Lowest tempering temperature for medium and high carbon simple steels

[S obsessed]

What temperature would be the lowest one could temper and have an effect on medium and high carbon simple steels? I know the end use has a lot to do with this. A knife edge may need a temper as it is small in section where as a square bit for the lathe may not need much or any temper. Some say that you don't need to temper 1045 at all. I am guessing since it doesn't have enough carbon to give 100% martensite that tempering may not be necessary in some instances. What about 1060 that should be 100% martensite or near it. Obviously not getting the proper heat treat will also affect this with getting pearlite or RA also. I tested a 10" nicholson bastard cut file and it was 65rc at the tip and an inch back it was 67rc consistantly in 5 different areas. So since those are supposed to be w1 or 1095 or something close to that it is likely not tempered at all to be that hard. Larrin mentions for 263c that one can temper at 250f for some instances. So at what carbon level does tempering even need to be done or start to have a greater effect and at what temperature does anything even start to happen as far as the martensite seeing some relief? I know alloying will change this also. I have read Larrins book and am at the early stages of reading "Tool steel simplifed" but some questions I am not figureing out. In the old machinist trade they say leave w1 full hard for a tool bit yet people have had an untempered knife crack up over night before a temper. I am guessing the heat treat with proper temps vs. heat with torch and dunk in water would have something to do with that method but again the nicholson file at full hardness raises questions. I hope I am being clear about this. Thanks

 

[Stacy E. Apelt - Bladesmith]

For a cutting edge, most low allow simple carbon steels do well with a 400°F/205°C temper. This will lower the hardness only a small amount, but increase the toughness sufficiently to make the edge hold up.
For steels with more alloying, 350°F/175°C is a good lower limit.

When you say "Some say 1045 does not need tempering", I am not sure where you read/heard that. 1045 is not used for knives, and for most other uses it should be tempered. In large/thick sections, some degree of auto-tempering may occur, which would be sufficient is some low stress uses, but those aren't knives. 1045 is generally tempered at 700-1000°F/370-540°C.

 

[shqxk]

Tempering doesn't only increase toughness but also increases yield stress which is very important as cutting apex.

 

[S obsessed]

Thanks for the responses guys. My question is not what tempering does or at what temperature to temper knife steels. It is rather a question of fundamentals. 1045 is used in a lot of axes and is run in the low 50s rc which is certainly tempered in the optimum range at the lower temper temps. Industry uses a lot of 1045 and 4140 at the higher temper range above tempered embrittlement in the 25-40rc range as that is where these steels shine for most uses. However 1045 (using that as it is simple) is also used for things like axle shafts and hydraulic rods at around 60rc which is right near quench hardness. I know that ideally a knife should be tempered between 375-425f as that is generally its peak toughness range at the higher hardness and if a steel does not work well in that range then you change the carbon content or alloying. However that is not my question. It is said that as quenched steel can contain up to 90% of its stress from internal stress in the martensite so we relieve that with tempering to move some of the carbon out of the martensite and relax it. If 1080 steel can have up to .8% carbon soaked up into the martensite and then we temper some out what percentage are we left with in the martensite at the higher useable hardnesses? If we took 1060 and did the same by putting the .6% into the martensite it would seem it could handle the full .6% without tempering and still have a useability even though it would obviously lack carbides and may still be a little brittle. 1045 would seemingly be able to have full .45% in solution and no temper as there would not be as much stress in the martensite as the 1080 or 1095. It is said that when hardness starts dropping in temper that it is an indication of carbon moving out of martensite so that would give us a temperature of likely 250f to start seeing that but what if we took a 1.3% carbon steel and quenched at 67rc started a lower temper of 200f and still have 67rc but perhaps some stress relief happened but it is just that the martensite is still super saturated with carbon. If a metal file can live at 67rc with all of its teeth and crack propogators then perhaps a 1.3% carbon steel can live happily at 67rc in same way without self destructing for some purposes. My actual question is at what carbon level does the martensite start to have such a great stress that it needs relief for simple steel? And at what temperature does the relief actually start to happen? We know the actual temps for austenitize but at what temp will a carbon molecule actually move out of martensite for tempering?

 

[S obsessed]

By the way the answer I have found for these questions over the years is that you use the steel you need and the hardness you need for your application by testing and if it works then it works. I just am looking for a scientific answer and not the it works it works method. I think the answer is out there but the question may not have been asked cause maybe it was never necessary to know. Till now...........

 

[Larrin]

Lower carbon steels can often show a bigger difference between untempered and tempered, rather than less. See this chart:

 

[S obsessed]

Thanks for the chart Larrin. I am guessing that a tensile strength chart like this starting from no temper to a high temper would reveal a lot also. Especially for higher carbon steels where maybe the strength would start to go up and hardness may not change much. Do you have any such chart? Also do you have any insight on the question of martensite and its carbon content vs. what it can handle? For example a full hard piece of 1060 vs. a tempered to equal hardness piece of 1080 would possibly be similar in toughness and tensile strength except maybe for the lathe vs. plate and a few carbides, or is that not even relatively close to possible? For example running a piece of 1055 at full as quenched 63rc hardness vs. 1095 at 63rc hardness for a extremely fine edge kitchen knife could have a similar geometry but wear resistance would be different because of carbides. I realize the carbides and different structure of the 1095 seems to make very thin edges hold out better but would they be similar otherwise as far as being chippy or rolling?