I have finished my Arduino based pulse width modulated (PWM) hot wire cutter controller. It allows precise USB or analog temperature control of a hot wire of length from 10cm to 2m for use with manual cutters or CNC based cutter. The PSU is extermal (battery or PSU) The 12v power is attached to the GND and + 12V terminals. The + to the hotwire is attached on the red terminal on the top. The current return thru the FET is connected to the black return terminal. An FTDI board can be attached to the Arduino thru the "programming window".

What do you think :) ?

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • By the way if you want to have a cheap power supply tip: buy 12V 7-15Ah lead-cell battery from a hardware store. A small one will do. A lipo will also work. They are cheap and also possible to transport outside (even to the field). I prefer lead-cell battery as LiPos present a fire risk.


  • Correction! Duty cycle is a number between 0 and 0,999
  • H!


    I took a review look on the datasheet of IRFZ44N and the threshold voltage is 2-4 volts. (The treshold voltage is the voltage where the FET starts to conduct and the depletion barrier is broken in the channel). To ensure full opening and low heating of the FET DS path, you should probably be better off with 12V on the gate so you are sure the channel is fully open (the FET is fully turned on).To be well above the treshold voltage is a good way to ensure that.


    Max VGS is 20 volts so 12V should be ok with a good safety margin.5V is probably too low (too close to the treshold voltage)


    54" should be OK depending on the thickness of the wire you use. I use nichrome wire as some foam plastics turn corrosive when melted. The nichrome wire withstans that corrosive effect pretty well. You can find resistance values per meter on the suppliers datasheet.Then you can calculate the power given off in the wire (approx) like this:


      Pwire=(12V * 12V * duty cycle) / (Resistance per inch * length in inches)


    Duty cycle is a number from 0-999 based on the PWM settings controlled by the analog pot meter.

    (I have not considered the Rds resistance in the FET that is in the order of 17,5 E-3 ohms)


    A couple of cautions:

    Do NOT breathe the fumes that the foam gives off when melted. That can cause serious damage to your internal organs. I use my setup outside and have a strong fan sucking (not blowing) off fumes.


    Do insert a circuit breaker that can handle several amps DC in series with the wire. This should be inserted on the positive side of the wire. In case the PWM shorts out (damaged FET) or if the software crashes and turns on the FET inadvrently or if someone grounds the cold side of the wire or simular faults- you don't want the battery connected directly without a way of turning off the current.That could easily start a fire.


    Also, be very careful as this equipment may cause fire hazards if not used and watched properly. If the wire gets red hot and breaks and a piece of wire falls on flammable material, you can easily start a house fire. Also, do not have the unit connected unattended.


    All use on your own risk.


    It will be cool to check out your eagle layout. Give any credits to UFO_MAN  :) 

  • on the FET's... it looks like the IRFZ44N is good for up to 55V?  So going above 12V shouldn't be an issue?
  • Maybe 12V won't be enough for this long of a wire?
  • I can add the resistor to the collector path on T1.  I'll prototype it first.  Yes I will release the eagle files when I have something working.


    For the 12V power supply for the hot wire...I am probable going to have at most 54" of hot wire.  How many amps do you think I'll need at 12V for a wire of this length?  Just looking for a 12V supply on ebay now...

  • Hi Brendin. That should work (I assume IC1 / IC2 are the FETs. Transistors are normally named Q1, Q2 and so on but I guess that is a technicality of your CAD package).


    What you need to do is to insert a series resistor in the collector path of T1. In case you don't have a current limiting device there, T1 may draw excessive collector current and damage may occur (and a little smoke haha) if base current is large enough.  R1 is necessary to limit base current but should be low enough ohmic value to saturate T1. Check the gate voltage of the FETS so you dont over and under drive them. You could either connect VCC to +5V or to +12V if you need higher FET gate drive. If you make a schematic and a board in Eagle, please publish it here. More people may need a hot wire controller.


    If you wonder why it is called "Ringnes Pils" controller, it is because that is a Norwegian beer brand.

  • oh and obviously the neg side of the 12V supply will be connected to ground.  The connection to the left of the resistor is attached to an I/O pin on the atmega 48.
  • How about this?

    VCC= +5V

    T1= 2n2222a

    R1= ~1K



     The hot wire would be connected between the outs from IC1,IC3 to +12V.





  • Hi Brendin! Thanks for commenting. An Eagle Layout would be nice. However be careful when milling fiberglass substrate. Dont breathe the particles from a milling operation as they are bad for your lungs.

    What I did in the last version was to actually replace the one FET with two in paralell. I just "piggybacked" one fet on top of the other (pin for pin). If you find a FET that can handle more current you probably only need one. The driver chip (EL7212) does have more than one output so you can use two outputs. Then you need to wire the pwm signal from the uC separately to two EL7212 inputs and then those outputs will need to be routed to the gates of each fet. I am not even sure you need the driver chip. An emitter follower and a resistor to ground probably would do the job. The reason EL7212 chips are used is that the gate input has some capacitance. This can reduce the swithcing speed. But in this application the switching speed is low so it is hardly a problem. I reused a circuit I used to experiment with ultrasound some time ago.

This reply was deleted.