HEAT TREATMENT- AIR COOLING RATE

[sydneyjongleur]

I was hoping that someone could give me some info on how to work out the air cooling rate of a metal to room temperature.

If there is a rule of thumb instead of a particular equation then that would be fine also.

I am tempering tool steel from 500°C and the components are Ø600 x 8mm thick and I was trying to work out how long it would take to air cool to room temp.

Any help would be appreciated

 

[davefitz]

At 500 C the primary heat transfer method is thermal radiation heat transfer. q/a=e*a*F*(T1^4-T2^4)* Planc, where T= abs temp

For an 8 mm wall thickness the part will cool very quickly- you can expect martensite or bainite to form unless you shield the parts from the ambient temperature . Refer to the TTT curves for that alloy. You can calculate the fourier number and /or the biot number and then estaimte the rate at which it will cool down, then refer to the TTT curve for that cool down rate.

By the same token , if you were processing a thick component ( say 35 mm thk) and wanted to form martensite ( as with P91) or bainite ( as with T23) then you want to ensure a fast cooldown , faster than 10 F/min. So make darn sure there is not shielding of the thermal radiaton from the part to ambient.

 

[blacksmith37]

The cooling rate you ask for will tell you when you can pick it up without burning your fingers, but nothing about properties. As Dave notes you need an alloy ,then TTT curve, or continious cooling curve.

 

[sydneyjongleur]

Thanks for the info guys.

I have the TTT and CCT diagrams for the tool steel(D2)but I am struggling to determine the rate of cool down from that( I am not familiar with fourier and biot numbers and how they relate to the TTT CCT diagrams)

I basically want to determine how quickly I can put it back in the oven to do a second temper.

This part has already been hardened and quenched to produce martensite

 

[metengr]

Use a heat transfer reference book below, not a TTT diagram.

http://web.mit.edu/lienhard/

http://www/ahtt.html

 

[redpicker]

It seems to me that for issues like this it is better to go by experience than try to calculate an answer. There are too many variables to control to calculate an accurate answer.

That is, take one out of the furnace and check it's temperature every 10 minutes. That should give you a good cooling curve and allow you to determine the heat transfer coefficients to be able to calculate the time for other sizes/thicknesses.

rp

 

[salmon2]

Heat transfer can take three formats: direct conduction, air convection, and radiation. Davefitz plundering at 500c radiation is dominant but when temperature drops, probably passed nose of Cct, convection will be the dominant one to room temperature. Direct conduction will never be big.

In theory, you can run fea to get an idea, but as redpicker, testing is still needed to confirm as fea can easily vary from testing by an order.

 

[sydneyjongleur]

Thanks guys, your help is much appreciated.