I recently had a discussion on LinkedIn about my definition of Pinch Analysis – “A largely academic exercise to minimise (rather than optimise) usage of certain resources”, which I decided not to include in my forthcoming dictionary of chemical engineering practice.
For anyone who wasn’t exposed to pinch analysis as university (which probably means you graduated before around 1975), it’s a way to analyse whether you can link heat exchangers together to reduce the use of primary heating sources.
They absolutely love it in academia: it forms a key part of the fairytale approach to process design which is all too often taught – and which then has to be untaught – to the maybe 50% of graduates who get to become chemical engineers. (I’ll come back to that issue in another article.)
Anyway, there’s nothing wrong with pinch analysis when used in the limited sphere to which it applies (post-construction redesign of unit operations, or at most subsystems in energy intensive industries), but when you make it the tail wagging the dog, as is taught in university nowadays, two problems inevitably follow.
The first of them is straightforward. As the graphic shows, you need to balance the cost of new heat exchanger capacity against the value of recovered energy, otherwise you lose track of the point of engineering, which is making money. The only time that the point of engineering is saving the planet is when there’s money in it.
The second is scarcely any less obvious, but it is too rarely priced in. The more connected my heat exchangers are, the less independent the bits of my plant are. This is most obvious when starting up the plant – either initially, or after maintenance or turnarounds. It’s all very well having a plant with a minimum input of primary heating when running at steady state, but how are you going to get it to steady state?
You have two choices. Either you install all of the primary heating kit which pinch analysis supposedly allowed you to do without, to be used solely during startups, or you hire in temporary kit for every startup.
Let’s hope that you allowed the ground footprint and utilities it will require! It will often be that case that if you price in these two things, the savings predicted by pinch analysis evaporate rapidly. All that will be left is the cost of the pointless pinch analysis exercise itself.
Professional engineers are supposed to know what isn’t worth investigating. Under all but a very limited set of circumstances, pinch falls under this heading. A few simple rules of thumb (which I cover in ‘An Applied Guide to Process and Plant Design’ will spot the low-hanging fruit of energy recovery in an operating context. In a design context, you are just making problems for the commissioning engineers. Which is never wise, commissioning engineers being who they are!
I have not personally worked on a plant which was unwisely integrated by a pinch enthusiast, but I have seen a similar lack of foresight in the hygienic industries. A decision was made to use UV as the sole disinfecting agent on a process water system before I became involved, supposedly to complete the fine detail of the design.
The problem with UV is that it only disinfects what it sees. The system was not protected against colonisation by bacteria, as it would have been if thermal or chemical disinfection was used. When I pointed this out, I was told that a yearly clean would be undertaken by a contract service. The person responsible for the design had talked to a supplier of such a service, and it was all quite do-able.
Leaving aside the problem of having a supposedly hygienic system which was only really clean once a year, it became clear that only the vaguest of chats had taken place, based on ‘back of a fag packet’ calcs, which wildly underestimated the system volume to be cleaned.
When I presented the supplier with the real volumes, it became clear that they didn’t have anything like enough kit to do the job. Furthermore, the site simply didn’t have enough space, water supply, power, or effluent treatment capacity to handle the necessary kit, even if someone could be found who could supply it.
The half-baked solution was devised by a green graduate, along with a more senior guy who had stretched himself too thin. No one took ownership for the design decisions, they all blamed each other and ultimately the entire plant – including many millions of pounds-worth of already bought kit – was scrapped. A few heads probably rolled too. That’s what happens when you try to be clever, but lose track of what’s important.
