Friday 22 March 2019

Aqueous Systems


Before reading this episode, I need to state that it comes with a caution. There is going to be some jargon and a small amount of science and chemistry.  Stick with it if you can, you might even enjoy it - I promise to keep it to a minimum.  I would also like to thank Mark and Phil for sharing with me some of the memories of days in Plymouth, various sites around London and of course, Bahrain.
If you have just dipped into this at random, you might want to first read my earlier post, Devon Interlude which will provide a little information as to how you arrived on a website devoting so much time talking about a small business in the south-west corner of England.
Aqueous Systems was a small privately owned company specialising in the design and installation of small to medium-sized water treatment systems.  Much of Britain’s water supply, particularly in the south-east corner, comes from water which is high in limestone and consequently the water is both hard and alkaline.  Anyone who has lived in a region with hard water will know of the effect that this has on kettles, irons and hot water systems, and of course taste.  The impact on industrial water systems is the same and the water therefore requires treatment to remove such hardness and alkalinity.   In the 1970s, the treatment methods varied depending on the end use of the water and types of boilers, but most cases involved the use of pressure vessels which contained tiny beads of material known as ion-exchange resin which capture the calcium (and magnesium) from the water and exchange it for the softer forms of sodium (water softening) or exchanges the hardness forming ions for hydrogen (dealkalisation).  It is not my purpose to provide an essay on water treatment processes.  Not only would such an attempt be dull and boring, but it would almost certainly be incorrect.  Having said that, I will step into uncharted territory and attempt a simplified chemistry lesson for those who may have an interest in such things.  Don’t try this at home!  For those who really don’t need all the detail, jump ahead three or four paragraphs – I promise you that you won’t have missed much.

Technical Bit

In water there are two types of hardness known as temporary and permanent hardness.  Temporary hardness is caused by dissolved calcium bicarbonate and can be removed by boiling.  It is not always the case, but temporary hardness is often referred to as alkalinity, and before the purists give me examples of where this is not so, let’s just say that it is.  I warned you that I am not going to write a thesis.  
Permanent hardness is associated with dissolved calcium sulphate and needs removal by other methods including water softening by ion exchange.  The water in the Greater London area is very high in alkalinity and hardness and, to provide the optimum water solution for most effective boiler performance, the water treatment method frequently used was a process known as dealkalisation.  This involves feeding the water first through an enclosed pressure vessel containing specific ion exchange resin which removes only the calcium ions associated with temporary hardness.  These are replaced with hydrogen ions which combine with the bicarbonate ions to produce carbon dioxide and water.  Are you still with me?  This has the effect of making the water more acidic so the next stage in the process is to pass the water through a degassing tower, where the carbon dioxide is exhausted out of the top of tower via a blower in the base and the acidity of the water is thus neutralised.  The permanent hardness is then removed by passing the water through a second pressure vessel containing resin which exchanges the calcium associated with the sulphates in the water for sodium.  This process is known as softening.  The treated water is then ready for delivery to the boiler water feed tanks. 
Eventually the resin in the pressure vessels reaches saturation and is no longer able to exchange calcium ions.  When this point is reached, a process known as regeneration takes place.  In the dealkalisation vessel, the unit is flushed with a dilute solution of acid (usually hydrochloric acid, but sometimes sulphuric) and the calcium ions attached to the resin are replaced with hydrogen ions from the acid.  Similarly, the water softener is regenerated using a salt solution (sodium chloride) where the sodium in the salt replaces the calcium and thus regenerates the resin ready for the next cycle.  The process requires management of a number of chemicals predominantly acid and salt, but also caustic soda which is used to neutralise the water from the acid regeneration process in the dealkalisation unit.  
OK, the chemistry lesson is over, thanks for re-joining us. 
You could be forgiven for asking why all the detail, but I think it is worth appreciating the complexity of the process.  OK, it’s not as complex as rocket science I know, but getting the process wrong can be quite unforgiving and I have a couple of great examples which I will share for you later.  That said, the real challenge with any water treatment plant is using the right materials of construction and getting the mechanical and electrical design correct – it is rarely the process.  
Many of the major hospitals in the Greater London region, where their boilers produced hot water and steam for laundries and heating were customers of Aqueous Systems.  They are all gone now, but establishments at Leavesden, Napsbury, Neasden and Cell Barnes all had Aqueous Systems dealkalisation plants.
At the time I joined the organisation, and presumably the reason for my recruitment, was that the company had recently experienced significant growth in sales and had been struggling to meet some of its contractual deadlines.  
The premises were on a small piece of land at Estover, an industrial estate near Plymouth Airport on the northern outskirts of the city.  The estate included major companies such as Wrigley’s and Fine Tubes – and smaller businesses such as ours.  The building was made of concrete block which included a reception area and offices for the managing director, his personal assistant and an accounts person.  The office area opened into a warehouse area where the treatment plants were assembled and dispatched.  The design office was housed in the car park at the back of the building in a demountable portacabin which had just about enough space for three people which included the engineering manager, a draftsman, a project engineer and now me.  The company started by making domestic water softeners and small filtration systems.  There was also a separate business based in Jersey, called Aqueous Systems International about which we knew very little, other than it being a frequent destination for the boss (“I’m sorry Mr Morris isn’t available, he’s in Jersey”). 
There were a number of talented individuals working at Aqueous Systems and I was to get to know and appreciate working with them all over the course of the next three years.  There are three or four people that I particularly want to mention. 
Mark Boardman was a project engineer.  A local Plymouth lad, who had previously worked in the building services industry.  With his National Health spectacles and straggly hair down to his shoulders, he looked more like a sound engineer for a Led Zeppelin concert than a water engineer.  He typically turned up for work in a tee-shirt and jeans and whether the shirt was on inside out or not depended entirely on whether he had remembered to leave it the right way around, the previous night when he had dropped it on the floor beside his bed.  His passion was small sports cars, and during the whole time I worked with him in Plymouth and later in Scotland, his project was re-building a scary little Lotus 7.  He may not have looked the part, but Mark was one of the most determined and dependable individuals I ever worked with.  He had very definite taste in music, enjoying progressive bands like Pink Floyd, Ten Years After and Soft Machine and abhorring what he called the pretentious sounds of Moody Blues and Emerson, Lake and Palmer.  I struggled to keep up, but he was many years ahead of his time in this respect, and much of the music he introduced me to then remains on my playlists today.  Mark had a dry and self-deprecating sense of humour which delivered in his Devon accent, made him a very hard person not to like.  Above all, Mark was a problem solver who would never give up – an invaluable asset when commissioning a troublesome treatment plant – and we had a lot of them.  He also had an extremely morbid fear of spiders.
Phil Perry was a design draftsman.  This was in the days when drafters worked at a drawing board with a tee square, triangle and a sharp pencil. Wide use of computer aided design with software programs such as AutoCAD were several years in the future.  Another cheerful fellow, Phil was also a musician, a folk singer accompanying himself on guitar.  He sold me a beautifully mellow Eko Italian-made Ranger 6 dreadnaught guitar, which I still have to this day, and which I am sure is now worth a lot more than the £25 that I gave him for it.  One of Phil’s major challenges was a fear of being in a confined space, which made even sitting in the rear seat of a two door car or travelling through the Dartford Tunnel an impossibility.  On the occasions when we worked on site and had to overnight, there was always a requirement to make sure that Phil’s room was on the ground floor, if there wasn’t one available, he would sleep in the car.  The combination of Mark’s arachnophobia and Phil’s claustrophobia made for some interesting scenarios.
Shortly after I had joined Aqueous Systems, whether by accident or design, the engineering manager decided to move on and took a position in another company.   I must have done something to impress the boss because shortly after this I was promoted and moved to the end desk of the hut, which made me furthest from the door, but closest to the heater which was handy when those cold northerlies blew in across the moor.
Based in Farnborough, about 40 minutes from London, and in the heart of the commuter belt, Aqueous Systems’ sales director, John Baker operated from a tiny serviced office.  John was exactly how I might have imagined him.  Portly, pink-faced, well-groomed moustache and outfitted more like a banker in his pin-stripe suit than the water expert he claimed to be but wasn’t, John was full of bonhomie and bullshit, but also engaging and a first rate net-worker.  It has to be said though, that the word “no” was not overworked in John’s lexis which no doubt put a smile on Mr Morris’s face, but proved to be a relentless strain on the resources and capacity of our engineering and project teams.  That doesn’t mean that John wasn’t good at his job.  His job was to make sure we won lots of work.  How we executed the work was our responsibility, but we might have appreciated the odd communication from him from time to time, perhaps enquiring whether we were able to add a few more deadlines to our already overworked program.  We should not judge John harshly though.  I have many times since those days been in comparable situations and have come to learn that the only thing worse than winning lots of work is not winning lots of work!
The solution to not having enough staff is to get more.  With my promotion, we gained a new Chief Engineer (me) but lost a project manager (also me), so we hired Bob Stone. 
Bob was about three years older than me, having previously worked in building services contracting before joining the company about six months after I did.  Originally from the Buckinghamshire area, he had lived and worked in Plymouth for several years and was weary of what he called messy little heating and ventilation jobs. He wanted to get his teeth into something more substantial – it wasn’t going to be long before that happened.  Bob was calm, unflappable, capable, a wonderful work-mate and yet another guitar player. Bob claimed to have once performed as an opening act at a pub in Slough for The Rolling Stones – I’m sure he did. 
Many a prickly contract issue was solved over a lunch time pint with Bob, Mark, Phil and other staff at the Lopes Arms in Roborough, about five minutes from the office and a thousand miles from deadlines and schedules.  There, through a healthy cloud of second-hand Panatella fumes, chucking down pots of Double Diamond and Carling, we focused on treble twenties and double tops while debating the relative merits of rubber-lined or ABS pipework for regenerating ion-exchangers or whether Status Quo twelve bar rollers could genuinely be classed as serious music.
None of this should suggest that we were not serious about our work.  We were constantly involved in design, fabrication, delivery, construction and commissioning of up to a half a dozen projects, each of which regularly demanded a three or four hours’ drive up the overfilled A303 and around the horrible North Circular Road to project sites which were neither easily accessed nor particularly welcoming.  
One of the most challenging of these projects (until the arrival of Derek Thomas and the Bahrain job) involved a contract to design and build a dealkalisation plant to treat the boiler feed water at Woolwich Arsenal.  I’m not referring here to the nineteenth century football club which moved to Highbury around the time of the First World War and became Arsenal FC.  This project was at the Arsenal itself, more properly described as The Royal Arsenal, where armaments manufacture, munitions testing, and explosives research were carried out for the armed forces until its closure in the mid-1990s.  Although they had long since stopped manufacturing armaments, it was still a major storage and testing site.  A critical requirement of armaments warehousing is temperature control and if ever we had to ensure that our equipment was fit for purpose, it was now.
We gained a lot of knowledge from this project, not the least being the fact that getting the process right is pretty much the most important part of a water treatment system design.  I have said many times in the past, and will continue to do so, that treatment plants rarely fail because someone got the process design wrong.  The most common reasons for failure are usually the result of bad mechanical or structural design, poor selection of materials, inadequate electrical and instrumentation or just plain bad installation.  For example, an engineer will correctly size a pump, basing his or her decision on flow and pressure requirements, only to neglect to adequately size inlet piping by not providing satisfactory suction head, producing an effect known as cavitation (a bit like sucking froth and bubbles through a paper straw at the end of milkshake) causing all kinds of damage to the pump and its connections.  Failure to provide flow control equipment which was fit for purpose in the treated water storage tanks was another great example of poor design and selection of materials, all of which had nothing to do with the process design.
However, every now and again, a poor process selection really can lead to a disastrous outcome.  I mentioned earlier that I would try to keep to a narrow path in terms of describing some of the technical stuff.  I will try again to describe our acid regeneration problems at Woolwich without diving too deep – I really don’t want to lose my readers.

Another Technical Bit

The dealkalisation process involves feeding the water through a pressure vessel containing resin which removes only the calcium ions associated with alkalinity.  Once the resin in the pressure vessel reaches saturation and is no longer able to exchange calcium ions a process known as regeneration takes place.  In the dealkalisation vessel, the unit is flushed with a dilute solution of acid and the calcium ions attached to the resin are replaced with hydrogen ions from the acid.  The acid used is generally hydrochloric acid, but sometimes sulphuric acid is used.  Sulphuric acid is much cheaper than hydrochloric although in a dilute form it is highly corrosive to steel.  This is not the case in the concentrated form and thus it can be stored in steel vessels which don’t require high levels of complex (and expensive) corrosion protection such as rubber-lining.  So why didn’t we always use sulphuric acid, if it is so much cheaper?  The reason will become apparent as this story unfolds.
As a chemical, hydrochloric acid is usually supplied at about 30 or 35% concentration which, when used to regenerate the resin, is diluted with treated water to about 5%.  The hydrogen (H) in the hydrochloric acid (HCl) washes the saturated calcium (Ca) from the resin which then combines with the chloride (Cl) part of the acid to form calcium chloride (CaCl2).  Written as a simplified formula where (R) represents the resin, the process looks like this: 
2HCl + Ca(R) 2H(R) + CaCl2
Calcium chloride is a benign compound which stays dissolved in the water used during the regeneration process and causes no problems during the process.
Sulphuric acid on the other hand is delivered at about 96% concentration and when used to regenerate the resin needs to be diluted to 1% or even lower.  This requires careful choice of materials since as I have already stated, the dilute acid is highly corrosive.  The nasty part about using sulphuric acid is that while the hydrogen in the sulphuric acid (H2SO4) does the same job as the hydrochloric by effectively washing away the calcium from the resin and replacing it with hydrogen, the sulphate (SO4) then combines with the calcium to form calcium sulphate (CaSO4), a compound which can very easily precipitate out of the solution to form a solid.  The solid is gypsum, a product formed naturally from hot springs where there are sulphur deposits.  The formula looks like this:
H2SO4 + Ca(R) 2H(R) + CaSO4
Thus, the dilution of sulphuric acid during regeneration needs to be at a level which unconditionally avoids any chance of the calcium sulphate precipitating during the process. 
Regrettably, this was not the case in the design of our plant at Woolwich.
Mark and Phil were on site during the start-up and commissioning process which began soon after the plant had been installed.  For the first twenty four hours everything worked perfectly apart from one issue which could have been disastrous were it not for their quick reactions (see sidebar).
As the resin became saturated the regeneration process began.  The sulphuric acid was not dilute enough, and within minutes, calcium sulphate began forming in the bottom of the pressure vessel. The vessel internals used plastic nozzles fixed into a thick polypropylene plate which supported the resin bed.  The nozzles rapidly became clogged with gypsum, the plate collapsed under the pressure and the whole process ground to a nasty halt.  The result was that the vessel had to be dismantled and the nozzles and plate replaced.  The ion exchange resin was replaced since much of it had washed away when the nozzles failed, and we installed a new acid dilution system and storage tank so that we could replace the sulphuric acid system with hydrochloric – but our troubles were still not over.   
Our sales team had obligingly added a premium component to our price, so we had no excuses for cutting corners and using shoddy or low-cost equipment.  Sadly, that didn’t stop our procurement team, always on the lookout for a deal, supplying the cheapest level controllers money could buy for our treated water storage tanks. 
After treatment to comply with the tough specification of the Department of Defence, the water was pumped to a high-level water storage tank on top of a gantry below the roof of the boiler house and accessible only by climbing a vertical steel ladder.  The water in this tank was used as feedwater to the boilers. This was where our float switch was installed.  If you are familiar with the old style ball-float operation of an old-fashioned toilet cistern you can imagine the design of this fragile little device.  When the level of water in the tank rose to a pre-set level the ball-float would activate a micro-switch which would turn off the pumps delivering the water from the treatment plant. Likewise, when the water in the tank dropped to a low level and the ball-float was low in the tank the float switch would flick the micro-switch in the opposite direction and the pumps would start again, ensuring that there was always a healthy head of feedwater in the tank.  At least that was the way it was supposed to work and I’m sure that in perfectly still water in the right environment, with little or no consequence of failure it would have been adequate – maybe.  But these tanks were in a barely accessible area, in a saturated steamy environment with water pouring into the tank though a three inch diameter pipe at about 120 gallons per minute.  At best, the consequence of failure of feedwater to the boilers, meaning no temperature control for the stored munitions, may have been to have rendered the ordinance inoperable.  At worst, an unplanned demolition of a significant portion of East London south of the river perhaps.  I’m exaggerating, but we were certainly under no illusion that keeping the boilers supplied with high quality water was critical to the success of the project.
None of this happened luckily, because Mark and Phil were still on site when the float switch became inoperable.  Of course, this was long before the days of mobile phones and the benefits of texting and I wasn’t made aware that there was an issue until I received a call at home about 11 o’clock that night after a pleasant hour or two at the Cornwood Inn supporting the local ladies’ darts team of which Pauline was an honoured member and I, a proud supporter. 
Hello Mike,” said Mark, when I answered the ringing phone as we walked in the door.  “We’ve got a spot more bother down here and could use some help directly” he said in his slow Devon tone, making it sound like there was no problem and we had all the time in the world.  Sure Mark,” I said, “what’s wrong?
“Only me ‘n Phil been takin’ it in turns squatting up on the treated water tank playing with that float switch, operating it by hand all night!  Phil aren’t too happy, and I’m not so pleased either as it happens.
It turns out that Phil and Mark had quickly learned while commissioning the plant that the float switch on the tank was sticking and the only way to stop it from either running out of water, or over-flowing was to operate the switch manually while crouching in a cramped sweltering space between the top of the tank and the roof as the water level in the tank rose and fell.  They had been doing this for most of the afternoon and all evening waiting for someone to relieve them with a control system that actually worked.  Mark had finally got through to someone (me) but the bad news was that we wouldn’t get a new float switch up there until sometime later the following day, so they were in for a long night. 
I’d had a few too many lagers at the Cornwood Inn to be dashing off to London than night so I woke Bob Stone up and explained the situation, asking him to organise a replacement part for me to pick up later in the day.  I left home about five the following morning and was ringing Bob again at about eight o’clock from a motorway service area outside of London.  We had decided to use level electrodes instead of a mechanical float switch and since the manufacturer was also in the south east area, near Lewisham, I was able to call into their factory and collect the equipment we needed and deliver it to Woolwich later that morning.  It had been a hell of a night for Phil and Mark, but fortunately we had arranged for a couple extra guys to join us on site.  We had the unserviceable switches out of there and replaced with reliable control mechanism by lunch time.
Come on guys,” I said, “There’s a new burger joint opened last year on Woolwich High Street where the old Wimpey Bar used to be.  It’s called McDonalds.  Lunch is on me.”
They didn’t complain at all.
Any financial contribution that this project was going to make to Aqueous Systems’ operating profit had long been reduced to zero, but an important lesson was learned.  We all agreed it was an expensive experience, and we had no trouble understanding why Mr Morris wasn’t exactly enthusiastic about the outcome, but then again, Mr Morris did not spend time crouching on top of a treated water tank in the middle of the night wrestling with a recalcitrant float switch, claustrophobia and spiders.
Of course, I describe these projects as though all the activities took place of a short span of time, but as anyone who has worked on similar contracts will know, that’s not how it happens.  The installation and commissioning period on most of these projects took place over many months and even when the plant was in operation, there was always a period of months, sometimes years, known as the maintenance period, or the guarantee period where although the equipment had been handed over to the client, we were still under an obligation to attend site and rectify any defects in the equipment or in the operation of the plant.
I have spent a lot of time here writing about poor design choices and inferior equipment selection – and indeed, clichés such as “you get what you pay for” only become so from frequent use and this is because they are usually true.  A similar such platitude is that while the memory of a few dollars saved on a project is fleeting, the pain it causes will last a very long time.
I'd also like to take a moment to talk about occupational health and safety practices in 1975. It's difficult to imagine how some of the activities which were carried out on worksites in those days would stand up to even a moment's scrutiny today.  Generally speaking, tradespeople and their co-workers were not more cavalier in their practices than today; they were not gung-ho rip, tear and bust individuals, nor were they ignorant and stupid. But without doubt work place behaviour, and with it the use of appropriate protective equipment and safe engineering played a less significant role than any of us would tolerate today.  Asbestos was still being used as a common insulation material and training and procedures in confined spaces was still some years in the future.  As a society, we continue to look at behaviour and practices which improve the health and safety of our people, whether this is at remote sites or in the office, and no doubt we will continue to find ways to improve in this area.  I am not here to judge the actions of our predecessors and those who went before them and I beseech you not to judge them either.  
Having got that off my chest, it’s important to mention some of the excellent suppliers that I had the pleasure of working with during this time.  One such manufacturer immediately comes to mind.  All treatment plant which incorporates a regeneration, or a cleaning process will require a timer if the process is to take place automatically.  A domestic water softener or a swimming pool filter is usually back-washed and regenerated manually using a single multi-port valve.  Large process vessels such as our dealkalisation units, were back-washed and regenerated using a series of separately plumbed valves, operated pneumatically and controlled by a process timer.  Today, a programmable control, about the size of a small finger nail carries out this work as part of a larger plant control process – easy, reliable and about as exciting as watching concrete set.
In 1975, such innovations were still a few years away.  Automatic process control was carried out using a mechanical timer and the masters of the art of manufacturing such equipment was A G Burrell & Co, Process Timer Manufacturers of Sheffield.  The company had been manufacturing mechanical process control equipment of various levels of complexity since the 1920s and their equipment were horological works of art.  Manufactured of high quality brass and stainless steel, precision ground to the finest level, once installed and in operation these units would go on for ever.  I had occasion to visit their works in Sheffield several times.  They were courteous and attentive and old Mr Burrell, presumably the son of the founder, made you feel as though you were the most important client they had.  If there are not some of these units still in place and operating somewhere, there should be.
Shortly after my promotion to Engineering Manager, the boss called me into his office.  Barry Morris was not an unpleasant fellow, but he was hard to read.  A very private and reticent man, he was given to holding numerous meetings behind closed doors, or in whispered asides, usually with John Baker.  He was slightly framed, in his early forties and judging by the furniture in his office, and the posh Volvo with the opening roof which he drove, he was a man who wanted his environment to be as comfortable as possible.  Perhaps, I was a little resentful of this, given that the rest of us were crammed into a small portacabin out the back, and we had all been requested by his Australian secretary, Val to be sure not to make use of the executive bathroom, but instead use the facilities in the workshop.  However, it did give me a sense of aspiration. 
On this occasion, BGM wanted to introduce me to a man I was going to be spending a lot of time with over the next three years.  Derek Thomas looked every inch like a man who would not have been out of place lecturing before an assembly of like-minded individuals on high energy particle physics or string theory.  In his late forties, or early fifties Derek had fine wavy grey hair covering his ears and most of his collar, thick black half-rimmed glasses, and a taste in ties that even in the 1970s, made you want to look anywhere else rather than at the thing hanging around his neck.  He had a fondness for jackets with leather patches at the elbows which he wore even on the hottest days and was never without his Hewlett Packard reverse Polish notation calculator, which like his avant-garde red Citroen DS, was he said, essential to survival.  Derek was an industrial chemist who had previously been the technical director at one Britain’s leading high-quality water treatment companies, Elga before starting his only small company, Aqua Conversion Engineering which specialised in small laboratory-sized reverse osmosis systems.
Until I met Derek, I had only the vaguest idea about reverse osmosis (RO) and other membrane water treatment processes such as micro-filtration (MF).  In 1976, RO was in its early years of commercial and practical development with large US companies including Dow and Dupont being major suppliers and manufacturers.  It would soon become the most widely used treatment process for purification of saline water.  It continues to be so today.
Our relationship with Derek and this exciting new technology was going to lead us to places that none of us had previously ventured – Bahrain was waiting for us.  Were we ready?
Sharing ideas and second-hand Castella smoke.




No comments:

Post a Comment