Furnace wall temperature 4

[ProcessJim]

I have a rotary furnace with a new lining (270mm brick, 70 mm brick, 10mm insulation board and 15mm steel liner. The supplier has given us some figures on what the shell temp should be at our tapping temp, 1300°C.

I am trying to work out how they have calcualted the shell temp as our observations don't match their calculations. I am struggling to understand how you can calculate the shell temp using heat transfer calculations. Can anyone help. Outside of vessel is at ambient temp 20°C or so.

 

[ProcessJim]

I have now calculated a figure for the temperature of the shell following the procedures described above. For an emmisivity of 0.94 and a h value of 2.5, I come to a shell temp of 260°C if the inside temperature is 1300 as in the suppliers assumption, the supplier calcualted 250.

Our observed shell temp is still higher (with a lower internal temperature), so further investigation required. Will start with compression of the insulation board as Corus suggests.

Thanks for all the advice!

 

[IRstuff]

Has the internal temperature been verified?

TTFN

FAQ731-376

 

[ProcessJim]

The supplier has assumed a figure of 1300°C as a worst case scenario. The temperature at which we tap our metal is 1050 - 1150°C (measured by dip thermocouple). I assume that this is the interior wall temperature in my calculations.

 

[ione]

IRstuff,

Please correct me if I am wrong. You have approached the conduction heat transfer through the multilayer wall as you were dealing with a flat multilayer wall.
But as the OP spoke about a rotary furnace I think it would be better to consider a cylindrical shape and so evaluate the thermal resistance of each layer accordingly.

 

[IRstuff]

My intent is not to solve the OP's problem, but merely to provide a tolerably close model that could be solved and used to evaluate potential issues in the parameters or assumptions.


TTFN

FAQ731-376

 

[corus]

If you're dealing with molten iron (with a freezing temperature of 1150C) then you could have wear of the brick lining reducing its thickness, as well as the reduction of the insulation board thickness. In general you should keep temperatures below 350C to prevent creep and significant loss of strength but hot spots can be acceptable (though worrying).

In general your calcuations are good enough (and simpler) using simple flat wall geometry as opposed to assessing a cylindrical shape as at large radii the heat transfer equations converge for the two coordinate systems.

ex-corus (semi-detached)

 

[ione]

Corus,
I see your point. In effect my mind was much more “pipe” oriented (small diameters) than rotary kiln oriented (big diameters), if you know what I mean.


ProcessJim,
At this point it would be opportune to know if the discrepancies you’ve noticed between calculated and measured temperature appeared from the very beginning (after the first commissioning) or after a period of operation. I am asking this because I presume thickness reductions, due to wear issues, require time to take place.

 

[ProcessJim]

The shell temp is higher than predicted from commisioning. We are trying a different liner than we usually use with a higher thermal conductivity. The observed temperature is 320°C for the new liner whereas the supplier predicts 250 and my calcualtions predict 260 - 276°C depending on emmisivity. I am currently using an emmisivity value of 0.76 for oxidised steel (Giving 276).

The concern is that the liner will not last as long as the previous liner due to the higher shell temp. We are trying this liner as it is supposed to have better wear resistance. Obviously there are significant cost implications to reduced liner life.

 

[corus]

The lining life will depend on a number of factors, as well as wear resistance. Strictly speaking, if your new lining has the same strength as before, then a higher shell temperature will reduce the stresses in the lining as the lining is less restrained in thermal expansion. This would be an advantage towards lining life.
Future problems may occur in the shell as you may have a greater differential thermal expansion to cooler parts of the shell and hence increased stresses in the shell, possibly leading to fatigue failure or gross distortion as stresses exceed design limits. In addition, as you now see temperatures of 320C then, as the lining wears, temperatures may exceed the accepted limit of 350C for carbon steel. Above 400C you might get distortion of the shell due to creep or general lack of material strength as the modulus and yield strength of the material decrease rapidly thereafter. Distortion of the shell would have implications on the lining as the bricks may no longer be held in place or hot material could penetrate between the bricks.
It's all a bit of a balancing act in keeping reasonable shell temperatures whilst minimising the stresses in the lining.

ex-corus (semi-detached)

 

[ione]

How have you measured the shell temperature?
If you have used the same type of thermo-couple you’ve used to measure your internal temperature, this could have led to misleading values as it couldn’t be suitable for the shell’s temperature range.