no longer in the loop so it can't protect against overcurrent anymore.

FINAL EDIT? The reason why a heat bed relay can heat up the bed faster is that it is more efficient at getting power to the heater. It basically uses a mechanical switch connection that will have less voltage drop than the on-resistance of the RAMPS MOSFET switch. The RAMPS board also has voltage drop in the "on" resistance of the polyfuse. There may also be more voltage drop from traces on the RAMPS board and the push-on connectors that are generally known to be somewhat imperfect. These voltage drops equate to power that never makes it to the heat bed.