Sciencemadness Discussion Board

heat tape

Magpie - 16-2-2005 at 09:50

Does anyone else have a need for a heat tape? I thought it would be useful to control the temperature of my fractional distillation column.

I've been wanting to construct a homemade heat tape that could supply variable power using a light dimmer. I'm wondering if you could tear apart an old toaster from Goodwill and use the nichrome heating element. For insulation of the wire I'm thinking woven fiberglass. Any suggestions or better ideas?

[Edited on 16-2-2005 by Magpie]

Twospoons - 16-2-2005 at 13:04

I suppose it depends on what temperature you want to operate at. Silicone is probably good for 200C, or teflon maybe. Actually there should be a small amount of fibreglass sleeving in the toaster. You can probably get more at an electricians supply shop.

neutrino - 16-2-2005 at 14:10

If you need nichrome wire and fiberglass tape, you can get them cheaply on ebay.

bio2 - 16-2-2005 at 14:16

Regular fiberglass (woven) will burn up in contact with the nichrome. Zetex fiberglass tape is ok or ceramic cloth. The toaster or hair dryer elements are way too much wattage for any normal sized column. One or two hundred watts is more than you need! Look for a post I made recently giveing a link to a coil calculator. The wire temp should not exceed about 450deg C at your operating amperage. The nichrome is also very brittle and for this application you need very fine wire or it will not flex properly. It's not so easy to make one of these. I took a heating cord out of a 60 watt warmer plate and it is more than adequate for what you describe and is very flexible.

Geomancer - 16-2-2005 at 16:59

Why won't steel work for low temperature stuff?

I'd forget about the toaster wire. Once I scavanged some; it was in the form of tape, not only brittle but prone to kinks if you don't handle it just so. I tried to weave it into a piece of fiberglass cloth; the operation was not successfull.

Twospoons - 16-2-2005 at 17:35

Just occurred to me that for lower temps you could scavenge the wire from an electric blanket. It will already be in a water-proof jacket, and be flexible.

Magpie - 16-2-2005 at 21:28

Geomancer: I have some 0.8mm diameter wire that I think is stainless steel. It has a resistance of 1.44 ohm/meter. So to get 30w with 1 meter I calculate that 6.6v is required. I would need a 30w step-down transformer and a light dimmer.

You can, of course, buy ready made heat tapes that give 42w/6 feet for water pipe heating. You can also buy just the tape by the foot (3w/foot). They are waterproof but a little bulky and stiff.

Twospoons: I think your idea has possibilities. I like that the electric blanket would already have its own variable power supply.

Quince - 17-2-2005 at 14:28

Uhm, does your calculation take into consideration that resistance is a function of temperature? For example, resistance of a light bulb filament jumps by a good factor when it's heated to a glow.

Magpie - 17-2-2005 at 15:18

Quince: No that calculation was prelminary and did not take resistance rise with temperture into account.

I did some scouting for materials today and found some likely candidates: door bell transformer (120/12.6vac, 37.8w, $10), light dimmer ($5), and fiberglass sleeve (1/16" ID, 700C, 5' long, $7.50) for "high temp" wire. BTW, this last item took a bit of searching in my small metropolis. You have to find a crusty old store with a crusty old owner that is in a timewarp and has a dark dingy old shop with all kinds of old goodies in it. :D These are becoming very rare unfortunately.

I will do some more calculations tonight taking into account the increase in resistance with temperature.

[Edited on 17-2-2005 by Magpie]

Quince - 17-2-2005 at 15:26

Some dude on eBay is selling the Zetex fiberglass loom tubing. I bought some, as well as 18 AWG nichrome. I plan to make the mantle temperature probe-controlled using a plain thermistor as a sensor.

Any ideas as to whether a magnetic stirrer will disturb the nicrhome heater coils due to the nickel content (nickel is ferromagnetic)? Constant vibration might eventually break them.

[Edited on 17-2-2005 by Quince]

Magpie - 17-2-2005 at 19:46

A rigorous heat transfer analysis would be too much work so I'm going to estimate that my wire will reach a maximum temperature on the low end of cherry red, i.e., 600 deg C. Assuming a resistance/temperature coefficient of 0.005/deg C (handbook value for steel) the resistance would increase by a factor of roughly (0.005/C)600C = 3. Since P=E^2/R my power would drop from 30 watts max at room temperature to 10 watts max at 600C. This would probably suffice for a 200 mm column with fiberglass insulation for most fractional distillations I'm guessing.

With the 12.6v transformer I have to use 2.9m for 30w at room temp. As my Hempel fractionating column is 30mm OD and the insulated wire is 3 mm dia I could accomodate [(200mm)/3mm](pi)(30mm)/(1000mm/m) = 6.4 m, wound helically. So there will be room for all 2.9m of winding.

I would appreciate any comments on this design, especially from those with experience. This would be nice to have prior to investing in materials which will come to ~ $30 plus tax.

Tacho - 18-2-2005 at 03:01

Magpie, take a look at

This has been working wonderfully for me for years. Too much heat (red hot) may really melt the glass fiber, but this hasn't been a problem for me.

A low temperature heating element that looks more like a tape may be done putting resistors in parallel on flexible wire.

Like this:

__________________flexible wire
.|__|___|__|___|__flexible wire

I have never done this, but I did something similar HERE

I don't know your toaster, but I found that home appliances usually use flat NiCr wire. This is cheaper per watt, but quite delicate for constant bending.

EDIT: Just for the record: this may sound funny, but you know where I found a thick NiCr wire bent in zig-zag (forming a tape)? Cotton candy machine's heating element! I don't know if all machines have the same kind of heater, but I thoght I should mention it. Over 3000W though...

[Edited on 18-2-2005 by Tacho]

Magpie - 18-2-2005 at 09:49

Thanks Tacho! I remember this thread now. How could I ever forget that inverted iron masterpiece! And I apologize for starting a new thread. Somehow my search using "heat tape" did not pull this up.

I think I have enough information and resources now to construct a decent heat tape. But ... I'm having 2nd thoughts about whether this will work for my intended purpose, i.e., regulate the temperature of a fractional distillation column. I think this is a tricky problem as efficient fractional distillation depends on liquid/vapor equilibrium over the length of the column. Since the concentrations of both liquid and vapor are changing continuously from pot to still head there is no one temperature to control to. For an indication of how this control has been done (or attempted) by others see Vogel, 3rd edition. So I will try 1st to just insulate the column very well. If that still (no pun intended) doesn't work I may end up trying this heat tape idea.

Quince - 18-2-2005 at 17:15

Has anyone tested this with a magnetic stirrer? Do the nicrhome wires wiggle from the varying field?

bio2 - 18-2-2005 at 18:53

Go to the helix calculator I posted. It will do your "rigorous calculation" in one click. 600deg is way to much for a fractionating column. Use some common sense here. Mine works fine w/ a dimmer control and a thermistor to monitor for manual control or just stick a thermometer under the lagging. 18 wire is way too big for any 120v set up. If you go low volts then put the dimmer on the transformer primary.

A 60 watt 18" heating cord from a warmer plate works fine for me. If your intent on DIY use 30 wire or something if you expect it to flex. Best is smallest you can handle in parrallel if necessary. This is how the commercial tapes are made, woven into the fiberglass. My home made versions woven into glass didn't last long for the obvious reasons already mentioned but it certainly is doable but not with big wire iimho.

No need to reinvent the wheel here.

The coil will vibrate a little and buzz pretty loudly at lower phase angles if in close proximity to the magnet. If properly secured it doesn't seem to damage the wire at least after maybe 100hrs use so far.

Quince - 18-2-2005 at 23:37

I also have 36 gauge wire, but the stupid program won't let me specify an arbitrary diameter, it just allows selection of the gauges already listed there.

Magpie - 19-2-2005 at 08:52

Thanks Bio2 for the recommendations and the results of your experience with DIY. In my experience DIY is rarely more economical or better, and certainly not more elegant. But on the fun scale and the instructive scale it can be very good. I don't know how dimmers work -wave chopping, potentionmeter? so I will have to look into that. You say that cheap ones are poor for this application - so that was good to learn.

Yes my stab at 600C for a coil temperature was not a good one. But I thought it instructive to see just how much power is lost when a wire heats to that temperature.

The coil calculator is very nice. Using my 20 guage wire with 12.6v I get an insulated temperature in the low 200C range. I don't know what kind of wire it is exactly but I'm going to try and find out. It looks like stainless steel and is very tough and flexible, and I have at least a hundred meters of it

Quince - 8-3-2005 at 13:29

Since I got no reply, I'm trying again. I have 36 AWG wire, which is not something I can input in that program. Thus, I'd like to know how to do the calculation manually.

I know that with this thin wire I have to do it in parallel; and it's too soft and won't stay in a spiral, so I have to put it in a needle and thread through the fiberglass. But it's all I've got, except the 18 AWG, which I'll probably use for a furnace (I'd go for an arc furnace for it can get much hotter, but when I tried a prototype a while ago there was too much ozone and NOx, and my graphite electrodes didn't last).

By the way, the Zetex fiberglass tubing is impregnated with some sort of plastic, I think acrylic. I wonder if I should burn it off beforehand.

[Edited on 8-3-2005 by Quince]