Author Topic: PSU requirement for constantan wire  (Read 4322 times)

0 Members and 1 Guest are viewing this topic.

Offline casper.bang

  • Frequent Contributor
  • **
  • Posts: 311
  • Country: dk
  • Pro SE, amateur EE.
    • BangBits
PSU requirement for constantan wire
« on: May 28, 2013, 08:03:36 pm »
As a kid, I remember my physics teacher hooking up ½ meter constantan wire to a lab power supply, and be able to cut through (melt) Styrofoam etc. I've also seen professional handyman versions, for fitting floor isolation panels the same way.

Logic would say that for ½ meter constantan wire (28?/m) at 14V, 1A will be drawn and 14W dissipated. How do I determine the temperature, so that I may calculate the power needed for melting i.e. get 100 centigrades? I assume I have to use the temperature coefficient and wire diameter (0.15mm)?
« Last Edit: May 28, 2013, 08:05:29 pm by casper.bang »
 

Offline ddavidebor

  • Super Contributor
  • ***
  • Posts: 1135
  • Country: it
    • Fermium LABS website
PSU requirement for constantan wire
« Reply #1 on: May 28, 2013, 09:38:45 pm »
Buy one meter of cable and try
Davide Bortolami,
Fermium LABS srl
 

Offline Paul Price

  • Super Contributor
  • ***
  • Posts: 1419
Re: PSU requirement for constantan wire
« Reply #2 on: May 28, 2013, 10:31:00 pm »
You may use your digital thermometer, but don't burn your fingers!

Wrap the wire upon a thermometer, slowly apply increasing current to whatever length of constantan wire you manage to have. When the thermometer reads 100 deg, you got it right. Some days, for some reason, may require some more current or some less wire.  It is funny how some things can be cold in the kitchen, yet hot in your bed. I guess there is no one blanket answer to this problem, live with it. Things are not always so easy as testing car batteries.

« Last Edit: May 29, 2013, 01:09:26 am by Paul Price »
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9584
  • Country: us
Re: PSU requirement for constantan wire
« Reply #3 on: May 28, 2013, 10:58:09 pm »
How do I determine the temperature, so that I may calculate the power needed for melting i.e. get 100 centigrades? I assume I have to use the temperature coefficient and wire diameter (0.15mm)?

The wire temperature is determined by energy balance: heat input = heat output.

The heat input is obviously V x I.

The heat output is where you will run into difficulty. Heat is lost from the wire by convection (air currents), conduction (wire contact with the material being melted), and radiation. The methods for calculating these kinds of heat loss have so many uncertainties that it is much better to do a practical experiment instead. Connect your wire to an adjustable power supply and slowly increase the voltage until you get the desired cutting/melting performance.

I'm not an EE--what am I doing here?
 

Offline codeboy2k

  • Super Contributor
  • ***
  • Posts: 1838
  • Country: ca
Re: PSU requirement for constantan wire
« Reply #4 on: May 29, 2013, 12:20:53 am »
How do I determine the temperature, so that I may calculate the power needed for melting i.e. get 100 centigrades? I assume I have to use the temperature coefficient and wire diameter (0.15mm)?
The heat output is where you will run into difficulty. Heat is lost from the wire by convection (air currents), conduction (wire contact with the material being melted), and radiation. The methods for calculating these kinds of heat loss have so many uncertainties that it is much better to do a practical experiment instead. Connect your wire to an adjustable power supply and slowly increase the voltage until you get the desired cutting/melting performance.

Ian is correct, and the equations are complex, usually involving calculus. And the power losses are your enemy..if you are trying to cut ice in a snowstorm in the arctic with such a hot wire, then it will need much, much more power (heat input) then if you are cutting foam on a still day in the Sahara Desert. 

As Ian said, it's often better just to try it and see emperically. Sometimes, however, it is useful to have at least a rough idea if you need a 10W PSU or a 1000W PSU..

So...

how much heat input (power) will be required to raise the wire to 100C, assuming no cooling and before doing any work with it (work will take energy out of the system)? This is basic physics 101 :)

 

This will give you a heat input requirement, in Joules. Remember 1 Joule is 1 Watt in 1 second, or a watt-second. You can use any combination of V x A for n seconds to get the Joules you need, as long as your wire can handle it. Of course you need to keep inputting this energy because of the losses mentioned already. If you exceed the Volts or Amp rating of your wire it will go poof :) If you don't put in more energy than your losses, it will never heat up.  If you can't get the Joules you need without exceeding some maximum of your wire (i.e. Volts or Amp ratings), then you will never get to that temperature., and you need a different wire.  In this case your probably OK, since 100C is easy for a resistance wire like constantan.

I took the density and specific heat capacity, S, of constantan from Wikipedia. r is the radius of the wire, and the area is thus the cross sectional area. delta-T is the temperature rise above ambient. So if you want 100C, you want to raise it by 75C, assuming ambient is 25C.

Because of the power losses you will need to input more energy then this equation indicates. But its a starting point. There will be thermal transfer to the surrounding air, so you need to input more energy to account for that... and when you go to use it, it will cool down because of conductive losses into the material being cut, so you need more energy for that loss as well; cutting a block of ice is different than cutting foam, its clear to see that cutting ice will require more energy input to the system.


 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9584
  • Country: us
Re: PSU requirement for constantan wire
« Reply #5 on: May 29, 2013, 12:31:28 am »
Actually, comparison with incandescent bulbs will give you a ball park figure in the 10's of watts. Depending on the length of wire and the thickness it could be as little as 5 W or as much as 25 W.

One thing I learned is that ordinary steel utility wire is not good for this. It is hard to get it thin enough and the resistance is too low. In my experiments with a 1 ft length of typical 26 AWG steel wire it needed far too much current at far too little voltage to heat it up sufficiently.
I'm not an EE--what am I doing here?
 

Offline casper.bang

  • Frequent Contributor
  • **
  • Posts: 311
  • Country: dk
  • Pro SE, amateur EE.
    • BangBits
Re: PSU requirement for constantan wire
« Reply #6 on: May 29, 2013, 06:32:25 am »
Thanks guys for taking the time to dive into this. :-+ Obviously the easiest is to just hook up the wire and gradually turn up the juice, I just wanted some theory behind since my PSU is a low-power electronics one (30W Agilent). Will report back on practical trial when I get the constantan-wire I ordered.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15117
  • Country: za
Re: PSU requirement for constantan wire
« Reply #7 on: May 29, 2013, 07:33:51 pm »
Having dealt with hot wire cutters and sealers you find that they tend to draw a large power. Typical for an impulse sealer is to draw 6-8A at 230VAC from the mains ( only for a short period though) and they use a 56VAC transformer ( running way overloaded so that the transformer is basically almost a constant current source) to give around 12V into the nichrome tape. Never measured that current, the tape goes to red hot in under 0.5 second if it is not in intimate contact via the PTFE tape with a massive cooling block.

For hot wire cutters you find that you use a 500VA or higher transformer with multiple taps, generally 12V, 24V, 36V and capable of supplying 20A. This is used for expanded polystyrene panels and blocks, for cutting higher temperature plastics the wire is a lot thicker, run at a much higher current and operated quite slowly.
 

Offline rolycat

  • Super Contributor
  • ***
  • Posts: 1090
  • Country: gb
Re: PSU requirement for constantan wire
« Reply #8 on: May 29, 2013, 10:23:54 pm »
Constantan isn't the best choice for hot wire cutting - nichrome wire would be preferable.

Nichrome is more resilient when placed under tension at high temperatures and can be heated further without softening.

Constantan is a better choice for current shunts, since it has a much lower temperature coefficient, and can be soldered easily, unlike nichrome.
 

Offline 4to20Milliamps

  • Regular Contributor
  • *
  • Posts: 248
  • Country: us
Re: PSU requirement for constantan wire
« Reply #9 on: May 29, 2013, 11:34:55 pm »
Constantan isn't the best choice for hot wire cutting - nichrome wire would be preferable.

Nichrome is more resilient when placed under tension at high temperatures and can be heated further without softening.

Constantan is a better choice for current shunts, since it has a much lower temperature coefficient, and can be soldered easily, unlike nichrome.

I aqree, and most commercial foam cutters I've seen use a variac.


http://en.wikipedia.org/wiki/Chromel








 

Offline jmacqueen

  • Contributor
  • Posts: 22
Re: PSU requirement for constantan wire
« Reply #10 on: May 31, 2013, 05:55:15 am »
Yup nichrome or kanthal wire would do it. I would go look at the manufacturers data sheets on the wire, they will likely have the data you need to calculate power needed for heating the wire gauge you want at the length you want.

Here's a start at this link for a kanthal products handbook..http://www.kanthalpalmcoast.com/C12570A7004E2D46/062CC3B124D69A8EC1256988002A3D76/D1D355F37C940491C12572B9003FD970/$file/1-A-3-3%20precision%20wire.pdf?OpenElement

 I was looking through this for a project I am working on, 20w through about 2" of light gauge kanthal in my case, it gets quite hot enough to cut things but I am using it to heat a liquid.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf