The KISS Principle

By Timothy Allinson, P.E.

Murray Company, Long Beach, Calif.

Surely everyone reading this article is familiar with the KISS acronym: Keep It Simple Silly (or similar second S adjective). This acronym holds as true for engineering as it does to the literary world.

 

For example, George Orwell has been quoted as saying, “In writing keep the spoken word constantly in mind. Never use a long word where a short one will do… One of the few writers of our time who has tried seriously to write English as it is spoken is T. S. Elliot. Lines like: ‘And nobody came and nobody went, but he took in the milk and he paid the rent,’ are about as near to spoken English as print can come… Never use the word eliminate when you mean kill.”

 

I was reminded of this passage while listening to the audio book of Anita Shreve’s The Pilot’s Wife while driving my long commute. The opening lines of the book are so simple and elegant that they are rare for literature today. Written in poetic cadence for emphasis rather than standard sentences, the book begins:

 

She heard a knocking,

and then a dog barking.

Her dream left her,

skittering behind a closing door.

It had been a good dream,

warm and close,

and she minded.

She fought the waking.

 

Hearing these words professionally read, giving them even more meaning than I would reading them to myself, sent me into a mental sidebar about why all writing can’t be so perfectly simple. The sidebar continued into thoughts of work, as much as I hated to be distracted from the audio book.

 

Lately my responsibilities at work have required that I review many design-assist hospital projects. I don’t much enjoy critiquing the work of another engineer, but it is a necessary evil of the design-assist process. In reviewing design after design I was reminded over and over of the KISS principle and the fact that many engineers disregard it.

 

One hospital, for example, had trap primers indicated for every single floor drain and floor sink on the project — even drains serving the discharge of prep sinks in a commercial kitchen. When I questioned the engineer about the overuse of trap primers I was told, “That is our office standard.” I wonder if it would be their office standard if they had to pay for and maintain all of those unnecessary trap primers? Probably not.

 

Another KISS violation I see frequently in hospital design is the over use of medical gas alarms. NFPA is quite clear that area alarms are only required for medical gasses in critical care areas; however, I see alarms provided at every single zone valve box on a regular basis. Valves serving regular patient rooms are not required to be alarmed, but many engineers indicate alarms regardless. It’s possible that on some projects they might be requested by the hospital, but in most cases I think it is lack of understanding on the engineer’s part, overcomplicating the design as a result.

 

Don’t forget to KISS when it comes to multi-floor typical fixture arrangements. Most engineers will know to provide vertical stacks to serve multiple levels of patient toilet rooms. In laying out the stacks, think “prefab.” More material is less of a concern that the ability to prefabricate the waste and vent trees such that each can be put into place as a single prefabricated unit. Labor costs more than pipe, and field labor costs more than shop labor.

 

Outside the patient rooms, the hospital “core” is often typical on each level, and many engineers fail to notice this. Frequently I see the engineer provide one vertical stack against a column, with fairly long runs of waste and vent pipe to the core fixtures on every patient room level. This solution is not only expensive from a labor and material standpoint, but it is difficult to coordinate with the other trades. Whenever the core fixtures align vertically, a stack should be provided behind each fixture or group of fixtures to again maximize the ability for prefabrication while simultaneously easing coordination.

 

Don’t be afraid to use ½" water supply pipe (or med gas for that matter – except vacuum). Many engineers and AHJs treat ½" pipe as taboo, but in reality it is economical, it eases in-wall coordination, and provides hot water delivery in half the time of ¾" branch lines. Of course pipe velocity and friction loss must be respected, but ½” pipe can flow 2 gpm - enough water for your average tempered shower or sink — at a velocity below 3 fps and a friction factor less than 4 psi/100', which is sufficient to satisfy most design conditions. An old engineer associate of mine once told me when I was only 21 years old, “A ¾" line is actually quite large — it probably provides all the water to your home.” He was and remains correct. So if a ¾" line is not small, a ½" line should not be treated like its poor cousin.

 

Anther “fear factor” in our industry is the use of 1¼" or even 1½" pipe for water closet flush valves. Many engineers immediately default to 2" pipe. This issue is hard to argue since Hunter’s Curve falls short of the zero Y axis, making the sizing of water pipe for one or two water closets quite difficult. We designed and built a hospital using 1¼" pipe for a single water closet and 1½" pipe for back-to-back water closets mainly because that was what the bridging (bid) documents showed. During construction the sizing was questioned by a third party engineer. Since the debate is hard to solve using Hunter’s Curve, we instead tested it in the field. We installed two WCs with a PRV to mimic the minimum pressure condition. When tested, the fixtures flushed perfectly and the issue was resolved.

Yet another KISS violation I frequently see, especially in hospital design, particularly med gas design, is the application of factors of safety. Many engineers possess this fear or belief that if a system is designed precisely it is somehow bad or potentially inadequate. This creates a desire to over-design with factors of safety to create a comfort level. The reality is that systems designed to Code or standards such as NFPA already have inherent factors of safety and are not “bare minimum” as some engineers suggest. Except for unusual circumstances, Code design and NFPA design is more than “just adequate” and will deliver a high quality and efficient design.

 

Unusual circumstances would include stadiums, for one. In the case of a stadium, where all the fixtures are used nearly simultaneously at half-time, certain “above the Code” precautions must be taken. Most stadiums today have “flushathons” before they open to the public. These are events where thousands of volunteers are invited to celebrate the completion of construction by manning every single fixture in the facility and simultaneously repeatedly operating the fixture. This simulates near zero diversity, and while the reality of usage will not be quite as demanding as this when you factor in fixture usage time, it is a performance test that must be satisfied. In such a case the engineer will have to take measures beyond the Code into consideration for the design of both the water and drainage systems. The details of these considerations are beyond the scope of this article and will instead be the focus of another, but suffice it to say that it is a fairly rare example of where the Code would not suffice.

 

Lastly, I would like to reiterate an issue I have mentioned many times, that of hot water circulation. This is without a doubt the most common mistake in engineering design, and the mistakes are twofold. First, never, ever, ever circulate water through a pressure reducing valve. The system will never balance. Second, be sure to use the KISS principle when selecting the circulation pump(s). Thorough circulation calculations are quite complex and equally unnecessary. Rather than calculating flow based on temperature loss, base it on the number of hot water risers or loops — at least 1 gpm each if they are small, and 2 gpm for large risers or loops. The reason for this is that systems engineered too precisely, with only a fraction of a gpm through each loop or riser, will never be made to balance in the field. This is one area where a little over-design is warranted. With the gpm known, the pump head usually becomes a simple function of the pump curve, and most are directly proportional: 30 gpm with a 30-foot head is a good example. But if your system is quite large, the pump head should be checked with a quick friction calculation through the longest loop.

 

George Bernard Shaw once said that writing is either simple or it’s impossible. I think engineering is the similar in this respect, and keeping the KISS principle in mind is a good way of putting this relevant quote into practice.

 

Timothy Allinson is a senior professional engineer with Murray Co., Mechanical Contractors, in Long Beach, Calif. He holds a bsme from Tufts University and an mba from New York University. He is a professional engineer licensed in both mechanical and fire protection engineering in various states, and is a leed accredited professional. Allinson is a past-president of ASPE, both the New York and Orange County Chapters.