• At our refinery we recently observed damage on the top of the radiant roof tubes of our VDU fired heater (2 parallel cabin radiant sections and 1 common convection section), clearly caused by long term operation at too high tube wall temperature. This is different from the issues we had in the past, where hotspots were mainly located on the bottom of the radiant tubes.

    During the run we can't monitor this location with IR camera, thus the high temperatures could not be detected.

    After restarting the unit the newly installed skin temperatures confirmed this area of the furnace is running at ca. 100 degC above the rest. Possible contributing factors are the presence of roof tiles just above the interested 8" and 10" tubes and the fact that the furnace was revamped to low-NOx burners 2 TARs ago.

    Do you have any experience of high temperatures on the top side of the radiant tubes and possible mitigation measures / modifications to be carried out?



  • Afshin Bakhtiari, Naftoon Arya Eng. Co,, bakhtiari_af@yahoo.com

    The flame length of LNB burners are longer than that of conventional ones (0.5 ft/MBH longer) and since the value of H/W (height to width of firebox) remains constant, flame length to firebox height ratio is increased. So the roof refractory is exposed to more heat causing by flame impingement (not necessarily visible) and more re-radiation to top of tubes is inevitable and as a matter of fact, top of tubes get hotter.

    According to API RP 535 the maximum value of F/L should be 0.66 and if this value is changed in new condition by installation of LNB burners, the burner design needs to be reviewed.



  • S M Kumar, Energy Environment Engineers, s.m.kumar@outlook.com

    If the roof tubes are on the path of flue gas exiting radiant section (firebox) they could get both radiant and convection heat. If they hang close to the roof refractory as you mention, they could get additional heat from hot unshielded refractory + refractory on the bridge between the 2 fireboxes.

    Presume the roof tubes are outlet tubes. VDUs outlets are generally at roof to minimize transfer line run to VDU column. You also mention 8” expanding to 10”. Usually near outlet. The outlet tubes run hot. In VDU, the fluid temperature in the last coils is higher than COT, Coil Outlet Temperature due to rapid vaporization near outlet.

    As others have pointed out, check about changes in (1) feedstock and turndown that influences flow regime — Baker Flow regime that can result in poor wetting of tube metal and hence coking. (2) Wet or dry vacuum — injection water / steam.

    Read the excellent paper, “Why Vacuum Unit Fired Heaters Coke, Tony Barletta”. Google Search or https://www.digitalrefining.com/article/1000319/why-vacuum-unit-fired-heaters-coke

    You can have a look at slides 50 thru 58 in Fired Heaters — Operations.pdf in my website https://sites.google.com/view/oilgasprocessengineers for additional pointers.

    As mentioned in my site under Design Tips/ Fired Heaters — (1) Equal Flow Vs Equal COT: Go for equal flow in all passes. Usually, pass inlet FCV keeps equal flow. Minor variations in coil outlet temperatures (COT) are OK, fouling or coking in one or two tubes. If you reduce flow to a coked pass on manual control to increase its COT, it will coke more, a vicious cycle. (2) A coked oil tip not regularly cleaned can result in directed flames and hot gas. Missing or broken burner tile directs flame towards tubes. Mostly, it is hot gas rather than flame impingement. You can observe with saw or coal dust thrown in air. (Courtesy: Dr RD Reed, "Furnace Operations"). Use sodium bicarbonate, Daniel Gonnet in LinkedIn. Sparklers show flue gas path and impingement, if any.

    This should help you find the path forward. If you don’t have in-house expertise, take the help of your heater supplier or consultant in your county. Good Luck


  • Jake Gotham, InSite Technical Services, jake.gotham@insitetechnical.com

    If the overheating is on the top of the roof tubes, it sounds like you have stratified flow, hence less heat-sink at the top of the tube than the bottom.  This would be the result of lower velocity or lower vaporisation.  Are you operating at lower throughput than before?  Or is there less light material in the feed (e.g. improvements in stripping at the atmospheric tower)?  If your furnace has velocity steam connections, this could be used to increase the velocity and move back into a dispersed flow regime but obviously this would add to the load in the tower and ejectors.  Some units also have the ability to send the overflash back to the furnace inlet — this might also help.

    If the top of the tubes has been running hot for some time, you may have formed some coke in this area, and should plan to decoke the furnace at the next shutdown.



  • Nagarathinam S Murthy, McKinsey & Company, Chennai, nssvdvr@gmail.com

    Two independent radiation sections with common convection section is indicated as per design. This implies that there is likelihood of flow imbalance between the two. Also, please look into the COT of both the radiant sections... any significant variations will lead to such from overfiring in one.


  • Michael Olesky, Consultant, mike.olesky@verizon.net

    Are you running with very low excess air? Is it possible that you are experiencing afterburning at the very top of the radiant section due to air leaking through penetrations in the roof? The afterburning could certainly cause hot spots on the tubes. My thoughts on this matter ...