Sciencemadness Discussion Board

Building a Heating Element?

elementcollector1 - 30-4-2016 at 19:45

I've recently been thinking about building a high-temperature nichrome heating element for those reactions that require elevated temperatures, but I'm in way over my head here. I have no idea how to start calculating what gauge of wire I need, or what amperage to apply for a given temperature, or how to insulate it from a metal container so it doesn't short out, etc.,etc.

Basically, my endgoal is to take one of these copper air chambers and combine it with the nichrome wire somehow to make a heating apparatus. There would probably be some amount of insulation in between the nichrome and the copper, but I don't know what it would be made of or how best to apply it. Same goes for the outside of the nichrome wire. Thoughts?

Magpie - 30-4-2016 at 20:27

Start with how many watts you want the device to have. This can be expressed as power, P. From Ohm's law P=EI =E2/R. R=E2/P.

So, for example assume E=120v and P =100w

Then R=(120)2/100 = 144 ohms

Then determine what length of nichrome wire you will need to provide 144 ohms of resistance. You can easily determine the ohmic resistance of your wire with a volt-ohm meter at room temperature. But you really need to measure this resistance when the wire is at full power, ie, when it is orange hot.

elementcollector1 - 30-4-2016 at 22:10

Hmm. I was thinking of powering this off of a 12V power supply, making the resistance 1.44 ohms instead. 22 AWG seems like a nice gauge to use, and from the formula R = pL/A (R=resistance, p=resistivity, L= length, A = cross-sectional area) I find that the length of 22-gauge wire required for this resistance is about 31.38 cm or 12.35 inches, which is reasonable to coil around a roughly 3/4" diameter (from memory) copper tube.

But how do I know the temperature this would output at 100W of power? Is there any way to calculate that in general, or are more specific conditions required?

Also, what would be a good insulation to use for the inner space between the copper and the nichrome? Preferably, it'd be electrical insulation only, and still conduct as much of the heat as possible.

The outer insulation would have to be something that keeps as much of the heat in as possible. Would concrete be sufficient for this task?

JJay - 1-5-2016 at 02:55

There are charts that relate temperature to amperage for nichrome. While I believe it is possible to calculate approximate amperage and wattage using such charts, the temperature charts are really only suitable as a rough heuristic; you'll need temperature control or a lot of experimentation to determine exact conditions to reach a given temperature in your application.

Here are some charts: https://www.heatersplus.com/nichrome.html

You'll probably end up using some kind of ceramic for electrical insulation. I don't think concrete would put up well with high temperatures (it contains a lot of water). Plaster of Paris might work ok but will probably crack a lot until it gives off nearly all of its water. Some kind of fired clay would work well I think, and more exotic ceramics are available. Quartz and fused alumina could also work well but are expensive due to the high temperatures required for working them.

WGTR - 1-5-2016 at 05:16

There are a number of different ways to go about this. I've made a couple of different configurations myself. One is below:

https://www.sciencemadness.org/whisper/viewthread.php?tid=26...

Another is here:

https://www.sciencemadness.org/whisper/viewthread.php?tid=55...

I like to keep my designs straightforward and repeatable when possible, even if some of them end up costing a little more up-front. The first design could easily be adapted to what you want simply by drilling a hole in the cover to insert your copper tube. An added bonus is that you can also use it to fire small ceramic objects, or to otherwise heat objects that won't fit into a tube, simply by changing out the lid. I added a ramp controller, with a thermocouple, to the kiln. This enables the kiln to fire things to precise temperatures. I've used it to make ITO coatings on glass slides, electrolyze molten salts, bond glass microchannel reactors together, partially oxidize nanographite, and to fire small ceramic crucibles, among other things. IrC has worked with some impressive stuff of his own, but I think it may be too big for what you want (at least for now).

Or, if you want to go with a simpler tube furnace-type configuration, I have some tips for you. You don't have to electrically insulate the heating element. If you wind your coil on a form that is the same size as the opening in your firebrick, when the winding is released it will spring back, fitting tightly to the inner diameter of the brick. If you're worried about bumping into the heating element with the tube, then add a copper sleeve that is supported on both ends, sized halfway between your reaction tube and the hole diameter. After winding the coil, let it relax on the form. Stretch it out to the width that you want, and then re-tighten it on the form. Insert the form into the brick, and then slowly release the winding, allowing it to expand to fit. Also, don't drill holes through the center of firebrick. The brick will eventually crack around the hole from thermal expansion. Instead, drill a hole between two bricks that have been put together. This allows the material to move somewhat freely during thermal cycling. The biggest investment overall (unless you add a controller) would most likely be the firebrick, which runs from $3-5 per brick, from pottery supply stores (it's a special insulative brick that is used to line kilns).

[Edited on 5-1-2016 by WGTR]

XeonTheMGPony - 1-5-2016 at 07:35

Start off by shooting a bit high, you need 100 at homeostasis so design it to 200w and use a pwm control to trim it to the target to maintain temp. Simple things like soldering irons that are cheap do self regulation that is why they fail so miserably when doing large solder pads! blow dryers and hair curling irons are all so self regulating

when it comes to heating it is tricky, to fix this most use higher wattage's and then regulate it down to what they want with a control, some are simpler cyclic (on/off) others trimmers (Rheostat/infinity switch) and better quality units PID/PWM.

Old hair dryers and old soldering irons are great sources for free nichrom as are hair curlers.

Insulation can be don with plain fine weave fiber glass sheet 1 layer on the pipe 2 layers over top the NiChrom. Soldering irons will give you fine Mica sheet that you can work as an insulator but it is tricky as they are very delicate.

Others have covered all the usual basses.

[Edited on 1-5-2016 by XeonTheMGPony]

violet sin - 1-5-2016 at 09:35

Soldering irons are not good sources of nichrome wire. Every one i have taken apart to fix after getting a short, had lots of super fine, brittle and oxidized wire. It is not suitable to rewinding after being used to the point of failure.

On another note, the copper "air chamber" is actually a "stub out". It is something plumbers use while running aquapex water lines usually, some times copper. Any how after the house is plummed the outlets for water at sinks, showers, washers, dishwashers etc. are left sealed like that under pressure for a couple days. Then the psi gague on the different lines are monitored for leaks. When done testing the closed end is cut off, and fittings are sweated on.

The point to this, some come with flat flanges for fastening to studs or sheewall(wood). Some come with brass fittings on the open end also. Of course the cost goes up a bit with the extra pieces

http://www.supplyhouse.com/Viega-44227-Zero-Lead-1-2-PEX-Cri...

http://www.supplyhouse.com/Wirsbo-Uponor-LF2945050-Copper-St...

These additional shapes, fittings and brackets could be usefull
Edited to make link work

[Edited on 1-5-2016 by violet sin]

XeonTheMGPony - 1-5-2016 at 09:41

Quote: Originally posted by violet sin  
Soldering irons are not good sources of nichrome wire. Every one i have taken apart to fix after getting a short, had lots of super fine, brittle and oxidized wire. It is not suitable to rewinding after being used to the point of failure.

On another note, the copper "air chamber" is actually a "stub out". It is something plumbers use while running aquapex water lines usually, some times copper. Any how after the house is plummed the outlets for water at sinks, showers, washers, dishwashers etc. are left sealed like that under pressure for a couple days. Then the psi gague on the different lines are monitored for leaks. When done testing the closed end is cut off, and fittings are sweated on.

The point to this, some come with flat flanges for fastening to studs or sheewall(wood). Some come with brass fittings on the open end also. Of course the cost goes up a bit with the extra pieces

http://www.supplyhouse.com/Viega-44227-Zero-Lead-1-2-PEX-Cri...

http://www.supplyhouse.com/Wirsbo-Uponor-LF2945050-Copper-St...

These additional shapes, fittings and brackets could be usefull
Edited to make link work

[Edited on 1-5-2016 by violet sin]


Why you don't use a failed one! Some times easier to buy a cheap one to dismantle then to order the nichrom its self, Usually I buy a newer one then use my older working on for parts..

violet sin - 1-5-2016 at 10:12

Trust me, i know. Have 5 or so dead ones right now and 3 working irons. Salvaged parts, scrapped the wire for digestion in acid later. Its just junk.

Also some of the nichrome was tapered... Thick at one end and thinner at the other end. Which would make it kinda hard to calculate a target temp/power input.

A small spool of nichrome is cheap off ebay. The finer the cheaper it is. Think i got a 100' roll for 3$, small enough a short length will easily glow yellow with a standard AA. It was from a USA shipper, so no three weeks wait. Blow dryers have much more robust clean wire in them, and cheap from a thrift store or yard sale. Better is a house rennovation that has dead wall heaters and scrap laying in the front yard waiting for a dumpster. The general contractor will almost always let you walk off with stuff like that for free. The units arent failed, just outdated, so you get a working switch, element support/element and all... For the lowly cost of a dirty spot on your back seat.

elementcollector1 - 1-5-2016 at 17:06

I think I'll just buy some wire instead of scrapping a soldering iron or two.

WGTR, I'd rather avoid firebrick if I can help it - I don't have the money to be lugging heavy stuff back and forth between campus and home during construction, even if it is cheap. Therefore, I think fiberglass insulation would be good for the inside, and the outside might be a thicker variety of fiberglass or even some sort of ceramic powder (that way I wouldn't have to worry about it breaking, just sintering).

Currently trying to figure out a thermocouple/PID setup for temperature monitoring - Arduino seems like a likely choice for this project. Shouldn't be too hard to control a logic-level MOSFET to switch on and off the power to the nichrome, right?

Twospoons - 1-5-2016 at 19:38

If I were you I'd just buy a PID controller - you get a nice display and controls in a nice case with all the software done and tested, including autotune. All you need do is pick the power control device - relay or SSR (i'd go with the latter). You can get the whole thing, including SSR, for under 20 bucks on Aliexpress.

JJay - 1-5-2016 at 23:18

Quote: Originally posted by elementcollector1  
I think I'll just buy some wire instead of scrapping a soldering iron or two.

WGTR, I'd rather avoid firebrick if I can help it - I don't have the money to be lugging heavy stuff back and forth between campus and home during construction, even if it is cheap. Therefore, I think fiberglass insulation would be good for the inside, and the outside might be a thicker variety of fiberglass or even some sort of ceramic powder (that way I wouldn't have to worry about it breaking, just sintering).

Currently trying to figure out a thermocouple/PID setup for temperature monitoring - Arduino seems like a likely choice for this project. Shouldn't be too hard to control a logic-level MOSFET to switch on and off the power to the nichrome, right?


I've been thinking about hooking up an Arduino to do the same thing.

elementcollector1 - 2-5-2016 at 02:34

Quote: Originally posted by Twospoons  
If I were you I'd just buy a PID controller - you get a nice display and controls in a nice case with all the software done and tested, including autotune. All you need do is pick the power control device - relay or SSR (i'd go with the latter). You can get the whole thing, including SSR, for under 20 bucks on Aliexpress.


True, but I also have access to a circuits lab with free components. Besides, Arduino's not so hard to set up in this manner, all I really need to work on is reading the thermocouple properly.

EDIT: Okay, so to flesh things out a bit more:

I'll be using a logic-level MOSFET to drive the main circuit, as I've had good experiences using these. It should be rated for the max amperage and voltage I'll be pulling, despite being in an absolutely tiny package. Might need a heat sink, depending on what intrinsic resistance it has.

Thermocouple will be a K-type designed for high temperatures. However, according to some tutorials around the Web, thermocouple voltage differences are typically measured in microvolts, so I'll run them through an op amp with a gain of 10000 or so to get things up to a nice, readable signal for the Arduino (though with that kind of gain, I'm worried about signal noise). I'll also need some kind of reference signal - would a thermistor in ambient temperature be able to produce that, or is using two different types of temperature sensor a bad idea?

The code for the Arduino, sans the thermocouple-interpretation bit, is very simple. Too high? Turn off the MOSFET. Too low? Turn on the MOSFET. Rinse and repeat.

I decided to switch to Kanthal wire, because it's rated for higher temperatures. The maximum temperature I'll be hoping for with this system is about 1000 C, because past that the copper and fiberglass insulation will both start melting (and even then, the copper's kind of pushing it with a MP of 1084 C). This will run 12V at 1.44 amps, for a power rating of 100 W.

I'm still a little fuzzy on the temperature output for a given power rating - wouldn't this just steadily increase up to some maximum in an enclosed space? The temperature charts from earlier in the thread were less of a ballpark estimate and more of a tennis-ball-in-a-football-field estimate.

Anyway, the whole thing will be powered from one of these, which is rated for 350 W, cheap, and gets better ratings than the Logisys stuff on Amazon.

Anyone see any gaping holes in this setup?

[Edited on 5-2-2016 by elementcollector1]

Twospoons - 2-5-2016 at 16:41

This is about the easiest way to interface a thermocouple to a microcontroller. It handles cold-junction compensation and gives you a 14bit conversion.

Building amplifiers with DC gains of 10000 is not trivial. If you do go that route do the gain in several stages, focus on low noise, very low offset, opamps in the first stage. If you are trying to measure uV then having mV of offset is not going to work. Watch out for feed back that could turn the whole thing into a big oscillator.
.
Your control method is known as 'bang-bang' control, and will not be very stable - if you're ok with that, then no problem.

There is no absolute relationship between power and temperature - its an equilibrium situation where T will rise until power in = power out. Better insulation = more T per watt.
If you can guess/calculate the insulation thermal resistance in your system you can get an idea of how much power you will need for your target temperature.

elementcollector1 - 2-5-2016 at 17:52

Quote: Originally posted by Twospoons  
This is about the easiest way to interface a thermocouple to a microcontroller. It handles cold-junction compensation and gives you a 14bit conversion.

Building amplifiers with DC gains of 10000 is not trivial. If you do go that route do the gain in several stages, focus on low noise, very low offset, opamps in the first stage. If you are trying to measure uV then having mV of offset is not going to work. Watch out for feed back that could turn the whole thing into a big oscillator.
.
Your control method is known as 'bang-bang' control, and will not be very stable - if you're ok with that, then no problem.

There is no absolute relationship between power and temperature - its an equilibrium situation where T will rise until power in = power out. Better insulation = more T per watt.
If you can guess/calculate the insulation thermal resistance in your system you can get an idea of how much power you will need for your target temperature.


That sounds like a much better idea, thanks. How does it handle cold-junction compensation - is there a thermistor inside the chip, or does it just 'know'?

So, in other words, maximum insulation. Probably want to go with ceramic powder, then - unless a solid piece would be better somehow? Personally, I don't think it would be.

Twospoons - 2-5-2016 at 18:50

The chip handles the cold junction compensation with an internal temperature sensor, so its important that the thermocouple connection is close to the chip.

For insulation just pick something light weight and fluffy. Fibre > powder > solid. Something like Kaowool would be ideal.

markx - 17-5-2016 at 23:18

First I would ask myself what is the goal of the current quest?
Is it :
a) to perform synthesis requiring elevated controlled temperatures
b) to enjoy the process and learning curve of building a temp control system from scratch

If the goal in mind is "a)", then I would not bother with building the whole system from scratch. It is kind of reinventing the wheel...nowadays there are very decent building blocks available on ebay and aliexpress for asocial prices. It makes little sense to spend time on trying to work out the control circuitry for a thermocouple (not a trivial task to be honest), if pid controllers and solid state relays can be had for a couple of dollars.

http://www.ebay.com/itm/320813025328?_trksid=p2055119.m1438....

For example this is a set that I've bought and used...can really reccommend. SESTOS makes very decent components with a lot of functionality and reasonable price. I would avoid the low end PID units, they do kind of work, but soon one will realise that there are reasons for the bottom end price.

Alternately there are also several types of heating elements available for different power ratings:

http://www.ebay.com/itm/8-50mm-220V-100W-Cartridge-Mold-heat...

These can be used at lower temperatures up to 600C, but probably not at 1000C. I use this kind of heating elements in my melting point apparatus and in a cold smoke generator to scorch the wood. Really compact and effective devices.

It also makes sense to work with line voltage AC if constructional safety is not an issue. One can omit separate power supply for 12V dc and use much thinner cables due to less amperage needed for the same power output. Plus standard components for commutating AC line power are dirt cheap and readily available (SSR-s and SCR control boxes for simple pwm). Alternately there are also dc ssr-s available, but these are a bit more expensive compared to ac variety. So it makes little sense to devise a separate control unit out of mosfets.

A little word on PID tuning is also appropriate if one decides to use these devices. Precise temperature control can actually be insanely complicated in small reactors that have little heat capacity and becomes even more complicated if one has to take into account possible exotherms from an reaction. In fact the mighty PID controller is a device that is dumb as a second coat of paint if taken out of the box and used without proper calibration for a given system. The autotune feature is of little use in small and dynamic systems....it will land you in a rough ballpark of paramaters, but it is lightyears away from precision. So to be able to use a PID controller effectively, one has to familiarize oneself with the inner workings of that control box and tune it by hand via trial and error. And even then the tuning works only for a range of setpoints. E.g. a set of tuning parameters that apply at 100-150C in a given system may become totally useless if the setpoint is pushed to 400C. Under some instances it is quite impossible to achieve a high precicion of temp control with PID...for example if the system has radically different rates of cooling and heating e.g a high power heating element in a small system that is thermally well insulated. This setup will heat up very quickly, but cool at comparatively low rate. Unfortunately PID assumes that these rates are equal and the compensation parameters apply in both directions on the heat-cool curve. Obviously this is not the case if rates are radically different, so whatever combination is used there may still be major fluctuations near setpoint.

Ok...long story short, I'm not trying to discourage anyone with this chapter and hoping that this information is of help :)

Melgar - 26-5-2016 at 12:10

If you want to get something up and running quickly, just use mains voltage through a dimmer switch. You can get a high-rated dimmer and a box for it at your local hardware store for $10-$20 depending on how many watts you plan on running at. Then just slowly turn up the power until your wire glows orange.

[Edited on 5/26/16 by Melgar]

careysub - 26-5-2016 at 13:09

Nurdrage has a video on making a high temperature heater using an Omega Engineering "Ultra-high Temperature Heating Tape", which according to the Omega website go up to 790 C (vs 450 C for a Glascol heating mantle).

https://www.youtube.com/watch?v=N5meDWAqYes

http://www.omega.com/pptst/STH_SST_SWH.html

He later uses this mantle to make sodium cyanide at ~650 C with cyanuric acid.

790 C is hot for a mantle, but certainly no where near "tube furnace" hot.

Is there enough interesting chemistry between the maximum temp of a regular (used) mantle and a 790 C mantle to make this worthwhile?

JJay - 27-5-2016 at 10:58

Quote: Originally posted by careysub  
Nurdrage has a video on making a high temperature heater using an Omega Engineering "Ultra-high Temperature Heating Tape", which according to the Omega website go up to 790 C (vs 450 C for a Glascol heating mantle).

https://www.youtube.com/watch?v=N5meDWAqYes

http://www.omega.com/pptst/STH_SST_SWH.html

He later uses this mantle to make sodium cyanide at ~650 C with cyanuric acid.

790 C is hot for a mantle, but certainly no where near "tube furnace" hot.

Is there enough interesting chemistry between the maximum temp of a regular (used) mantle and a 790 C mantle to make this worthwhile?



I think so... I personally have not bought a heat tape because I can use flame or nichrome wire for many experiments that would require such a mantle, and there are lots of other things I would rather have. But getting one of those heat tapes is on the list.

It should be pointed out that 650 C is above the regular working temperature of borosilicate glass. Making cyanide is dangerous, so I really can't say I recommend following his procedure - NurdRage advises against trying to reproduce his experiment rather vehemently in his video, including such admonitions as, "YOU WILL DIE," and "NO, SERIOUSLY."

I've been thinking about trying a cyanide synthesis in a small crucible, but I'm not sure I really want cyanide around the lab. It would be pretty trivial to wrap a crucible in nichrome with a triac dimmer and measure temperature with a temperature probe, though....


[Edited on 27-5-2016 by JJay]

careysub - 27-5-2016 at 11:19

Quote: Originally posted by JJay  
...

It should be pointed out that 650 C is above the regular working temperature of borosilicate glass. Making cyanide is dangerous, so I really can't say I recommend following his procedure - NurdRage advises against trying to reproduce his experiment rather vehemently in his video, including such admonitions as, "YOU WILL DIE," and "NO, SERIOUSLY."

I've been thinking about trying a cyanide synthesis in a small crucible, but I'm not sure I really want cyanide around the lab. It would be pretty trivial to wrap a crucible in nichrome with a triac dimmer and measure temperature with a temperature probe, though....


[Edited on 27-5-2016 by JJay]


It is way above the service temperature of borosilicate (according to Corning the normal limit is 230 C, with 490 C the "extreme service" - about where regular heating mantles conk out) - you would have to use some sort of crucible (metal or ceramic) in this regime.

Nurdrage used a steel can -- which makes a nice disposable one (once you have decontaminated it).

Sure, you shouldn't make cyanide unless you really have a particular use that calls for it.

My point is: are there other interesting/useful processes in this temperature range, besides cyanide. I know over several that require >1000 C, but none in the particular 500-800 C range that the heating tape provides.

I liked his video where he made a mantle using the tape. I like making things (when they are much cheaper than commercial or provide unusual capabilities) but I wondered how useful would that be, really?

[Edited on 27-5-2016 by careysub]

Fulmen - 27-5-2016 at 12:06

At one lab we gave all glassware a 400°C "roast" after regular cleaning, everything from flasks to beakers. This was a large oven going through a slow cycle (overnight), and we still had a fair amount of breakage. I wouldn't trust regular borosilicate at these temperatures for anything toxic and volatile.

careysub - 27-5-2016 at 16:57

I did a little research and came up with four useful processes that require or run best in the 450-800 C range:
Magnesium Oxide (MgO) - Regenerate at 800 °C for 6 hr (not a great dessicant though)
Calcium Oxide (CaO) – From Ca(OH)2 at 580 °C.
Barium Oxide (BaO) - Can be prepared by decomposing Ba(NO3)2 at 800 °C.
Boron (III) Oxide (Boron Trioxide) - To prepare in molten form above 500 C

elementcollector1 - 1-6-2016 at 16:08

My main plain is to use this for a cesium reactor (once I get the metal-to-glass joint down) and to decompose oxalates without calcining them.

I've since bought some black iron pipe lengths instead of the copper, as it doesn't hold up very well at high temperatures.

elementcollector1 - 7-6-2016 at 12:28

Just received my kanthal wire, currently figuring out how much to use. If I step up the ATX 12V power supply to 60V, it will most likely provide 2.8 A and thus 168W (same as the 12V/14A it's rated for). This requires 6.375 feet of 26 AWG kanthal wire, which is a reasonable amount to put on a 5" long, 1.25" diameter pipe (this is after electrical insulation with fiberglass).

What DC boost converter will be able to stomach 2.8 A at the output and 14A at the input, though?

[Edited on 6-7-2016 by elementcollector1]

Metacelsus - 7-6-2016 at 14:43

Why do you even need to use DC? Why not AC? Transformers are easier to deal with than DC/DC converters.

elementcollector1 - 7-6-2016 at 15:23

Does AC work with kanthal? I tried looking it up, but didn't find much.

Twospoons - 7-6-2016 at 16:16

Kanthal wont know the difference. AC is much easier. You could then use a variac for power control. Simple, and very robust - which switching supplies are not.

LookingForAnswers - 10-8-2024 at 23:06

chatgbt sugest me to use silicon carbide as isolation. I checked price and it is very cheap. What do you think guys?

macckone - 13-8-2024 at 10:19

Kanthal is usually insulated with magnesium oxide inside heater elements. I cannot think of a cheaper material. And the kanthal will fail well before the magnesium oxide.

fx-991ex - 13-8-2024 at 11:27

You cant push a 12volt atx power supply up to 60 volt.
First you will need some kind of load on the 5 volt line because the overvoltage protection will cut off(even if you fool the 12volt OVP part with trimmer/pots or zener diodes)
Also, the electrolytic capacitor inside that power supply are not rated for this voltage, so if you manage to get this kind of voltage then it will blow up.

Running 2 smps in series is usually a bad idea(but it do work).

As for calculating wire length and voltage/current, i like to use this calculator
https://jacobs-online.biz/calc2.html

macckone - 13-8-2024 at 13:04

a half wave rectifier with a motor start capacitor will give a 60v output up to about 7 amps assuming 120vac input. Using a high wattage light dimmer or using a regular light dimmer to drive a triac ahead of the rectifier will do fine. Honestly for a heater application you can just use a half wave rectifier and high wattage light dimmer.

You can use an arduino to drive a triac but it requires a few other components.
Search for arduino kiln controller. It will provide a number of hits.

Most SSRs are just a triac with driving circuitry to allow 3 to 24 volt drive.
Look up SSR tear down on you tube.

LookingForAnswers - 14-8-2024 at 15:09

Quote: Originally posted by macckone  
Kanthal is usually insulated with magnesium oxide inside heater elements. I cannot think of a cheaper material. And the kanthal will fail well before the magnesium oxide.


On another forum I asked about this and one guy replied that I need burnt magnesia to fill the heating element. To get this element I have to heat MgO to 1700-2000*C so I decided to leave it. He suggested me to use talc, water and phosphoric acid to make a paste but after I asked about the proportions he didn't answer so I started looking for another material. Before I ordered SnC I did some good research and it looks promising. Now I have both (SnC and MgO), only one detail is that I use Ni-Cr wire and not Kental so maybe this guy is right about overheating if I would use MgO with nichrome wire. Does anyone can confirm or decline this info's?