Friday, May 22, 2015

Prusa i3 Rework- MOSFET Repair, Part 3

Now that the modifications to the original RAMPS 1.4 board and the MOSFET breakout board were complete, it was time to build the cable harness that would connect the two parts. This was no ordinary cable harness. Off-the-shelf cables for 3-pin 0.1" headers are not rated for the current needed in this operation. Here is how we know:

- Wire capacity for electrical power is determined by the current flowing through the wire.

- Current flowing through the wire can be determined by Ohm's Law: Voltage is equal to current times resistance, or V = IR.

- We know that 12V should be supplied to the RAMPS board. MOSFETs are basically switches for electrical power, so 12V should also be supplied to the heated bed and the ceramic heating cartridge (source: RepRap Wiki). I checked, and I supply 12V to those circuits on my printer.

- The resistance in each of the three MOSFET circuits can be determined by the resistance of the heating element.

- For two of the circuits (the hotend circuits), the heating element is a ceramic heater cartridge. It has a resistance of approximately 3 ohms (source: RepRapPro). My ceramic heating cartridge also has a resistance of 3 ohms.

- For the other circuit (the heatbed circuit), the heating element is the MK2B PCB Heatbed, with a resistance of 1.0 to 1.2 ohms (source: RepRap Wiki). My heatbed has a resistance of 1.2 ohms.

- Back to V = IR. We're solving for current, so let's rearrange the equation to get I = V/R.

- First, let's find the amperage traveling through the wires for the hotend circuits: 

V = 12 volts
R = 3 ohms

I = V/R

I = 12 volts/3 ohms
I = 4 amps

- Next, let's find the amperage traveling through the wires for the heatbed circuit:

V = 12 volts
R = 1.2 ohms

I = V/R

I = 12 volts/1.2 ohms
I = 10 amps

Consulting a wire gauge chart allows me to figure out how thick I need to make the wire to handle a maximum of 10 amps. 14 AWG wire that has 7-24 cores seems like a good fit.

But will off-the shelf 3-pin extensions work? According to Pololu's specification for their 3-pin servo jumpers, the wire size is 22 AWG. Not even a solid core 22 AWG wire can handle 10 amps. So the off-the shelf 3-pin extension will not work. Therefore, I would pick out my own 14-gauge wire for the harnesses. Now onto the build!




First, I cut the 3-pin connectors for the cables. Next time, I would definitely use crimp connectors. I didn't at this point because I didn't feel like spending a couple more days without my 3D printer, stuck waiting for a crimping tool.


One of the connectors, ready to be soldered onto the wires.


Here are the nine 14-gauge wires that make up the harness.


I was a bit concerned about melting the plastic on the male headers with the heat from the soldering iron while getting them soldered to the wires. So I put the male headers in female headers, in order to stabilize them.


Getting the wires soldered to the headers was quite simple.


Another shot of the soldering process.


One male header, completely soldered on to the wires.


I put heatshrink tubing on each lead so that they won't short out when they touch.


The heatshrink tubing, now shrunk.


One of the female connectors. Heatshrink tubing tends to throw people off when they realize that they can't get it on a closed wire after they solder it. Fortunately, I didn't forget that.


One of the harnesses, with a ziptie on it to keep it from getting tangled up.


The breakout board is finally connected to the RAMPS board. I actually ended up marking one of the leads on each harness with a Sharpie, to make sure that it was not lined up in reverse on one of the boards.


A closeup of the breakout board. Looking good!


And we have success! At last, my 3D printer was working again.

Everything was working perfectly... but a couple minutes later, I realized that I should check the max amperage for the pin headers. It turns out that it's 3 amps. However, I decided that I would test and see if the 3-pin headers would stay below their max operating temperature (105 degrees Celsius) when in operation. They have never gone over 60 degrees Celsius, even on 100 degree bed heating, so I'm not worrying about the extra current for now. Of course, I would never do this on any sort of production build, but it will work for the next couple months, until I find a better fix.