Ingenuity in design

Given the current penchant for IT and proclivity towards management, it came as a surprise that my recent “technical” articles were so well received and elicited favorable responses. World’s moving at AI pace, and we are still trudging in the engineering & design quagmire. Well… it’s a quagmire I like, because of its exclusivity and the creativity it offers. So must be the case with my other appreciators I guess.

So many people reverted and told that they want to read more of such articles. Why ? Has the interest & quest for domain expertise in core technical engineering, not dwindled yet ? Would our younger lot still consider delving deeper into core design intricacies ? To me, it seems yes… given the opportunities that truly challenge the self, and if the learning path is made a bit more interesting to tread.

Here, I’ll share another troubleshooting experience at Melaka Refinery, where I was deputed as independent consultant from Foster Wheeler-Singapore. This was second of more than half a dozen issues I was given to solve, and I’ll share more later as becomes possible. Pardon if my writing appears staid, I haven’t yet been initiated into writing skills.

Two of the Stabilizer columns at the Refinery were under-performing. Stabilizers are last in line of product purification steps, as you know, and resulting product was effectively no more saleable. Offspec product was eroding Refinery’s profitability… the very purpose of business thus compromised. Refinery was having to manage the pain of off-spec product by blending, the least desirable resort… as it reduces make of another premium product.

Refinery had already consulted their regular experts, and had got (undisclosed) solution. Refinery wanted independent & unbiased set of solutions from me as their new consultant… who doesn’t like to play with their new toy ! Seriously, its a very prudent approach in my opinion, because though the Refinery was spending more on getting more sets of independent design solutions, it was cutting the Capex & downtime of future modifications by selecting best possible revamp option.

These two Stabilizer columns were in different units, meant to remove the lighter components from incoming feed, to achieve the desirable RVP in the stabilized bottom stream. Lighter components were getting out from top of the column, as is quite conventional, and stabilized bottom stream was going out as required onspec product containing only the allowed lighter content within it.

Here, problem was that the specification of the bottom product was not being met. Lighter component in bottom stream was exceeding the stipulated amount for which the columns had been designed, and hence bottom product’s RVP was more than acceptable. So the bottom products from these Stabilizer columns were offspec, and columns were under-performing.

Feed was much the same when compared to original design intent, so the possibility that this was happening due to change in feed composition compared to design intent - was discounted. Refinery also informed that the columns’ performance went down gradually over period of time.

To solve the performance issues of a distillation column, normal approach is to simulate it, analyze the vapor liquid traffic and compare actual with original design intent, compare with gamma ray scan results to see what trays can be underperforming leading to overall reduction in performance. Most typical solutions that emerge are to replace the trays with better trays or to retrofit structured packing. But I suggest that before getting on to sift through the chaff, spend time pondering over the scheme on paper. Most of the time, simple solution emerges in mind rather than on screen.

This one didn’t take long to solve, and so was bit surprising why wasn’t it solved earlier.

I’m giving here the indicative sketch of the configuration, and encourage you to first think of the solution yourself, before moving on to next paragraph for solution.

Most typically in such problems, issues are presumed to be with trays – that may be broken or valves might be stuck or downcomers might be leaking or weirs might have eroded etc etc.

But an important observation is that same problem was happening in both the columns. You could expect broken trays in one column at a time, but its simultaneous happening in two columns is something like double jeopardy in Process engineer’s parlance. Refinery confirmed checking them during previous inspections and there wasn’t any damage to trays.

Another important observation here is that as the Refinery had informed - columns performance went down gradually over time. Gradual reduction is the key here. Gradual reduction is definitely not breakage, rather would happen due to something loosing its functionality…. more indicative of something like deposition getting build-up & accumulating with time to cause progressive deterioration – something like fouling.

Yet anther important point here is that the reboilers are Stab-in type. Stab-in reboilers see largely stagnant flow on their outer surface, because they are immersed in the pool of liquid in column bottom sump. Fouling loves to live & thrive in stagnant flow – as all Process engineers know.

Underperforming reboiler is usually associated with reduced reflux, reduced temperature profile especially in the stripping section, and reduced steam flow to reboiler, as also appeared to be the case here.

So, the problem analysis was that the reboilers were getting fouled gradually… more & more over time due to stagnancy of flow on their outer surface – leading to reducing heat transfer – leading to reduced vapor generation in column bottom – leading to reduced vapor liquid traffic inside the column – leading to reduced reflux – leading to reduced stripping & rectification – leading to reduced mass transfer – leading to slippage of light components into the bottom stream and hence the off-spec bottom product.

Where should this fouling coming from ? You might have heard of polyolefin polymerization concept. At elevated temperatures beyond approx 125 deg C, olefinic molecules start joining at their ends – that’s polymerization and start creating bigger & bigger molecules – and these bigger molecules deposit on hot surfaces. This is the reason why in olefin services, temperatures of heating medium is kept low. This is the reason why in olefin services, compressor discharge temperatures are contained below 125 to 130 deg C – to prevent polyolefin polymerization.

Here, due to stagnant flow on reboiler’s outer surface, the tube metal temperatures would be high – close to bulk temperature of the incoming hot fluid. That’s what was causing polyolefin polymerization, leading to fouling on the reboiler’s outer surface. This is the explanation I gave to the Refinery, which they bought. I think this is also the reason why the solution had remained elusive so far. Earlier trouble shooters thought the fluid is clean, so discounted fouling. They probably did not think of polyolefin polymerization as the cause of fouling – that’s my guess.

So I proceeded to suggest the solution : To replace the stab-in reboilers by thermosiphon reboilers. Thermosiphon reboiler has multiple times more heat transfer coefficient by comparison, and would do the job well without getting fouled over time as it sustains high turbulence on its cold side due to two phase flow. Reconfigured sketch is at the bottom of article. Solution was presented to the Refinery, and they unhesitatingly agreed. Afterall, the payback was less than two months ! You see now, the point I made initially, regarding Refinery’s intelligence in spending more on getting design alternatives from various independent consultants – it paid them back handsomely.

Reboiler change as suggested was implemented and that solved the issue for good.

While thermo-siphon would be the first choice in such situations, sometimes the physical constraints may not allow installation of thermosiphon reboilers, in which case forced circulation can be a viable option too.

I reflected upon what prompted me to write down such instances after all these years. Perhaps to share ingenuity in design - to share that solutions can be quite simple for seemingly unsurmountable issues. That applies for job and life both. Or perhaps the realization that we are ephemeral and so is the knowledge unless its documented. I have shared these often with people directly in touch with me. But penning these down may disperse such interesting cases more pervasively for wider consumption. You may repost my articles if you like, to widen disperse amongst your network.

Would love to read your thoughts, lets make it two way.