Calibrating propane forge using K-type thermocouples and optical thermometer

[zambezi]

I have some exciting stalking booked for the moment we are given the all-clear to travel further afield.

And before that I have a range day booked per NRA guidelines. I will get to fine-tune some loads for the stalking.

Pending that, I crack on with my primary lockdown hobby: knife-making. Some early successes and some heat treat fails. So I decided I need to be more certain of forge temperatures.

Inexpensive K-type thermocouples abound on the internet but very few are rated to the 1100℃+ needed to HT stainless steel. And many that are can only survive one or two runs at that temp So my quest was on to find out whether there are any thermocouples out there that can give accurate results and survive multiple heat cycles. The latest addition is a stainless steel probe of 7mm girth. It looks the part

I have some Orton cones from Utecktok which I will test drive as soon as I can source some appropriate clay to bed them. Alantoo recommended an optical thermometer, so I got one of them too. Old and new stainless thermocouples in forge

Data compared with optical readings suggest new stainless probe is probably in the right ball park [upper of the two readings on the DMM at left]. The difference between that probe and the optical may reflect the actual temp gradient between probe tip and forge epicenter.

So what I think I will do going forward is use the stainless probe to warm up and stabilise the forge temp and acetlylene flow rate, and then switch to optical surface readings on the work piece to ensure fine control of temps.

 

[ejg]

We used to run ovens at 1600-1730 C in air atmosphere with Type B and had some furnaces lasting 20 years. I wouldn't trust optical measurement.
Plenty gear available from the pottery crowd.
edi

 

[williamwansbeck]

A 7mm sleeved thermocouple will have a heat sink effect giving delayed responses,

 

[WH308]

I have worked with thermocouples for most of the last 20 years. A word of caution - beware the potting in the end of them, if they are overheated the air behind the potting can expand forcing the potting out at great speed. It makes a very hot, sticky projectile not the sort of thing one wants near an eye.

This company are excellent for both range stocked and advice. (No affiliation, I use them for my TC’s at work)

TC Direct for Thermocouple & RTD Pt100 Sensors, Cables, Connectors and more

TC Direct for ex-stock Thermocouple Sensors, RTD Sensors, Pt100, Thermocouple Connectors, Thermocouple Cable, RTD Cable, Temperature Instrumentation and more

www.tcdirect.co.uk

 

[zambezi]

A 7mm sleeved thermocouple will have a heat sink effect giving delayed responses,

I am aware that all of the thermocouples I have trialled so far take a while to come to temp. I do not take reading until forge has been in a steady state for 10 mins ish and thermocouples are inserted for that full duration and not moved. [no air-fuel change]

 

[Utectok]

You could just stand the cones up by putting a couple of bits of off cut metal round them

 

[zambezi]

I have had something of an epihany tonight: temperature is not the whole deal in an propane forge.

The chap in the vid below points out that an additional consideration is the amount of available carbon in the forge atmosphere.

His take is that by reducing the air in the air-fuel mix immediately prior to introducing the workpiece creates a greater amount of uncombusted acetylene [yellower flame]. That produces a carbon rich environment and thus improved knife finish.

Any learned input on this strategy?

 

[Utectok]

Reduction and oxidisation is a big part of ceramic firing including reducing at some temperatures and oxidising at others. Beware heavy reduction will drop temperatures and oxidisation will raise temps quicker

 

[Sharpie]

Inexpensive K-type thermocouples abound on the internet but very few are rated to the 1100℃+ needed to HT stainless steel. And many that are can only survive one or two runs at that temp View attachment 198431 So my quest was on to find out whether there are any thermocouples out there that can give accurate results and survive multiple heat cycles. The latest addition is a stainless steel probe of 7mm girth. It looks the part View attachment 198432

Yes there are proper industrial grade thermocouples that will be rather better. K types are usually rated to a max. of 1200C.

Take a look at Turning up the heat: 12 MI thermocouple probes get tested in extreme conditions

Then see 1.5 to 6mm Diameter MI Construction Thermocouples Terminated With A Pot-Seal & PFA Lead Wire to see if they might have one that suits your application. They cost around £30, I expect that is plus VAT etc.

Other makes are also available.

I used to these for measuring exhaust gas temperatures in motorsport applications, custom made to our requirements, and they worked well for us at these sorts of temperatures, or even higher.

From your 'photo your meter appears to be selectable between various types, if so you might want to research alternative types than K, in case another type is better suited. If it is fixed for type K then obviously that is what you need.

You are certainly getting more scientific than the "heat it up until a magnet doesn't stick then quench it in used engine" oil types.

The tempering is also a key part of the final treatment as I'm sure you know, and are on top of. An old fan oven with the electrickery and thermostat bypassed and a thermocouple and PID controller used instead could be a good lockdown project.

TBH, If I was thinking at having a go, I'd make myself an electric furnace out of resistance wire and ceramics, with PID control, and install a cooker point in the garage to supply it. I think it might even be more economical to run, much cooler work environment, and none of all that exhaust gas to breathe in. Run it whilst the wind is blowing over our windmills and feel even more eco.

PS: why are you using acetylene, when AFAIK propane is sufficient ? A builders blowtorch should be up to the job for a small knife furnace.

 

[Alantoo]

I have a fixed speed three phase fan with a simple slide air valve on one of my propane furnaces. The other furnace blower is a single phase brush motor fan with a rheostat.

On a very basic level, excess gas gives a
carburising atmosphere, and too little gas gives and oxidising one which will create much more scale. Sometimes a heavy layer of oxide is useful if you want a texture in the surface but I guess in your blades would be unwelcome.

I set my home built furnaces to a neutral atmosphere by tweaking the gas valve to trim back the visible (in shadow) flame so it just doesn’t project through the gaps outside the furnace. I reckon if all the gas is consumed within the chamber and the chamber is full of flame that is neutral.

I wouldn't trust optical measurement.

The big advantage of it in this instance is that you can take a reading directly off the bit of metal you are heating in situ, rather than just the general atmosphere of whichever part of the chamber you put the thermocouple.

The emissivity adjustment on the optical gun should enable you to calibrate it with the thermocouple.

It should be accurate enough to allow one to take readings from different parts of the blade to ensure they have all reached the correct temperature.

Alan

 

[Sharpie]

I have had something of an epihany tonight: temperature is not the whole deal in an acetylene forge.

The chap in the vid below points out that an additional consideration is the amount of available carbon in the forge atmosphere.

His take is that by reducing the air in the air-fuel mix immediately prior to introducing the workpiece creates a greater amount of uncombusted acetylene [yellower flame]. That produces a carbon rich environment and thus improved knife finish.

Any learned input on this strategy?

Yes. To some extent I would agree that a neutral atmosphere inside the furnace is better than an oxidising one.

But why on earth are you using acetylene. He isn't, its not necessary, OTT, expensive, and dangerous.

It is just a rather unstable carbon hydrogen molecule. C2H2. Propane is C3H8. Butane C4H10. Methane (natural gas) CH4., etc. All with different potential burning temperatures with the right ratio of oxygen. Feed acetylene with pure oxygen and it is fantastic stuff. But acetylene has to be synthesised. Unlike all the others. And cannot be stored safely as a gas, your cylinder will be stuffed with a sponge soaked in a solvent. The double bonds between the carbon and the hydrogen in it are unstable and release masses of energy when broken, no oxidiser required. Unlike the others, linked together by single bonds.

Even then, mostly dissolved in a solvent inside the cylinder, if it gets dropped, and there is some problem inside it can still explode. Shock, not much, can set it off. Much as I like the stuff, so flexible in use, and applications, and am qualified in its use and have passed stringent welding tests with it, (part of my apprenticeship) I wouldn't dream of using it outside of a properly set up workshop. Chained up.

There are still some things that just can't be done efficiently without oxy-acetylene.

Perhaps you are over-thinking a little. At this stage.

Were I hypothetically, to make a little electric furnace it could be purged with neutral argon. Or CO2. Or Argon/CO2, Or maybe even nitrogen, at some point in the process. Nitriding for example way of putting an extremely hard surface onto some types of steel. All readily available welding gases.

Specialist steel is, hopefully, precisely formulated to work as it has been made. Not oxidised and the surface carbon perhaps burned away., neither carburised on the surface in a reducing atmosphere.

It really rather depends on whether you are using specialist steels, in which case the manufacturer might be able to advise about appropriate heat treat profiles and atmospheres. Or even doing it old-school with old files or axle springs on an open forge fuelled with charcoal or coke, as is traditional in parts of Scandinavia.

And what blend of hardness, toughness, and wear resistance you are seeking to achieve.

But every knifemaker has their own way of doing it. Some may even be trained metallurgists. And the best ones won't tell you how they do it. A lot of mystique seems to go on about this, in the knife world. I'm not sure that it has to be particularly complicated to get a decent result, for a simple knife blade.

 

[ejg]

I never had interest in metals and did not do any deeper studies apart from uni and toolmaking. Saying that, my choice would be electrically heated furnace preferably from Nabertherm. Acetylene would not even cross my mind for that temperature range. Just recently found one of my fathers patents of a gas burner for a furnace. I was along when we fired an at least 50 liter porcelain furnace to 1450 deg C only with a household gas bottle. Oven could be run rich or lean and was very efficient.
edi

 

[zambezi]

your meter appears to be selectable between various types, if so you might want to research alternative types than K

Yes. And I have now discovered that some "N" types might be better suited.

Mineral Insulated Thermocouple with IP67 Micro Weatherproof Head - TC Direct

Mineral Insulated Thermocouple with IP67 Micro Weatherproof Head These semi rigid thermocouples are ideal for all applications from simple temperature measurement to more arduous environments. They have insulated junctions to prevent earth loops and can be formed to shape to suit particular...

www.tcdirect.co.uk

why on earth are you using acetylene

My typo. I had been watching a bunch of YT vids on a variety of projects [including automotive chassis re-modelling] in which folk were using acetylene and that word stuck in my brain as I typed late last night. Actually, I am using propane.

The emissivity adjustment on the optical gun should enable you to calibrate it with the thermocouple.

It should be accurate enough to allow one to take readings from different parts of the blade to ensure they have all reached the correct temperature.

I agree that the optical measuring strategy will enable me to compare the temp of all parts of the workpiece. But since I am not yet 100% confident that I can trust any of the thermocouples, I am loathe to calibrate against any one of them.

 

[Alantoo]

I agree that the optical measuring strategy will enable me to compare the temp of all parts of the workpiece. But since I am not yet 100% confident that I can trust any of the thermocouples, I am loathe to calibrate against any one of them.

I only mentioned that in response to @ejg 's comment about mistrust of the optical gun. I would have no such qualms for your use providing you set the emissivity correctly. I would not have recommended them otherwise.

It is a much better tool for the job as it gives you instant spot readings. You want to know the temperature of the blade, not the temperature of the furnace which has the inevitable hotspots from the top flame heating.

I built all my later furnaces with a tangential flame so that it swirls / rotates around the chamber and the workpiece is heated largely by the radiant heat from the ceramic walls of the furnace. Light the larger one in the morning and by middle day it was getting really efficient heating 1200mm lengths of 50mm or 80mm square bars! The first hour or two of the day the heating took too long and was a constant attendance of draping wet rags over the bits sticking out to prevent the handles getting too hot. We even managed to get some 100mm square bars up to forging temperature with it on occasion.

You can calibrate it/check the optical in your furnace with either the ceramic cones you have acquired or by melting a bit of Aluminium, copper or brass and reading the temp when it melts if you have any doubts about it.

I had the instructions for a home built optical thermometer which used a 12volt flasher/indicator bulb and a rheostat...you calibrated that by noting when the filament of the lamp disappeared / matched the glow of various materials of known melting point. I might have a copy of the instructions somewhere.

But I would just use the optical gun as-is on an offcut piece of your blade material to check the process.

Alan

 

[Triffid]

I like Alan's idea of calibrating your thermometer/thermocouple by the melting point of a metal as these are more-or-less fixed. I'd use copper (at 1084 C) as the brass & aluminium alloys have varying melting points.
Triffid

 

[Alantoo]

I like Alan's idea of calibrating your thermometer/thermocouple by the melting point of a metal as these are more-or-less fixed. I'd use copper (at 1084 C) as the brass & aluminium alloys have varying melting points.
Triffid

Yes, good to make that point clear...'tis why I mentioned "known" melting points...as long as you have the spec. of the actual alloy/materials you are melting.

Alan

 

[zambezi]

You can calibrate it/check the optical in your furnace with either the ceramic cones you have acquired or by melting a bit of Aluminium, copper or brass and reading the temp when it melts if you have any doubts about it.

Really useful!

 

[zambezi]

I like Alan's idea of calibrating your thermometer/thermocouple by the melting point of a metal as these are more-or-less fixed. I'd use copper (at 1084 C) as the brass & aluminium alloys have varying melting points.
Triffid

Sounds like a plan. Is domestic plumbing pipe [15mm] pure copper? I could use that to calibrate the optical meter perhaps.

 

[Alantoo]

Current spec of plumbing pipe is 99.9% copper with a bit of silver according to Yorkshire Copper .com

24ct Gold and Sterling Silver would be fairly close to your desired temperatures of 849˚C and 1060˚C

But just experiment by heat treating an offcut, it will tell you so much more... handling, temperatures, oxidation....

If it doesn't work that is the time to faff with calibration.

The thermocouple readings discrepancy/variations you have had should indicate that empirical results and observation may be more reliable at this level of operation....A lot of metal work heat treatment has been done over the years with out any temperature readouts...most of the industrial revolution in fact, let alone before.

Interesting I have just looked at the video you posted earlier and was amused to see that even though he set the furnace temperature with his thermocouple...he still relied on matching the colour of the blade with the colour of the furnace lining by eye!

Alan

 

[zambezi]

video you posted earlier and was amused to see that even though he set the furnace temperature with his thermocouple...he still relied on matching the colour of the blade with the colour of the furnace lining by eye!

I suspect that his thermocouple junction was nestled in the forge lining, or at least did not protrude into the furnace very far. Thus he knew that what he had measured was the temp of that lining.

So yes, once the workpiece was at same colour he could be confident it was at temp he had measured.

I realise the wisdom of your words: testing an offcut of knife metal is more likely to give me a true idea than any equipment calibration since it informs on the totality of the metal's journey.

If I harden a few pieces of AEB-L stainless at an optically defined 950℃, 1050℃ and perhaps 1150℃ and then apply Tsubosan files to the products, I will have a good idea which is baking the cake correctly. [and compare finishes]

I will do some test pieces later...