Case study

A 200,000bpd refinery in Germany wasintending to change from oil to gas firing and
sought to replace its fired heaters due to extreme fouling issues. It was faced with a decision:
replace both fired heater sections or approach Tube Tech to develop a new cleaning solution, as an alternative to replacement.
It was important to have the fired heaters free from hydrocarbon fouling to ensure maximum heat transfer to reduce fuel consumption and increase asset life. The refinery operator opted for the latter, giving Tube Tech’s engineers only three weeks’ notice to build three multi-functional robots capable of achieving 80+% production standard clean or better, while not damaging the refractory.
The refinery had previously trialed many traditional lancing systems including dry ice,
chemicals, water jetting and abrasive shot blasting, all of which were limited in their success as they could only clean what the operator could see. Access deep inside and between the furnace rows was never achieved and as a result, the units had never been cleaned properly for 20 years – as can be seen by the heavy fouling shown in Figure 6.
The cleaning robots (shown in situ in Figure 7) took just 48 hours to clean three levels
on two fired heater containing six 15m long convection banks on three levels. This was despite expectations that 90% of all the furnace tubes had some fouling, with the deposits being 1-5mm thick and composed of hard scale and compact gummy residue between the finned tubes.
Tube Tech used remote CCTV to inspect the units before, during and after cleaning. No
refractory was damaged. The robots had no difficulty in obtaining access through 150 x 225mm doors.


Figure 6. Images showing the extent of fouling on the refinery’s fired heater’s convection

banks prior to robotic cleaning.
Figure 7. A Robotic cleaning unit positioned on the fouled finned convection bank



Figure 8. Videoprobe inspection of finned convection bank tubes after robotic cleaning

The robotic cleaning resulted in the inspection standard shown in Figure 8. The refinery
recorded 50°C and 80°C reductions in HVU2 and CD2 stack temperatures, respectively and
stated that the cleaning procedure had delivered several megawatts of energy savings.
Assuming that the loss of 1MWhr costs US$33.60 at an oil price of US$56/barrel, then it can be assumed that the energy savings alone result in financial savings of at least $588,672/pa.