Fired Furnace Residual Heat

[Will Qin]

Hi all

I have two questions about fired furnace residual heat when calculating relief rate.

Fired furnace upstream of crude unit, power failure scenario, furnace goes offline.

1.For dynamic simulaiton, the holdup volume is equivalent to radiant & convection section furnace coil,
or just equivalent to radiant section coil?

2.One article mentions use of 50% of normal duty as refractory residual heat, so normal duty is total duty of
radiant & convectionn section? or just duty of radiant section?

 

[georgeverghese]

1. Should be volume of both radiant and convection section coils
2. Would agree that we use Q of radiant section only, since bulk of the heat transfer in the convection coils in normal operations comes not from the bricks, but from the flue gas, but only if the FD and/ or ID fan is stopped. But typically, the fans are kept running running when furnace is stopped, to prevent the coil contents from coking up, so then 50% of Qconvec would also apply.

 

[Will Qin]

Thanks for your share,georgeverghese.

As I know, volume of convection section coils is very big sometimes, so is it too conservative to use total volume when calculating relief rate caused by furnace residue heat?

Feed pump stops when power failure, so there is no pressure to charge liquid feed from convection section coils to radiant section coils. And air fans stops at the sametime when

power failue, in this case, is it required to consider convection section coils?

I‘m confused by this point, thanks.

 

[georgeverghese]

Feed pump will continue when there the furnace is blocked in, for non power failure case ?

Also combustion and draft fans will continue running when failure is not power failure ?

 

[Will Qin]

thanks georgeverghese.

What the case assumption is:

All pumps and fans stop for general power failure case.

And fuel gas/oil for fired heater is shut down.

 

[MPop]

As per some dynamic simulations you can use for the heater relief rate 45% of normal heat duty as residual refractory heat after power out. Earlier when static short-cut calculations were bused 50% was used.