NEAT bike!

I’d been sedentary for most of my adult life. From 2004-2010 I worked in a job where I was mainly sitting. I gained a considerable amount of weight over those years, as a result of bad eating habits and little activity. Not quite to the extent that I could appear on a reality weight loss show, but enough that the textbooks would classify me as obese.

The bigger I got, the less I wanted to go outside and exercise, which only made things worse. I was getting about 30 minutes of exercise a day commuting back and forth to work on a bike, and though there were the occasional trips to the shops or town, I was not moving much at all, and certainly not enough to compensate for my love of food.

Last year I moved to a small town with my wife, and started working from home. This exacerbated the problem, as I no longer had to commute to and from work. My days were spent in front of a computer, and save the occasional outings and trips to the (close-by) shops, I was getting no exercise at all.

I had dieted in 2008 and dropped a fair bit of weight, but I slowly regained most of that over the next two years, and in 2011 with my even more sedentary lifestyle, I was back to my highest ever weight. My body fat was higher than it had previously been, as a result of the yo-yo diet and zero activity. I had high blood pressure too, and knew that if I didn’t do something about my lifestyle I would be heading to an early grave. 28 years after being brought into this world, it was time to finally grow up and start looking after my body properly.

Last year there were reports in the media about a study that showed the amount of time spent sitting per day had a direct influence on the likelihood of death. What was surprising about the study was that it showed the amount of supplementary exercise was not a significant factor - so even if you try to get a few hours of exercise in a day, if you spent the rest of the day sitting then you’re apparently heading for an early grave.

This news was disappointing. I need to be in front of a computer to get my job done, and while I could justify taking a bit of time out of each day to exercise (not that I was doing so at the time, though), I didn’t have the luxury of cutting my computer time down by a lot in the interests of health. I sighed and went on with my unhealthy ways.

Earlier this year, I stumbled across the concept of working while walking on a treadmill. I later learned that working while exercising has been championed by a Mayo researcher by the name of Dr Levine. In his paper "NEAT - liberating the life force", he explains how movement-based energy burning is comprised of two parts - deliberate exercise like playing tennis or doing weights, and incidental exercise such as standing, walking to the fridge, and so on. The latter is referred to as non-exercise activity thermogenesis, or NEAT. He shows how the NEAT portion of our energy burning is dramatically lower than it was in the past, as the result of our more convenient and sedentary lifestyles. For anyone who’s seen "The gods must be crazy", one of the early scenes where a woman drives a few houses down the street to post a letter is a funny, but not so off the mark example of this.

Levine argues that if a workplace can be designed to encourage low intensity exercise in the course of one’s duties, a person’s daily energy consumption can be significantly increased. Depending on their food consumption, this could mean a slowing of weight gain, or even result in weight loss. And time spent walking on a treadmill is also time that’s not spent sitting down.

There are companies that provide treadmill and desk sets that are designed to work together, like the one pictured in the previous section. With quiet engines and controls integrated into the desk, they look like nice if somewhat expensive options. I couldn’t find anything like that available over here however, so I decided to try put something together myself.

The next time I was at the local sports store, I had a look at the treadmills they offered. There wasn’t a huge selection, and I ended up opting for one that wasn’t too expensive, as it was a display model, and somewhat dusty. It was powered, which I figured would be important while at a computer for balance reasons, and had hand rails which I figured I could bridge with something to make a laptop mount.

In retrospect I really should have done my research first. I did not test it plugged in at the shop, and when trying it at home it was a lot louder than I’d expected - to the extent that if I wasn’t careful about the hours, the neighbors were likely to complain.

The display has a section which a magnet on a cord sticks to. You’re supposed to attach the other end of the cord to your person, and in case of an emergency you pull on the cord and the machine shuts down. This might be a nice feature for the elderly, but it messed with my plans of building a treadmill desk. I didn’t want a magnet getting anywhere near my laptop, which meant that the laptop would have to be mounted with adequate clearance, making it cumbersome to use.

I decided to give it a try without a laptop. I stuck an iPod in the drink holder area to the left of the magnet, and started the machine up. I was getting terrible static in the earphones, which was likely a result of the poorly shielded motor below. This made me even more reluctant to put any electronics near the thing. I was disappointed.

The final nail in the coffin was the 30 minute usage limit. After 30 minutes the motor and belt get hot, and needs to cool down for some time. The machine is designed with the assumption it’ll only be used once a day, and thus using it for hours a day was going to drastically reduce its lifespan. This was the case even if I was careful to let it cool down between uses. I gave up on the project, and the treadmill started gathering dust.

Later on, when I hit my highest weight, I decided to at least try to get 30 minutes of exercise a day in. This wasn’t going to save me from death by sitting, but I figured it was better than nothing.

I’d given up on the laptop idea, and instead tried power-walking while holding an iPad. This was cumbersome and didn’t work out well. My next idea was to mount the iPad above the magnet, and hope that the electrical interference didn’t kill it. I made a mount with some cardboard and gaffer tape, which while not winning any style awards, was pretty effective. It consisted of a lower pouch, and upper stand area, so the iPad could be slipped in and out easily.

Levine advocates the use of low intensity exercise over a long period. As I had only allotted myself 30 minutes a day, I bumped up the speed to compensate. I was getting a better workout, but this made reading webpages a bit difficult as my head was moving about, and keeping my hands still enough to scroll the page was difficult.

I later stumbled across a Mac app called Air Video that allows you to stream video from a computer to an iPad without any cumbersome conversion steps - you simply load the app on the iPad and can remotely browse and play videos that are on your computer. This worked better that browsing websites, but I rarely watch videos, so while it was better than nothing, I felt frustrated that I was zoning out to tv shows instead of doing something more productive - even if I was getting some exercise in at the same time.

I came across a discussion on the net about standing desks, and decided I was going to try them instead of the treadmill. I piled a bunch of boxes and books on my desk, and put the laptop on them. I used it for two days like that, and was pretty happy with how it was working out, so I decided to try something more permanent. In retrospect, like before, I probably should have done more research before jumping in.

I determined the optimum height to be able to type comfortably on my laptop without craning my neck, and I went to the local furniture store in search of something appropriate. I found a set of shelves that were pretty close to the perfect height, and brought them home and assembled them.

The first two days were great. I found using a standing desk to be really liberating. You can pace about while you think, and if you suddenly need to fetch something there’s no burden of energy to overcome. When I used to spend the day sitting, I’d find myself putting off fetching things until later because I was lazy.

Unfortunately, after a few days my feet started to grow sore. I knew it might take a while to adapt, and I tried to just grin and bear it, but the pain kept getting worse. I quickly switched from standing barefoot to wearing indoor sneakers, which provided a lot more support but weren’t able to stop the pain. Within a week it had gotten to the point where I could barely walk and my feet would ache even when lying down, and I was worried I was going to do permanent damage.

I also realized that when concentrating, I was stationary, and had a tendency to lock my knees (overstraighten them to the point they can’t bend anymore), which is bad for circulation. As I wasn’t moving, there was no periodic relief of pressure on the heel of my feet. And while standing up uses slightly more energy than sitting, without any movement there’s not a huge difference. I tried to make a conscious effort to shift my weight about, but it didn’t seem to help much, and my movements were limited by what I could do while still being able to type.

The fact that I was overweight no doubt made things worse. Someone at their optimum weight would have less pressure on their legs and feet.

At least until my heels healed, I was going to have to go back to sitting. I bought an exercise ball and used that instead of a chair, something I had done back in my university days.

My feet did get better after a few days, and I tried to start moving the computer back and forth between a sitting and standing position. This worked out reasonably well, though I frequently caught myself being lazy and spending too much time sitting and not enough time standing.

I continued this way for a little while. I was no longer in pain, and I’d improved upon my old glued-to-the-chair lifestyle, but I also found myself wishing that I could be walking instead of just standing. I wished I’d done more research before buying the treadmill, and I was very reluctant to think about buying another one.

I briefly considered getting a stepper to use instead of just standing up. I wasn’t sure how easy it would be to type while using one. Unfortunately when I looked into them more closely, they all had an upper weight limit a fair bit lower than my weight at the time. The sturdier ones had handles and other stuff sticking out the front which would have made using a laptop difficult, and I gave up on the idea. I later saw a photo of Dr Levine with a stepper in his office, but I’m not sure whether that was used for computing or not.

One day I came across a post on reddit, where a guy had raised his desk with some cinderblocks, and stuck a spin bike underneath. The thing looked like a death trap, but he claimed that in one month he’d considerably increased his fitness and been able to do a marathon ride after only that time. He’d graphed his weight and fat% changes over the time, and they looked impressive. It rekindled my interest in something slightly more intense than periodic standing.

I decided to do a bunch of research before diving in this time, given the problems I’d had with previous attempts. I searched for laptops and exercise bikes, and looked at the options available. One option would be to buy an under-the-desk set of pedals. Online opinions of this were largely negative however, with people citing the noise, propensity to move about, lack of durability and an inability to put up anything but mild resistance.

Another option are exercise bikes specifically designed for laptop use, like the FitDesk. It’s not available where I live, but when I looked at videos of it in use it didn’t look very appealing anyway - typing looked difficult and it looked flimsy.

It looked like my best option would be to get a regular exercise bike, and then try to mod it. I couldn’t find many examples of people doing that online (which is why I decided to publish this page), but there were a few videos such as this one and this one which gave me a general idea of what it would be like ergonomically.

Anything that can be fixed with gaffer tape, I’m your man, but that’s as far as my handyman skills extend. I lack both the tools and experience to be cutting/welding/drilling, but I figured that I would at least be able to tape something together. I sat on the sofa, trying to emulate the posture of pedaling on a recumbent bike, and imaging where the laptop would sit. I would need to think about how high my knees come up so I don’t end up hitting them, and whether the control panel of the bike would get in the way.

With those things in mind, I started to search for bikes online. Wary due to my previous experience with a 30 minute time limit, I hunted for recumbent bicycles without such restrictions that looked like they would last when being used for multiple hours a day.

The search was disappointing. All of the moderately priced bikes had time limits on their use, and did not look at all durable. To an extent I guess it’s a case of "you get what you pay for", but I’d seen videos of exercise bikes in America that looked durable while being inexpensive, so I was disappointed to not find anything similar over here. There was also a dearth of recumbents.

It looked like my only option would be get something designed for the commercial market instead of the home market. Bikes from companies like Life Fitness and SportsArt are available here, but only particular models are imported and the importers seem to only bother with the high end models. An intermediate residential or light commercial use bike would have done me fine, but that didn’t seem to be an option, and so I was faced with the prospect of paying the premium for a bike designed for commercial use, or using something cheaper but fragile and/or uncomfortable.

I thought about it for a few days, and finally decided that I was better off paying the premium than getting something I’d be disappointed with like the treadmill. This was something I’d be using every day for the foreseeable future, and if it was going to mean a big improvement in my health, the cost per day over a few years was not too expensive.

I’d originally been thinking of getting the R3 or R1 from Life Fitness. They looked sturdy and the seats looked comfortable. I couldn’t find much in the way of manuals or technical docs about the bikes, so I emailed them asking whether the bikes would be suitable for continuous use and a low speed. I got a reply fairly quickly pointing out that if I was going to pedal slowly, the R1 (wall-powered) option would be the better bet, as the self-powered R3 won’t operate unless you’re peddling fast enough.

I then came across the SportsArt C532R. I didn’t like the aesthetics as much as the previous bikes, but I was impressed by the documentation available on their site, which was much more comprehensive than LF’s site. This was important because I suspected I would need to remove and rewire the display to get it out of the way of a laptop, and if the front section was welded on then I’d be in trouble. I looked through the manual and repair manuals, and it looked like I would be ok with the display - it was bolted on and the wires were connected by a clip connector, so in the worst case I’d be able to extend the wires with a soldering iron.

One other appealing point was that the bike could operate both in self-powered and wall-powered modes. The treadmill guzzled power, so a bike that doesn’t cost anything to operate was appealing. I wanted to use the bike disconnected, but I wasn’t sure at the time whether I’d be able to peddle fast enough without it interrupting my work, and the knowledge that I could buy an external power adaptor if it turned out to be a problem was reassuring. The minimum speed turned out to be slower than I expected though, so I didn’t end up needing the adaptor.

The SA bike also offered a bunch of features that cost a premium on the LF bikes or were just unavailable, such as wireless heart rate monitoring and controls on the handlebars, which meant I wouldn’t have to reach over to the display. Heart rate monitoring isn’t particularly useful for low intensity exercise while working, but I also wanted to use the bike for small amounts of higher intensity cardio work, and figured it would be a useful option to have. The SA bike also had a very generous warranty, including a lifetime guarantee on the frame and mechanical parts. I deliberated some more, and finally hit the buy button.

The bike arrived a week later, and took me about 1 1/2 hours to assemble. It was really heavy, which was a pain while assembling it, but gave me confidence in its durability. When I got it all assembled, I jumped on to try it out, and was quite pleased. It was quiet, comfortable, and completely stable.

I rolled it into the next room and pondered how I’d use a laptop with it. I decided that rather than trying to secure something to the display stand (which would both hit my knees and be uncomfortably close), I’d try the cinderblock approach. I didn’t have any cinderblocks, but I did have a bunch of styrofoam from the box the bike came in. I selected some of the more durable-looking pieces, placed them on either side of the bike, and then hefted a small table on top of them.

At this point the display was still attached as intended, which meant that the table couldn’t be pushed too far to the front of the bike without it hitting the display. While it wasn’t a particularly deep table, with the display getting in the way the table came right up to the seat, which made getting on and off a bit difficult, and typing uncomfortable. The setup also wouldn’t have withstood a moderate earthquake, but it was acceptable as a prototype, and gave me a good idea of what it would be like to work while cycling.

Despite the prototype’s limitations, I was extremely satisfied. Unlike a treadmill where your whole body moves about at all but the slowest speeds making typing difficult, I could type over a wide range of speeds and resistance settings.

The next step was to get the control box/display area out the way. The metal pole was attached from the inside of the control box, which meant I had to disassemble the box in order to separate the two. The bolts on the pole were extremely tight and I was worried they wouldn’t budge, but I finally got them moving.

I’d originally thought of mounting the control box up on the table where the laptop was, but I realized if I just attached the control box to the body without the pole, then I wouldn’t need to chop up cables and extend them. So I pulled out my trusty gaffer tape, and oh-so-elegantly attached the control box directly to the body. It cleared the pedals/my feet by about 3cm.

With the pole removed, I could move the table so that the laptop was more ergonomically placed. The table was a tad high in order to avoid hitting my knees, but I was really happy with how it turned out.

A few days later I headed back to the home center, measuring tape in hand, looking for a set of shelves that I could use without needing to put styrofoam underneath. I found a sturdy set of metal shelves. They were longer than I’d been hoping for, but that worked out to be a good thing in the end.

Constructing the shelves was an adventure, as I didn’t have the right size spanner. I set up the three top rungs and left the bottom two empty - that’s where the bike would go. With the bolts only hand-tightened, the shelves wobbled about like nobody’s business, and I was worried the lack of the bottom two rungs had dangerously destabilized them. But I finally found that a grip for a set of screwdrivers I had was the right size, and when I used that to tighten things up, the shelves got a lot more stable. If I try to move them they wobble a little, but when I’m typing on the computer while cycling, they don’t move.

I slid the bike under one end of the shelves, such that when I’m cycling my knees at their highest rise to the bottom part of the shelf, but about 3cm in front of the shelf so there’s no danger of them hitting anything. This allows me to extend my arms forward at a comfortable distance. My hands are about 10cm lower than my shoulders, and elbows slightly bent.

Another reason I’m glad I went with the SportsArt bike instead of a LifeFitness one is because of their "step through" feature. I thought LF’s downwards sloping chair was more attractive, but it means you have to lift a leg over when sitting down and standing up. Because of the proximity of the shelves, handlebars and wall, this would be hard to do, so I really appreciate the low frame on the SA bike.

Because less than half of the shelf is taken up by the bike, I can use the rest of the shelf as a standing desk. The laptop sits on a linen placemat, and I can easily pull it back and forth between a cycling and standing position, which means I don’t have to interrupt my workflow to transition from one position to the other.

To the right of the standing desk is a small desk & swiss ball, as pictured in a photo above. This gives me three different positions which I can move between, which minimizes fatigue. I tend to limit sitting to 1 or 2 hours at the start and end of each day, and meal breaks.

Before purchasing the bike, I’d assumed I’d be sitting down for 8 hours a day, cycling away. This proved to be more difficult than I thought, and my tail bone would get progressively more uncomfortable the longer I tried to ride. This was partly mitigated by adjusting posture, and it’s possible to tilt the seat back to place more of the load on one’s back, but this comes at the expense of being able to type.

The bike defaults to 30 minute sessions, and I’d try to get through them without taking a break, but I was pretty uncomfortable by the end, and needed to stand for a while to give my butt a rest. This tended to get worse the further into the day I got, and I ended up reducing the time between breaks to compensate.

I also tried placing a memory foam cushion down on the seat, which did help a bit, but still wasn’t perfect. I tried that for a few days, and then went hunting online for something better. I was hoping for a cushion with cutouts to relieve pressure on the tailbone, and I found something that looked like it could work. It ended out working well - the foam is strong and elastic (much more so than the cushion I was using previously), and the cutouts seem to do a good job of keeping pressure off the bone. I’m back to 30 minute sessions before I take a break now, with very little discomfort.

The only remaining issue is that occasionally my toes start to go numb. This is easily rectified by adjusting the position of my feet on the pedals, and perhaps I could improve the situation if I loosened the straps a little. Apparently it’s a common issue on recumbent bikes.

For the first few days I kept the bike at the minimum resistance setting, and tried to pedal at 30-35rpm, which is the bare minimum to keep the machine on. I tried to keep it at that minimum because I was worried that the harder I made it, the more tired I’d get, and I didn’t want to end up working hard for 30 minutes and then be chair bound for the rest of the day as I was too tired.

30rpm is quite slow, and combined with the minimum resistance it was very easy to keep up that pace. On the first day of setting up the prototype, I cycled for 5 hours, in 30 minute sessions. Within a few days I was cycling at 40-50rpm, simply because it was a more comfortable cadence. After a few days I started bumping up the resistance by 1 or 2 notches every few days. Within two weeks, I’d gone from putting out about 20W of power to about 90-130W, and the muscles in my legs had grown noticeably bigger.

One of the advantages of a NEAT bike over a treadmill desk is the fact that you have a lot more room to ramp up the intensity without affecting upper body stability. Even at a high intensity, the upper body is quite stable, and so reading and even typing is possible. That said, the higher the intensity, the harder it becomes to concentrate and accomplish intellectual tasks, so it’s not possible to write software while doing a workout from hell. But if you want to browse your favorite websites while doing so, that is a possibility.

I found that up to a moderate level (heart rate in the lower 110s), I could maintain my concentration and work just like I’d work if I were sitting in a chair. From what I can tell this is a considerably harder workout than you’d get from a treadmill desk at the recommended 0.7-1.2mph speed.

Of course the harder you work, the more heat you’ll be putting out. I leave a fan sitting on a nearby desk, pointed at me to keep me cool, and the air conditioner is running in the background. It’s the middle of summer at the moment, so hopefully as it cools down the aircon can be left out of the equation. If I weren’t working from home and had other people in the same office, and using a fan was not an option, I suspect I’d need to reduce the intensity somewhat to avoid sweating and becoming uncomfortable.

For the first two weeks I pushed myself to use the bike for 4 hours a day, every day. Combined with standing breaks (which can vary from a few minutes to half an hour or more if I lose track of time), and other distractions, the 4 hours ends up taking up most of a working day.

At higher resistance levels or speeds, while you may not notice perspiration if you have a fan pointed at you, you can lose water quite quickly. On higher intensity days I go through a number of jugs of water a day, and I sometimes find myself having to push myself to consume enough water to stay hydrated - especially after meals, where your stomach may be full even though the body needs hydration.

In my enthusiasm for the new toy, I pushed myself pretty hard for the first few weeks and rapidly increased resistance levels. My muscles and strength were noticeably increasing, and my weight loss had accelerated by an impressive amount. But after about 2 weeks, I stopped losing weight and wasn’t feeling as well as I had before.

What probably caused this is a hormone called cortisol (pictured right), which is released when the body is under stress. I was on a fairly restrictive calorie diet, and while I did increase my calorie intake somewhat when I started using the bike daily, I increased it by only about 1/4 of the extra calories I’d started burning.

The stress of the higher energy deficit combined with the exercise had likely lead to a mild case of overtraining syndrome, where the stress put on the body exceeds the body’s ability to recover.

Cortisol has all sorts of negative effects when released due to prolonged stress. It encourages the body to put on fat, which sabotages diets and pushes healthy people towards excess body fat. It also disrupts sleep, and suppresses immune system function and testosterone, leading to a state where your body is more likely to burn muscle than build it. In my desire to become healthy, I was making myself sick!

To address this, I decided to tone things down a bit. I added in 2 rest days a week where I avoid the bike and alternate between sitting and standing, and I reduced the resistance and speed when I do use the bike. This way I keep moving, without making the muscles work so hard, and I avoid building such a large calorie deficit. On days I cycle, I limit higher intensity exercise to 30 minutes, and don’t force myself to do that every single day. Any other time spent cycling is at a slower pace.

When I’m finished restricting my calorie intake and I’m back to a normal amount of food, I should be able to ramp things up somewhat without causing any harm, but with recent studies showing that too much intense exercise could be worse than none at all, I’ll need to be careful if I start trying to push higher again.

Exercise bikes can use a number of different types of resistance. The one I use uses electromagnetic resistance, which is reasonably quiet, but not silent. The pitch gets higher as you pedal faster, and feels like it’s louder - but not to the point where it’s going to make your ears uncomfortable. The bike can’t be heard from the next room if the door is closed, so I presume it’s not irritating the neighbors.

At 30rpm the sound is very quiet and barely audible. A phone conversation partner probably wouldn’t notice. At 45rpm the sound is a bit louder but still quite quiet. By 60rpm the sound makes it harder to enjoy quiet music, and it would make a phone conversation harder. Since I got the bike I’ve had a few video conversations with people, but I haven’t tried to use the bike while doing so. As the mic is on the computer rather than next to your mouth with a phone, I imagine the noise would be picked up and be a distraction.

Unlike a treadmill which tends to put out a consistent noise when you’re walking slowly, the pitch can vary up and down if like most people you tend to put more power into one part of a pedal stroke. The variation in pitch makes the sound more noticeable than if it was at a consistent pitch. The pedals have straps though, and if you use proper form and distribute the load more evenly over the stroke, the sound becomes more consistent. The variation is more noticeable when you’re pedaling slowly.

As the bike I use is self powered, it has a built in alternator to top up its battery. I suspect an electromagnetic bike without an alternator could be a bit quieter, but this is merely a suspicion as I have not tried any other bikes.

While I’ve only had the bike for a few weeks at this point, I’ve been impressed with the results. Combined with the 30 minutes on the treadmill I was doing before I got the bike, and dietary changes, my blood pressure is back to normal. My resting heart rate is about 60bpm, indicating that my cardiovascular system has become more efficient. I feel a whole lot healthier.

An unexpected but positive change has been on energy levels. I was worried that by moving about I’d become exhausted, but it actually had the opposite effect - moving about actually keeps you alert. When I was sitting around all day I’d experience drops in energy, and those are nowhere to be seen now.

The bike helpfully keeps a running total of time used, estimated distance cycled, and the amount of calories you’ve burnt which is based on your weight and age. I record these values each day to see how hard I worked over the day. During 10 days of the 2 week period where I was pushing myself, I cycled for 39 hours and 822km. At that rate I would have virtually cycled from the top to the bottom of the Japanese mainland in the course of a month! Though as I’ve dropped my cadence from 55-60 to 45-50 recently, my pace has slowed down a bit.

At my peak I was working at 500-750 calories/hr for 4 hours a day, so about 2000+ calories a day (illustrated in the snapshot to the right). I’m aiming to keep the majority of the time I’m on the bike at about 200-350/hr now.

When I went on a 1 hour ride outside a few days ago, I could definitely tell the difference compared to a month prior. I was able to pedal considerably longer and harder than I used to be able to, without getting out of breath.

As it’s only been a few weeks so far, I guess you’re wondering if I’ll keep it up, especially when I’m no longer dieting. I can’t see into the future, but there are a few reasons why I think I will be able to:

Of course, this page was written while alternating between cycling and standing up.

Some of the images have been copied from other sites on the net to avoid broken links in the future, and some have been resized and cropped.

At one point I’d like to write a page about the weight loss resources that I’ve found helpful, but I haven’t gotten around to that yet. In the mean time, I have some amateur photography and open source apps on the home page if you’re interested.