Heatbed broken...?

Hi all,

After reading a couple of topics of other people that also have issues with a heated bed and reading the wiki over and over and bothering a LOT of people on the IRC channel I can not figure out what is wrong with my heat bed set up. Also burned a MOSFET on the way trying and had to buy a new RAMPS. "Luckily" no other electronics was damaged.
I try to be very detailed as possible.

The electronics that I use is a RAMPS 1.4 with a MEGA 2560. All the motors, sensors and endstops works correctly.
I have a MK2a bought from ebay and when I bought it I did not tried it yet because my printer was not 100% finished.
Without anything connected and soldered the heat bed reads 0.8 Ohms from the solder paths.
I soldered two leds on the heat bed with a 1K ohm resistor (same set up as show on the wiki [reprap.org]).
After soldering the LEDs it reads 1.4 Ohms (two LEDs and one 1K Ohm resistor). I also checked the continuity of the wires to and from the heatbed and all ends reads .001

My PSU is a ATX 350W cannibalized from an old PC. I've put an 50W 12v halogen lamp on the 5V jack for the normal load.
The PSU has a 12V 10Amps (yellow wire) and a 12V 15Amps (Yellow/black wire). I connected the 12v 10A to the 5Amps input of the RAMPS and the 15Amps to the 11Amps input.
D8 / D9 / D10 reads 12v when I turn it on but without attaching the hot end and the heat bed.
The hot end works perfectly and heats up if I send the commands in pronterface, but when I connect the heat bed and give the command M80 the PSU directly shuts down.

Someone on IRC suggested to connect the heat bed direct to the PSU to check if one of the LEDs light up. To prevent any damage he/she suggested to put a lamp bulb in serial between the PSU and the heat bed.
I've put a identical light bulb (I bought a second bulb and did not use the one I use for the normal load) that I use for the normal load and put that in serial between the PSU and heat bed.
The bulb lights up very brightly but the heat bed LEDs did not and the voltage of on the heat bed reads a 0.02V. After checkin and measuring the cables again it still reads 0.02V and the LEDS on the heat bed did not turn on.
After a lot of chatting on the IRC they suggested to try to connect it directly to the PSU without light bulb in between, after doing this the PSU directly shuts down after turning it on.

A lot of people on IRC said it was the PSU that could not handle the high current or the current limiter is very sensitive so I bought myself a PSU 12v 30Amps (this PSU [www.ebay.com]). I've connected the new PSU to the RAMPS and connected to the heated bed.
I send a M80 command and I did not see the LEDs on the heat bed light up but suddenly there was smoke coming from my RAMPS. I directly shut down everything and saw that my MOSFET was fried.
So I think my ATX PSU can handle the current but there is somewhere a short or is my heat bed broken?
I've checked the LEDs on the heat bed it is soldered correctly (also had the local "electrical" guy look at it, those guys are not that helpful anyway). I also tried to connect it without any LEDs or resistor soldered on it but still the PSU shuts down.

Am I missing something or is my heated bed broken?
I hope anyone can help me with this.

dvdx

Edited 2 time(s). Last edit at 11/10/2013 09:29AM by dvdx.

Light bulbs are not a real good load for a solid state power supply. They pull a lot of current when cold (not lit) and the current drops back as they heat up (light up).

The power supply you got from eBay *should* be ok, but who knows. It's a clone of a known good design. If they did it right it will work ok. I would check it's output voltage both at idle and under load. It should hold 12.0V +/- 0.1V. It also should be adjustable over > +/-1.5V. I would trust it far more than an typical 350W PC supply. In many cases the current output numbers on PC supplies don't all add up.

It sounds like you have a short somewhere. It could be a lot of different places. Bolts on the heated bed to an aluminum frame, solder joints to the frame, wire connections to frame or to each other and solder blobs on the Ramps board - all happen fairly often. When things vaporize it's a dead short you aren't looking for a 0.8 vs 0.75 ohm difference. Some quality time with an ohm meter and a bit of flexing all the wires should track it down. There are a few odd things that can happen, hot end heaters can short only when hot. Heated beds move a bit when they warm up, bolts that don't short cold could short when hot.

Hi, u need to check ur heat bed. Normally is 1.2-1.3 ohm. There is some seller in ebay selling those china clone mk2a heatbed that have a short on the underside go the bed. The short is from the two wire soldier point and the two hole to the back of the board that have copper. The correct design will have two cut box on the back. Hope this will help

Thomas.

Definately a short. Also, why did the resistance of your heatbed increase after installing the LED+resistor? It should be connected electrically parallel to the printbed, so if anything the resistance should decrease. And why replace the entire RAMPS just because of a bad MOSFET? When you connected the light bulb in series with the printbed, and the light bulb was bright and the LED on the printbed wasn't on, that was an almost certain indication that there is a short. The person in the IRC channel should not have suggested you connect it directly to the PSU at that point.

If the heated bed is made to the Prusa specs, it should come out between 1.0 and 1.2 ohms. They do show up down to 0.8 ohms, but that's a bit unusual. Once you to over 1.5 ohms you will have trouble at 12V getting your bed to heat as fast as it should. With an adjustable supply, that's not as big an issue as with a fixed supply.

--------------

None of that relates to the problem you are trying to solve right now. It can all be worked out after you find the short.

----------

Your heated bed PCB should not have metal in the mounting holes or near them. If it does, it's been made wrong. That can be a problem when it comes to shorts.

Edited 1 time(s). Last edit at 11/10/2013 01:36PM by uncle_bob.

Thank you all for the reply,

I tested all the wires and measured them again but could not find any possible shorts.
Then I disassemble the heat bed and measure it again and it gave me 0.8 Ohm.
So I read the post of Emadine about maybe a short on the bottom of the heat bed.

I made a picture and put some comments to make it more clearer.
I found out that the whole bottom side is full copper layered. The wholes where the cables should be mounted is layered with coper until the edge.
I removed the paint a little on the edge of point A1 and A2 to reveal the copper. I then measured for continuity on A1 and B2 and it showed .001 and on point A2 with B1 also showed .001.
This means that the whole bottom copper layer is connected to the top pads.

Then I removed the copper on the edged of A1 and A2 and measured for continuity on A1 and B2 and between A2 and B1.
Now it reads 1 so the top layer is not connected with the bottom layer anymore. The top layer is now measured at 1.6 Ohms while it was 0.8 Ohms before.

Hooked everything up and tried the M80 command and heat it to 60 degrees and it works flawlessly.

Thank you guys for helping me out here, case closed

Edited 1 time(s). Last edit at 11/11/2013 10:13AM by dvdx.

That's a plated through hole board. It's not been made the way it should be. All the information on why it's wrong (and how it's cheaper to make it wrong) is in Prusa's notes on the design.

@uncle_bob , yes it's made wrong I think it is hard to see if it's wrong made when you buy it from ebay.

Maybe there must be a warning on the wiki page of the heat bed saying becareful if you buy from ebay and it measured 0.8 Ohm.

OK this is a new one, I just tried heating up the heat bed to 60 degrees and at 42 degrees my MOSFET started to smoke ....

The heat bed is getting hot and I noticed that the LEDs on the heat bed is now not lighting up also.
Possible short somewhere else???

There is a note on the design notes page for the heated bed about the plated through hole heated bed pc boards. The whole updating the main wiki pages issue is an emotional one, you can dig into the threads about it if you wish.

I'd guess that you have another short. The MOSFET should not smoke. It should not smoke for long before either it melts or the fuse cuts out. The fuses are not very reliable in my opinion they are better than nothing at all. If the MOSFET gets to hot for to long you can burn it out.

The only shorts that will do this are "downstream" of the MOSFET. Either the pc board, the wires, or the Ramps right at the connector to the heated bed.

How long did it take to get to 42C? It should get there in about 4 minutes.

Well I did not time it but I do not think it was 4 minutes before it started to smoke.
It started to smoke after approximately 1 or 2 minutes.

Just for the info the heat bed was 30 degrees when I started the heat bed.
The room temperature is 35 degrees.

I think I go get another heat bed and replace the MOSFET.

Heat bed MOSFET that was used on my RAMPS was a P55NF06 7SAKC 6V PHL 222, I have the feeling that this is not the right MOSFET.

EDIT:
Does anybody tried this as a heat bed: [www.ebay.com]

Edited 1 time(s). Last edit at 11/12/2013 08:22AM by dvdx.

The proper part number for the FET is P55NF06. The part you have should be correct. The standard schematics show a part made by ST (that gives ST P55NF06). There are other people who make the same thing.

The eBay heater you found is a 14W part. Your heated bed needs to be a 140W part. In this case, more power does matter. Compared to the rest of the printer (or the price of 2KG of filament) the real heated bed from a reputable dealer is well worth the $20 to $30 price.

Hi, I have recently bought a prusa i3 rework kit from an official reseller listed on the reprap wiki.

I assembled it and everything was working fine for a few days. 3 days ago I noticed some smoke coming out from the connection close to Mosfet Q3.

The 2 wires going to the PCB Heatbed started to burn in proximity of the PCB Heatbed connection on the RAMP board. The thermistor connection on the RAMP board smells a bit too. Wires were hot.

I do not know what happened. Any clue? I immediatly disconnected the main plug. I cut a bit the 2 wires where burnt and reconnected them.

I disconected the RAMP board from the Arduino board to check if something was visibly burnt. But nothing seemed broken.


I tried again and the printer worked for a while.

I disconnected everything and now the wires are burning again.The PCB Heatbed led is always on and the PCB Heatbed is on all times now. I do not know how to turn it off without disconnect the plug. From the software is not working.

How to mend it? Have you any suggestion? There is a way to check if there are shorts somewere? I am not an electrician so some drawing or picture could be useful.

Thank you.