Hi,

I'm trying to solve a problem I am having with a small project. Basically I am looking to heat a small length of stainless steel tube approximately 3 inches long to a temperature of approximately 500 degrees F (0.194" ID x 0.25" OD). My main concern is doing it simply and, as a result, I was am trying to heat it using "impedence heating" i.e. running a current through the pipe and having the pipe act as the heating element itself. The reason this would work well is because i need the INTERIOR of the pipe to be hot to vaporize a liquid. The pipe itself is electrically isolated and I am using a variable AC transformer but as soon as I approach approximately 1 volt and the pipe becomes "warm" I blow the inline 15A fuse I incorporated. The entire circuit wiring becomes warm as well (12 gauge wiring). I understand the problem I am having is because I am basically just making a short circuit due to the extremely low resistance of the pipe. I understand they use do this on the industrial scale but can it be done on a small scale? Can I use "dummy loads" along the circuit so as to not cause a short circuit? Not sure if I'm on the right track and a little confused about how I can do this if I can... I'm hoping I can because it really would work beautiful if possible.

Any help/suggestions would be appreciated....

 

I'm a total beginner at this so take what I say with a grain - or whole shaker - of salt, but wouldn't adding a dummy load just cause that load to get hot at the expense of the pipe not getting as hot? I would (again, naively) think that you would need a higher resistance pipe. Would putting a heating element inside the pipe work instead of electrifying the whole pipe?

 

I'm a total beginner at this so take what I say with a grain - or whole shaker - of salt, but wouldn't adding a dummy load just cause that load to get hot at the expense of the pipe not getting as hot? I would (again, naively) think that you would need a higher resistance pipe. Would putting a heating element inside the pipe work instead of electrifying the whole pipe?

I agree having a higher resistance pipe would probably do it, problem is it is so small there's no room for mass (increased resistance) and I can't think of a conductive metal with a high enough resistance to make it work....

 

Is you liquid electrically conductive?

 

Do the right thing and use a Band Heater.
You can find lots of surplus units online for cheap.
http://www.omega.com/heaters/pdf/HEATER_INTRO_BAND_REF.pdf

The resistance of that pipe is far too low for a heater element. It's almost a direct short for a normal 110 circuit.

 

Is you liquid electrically conductive?

Nope, it is mineral oil actually. The only other option I can think is having a non-conductive tube, wrapping it with nichrome and potting the exterior of the tube (and nichrome) with something like a liquid ceramic. Problem with that is A) it is more work than I'd like to do and B) what would a good medium that is a hollow tube of that size that is non-electrically conductive but still a good thermal conductor. hmmmmm....

 

Yes, this is quite obvious. Adding an external resistor will just give you more heat elsewhere, unless you can fix the resistor onto the pipe so that both heat together, in which case you might as well use the external resistor for all the heating, if you can get or make one that will take the heat.

To make this sort of scheme work with just he very low resistance of the pipe, a special kind of transformer with a secondary winding capable of very big current output would be needed. I think that in some such schemes the pipe would actually form part of the transformer secondary, lying in the magnetic field.

Your problem is basically that the power P watts dissipated in a resistance of R ohms by a current of I amps is equal to P = R * I squared.

You require a large value of P to give the necessary heat. The resistance R of the pipe must be very small, thus I needs to be large.

My guess is that this might need from some tens of Amperes upwards, but don't really have much idea of the magnitudes involved. You might start by getting a value for the power input required from the heat capacity of the ( gas? ) going through the pipe and the flow rate. You probably need to factor in a bit more for losses. Then work out the pipe resistance from its dimesions, and turn the handle, and probably get a bit depressed about the answer.

Good luck!

 

Most impedance heating systems work in the 1 to 30V AC range.. Lengths under 3 ft are typically impractical because the resistance is so low like you have found..
Time to move on to a different heat source.

 

Do the right thing and use a Band Heater.
You can find lots of surplus units online for cheap.
http://www.omega.com/heaters/pdf/HEATER_INTRO_BAND_REF.pdf

The resistance of that pipe is far too low for a heater element. It's almost a direct short for a normal 110 circuit.

Unfortunately that won't work, the pipe is only 1/4" OD. I was just wondering if I could add resistance, either before or after the pipe (or both) and if that would acheive the effect...

 

Most impedance heating systems work in the 1 to 30V AC range.. Lengths under 3 ft are typically impractical because the resistance is so low like you have found..
Time to move on to a different heat source.

That's what I was thinking, but maybe didn't want to face the facts. i figured the large mass and length involved make it work and on such a small scale you really can't do it......

 

Yes, this is quite obvious. Adding an external resistor will just give you more heat elsewhere, unless you can fix the resistor onto the pipe so that both heat together, in which case you might as well use the external resistor for all the heating, if you can get or make one that will take the heat.

To make this sort of scheme work with just he very low resistance of the pipe, a special kind of transformer with a secondary winding capable of very big current output would be needed. I think that in some such schemes the pipe would actually form part of the transformer secondary, lying in the magnetic field.

Your problem is basically that the power P watts dissipated in a resistance of R ohms by a current of I amps is equal to P = R * I squared.

You require a large value of P to give the necessary heat. The resistance R of the pipe must be very small, thus I needs to be large.

My guess is that this might need from some tens of Amperes upwards, but don't really have much idea of the magnitudes involved. You might start by getting a value for the power input required from the heat capacity of the ( gas? ) going through the pipe and the flow rate. You probably need to factor in a bit more for losses. Then work out the pipe resistance from its dimesions, and turn the handle, and probably get a bit depressed about the answer.

Good luck!

I am thinking of going the nichrome route and wrapping a non-conductor with it and potting that. Maybe pyrex?.....although if you have any suggestions as to tubing I could use that wont short the nichrome out that would be great.....

 

OOps, just spotted a possible safety hazard. Most variable AC transformers do not give isolation from the mains. For safety (and to permit discussion on this site), this is really a requirement.

In any case, you are going to need some kind of voltage step-down, current step-up transformer to get enough current to heat a pipe if you go on with this method. The transformer used should then be of the double - wound kind for safety isolation.

 

I think the likes of soldering iron elements use things like thin sheets of mica for insulation. There are also various woven ceramic or fibreglass materials used for high temperature insulation, but maybe your application is too hot for glass?

 

This can be done, but the last time I did it I had a transformer factory at my disposal. Without that, it is nearly impossible to find high current step down transformers in the 1 volt range! Perhaps a 1200 watt microwave oven transformer connected backwards?

I like the idea of wrapping nichrome on an insulation layer, on the tube.
I'm afraid of using glass for the tube because it might crack when you add the oil.
I have used hollow ceramic resistors for furnaces. You would put the stainless steel tube completely inside the resistor. Next problem: buying a ceramic resistor that can survive 500 F.

I haven't really answered exactly how to do it. Just some ideas.

I remember a TV show called, "Breaking Bad" where Bryan Cranston ordered a "50 millimeter furnace". Possibly your answer lies in the Chemistry department?

 

Unfortunately that won't work, the pipe is only 1/4" OD. I was just wondering if I could add resistance, either before or after the pipe (or both) and if that would acheive the effect...

That is tiny. We would bend the tubing on top of a small form (section of pipe) and clamp the heater to that or use a clamp-on split heater block.

 

This can be done, but the last time I did it I had a transformer factory at my disposal. Without that, it is nearly impossible to find high current step down transformers in the 1 volt range! Perhaps a 1200 watt microwave oven transformer connected backwards?

I like the idea of wrapping nichrome on an insulation layer, on the tube.
I'm afraid of using glass for the tube because it might crack when you add the oil.
I have used hollow ceramic resistors for furnaces. You would put the stainless steel tube completely inside the resistor. Next problem: buying a ceramic resistor that can survive 500 F.

I haven't really answered exactly how to do it. Just some ideas.

I remember a TV show called, "Breaking Bad" where Bryan Cranston ordered a "50 millimeter furnace". Possibly your answer lies in the Chemistry department?

i'm looking into a machineable hi-temp coating/paste (something like Lab Metal) i can use, then turning it down on my lathe after it cures so as to have a thin smooth, non-conductive layer surrounding a thin wall metal core then wrapping with nichrome and then potting the resistance wire. little more work than I wanted, but I KNOW for a fact that can work.

 

If you could get by with a 1/16 ID, the heating element from a solder sucker would possibly work for you. They usually operate on 24V AC and have a feedback device for temperature control built in.

 

Responding to post #16: The only thing I would recommend is that you realize a very thin insulation layer is not important. The only thing a thick layer will affect is the time constant of the device. Heat, properly insulated, will arrive at the temperature you want. The only thing left to consider is, "How fast does it have to be?" This usually is not a problem.

 

Responding to post #16: The only thing I would recommend is that you realize a very thin insulation layer is not important. The only thing a thick layer will affect is the time constant of the device. Heat, properly insulated, will arrive at the temperature you want. The only thing left to consider is, "How fast does it have to be?" This usually is not a problem.

If the mineral oil is flowing, then the thickness and thermal conductivity of the insulation does matter.

 

Why not 'piggy-back' the sample tube to the heating element of a soldering iron?

Or even use a soldering gun like a Weller style to provide the current to heat the sample tube. In place of the soldering tip on the gun, wire the ends of the tube. The soldering gun transformer is made for that type of use and the stainless tube probably has more resistance than the copper of the tip.