Hot wire

Thread Starter

cmartinez

Joined Jan 17, 2007
8,218
Hi,

I'm new to this forum, I know some digital electronics and I'm rather good at programming 8051 microcontrollers using assembly language, I'll be more than happy to help anyone if it's within my power. However, my knowledge of analog electronics leaves much to be desired.

My problem is this:
I want to heat a Nickel-Chrome wire to a temperature of around 750°C and keep that temperature constant as the wire is being used in a cutting application.

To do that, I only have two options:
1.- Monitor the wire's temp using an IR thermometer connected to a temperature controller. (rather expensive).
2.- Monitor the wire's resistance value (it increases as the temp goes up, check this page: http://www.heatersplus.com/nichrome.htm) and adjust the voltage accordingly (AC voltage) using a solid-state variac with an analog input until it reaches the desired resistive value, thereby also reaching the target temperature.

I have a hunch this has something to do with differential amplifiers, but how do I measure the wire's resistance while it's being being heated by an AC source? How would the circuit look like?

Any suggestions? or would some other approach be more practical?

More data on Nichrome 60 wire:
http://www.pelicanwire.com/alloyproperties.htm#n60
http://www.resistancewire.com/Html/Technical/CurrentTemp/CurrentTemp.php

Thanks
 

beenthere

Joined Apr 20, 2004
15,819
Hi,

You're probably pushing a few amps through the wire. If you place a big .1 ohm resistor (50 watts or so) with one end at ground, then the voltage across the resistor will give you the circuit current (E = IR). Even if AC, you can rectify and change it to DC. Then you can scale it and display however you wish. You should be able to use an ammeter to calibrate the measuring circuit, and then convert to temperature with the nichrome resistance & temperature chart.
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,218
My intention is to power it with AC, but I'm more than willing to pay the price for DC power if that's the only way around it.
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,218
Hi,

You're probably pushing a few amps through the wire. If you place a big .1 ohm resistor (50 watts or so) with one end at ground, then the voltage across the resistor will give you the circuit current (E = IR). Even if AC, you can rectify and change it to DC. Then you can scale it and display however you wish. You should be able to use an ammeter to calibrate the measuring circuit, and then convert to temperature with the nichrome resistance & temperature chart.
I have to push around 17amps through the wire to get the temperature I need, the length of the wire I need to heat to 750°C is around 1 inch, and assuming that I'll be using an #18 gage wire (0.040" diam), it's resistance will be 0.0356 ohms at room temperature, increasing to 0.0398 ohms at the target temperature, which means only 11.85% change in value!
 

beenthere

Joined Apr 20, 2004
15,819
Hi,

The resistor method isn't practical here - it would have to dissipate as much as the nichrome wire. The small change in resistance is not an issue as a couple of op amps could expand that up to many times as great.
 

n9352527

Joined Oct 14, 2005
1,198
Why don't you just measure the voltage across the wire and with a calibrated room temperature resistance you could arive at the set point voltage for the intended temperature.
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,218
Why don't you just measure the voltage across the wire and with a calibrated room temperature resistance you could arive at the set point voltage for the intended temperature.
Thanks for the suggestion, but my intention is not only monitoring the wire's temp, but also to keep it constant as it cuts through a material. That would require a circuit that could increase or decrease the current being fed to the wire as it goes.
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,218
Thanks Beenthere and n9352527, I've studied the file pointed at by thingmaker3:
http://www.circuitcellar.com/library/print/0606/Malik191/Malik-191.pdf
and I believe he's on the right track.

I'm sure I can substitute the Power Switch circuit shown in Figure 4 of the article using a simple industrial solid-state relay with the appropriate capacity.

My questions at this point would be:
- how do I connect the wire to be heated to the circuit provided by the articles?
- Do I just connect the wire to poles 1 and 2 of J1 as shown in Fig 4, or will I need a third wire and a resistor, as shown in Fig 2?
- Will it work for such a high current and low resistance?

You guys are great, thanks for all the help.
 

thingmaker3

Joined May 16, 2005
5,083
R2 of Fig 1a is the wire to be heated. It goes between J1-1 and J1-3 of Fig 4.
R4 of Fig 1a goes between J1-2 and J-3.

For R4, you can get 0.06ohm 35watt resistors via eBay. Two in parallel should let you balance the bridge.



Shoot, now that I think of it, you might just put a few of the 35watt resistors together in series with the nichrome. You'll want about half a volt across the nichrome, yes? Then you simply need to adjust voltage across the stackwith a resolution of a millivolt or so across the nichrome...
 

Thread Starter

cmartinez

Joined Jan 17, 2007
8,218
R2 of Fig 1a is the wire to be heated. It goes between J1-1 and J1-3 of Fig 4.
R4 of Fig 1a goes between J1-2 and J-3.
For R4, you can get 0.06ohm 35watt resistors via eBay. Two in parallel should let you balance the bridge.
Shoot, now that I think of it, you might just put a few of the 35watt resistors together in series with the nichrome. You'll want about half a volt across the nichrome, yes? Then you simply need to adjust voltage across the stackwith a resolution of a millivolt or so across the nichrome...
Ok, please pardon my kindergarten level, but I need to verify calculations.
First, I need to push around 17 amps through one inch of wire, equivalent to 0.0398 ohms at working temperature, in order to get the desired (750 °C) temperature.
That leaves us with V=IR, V=0.677 Volts, which is in agreement with what you're suggesting.
Next, assuming that I'll be using a 24V power supply I place a resistor in series with the nichrome to guarantee the appropriate current flow. So I'm left with V=I*(R1+R2), in which R1 is the heating element (0.0398 ohms) and R2 the resistor in series I'll be using, in this case, V is 24V and I remains at 17amps.
Solving for R2, R2=(V/I)-R1, R2 is 1.372 ohms.
I the real application, I'll be needing more than 17 amps, since the wire will loose temperature as it cuts (17 amps is what I need to heat the wire in free air), thereby consuming more current. So let's say that I'll be consuming close to 50 amps constantly at peak times (worst case scenario)

What should I do?
 

thingmaker3

Joined May 16, 2005
5,083
Consider: 50Ax50Ax1.324ohm is well over three kilowatts. I've never seen a 3.5 kilowatt resistor. (I've never seen Paris either, so 3.5kW resistors might indeed exist.)

If the wire's resistance changes by less than 12% over temperature range, then current will also change by less than 12% over temperature range. 19 or 20 amps should be plenty. That still works out to more than half a KW. Beenthere is right (as ususal) - resistor in series may not be practical. At least not with your big 24v transformer...

One foam-cutting lathe used a home-wound secondary to get the voltage down to desired value. Control is provided by a triac "dimmer" feeding the primary.

Another thing to consider would be using longer wire. Most of the foam cutters I've seen use eight or ten inches of wire and run at around five watts. Your application may require otherwise, of course.
 

thingmaker3

Joined May 16, 2005
5,083
Um. Wow. 17A through the AWG18 wire yields over 1K centigrade. For 750C the current only needs to be 10A or so.

By the way - what is being cut here? Foam only needs 500C - and that hot only when self-cleaning the cutter.
 
Top