printer bed

I wonder if someone might be able to shed some light on the problem I am experiencing. What would cause the printer bed to heat up to about 50 degrees and then just suddenly start cooling down. If I switch everything off and back on again the same thing happens.

All help is greatly appreciated.
Jannetta

Hi, welcome to the forum.

Are there any messages written to the console/log when it shuts down? There are some settings for detecting faulty thermistors that can cause a shut down, but there should be a message when they kick in. A faulty power supply might also do it, so it would be worth checking what the voltage is before and after the shutdown.

Hi James

Thank you very much for your response. I use Pronterface and I'm not aware of any log files other than the console. Nothing is displayed on the console when this happens. I have just checked the voltages. When I press the button in Pronterface to heat the bed up I measure 10V over the connectors, when it cuts out it drops to mVs.

Regards
Jannetta

OK, that sounds like a faulty power supply then. What sort of supply is it, a dedicated 12v unit (typically used for LED lighting) or a pc atx unit? If it's a LED style 12v unit then it's just plain faulty, if it's a pc supply then it might be tripping over/under voltage protection.

Hi James

It is a PC supply. I bought the parts as a kit from Greg Colbourn on eMaker a couple of years ago. I think the power supply should be more than able to handle the load. Why would it be tripping? How do I solve the problem?

Regards
Jannetta

Ok, have you got some load on the 5V line? PC power supplies are rather crudely regulated by combining the 12 and 5 lines to create a single feedback signal. If you put a lot of load just on the 12V line the 12V voltage drops (which might explain only measuring 10V on the 12V line) and the 5V rises. If the 12V drops low enough or the 5V rises high enough it can trip out the protection circuit and the power supply will shut down. The easiest way to fix this is to add some load to the 5V line so that the feedback signal given to the regulator is a bit more sane. I have an old car headlight hooked up to mine - it looks very fetching, if slightly steam punk mounted in the back corner of the printer.

You should probably do some more checks to make sure we're on the right track though. When you measured 10V was that at the power supply or at the heat bed? There are other reasons for voltage drops at the heat bed (too small a gauge wires or a bad mosfet). Also, it would be worth measuring the 5V output from the supply while the bed is heating to confirm that it is rising above the expected voltage. Mine sits at about 5.4V even with the load.

Edited 1 time(s). Last edit at 03/13/2016 06:32PM by JamesK.

Following the instructions on Greg Colbourn's site, this is how the power supply is wired:
cut off all connectors, then wire:
4 black, 2 yellow to 5A; 8 black, 4 yellow to 11A (black to negative, yellow to positive)
1 orange 1 brown & 1 green 1 black to 4R7 power resistor
1 red & 1 black to 10R power resistor

I can't get to the 5V wires because they have been wired to the power resistors which are now mounted inside the power supply. Unless really neccessary I didn't want to open the power supply up tonight.

The thing is that everything was working okay until a few days ago when suddenly the bed started cutting out. The 10V I measured was at the connectors that go to the bed. If I measure the yellow wires (as above) then I get pretty much 12V.

Ah, ok. Sounds like I was leading you in the wrong direction then - sorry about that. The power resistors are the extra load I was talking about, so you're already good there (as long as nothing has come disconnected). If you keep the meter on the power supply when the bed cuts out do you still get 12v? If so, then we can put all of my power supply ideas to bed and go back to figuring out what your ramps/firmware is doing

Yes, I measured the 12V from the power supply before and after cut-out and it was stil about 12V.

Back to the firmware and control board then. What sort of control board is it, and what firmware are you running? I'm surprised it didn't print anything to the console when the bed turned off. Do you have an lcd on the printer, and if so is it still responsive after the bed shuts down?

It is an Arduino Mega with Marlin. No lcd on the printer. Everything else works after the bed shuts down.

Given no messages from Marlin to indicate it meant to turn off the bed, it sounds like an intermittent fault in the wiring. That might mean an actual broken wire, a problem at either end of the wire, or a dry joint on the mosfet, and includes the wires from the psu to the 11amp input of the ramps, as well as the power connector - you could check that it hasn't overheated and melted. The mosfet swtiches the -ve line, so when the bed is off you can expect to see +12V at both of the terminals on the ramps (that the bed connects to), when the bed is on the - terminal should fall to close to 0V. If you set the bed to heat to a high temperature you should have time to measure the voltage between the - at the power supply and each point around the circuit (+ bed terminal on the ramps, + terminal on the heatbed, - terminal on the heatbed and - terminal on the ramps again). Then when the bed fails, go around each point again and compare, hopefully that will be enough to work out where the fault is.

Hi James

The fault is not intermittent, it is now pretty consistent. Between, more or less, 50 and 75 degrees the bed will cut out. I would have thought this would mean an error with the bed. What do you think? Or could it be something with the control board? I still think bed. What can go wrong with the bed itself? Can it be fixed or does it have to be replaced - if it is the bed?

BTW thank you very much for helping me fault-find. I appreciate your time and effort.

Regards
Jannetta

The bed might do that if one of the pcb traces was cracked; as the bed heats up and expands the crack would open and break the circuit. But, if that was the case you'd see 12V across the terminals at the bed even when it cut out, and I think you said that when it fails the voltage at the bed falls close to zero. That suggests the bed is fine and makes it more likely to be a wiring or ramps problem. Most problems can be fixed, but sometimes it takes more effort than makes sense given the fairly low price of replacements.

i had this problem last week. it ended up being the Mosfet on my Ramps 1.4 board, the temperature shut off was being triggered and when i touched it on the board right when it started cooling down the board was fire hot. i fixed this by installing a Relay to power the heat bed straight from the power supply and not the print board.

try that

Hi siderala

Could you explain a bit more about how you wired up the relay please?

Quote
jannetta
Hi siderala

Could you explain a bit more about how you wired up the relay please?

I will certainly try. warning though, I am very new to all this so im just saying that this is what works for me and my house is still standing.

two files I attached. the Relay wire guide is the relay i used. it cost like 6 bucks online. it also has the pin out for the wiring. the two tabs at the top of the drawing are the red and black that will go to the "coil" side of the Relay. these will then be wired to the RAMPs board in the position on the second drawing attachment. now connect a wire (18awg min) from the PSU to the tab on the first drawing circled in Red with the arrow pointing Right. then connect another red wire from the heatbed to the tab on the relay with the arrow pointing left. wire the power to the ramps board just like it currently is and then it should work.

hope this helps

I'd recommend you put an inline fuse in the line from the psu to the bed.

Quote
JamesK
I'd recommend you put an inline fuse in the line from the psu to the bed.

thank you for reminding me to "borrow" from our warehouse! and that is to just protect the Ramps board in case of a surge correct?

More to protect the PSU and your house in the event of a short at the heatbed.

I had the same thing happen with my Ramps controller, basically it would cut out after a while, with no errors or warnings. Act like it was heating the bed in repetier host. I found on a forum talking about fuses that connect to the mosfet, if the mosfet or these fuses get overheated it shuts down until cooled. Placed a heat sink on the mosfet, and then added a pc cooling fan to blow on the controller board, and haven't had the problem happen again. Didn't want the hassle of wiring up a relay, however you probably would get better heating control from a relay

Edited 1 time(s). Last edit at 03/29/2016 01:29AM by Nate523.