The life & times of an HVAC Engineer

{May 9, 2012}   Happy Happy Joy Joy

The irony of blogging is that when you have things worth blogging about you’re too busy to blog, and things have been pretty busy for me lately. After about 8 months of working on oil & gas projects as a project engineer, and a couple more months working on technical audits & quality procedures, I’m now back in the familiar territory of pharmaceutical HVAC. I can’t talk specifics as, like so many other pharmaceutical projects, it’s confidential. What I can say though is that I’ve loving it, I’m back being involved in my favourite part of a project: identifying all of the requirements, and starting to find concept solutions.

The last couple of weeks have been jam-packed (not literally, it’s not a preserve factory) with site visits to discuss the client’s needs, reading through URSs (User Requirement Specifications) to glean as much information as possible about what we need to provide and going over Part L of the building regulations and the BREEAM (Building Research Establishment Energy Assessment Method) guidelines with a fine tooth comb to ensure our designs will be compliant. It makes for a steep learning curve on every project to get the whole team up to speed with what the purpose of the project is, what the client wants and what the constraints of the site and the funding is, it can be (and certainly has been on this occasion) quite an intensive time.

ImageFor me though, it’s a happy time, I like working out a precise brief, and I really love figuring out how we’re going to meet that brief. Also for this project I’ve had the mixed blessing of working with both a highly experience Pharmaceutical HVAC consultant, and having an equally highly experienced Sustainability & HVAC lead engineer. I say it’s a mixed blessing because though I’m benefitting hugely from their knowledge, they’re never in the office at the same time. This can be an interesting problem as they both approach the project from very different view points so I’m not just an engineer – I’m the diplomat & go-between for our talented founts of information. Hopefully though, by having my consistent presence to play the go-between role we’re able to take on board both knowledge sets, giving us a final solution that is both pharmaceutically sound and environmentally friendly – ideal since that’s the career specialism I aspire to!

Life has been a little busy of late, I’ve moved house, changed job role (I’m currently on a secondment as a ‘Project Engineer’ on some BP projects) and had a few other things on my mind. I hadn’t realised quite how long I’d been away from the blogging though until a colleague from another office commented on how they were missing my blog. So, this one is for you Stuart.

One of the challenges on my previous project was to ensure that no air could pass from room A to room B or vice versa. Not especially difficult you might think, considering there was a wall between the rooms without any doors or windows in. What there was though was a large hole in the wall between these rooms for a conveyor belt to pass through. This was made into even more of an interesting challenge given that the objects on the conveyor belt were very lightweight, meaning that they could easily be sucked up if you had very low pressure air.

Here’s a little sketch to explain:

Don't let the air move between A & B

To make sure that no air could ever go from room B to room A we just made sure that room B was at a lower air pressure. You can see this on the sketch above, room A is at 15 Pascals and room B is at 0 Pascals.

But how do you make sure that air isn’t going from room A to room B? Especially now that you have higher pressure in room A…this means that the air is pushing to get into room A. Normally you would deal with this by having a small airlock room between the two, but here we have a conveyor belt that is constantly running so you can’t do that. What we did instead was to install an extract duct within the wall so air was being pulled into the duct from both rooms, like this:

Use an extract duct to capture air in the opening

It’s a simple idea, but one that is difficult to get right. If suction airflow is too small then you will still have air flowing between the two rooms, if it is too high then you will suck all of the lightweight items on the conveyor belt into your duct. And of course it’s not just about the speed and volume of the airflow, it’s also about the shape of the opening and what airflow patterns that makes – much like the aerodynamics of a formula one car. And, much like the aerodynamics of a formula one car we had to model our designs using ‘Computational Fluid Dynamics’ (or CFD).

Doing this modelling allows us to see what the patterns of the airflows around our design will be, as well as how fast they will be. Usually the results are displayed using colours to indicate direction, pressure, speed or temperature depending on what you’re trying to find out. As it turned out, it was a very good job we did do the modelling, because the first design gave us this result:

The direction of the arrow shows the direction the air is flowing in, and the colour shows you how fast it is. So the red arrows going straight across the middle of the picture from left to right show air flows going straight through the opening at nearly 10 metres per second! This obviously isn’t what we were trying to achieve. We tweaked the design of the opening, and the volume of the air being extracted, and ran the model again. The final version shows flows like these:

Final airflows - x cross section view

Final airflows - y cross section view














So now we can see that the design should work. The arrows/airflows come into the opening from the sides and upwards into the extract duct as we want them to. Though of course a model is only as good as the information you put into it, so we need to make sure that the construction, installation and commissioning makes the reality as close to the modelled design as possible.

With thanks to Richard Ozaki at Mentor Graphics for the modelling and the images

When I’m asked what I do, and I say “I’m a building services engineer”, people often look at me blankly – it’s not a role they’re familiar with. So I explain it in brief by saying I design air-conditioning…this usually illicits a response akin to being told they’ve got to do the washing up – bored acceptance. People are aware that air-conditioning is sometimes necessary, but they assume it’s a dull job to be designing it.

That wasn’t always the case though, air-conditioning, cooling and ventilation designers were once so respected that their names, or at least their designs, live on hundreds of years later. These are a few of their stories;

The first air-conditioning design on record is a rotary fan intended for cooling. It had 7 wheels, each with a 3m diameter, and was manually powered. This was designed by a Chinese inventor named Ding Huane…right back in the 2nd Century, yet we still know his name now. Cooling designs, such as water powered fans and fountains were then used in Chinese palace design…one particular example being the ‘Cool Hall’ of Emperor Xuan Zang’s court in 747.

Of course the Chinese were not the only culture to have incorporated cooling into their architecture. The Ancient Romans’ ran aqueduct water through the walls of some buildings and the Medieval Persians used wind catcher towers, cisterns and water towers to provide cooling in their buildings – as can be seen in the diagram:

Persian Air-Conditioning - image credit: 'Cyrus' from

Getting gradually closer to our time, one of the first working air-conditioning systems was developed by Cornelius Drebbel and was demonstrated at Westminster in the 17th century. At the time Drebbel and his inventions were considered so exciting that he was employed by King James I & given rooms at one of the palaces so that he could entertain and astonish the court! His inventions also earned him an invitation to the court of the Holy Roman Emperor Rudolph II in Prague.

Last, but by no means least, the inventor of modern air-conditioning was Willis Haviland Carrier. He designed a machine using the same kind of refrigerant circuits as every air conditioning system uses today. His contribution was just over 100 years ago (his invention was in 1902), and his name is still known and for his work he has been inducted into America’s National Inventors Hall of Fame, received an honorary doctorate and been awarded the Frank P. Brown medal.

For me, being part of an area of engineering that has been going for hundreds of years and yet still keeps pushing technology further to remain on the cutting edge, makes my job really exciting. I also love to know that my predecessors made emperors comfortable and entertained kings, that tells me that whilst it may be a hidden art these days there was a time when being a building services engineer was a truly glorious role. Perhaps I can make it a little more glorious again, and my blog is where I shall start…

Snow Scientist by Million Moments

“December 2010 is every bit as cold and snowy as the worst December of the 20th Century – 1981 – and it could well turn out to be the coldest December since 1890” says Philip Eden, Vice President Royal Meteorological Society 2007-09.

The unusually cold pre-Christmas weather is affecting many things, airports and roads have closed, people have been unable to get to work and businesses have been unable to deliver presents. So amidst all the chaos, how is the HVAC coping?

To choose the outside temperatures we’ll use for designing the heating and cooling systems for a building we look at weather data (I, alike many other building services engineers, us the CIBSE & ASHRAE world-wide weather data & design conditions). This advises us of the minimum and maximum likely temperatures in a given area. The nearest weather station to my current project is Manchester & the ASHRAE data for there say that the temperature goes below -3.8 degrees Celsius for less than 99.6% of the time.

So that means that it will be colder than -3.8 Celsius for less than 1.5 days per year. But engineers always like to have a bit of a margin or buffer, so we chose a design temperature of -10 Celsius. All our air-handling units are designed to be able to operate, and provide comfortable working conditions (18-22 degrees Celsius) even at this low temperature. However, with local temperatures plummeting to -16.4 degrees Celcius even our design wont be able to function properly.

But what happens when the temperature is colder than you designed for?

    Letting the air handling unit (AHU) keeping sucking in air in extreme temperatures can actually damage it, or its components. If the AHU can’t heat the air up enough, or fast enough, ice can form inside it and destroy filters and sensors.

    The coils inside the AHU (these are like radiators, they have hot water running through them and they heat the air up as it passes over the many tiny ‘fins’ on the coil) wont be big enough to heat up the air to the temperatures needed inside the building. If you let the AHU keep going then you’ll be blowing cold air over the people inside…soon you’ll have ‘snow scientists’ like the picture above!
    switch AHU off

    This leaves you with a dilemma…you can’t leave the AHU on or you may damage it or the people in the building. But if you switch it off then you may not have enough fresh air (people need about 10 litres every second to feel comfortable) and recirculated air may not be clean enough to be able to carry on manufacturing. Also, if your ventilation was providing all of your heating then it will slowly start to get very cold inside.

The usual answer is that during very bad weather, such as the stuff we’re currently getting, it’s not possible to keep running the ventilation systems unless you’ve had the foresight to design them to cope with such low temperatures…but then if none of your staff can safely get to work perhaps the HVAC isn’t your biggest problem!

One thing I have to take into account as an HVAC engineer is how noisy the air distribution is going to be. You can’t usually see or feel air, but if it’s got some force behind it you can sometimes hear it – think about the sound of the wind howling around tall buildings, the squeal you can get out of a deflating balloon, the background noise around the car on the motorway (it’s not all from the tyres & engine!) or even the beautiful noise you can get from flutes, clarinets & the like.

Well since my job is often to make rooms more comfortable, by providing them with fresh air and keeping them at a nice temperature, it would somewhat defeat the object if I then made them really noisy. So to avoid the whoosh & hum of fast moving air it’s important for me to design the ducts to be big enough to keep the velocity low, and to avoid obstructions & corners (much like in my earlier post: Die Hard School of Ductwork Design). The lower the air speed, and the smoother the ductwork, the quieter the air will be…but the ductwork will take up more space – so there is a balance to be had. I also need to make sure there are attenuators before and after the fan to make sure the noise generated by the fan isn’t passed into the rooms. On top of this if there are quiet/private meeting rooms I need to consider installing ‘cross-talk’ attenuators to stop conversations from travelling from one room to another via the ventilation. If there is noisy equipment in the plantroom (the room that contains all the boilers, air handling units & other equipment to keep the facility working) then I might also need to consider acoustic louvres to make sure air can get in, but not much sound can get out.

But where else in engineering is sound critical? There’s some relatively obvious roles like sound engineer or telecommunications engineer…but have you thought about the impact of acoustics on car design? Many automotive engineering companies now employ NVH (noise, vibration & harshness) engineers to make sure cars give good feedback and a safe, enjoyable experience to their drivers and passengers.

But what if your car makes no noise? There is a lot of debate at the moment about what noise electric vehicles ought to make. There are concerns that if cars are silent they’re a far greater risk to pedestrians, especially blind & partially sighted pedestrians, who may not be aware that the vehicle is there. Also, many drivers actively enjoy the sound of their engines, so may not wish to drive silent electric vehicle. Not to worry, Elvin from Warwick University has come to our rescue. “Who is Elvin?”, I hear you ask. Elvin is and ELectric Vehicle with Interactive Noise, and he looks like this:

image credit: Warwick University

He’s a bit of a test project to gauge opinion on the acoustics (whether artificially added, or left ‘silent’) of electric vehicles and he lives on the campus of Warwick University. If you’d like, you can go and listen to Elvin and give Warwick your feedback on what you think he should sound like. As electric vehicles are likely to be the transport of tomorrow I think this is a brilliant opportunity to be involved in a little bit of engineering design. I’ll certainly be taking part!

{November 8, 2010}   Is this thing switched on?

In the facility I am currently working on, part of the process is to spray the product with a fine mist of 70% Isopropyl Alcohol, ‘IPA’. As you can imagine, that poses something of a hazard. To paint a picture of how much of a hazard, here are a couple of facts:

Lower Explosive Limit of IPA = 2%
[i.e. only 2% of the air volume needs to be IPA for it to still be flammable]

Flash Point of IPA = 12oC
[i.e. the room temperature only needs to be 12oC for the gas to vaporise & be ignitable]

Image credit: bruce7 from istockphoto

So, as it’s critical to spray the product with this hazardous substance, how do you go about making sure the operators don’t get blown up? Well there are a variety of different ways, so to name just a few;

  • Minimise the amount of spray used
  • Ensure all equipment within the hazardous zone created is safe for that environment (i.e. it is non-sparking / intrinsically safe / ATEX rated)
  • Provide extract ventilation to keep the amount of IPA in the room below the lower explosive limit

Well as a building services engineer, and thus a designer of ventilation systems the latter is the most relevant to me. So off I went & designed the ventilation to remove the IPA and protect the operators. Brilliant, Chloe saves the day…just one problem though…how do we know it’s working? And if it’s not working, how do we stop the machine from continuing to spray IPA into the room? Aah. Yes. Well…best do something about that hadn’t we.

So to make sure the machine doing the spraying knows that it’s safe to spray, we’ve included a flow sensor in the extract duct. The machine receives a signal from the sensor to say there is air flow, and then it can safely spray the product with IPA. We can all breath (an IPA free) sigh of relief. But no…what if the sensor is broken?! Okay guys…we’re getting into double jeopardy here, but as it’s for safety then the more the merrier, what do you suggest?

A couple of process engineers later and to ensure we have a double layer of protection to check the ventilation is working we are installing a sensor on the fan motor – that way we know it’s running. If the fan motor isn’t running then you know it’s not safe to spray the IPA.

I can’t help but thinking though, just because the fan motor is running doesn’t mean that there is extract ventilation…the fan or drive shaft could be broken. A little bit of me thinks that a few ribbons (perhaps that’s giving way to my girly side though) around the ventilation intake would be a visible indicator of the extract working that could never give a false signal. It would be reliant on the operators stopping the machine from spraying though, as ribbons can’t give a signal directly to the machine!

Image credit: The Seattle Times

I came to the realisation recently that engineers are very good at thinking everyone knows what they’re talking about…and I’m sure that criticism can be applied to me too. For all I write about what I’m thinking and doing as an engineer I don’t always remember that most people outside the process industry have never seen the innards of a process facility, and people outside of engineering generally haven’t had the chance to stick their head into any ductwork. So, for many folk their only experience of what an air-conditioning system looks like on the inside comes from Bruce Willis in Die Hard, Tom Cruise in Mission Impossible, Milla Jovovich in Resident Evil or even Homer & Bart escaping from Willy after stealing grease in the Simpsons. Now those scenes are not entirely accurate, though there was an escape from Alcatraz that utilised the ventilation shafts, but they can still be very helpful in explaining a few fundamental bits of ductwork design. So, without further ado, let me begin the Die Hard School of Ductwork Design:

From an HVAC engineer’s perspective it’s really important when designing ductwork layouts that you ensure air flows are as smooth as possible. The smoother they are, the more energy efficient and quieter the system will be…and the more likely the system is to work properly! The same goes for designing the ductwork from Bruce Willis’ perspective though. After all, whatever gets in the way of air is bound to get in the way of Mr. Willis, no matter how much of an action hero he is! So…if you were clambering around in air-conditioning ductwork, trying to escape from the bad guys, what might get in your way?

1) Corners

Right angles bad, curves good

Obvious as it may seem, it’s still worth a mention. It’s never really possible to lay all the ductwork out in straight lines with no corners, so they are a necessary evil. However, putting yourself in John McClane’s shoes (or lack thereof), how would you like the corners to be designed? Personally I think a nice gentle curve would be alot easier to get around than a sharp right angle, and from the look of this I think Mr. McClane agrees:
It’s certainly the case that airflow is alot smoother around a curve, which means it looses less pressure so less power is needed to get the air to wherever its going.

2) Joints

Internal flanges bad, smooth insides good

Anything that gets in Bruce’s way, and makes his life more difficult when navigating buildings via the ventilation will get in the way of the air. So when joining the lengths of ductwork together it’s best to put the joints on the outside. The same goes for any other obstructions in the duct work – if Mr Willis would have to put in extra effort to squeeze through then so will the air.

3) Access Hatches

Obstacles bad, access good

When trying to sneak up behind the bad guy through cunning use of ductwork the last thing you want is to be stopped by some impassable obstacle. So to make it possible for Bruce Willis/John McClane to out manoeuvre his enemies you should always put in an access hatch nearby. These access hatches are also rather essential for maintenance staff to keep everything in order without having to take down the duct work to access moving parts – in this instance a damper.

You can also help Bruce, Milla, Tom & Homer out by making ducts large with nice smooth inside surfaces. The less of a squeeze it is for Hollywood stars or air then the less energy it takes, and the same is true for keeping the friction low.

So if you’re ever asked to design some ductwork, bear Bruce in mind and think “What would John McClane want?”.

[Artwork created by my fiancé James Agg from my terrible sketches]

Ingenious Donkey

Making an ass out of me

A week or so ago I thought I’d finished writing the specification and producing the drawings for the air handling units for the project I’m currently working on. Then I sat down to review all of the specifications with engineers from other disciplines, and with the buyers. It quickly became apparent that I, and many of the other engineers, had made assumptions about what items had been included in other department’s specifications. It’s really not a problem discovering these things at the current stage of the project – we just add, or occasionally delete, items into our specifications to make sure that all the interfaces are covered. If we hadn’t stopped to have that review though, there would have been a few gaps that would have left us looking pretty silly once items were installed on site. After all, it’s no good specifying, paying for, and installing equipment if no-one provides a power supply to it!

One of the items that had been left out for example was the mesh in the low level extract scoops. You normally extract air from a room via grilles in the ceiling and the ductwork contractor provides all the necessary items. However, in clean rooms it’s often preferable to extract air at a low level, in which case the architectural contractor forms the ducts within the room as they’re making the rest of the room. That’s fine so long as they’re aware of all the bits you need within that ductwork – like a mesh to stop pieces of paper or rubber gloves or cleaning cloths being sucked up into the air extract system. Thankfully we found out that they hadn’t included the mesh in their specification and now it is in there. I wouldn’t have envied the commissioning engineers trying to figure out what was wrong with the new system only to discover the filters, which are intended for very tiny particles, were covered in rubber gloves!

Another assumption which has been made a few times recently is that I’m a secretary or document controller. Or more simply, when people haven’t seen me in a room, they often assume the meeting room or office is going to have no women in it. It can be a little frustrating having to regularly put people straight & explain that I’m not just there to take the meeting minutes but can also make useful contributions to the discussions as well. That said, I’m sure part of that is my age & youthful looks rather than just my gender – I have been ID’d when buying alcohol within the last 6 months after all! The second assumption, that there will be no women in any given engineering office/meeting room can actually provide a certain amount of amusement. The mischievous, mould-breaking streak in me rather enjoys seeing people blush beetroot red as they’ve said ‘morning gents’ then realised I’m there. I also find it rather curious how embarrassed many male engineers become having realised they’ve sworn in front of a woman as well. It’s not like my delicate donkey ears haven’t heard such words before after all…

Last but not least, I went on mentor training last week in preparation for having a 16-19 year old mentee from the local school’s Engineering Diploma programme. One of the main aspects of the training was about not assuming the mentees will know what we consider to be the most basic of work place behaviour. It was fascinating listening to previous students’ testimonials. Prior to the help of a mentor they had made mistakes on work experience such as answering a work phonecall by saying “yo”, or accidentally making a cup of mixed tea and coffee and then being too embarrassed by their error to do anything but drink the horrible concoction.

As you can see, over the last week it’s becoming increasingly clear to me that the saying, and title of this blog, “when you assume, you make an ass out of you and me” really is rather true. If I can try and make a few less assumptions perhaps I’ll avoid some embarrassing moments, not just for myself but for others too. After all, who wants to be an ass?

{October 5, 2010}   What’s behind door number 3?

As I’ve said on this blog before, it’s really important to keep pharmaceutical facilities clean. That means that when they are in production you can’t take down any of the ceiling tiles, and if you take one down outside of production hours then there must be a full clean down afterwards…which can take as long as 3 days, which is a lot of expensive lost production time. That might not sound like a particularly big problem, after all how often do you need to take a ceiling tile down? If you design the facility well, and keep items requiring maintenance out of the ceiling void wherever possible then there’s usually no need. Until you decide to refurbish or upgrade the facility that is. “What’s the problem with that?” I hear you cry – “Surely if you’re carrying out a refurb you’re going to stop production & take down the ceiling tiles anyway?!”. The problem is with doing the design.

Of course most facilities have “as-built” drawings of all the services that are in the ceiling void, so you should already know what is up there, and that is what you base your design around. More often than not though, until production is stopped and construction begins there are no opportunities for a full survey and there are almost always a few surprises along the way. Sometimes it’s little things like a cable tray where you didn’t expect one, sometimes the ducts or pipes have taken a slightly different route than is shown on the drawings. Or perhaps you just can’t find the smoke detectors, or the control panel for the doors. All of these things result in needing to tweak your designs, and tweak them fast…you probably have 3 months worth of work to fit into a 6 week shutdown (if you’re lucky) and now your lovely, simple, quick to install designs are out of the window.
That though, for me, is when things get exciting, I love a bit of a challenge, I enjoy solving problems…and when the team pulls together, and everyone from the client, to the contractor and ourselves in between is trying to make something new work as quickly as possible it’s actually quite a thrill. It’s even more pleasing when your fast maths and new layouts are being approved, then installed and before you know it you’re watching them operating successfully.

Yesterday morning the unexpected find revealed was an entire fan coil unit, moving over one thousand metres cubed of air each hour. By the afternoon I was busily working on the maths for a new solution, this morning the layouts were completed, now I’m working on client approval and getting contractor buy-in. It may only be a matter of days before the design is being installed, and right now I have a huge smile on my face.

et cetera
%d bloggers like this: