The Doty

Subscribe via RSS or head to the first post.

Our awesome iPhone and iPad game Crazy Shapes just went live in the App Store this afternoon and I'm so excited about it. The setback from the previous post only set us back 24 hours in the process, I think. It caused some stress and added some additional suspense, but it wasn't too serious is the end. The good news is that the game is out! Hurray!

Right now, I'm spamming the refresh button, waiting for the first reviews to show up in the App Store. This seems to take anywhere from a half hour to many hours. In many ways, what happens in the next 48 hours are key to the life of the game. During the first 24 hours (this afternoon through tomorrow afternoon), the App is at the top of some of the “Recent Apps” lists which lets people know about it, but that doesn't last long. If the game has enough momentum (sales, reviews), it can then get vaulted into other more permanent lists (Featured, What's Hot) which will let even more people know about it.

Sales reports only come in once a day, in the morning. The reports that come in on Saturday morning will contain Friday's data and will tell me how the game was received. Before that, I'll be reading all of the reviews, either from the App Store or online and hoping for the best. It's an exciting and nerve wracking experience. We put so much work and love into this game. We tried so hard to make it awesome and fun. We were our own judges and made sure that we enjoyed playing the game. I can't wait to see how it fares in the wider world. I hope people really enjoy it and have a great time playing!

The bad news today has to do with the iPhone game we've been working on: Crazy Shapes. When you want to sell an app for the iPhone, you have to sell it through Apple's app store, and Apple has a complicated process for handling this. You have to sign up for their developer program (It's $100 a year) and jump through a lot of hoops before you can even install an application on your own iPhone. In many ways, it bothers me that I have to register each and every application that I wrote with Apple in order to run it on hardware which I own. For widespread distribution, Apple's approach starts to make more sense. Everything is strongly controlled by cryptography so that Apples knows where each and every bit and byte came from. This strongly discourages any sorts of funny business and it allows Apple to review everything before it's allowed in the store.

Apple uses this opportunity to do a number of things. The first (and only one which I'm going to talk about today) is to make sure that the app simply works. A lot of software one downloads from the internet can barely qualify as functional. Apple has decided that apps which don't do as they advertise will not be sold. I mostly applaud this. A corollary I like to cite is Nintendo's approach to the video game market back in the early 80s. In the late 70s, games for the Atari and similar platforms (totally open to any developer that cared to bother) had become so bad that people were getting very hesitant to even buy software for fear of getting burned. There was no internet to look to for reviews and there was no quality control. Nintendo created their gaming system so that each and every game had to have a little microchip in it, otherwise that game wouldn't play. These chips only came from Nintendo, and they didn't freely give them out. They reviewed games for quality and they only allowed a software house to produce a limited number of games per year. This forced software houses to use their best ideas for the Nintendo platform and to make sure that they delivered good quality software. What happened? Nintendo controlled the industry for 15 years, of course.

There's another interesting corollary between Apple vs. Android and Nintendo vs. Playstation. One of the reasons that Sony (and later Microsoft) were able to overtake Nintendo had to do with their open natures. By allowing more developers to participate, they were able to succeed with a wider market than Nintendo. I would actually like the smartphone market to end up similar to the video game market. If there is one company that strongly enforces quality(Apple, I'm looking at you), it makes everyone to compete on quality, which improves the software for everyone's phone. If other companies provide a more open interface(Google, and probably Microsoft, although they'll probably try to subvert things to lock you into their platform while they are at it), then there's also an outlet available for software which doesn't fit in a walled garden. Right now, I'd say we are in the mid 80s with Nintendo running the show. I hope Android catches on to compete with Apple and everyone will win by it.

Where am I going with all of this? Well, we submitted our game to Apple for approval last week. This afternoon, they rejected our build for something so minor that I had overlooked it: if you turned the device upside down and then launched the application, the game launched upside down. Apple's interface guidelines state that iPad apps should launch in the orientation of the device: if the device is being held upside down, the application should still be rightside up for the user. This is great for the user, but frustrating for the developer. I want to make great software. However, being able to launch upside down is not useful to the gameplay. It doesn't make the game any more fun. It doesn't make it any more humorous or addictive. It's not something which makes me stay up until dawn fiddling with in order to make sure it's perfect. It's actually only a 10 minute change to make it work to Apple's liking. What feels so discouraging about this is that they almost seem to have missed the point. For you cyclists who are reading this, it would be like someone telling you that you shouldn't ride your carbon fiber Dura-Ace time trial bike (For you non-cyclists out there: this is something like Ferrari making a bicycle) because the shade of blue is too dark. Sure, it might be closer black than blue, but that's not really the point, is it?

I actually think Apple is correct to make us fix this, but it is still causing me quite a bit of stress. By enforcing these simple rules, Apple is able to make applications to follow a basic level of standardization which significantly aids the user. My fear is that I will have to wait another week before they look at the fix again and that they will find another trivial thing wrong with it. There are stories of apps languishing in the approval process for months. I am very eager to see this game in the hands of users. I want to see if they enjoy it, who gets addicted, who doesn't. Erica and I didn't weren't targetting a particular market when making Crazy Shapes. We did it as art: make it beautiful and make it fun. It's something that we enjoy playing, a lot. I really want to know how it fares in the consumer market. You may notice that I'm not complaining (much, anyway) about Apple. I'm frustrated and disappointed because I was hoping to launch tomorrow, but I don't think they are wrong and I don't think they need to change anything. They are doing something wonderful with the smartphone market which will benefit everyone for years and I'd like them to keep on doing it. That doesn't change my disappointment, but oh well. The only thing I can do now is to wait.

This post will cover the surprisingly time consuming installation of SKS Fenders. The fenders come free courtesy of my previous attempts to install them on my road bike. This is cheating the budget (of $400) a little bit, but I figure anyone attempting to build a $400 touring probably has some parts lying around that they can recycle, so it's not an unreasonable thing to do. I actually have two sets of fenders: 35mm and 45mm. The 35mm will, I believe, be just a touch too small to fit on the 27 1/4 inch (roughly equivalent to 32mm) wheels that I roll, so I'll be using the 45mm. This is also in the hope that at some distanct point in the future, I will have wheels smaller than 32mm on my road bike and will be able to use the 35mm fenders (they clear my caliper brakes! huzzah!). Either way, my rusty new Heavy Bike will be rain worth in just several short hours!

Saying that the process was “surprisingly” time consuming was a bit of a white lie. If you've ever looked at reviews of these fenders, the conclusion is nearly universal: fantastic and super hard to install. As such, you couldn't say that I was surprised, but I didn't understand the magnitude of difficulty until I set about installing these myself. The difficulty comes from some different reasons. The first is the instructions sheet which is included. Do you remember Legos? How the entire instructions were done without a single word and there were lots of pictures which told you exactly what you needed to build next? Each page you turned, you could look to see what was new and build accordingly. Sometimes this involved a little bit of thought, but if you messed up, by the next page, it was clear, and then it was easy to fix. Similarly with IKEA instructions. Here they don't even have color and the commitment to quality is quite a bit lower than with Legos, but the word-free instructions are dead simple to follow.

Alas, the SKS instructions seem to be afraid of both words and pictures. You get one black and white drawing for the front wheel and one for the back. You are also provided with about 1.7 sentences of description to help you along. By searching for images of bicycles that have the fenders installed, one can get an idea of how to do it, because it's not at all clear from the meagre drawings. Once you've done that, it's time to get out the hacksaw.

Yes, I said the hacksaw and I meant it. The metal rods that connect from the eyelets by the hub to the fenders will need to be cut to length. I had (stupidly) hoped that these metal bits would be sized for your average 700c wheel and frame, but SKS seems to like one size, and that size is long. The simplest way I found to deal with this was to attach these “fender stays”(I'm making up the term...) to the eyelets and then put the fender proper (with plastic retaining bits installed) in the appropriate spot. I then put a bit of tape at the place where I would need to cut the fender stays to have them end at end of the plastic retaining bits. I did this for the right side of both front and rear and then marked the left side by looking at the marks on the right in the hope that it would help me center things later on.

Then came out the hacksaw. There were 8 steel wires to cut through. I found a piece of scrap wood from a recent IKEA project that had a groove cut it in that was just about the same guage as the wire. I'm sure a vice would have been better. I then started to cut. After you get a little way through, the wire will start to grip the saw and seize things up. You can deal with this by bending the wire so that the opening you are cutting is wider. After I got about halfway through, I would then bend the piece back and forth a half dozen times until it snapped off (much easier than continuing to saw). Sanding down the sharp edges was relatively easy since this is pretty soft metal. You have to do this for all 8 stays and it took me probably 2 hours.

The front fender attachement is relatively simple. Two of the stays have plastic “thingies”(made up term, again) that will theoretically allow the front fender to fly off if debris gets caught between the tire and the fender, hopefully preventing a nasty crash. They look a tiny bit goofy and although they seem to be optional, I decided to leave them on. The fender has a little tab that attaches to the top of the fork. If you use caliper brakes, this is the same bolt that the brakes use and the bolt will be doing double duty. If you have cantilever brakes, this hole seems to be open and free for fendery use. Aligning the fender seems to be not disimilar to tensioning a wheel where the fender stays do a similar thing as spokes. I suggest tightening the stays by the hub all the way before attempting to tighten down by the fenders. It will take a bit of fiddling to get everything aligned.

The rear fender is a bit more complicated. Included in the parts bag is this metal claw like contraption. This attaches to a bridge on the seat stays and graspes the fender. There are also a myriad of small metal bits which are supposed to help you attach the forward-most bit of the fender to the bridge between the chain stays. I screwed up both of these things on the first install. The claw seemed relatively simple, but I failed to notice that it's super useful for helping you align the fender to be centered on the wheel. You need to bend it to be the correct width (another instance of where SKS seems to be using one-size-fits-all pieces that need to be modified for your size fenders and bicycle) which requires you to bend it to shape.

The last part is where the fender connects to the bridge between the chain stays. This is nearly an inch further away from the tire than the bridge between the seat stays and the metal bits didn't seem to help with this. SKS seemed to be recommending using two metal bits to clamp the fender directly to this bridge, but that puts that bit of the fender much further away from the wheel than anywhere else. I didn't like this, so I spent another hour rigging these metal bits to attach to the seat stay but provide uniform distance between the tire and the fender. This was shaky and best and I only learned why it was incorrect when I went to take the wheel off. The bicycle has forward facing horizontal dropouts. If the fender is only 1cm away from the wheel and your dropouts are 3cm long, you can't the stupid wheel out! I then detached all of the metal bits, bent them back to how they came, and re-installed them so that the fender comes away from the wheel in the front. This allows you to remove a wheel without having to deflate it.

So after a lot of work, I seem to have a set of very solid fenders. I can't wait for rain to see how they do!

Back in the Frank Discussion of Gearing, Part II and my last post about the Heavy Bike, I discussed in general what I was looking for in a drivetrain. Today I want to talk about what to do about it! So let's start off by looking at my current road bike. This has a compact double setup with a medium range 10 speed cassette in back. For the road, this is great.

My favorite two “cruising” gear ratios are on the 34 tooth chain ring using the 14 and 15 tooth sprockets. Using the calculator, we see that this gives us gear ratios of 61.2 and 65.6 gear inches with only a 7.1% change between them. In practice, these I use 65.6 more when I'm feeling like making an effort and 61.2 when I'm slacking. Of course, head-winds, tail-winds, hills, Nazgul, mountains, and slalom courses will all cause me to deviate from this cruisin' gear. When I did my one semi-loaded tour, I found myself almost always at 57.4 GI or lower. This will be useful information to know when setting up the drive train for the heavy bike.

I will be riding the bike in three primary ways: about town, loaded, and loaded with steep hills. I also want a dead simple shifting pattern, the bike has three chainrings. If possible, I'd like the large chainring to be for cruising around town, the middle chainring to be for loaded touring, and the granny to be as low as possible. With this setup, any given ride should be brainless: just shift the in the rear. In order to make that happen, I will need a reasonably big range in back, too. Think of it almost as having two single chainring setups next to each other with the addition of some mountain bike gearing. Before I bought the Schwinn, I was thinking of setting up something like the gear chart on the right. Using an 8 speed cassette is super cheap and gives me exactly what I'm looking for. Notice that the chainrings are those of a mountain bike. This particular setup is a cheapo Shimano crankset that can be had for $25. In the top chainring, I have a perfect cruising gear of 62.5GI right in the middle of the sprockets. In the middle chainring (for loaded), I have the perfect 57.6GI gear right next to the top. I have one more if I get frisky, but mostly I want lower gears, of which there are a ton. The granny then goes down to a insanity inducing 17.5GI!

Problem solved, let's order some parts, right? Alas, there is no such luck :( The 1986 Schwinn Passage uses a threaded freewheel instead of a freehub cassette, which is absolutely the norm nowadays. That can be changed, right? Sure! I need a new rear hub ($30 and up, mostly). I then need to install that new hub, which means cutting all of the spokes on the rear wheel, buying a new set of spokes, and rebuilding the wheel. While I'm at it at, I should switch from 27” rims to 700c rims. All told, this would cost at least $100 which I think would be better spent elsewhere. The existing wheel is fine (needs some truing, but it's sound). So I start looking for 8 speed freewheels only to discover that they don't really exist. The freewheel threads onto the hub all the way at the left of the sprockets. By the time you've got 8 gears on there, the point where it connects to the hub/axle is so far in the center that you start bending axles! The freehub doesn't suffer from this problem since it has the freewheel mechanism built in, and it's built in all the way at the right. This was one of the reasons why we all made the switch from freewheels to cassettes. Unfortunately, I'm stuck with a freewheel, so I started looking at 7 speeds....

The first thing I found was a shimano 11-34 tooth “Mega-Range” freewheel. Combine that with another cheapo 42-32-22 welded crankset ($20!) and you get a beautiful setup as seen at the left. In the cruising chainring, I would have a 63.2GI sprocket right smack in the middle, and in the touring chainring, a 57.8 nearer to the top with plenty of downward range. Awesome! I went to go order and found out that you can't actually buy these freewheels anymore, even if you went to Mordor itself. They are still carried on some online sites, but they are universally out of stock.

Next up was the 14-34 tooth Mega-Range freewheel. This one was cheap and widely available and in stock. Alas, the gear ratios you get with a cheap mountain bike crankset aren't what I'm looking for. A high end of only 81.2GI? I think I can do better than that! Also, on both the top and middle chainrings, my "ideal" gear ratios aren't there. Sticking with the existing chainrings on the bicycle (50/46/30) is even worse. This one stumped me for a little bit. It actually took a day before I hit on a new idea: the existing crankset has swappable chainrings. Buying individual chainrings are expensive, so I had dismissed it before, but I thought of something neat. By taking the existing 46 tooth middle chainring and putting it on the outside, I get an almost perfect cruising ring. I did a quick look at the bike and confirmed that the BCD (Ie, bold crank diameter. Just like every other bicycle component, there are 18 different variations here. You get 4 bolts or 5 and then common spacing is 74mm, 110mm, and 130mm, but 94mm and 58mm aren't uncommon either) was the same for both the outer and middle chainrings and that the teeth weren't shaped (Shaped teeth like on hyper-glide chainrings can't be flipped around and still shift well. A lot of chainrings are still made with straight teeth, though). I'm almost certain I can do this swap.

A little playing around with the gear calculator and it turns out that a 38 tooth middle will fit that freewheel perfectly. The granny chainring, at 30 teeth, is also a little large for my needs. It has a 74mm BCD which has a lower limit of 24 teeth (the larger the BCD, the high number of teeth it has to have before it just runs out of room. If you use too small of a BCD, the chainring is going to be too bendy because it's far away from the bolts. In general, there should be different BCD sizes, I would just like fewer of them and no variation in number of bolts!). The prices of these chainrings aren't too much more than swapping out the entire crankset. The 24 tooth chainring is cheap at $16 and the 38 tooth is $21. This is also a better solution than the 42/32/22 crankset I was looking at for $25 since that would have been stamped chainrings. If something got worn, the entire crankset would have to be replaced. It was ultra low cost. By spending just a little bit more, I was able to get a much better setup.

The cruising ring ranges from 89GI to 51.9GI (don't want to use the big ring with the big sprocket) with a 62.3GI right in the middle. The spacing is also pretty tight at 14%, 13%, 11%, 10%, and 9.1% (We'll get to that 42% jump in a second, just you wait...). The top end is almost big enough and the bottom is almost small enough (Ideally I'd want something like 45GI - 95GI, but you can't have everything in life, right?). The touring gear is great with a 57.2GI ratio near the center. I've got two gears above for a tailwind or a hill and I can get down to 30.3GI if I want to, but realistically, anything lower than the 42.9, I'd change to the granny. This is where I'd finally like to address the jump from the 24 to the 34 tooth chainring. In reality, this is more like a 3x6 +1 setup. You have 6 usable gears on the top chainring, 6 usable gears on the middle (the 7th is a huge jump away), and then you go to your granny gear. At the bottom of all of this, is what I'm calling

At 19.1GI, it should be able to take you anywhere on Earth, Middle or otherwise. This Mount Doom Gear is the +1 of the setup. Its use will be so rare and in such dire straits that the 42% jump is irrelevant. Here's a chart with the unusable gears occluded to show the complete useful setup.

For the past couple of months, I've mostly been working with Erica to develop the game Crazy Shapes for the iPhone and iPad. Great news! We are finally done with the release version and have submitted it to Apple! I'm so proud of this game and so happy with the results. It looks beautiful, is extremely addictive, and is a lot of fun to play. It was a lot more work than I think either of us expected it to be when we set out, but it was completely worth it.

Since Crazy Shapes is about to launch, we created this awesome trailer of the game. We spent so much time carefully making sure that every aspect of the game is polished and worked great and it was all in the name of making a fun game. I think you will really like this game!

We are also doing something which I think is pretty cool. The game is being sold for $0.99 as a Universal App. This means that you can buy it once and it'll work on your iPhone, iPad, and iPod Touch. Most game companies nowadays are doing something which I don't like. They create the iPad version as a separate application and then charge $4.99 for it! I don't want to pay for the same game twice! By making it a Universal App, Crazy Shapes will work on iPad at full iPad resolution with beautiful artwork: no scaling. And we aren't going to price gouge you for it: we want you to play at full resolution and we want you to enjoy it!

In the last post, I talked about some of the things my rusty new 1986 Schwinn Passage was good for. Today, I want to talk about the things which are going to need some work. I've declared it a formal project to restore this bicycle on a total budget of $400 ($225 remaining). Today I want to talk a little bit about the gearing that came with the bicycle, the handlebars, brakes levers, pedals, fenders, and racks.

I'd like to start with the Gearing. The bicycle has three chainrings and a five speed freewheel in the back. It uses a scheme that wasn't brought up in the Frank Discussion of Gearing because it's not much used anymore: the Half Step Plus Granny. The chainrings have 50, 46, and 30 teeth and the sprockets run 14, 16, 19, 23, and 28. At first blush, this is a very strange set of chainrings. The 50 and 46 are so close together that they are giving you almost the same gear ratios. Why could this be?

The explanation for this seeming enigma has to do with the large steps you see between gears. These are the 19%, 21%, and 22% you see colored yellow in the chart above. These are some pretty wide ranges to deal with and would cause some annoyance to ride. However, at the time, five speed freewheels were very common (7 being the cutting edge: think 11 speed today). With only five speeds, you can either provide a reasonable range with wide spacing, or you can provide a very narrow range with tight spacing. Neither of these are great comprimises on something like a touring bike, so a creative solution clearly had to be devised.

The solution was to have a pretty much constant percentage step between sprockets in the back. In this case, it's roughly 18%. With that in place, you use your two highest chainrings to try to split the difference. The 50 tooth chainring is just about 9% larger than the 46 tooth. By using both chainrings for your normal range, you can get a pretty good range and have small spacing between the two. The gears we get come out to: 44.5, 48.4, 54.2, 58.9, 65.6, 71.3, 77.9, 84.6, 89.0, and 96.7 gear inches. This gives you changes of 8.8%, 11.9%, 8.7%, 11.4%, 8.7%, 9.3%, 8.6%, 5.2%, and 8.7%. All of the steps are low, so low that only one is above 10%! That's better than a lot of cutting edge 10 speeds. Someone clearly spent a lot of time with a bicycle gear calculator to get this so good.

However, if we've learned anything about gearing by now, it's that nothing is free. You always have to give something up to get something. You want a 10 speed cassette in back? Can do, but the chain is going to be thinner and wear faster (It'll also cost twice as much!). We managed to get 10 good gears ranging from 44.5 to 96.7 gear inches with a very tight spacing. What did we give up? We gave up ease of use. Half of the gear changes in that sequence require a shift of both a chainring and a sprocket. To shift up from 58.9 to 65.5 you would need to shift from the 50 tooth chainring to the 46 whilst simultaneously shifting from the 23 tooth sprocket to the 19. The same inconvenience exists when down shifting. The third chain ring is the granny: 30 teeth. It's not part of the sequence, but is provided to give you some low gears for big hills. In this case, down to 38.7GI.

As good as the half step plus granny setup is, I will be changing for a few reasons. The first is that I don't have the patience to deal with the complicated shifting it requires. I'd really like something pretty brainless so that I can look at the trees when actually on the bicycle. The second is because the particular freewheel that I have just plain sucks. Hyperglide and other shaped tooth systems weren't the norm back then, but that doesn't excuse the tooth shape this freewheel uses. The picture above has an inset (in red) that shows the edge profile of the tooth. It's almost as if they put a notch in the middle of the tooth. This causes some fun times when shifting: you push your friction shifter and not only does it disengage from the sprockets (I hope you weren't out of the saddle when this happened, because now you are pedalling air), but the side plates of the chain will now get caught in that little notch in the edge of the teeth! It engages halfway through and your chain is skating merrily away without moving your wheel. Thirdly, a low gear of 38.7 gear inches is not low enough for what I want. This is another instance of a touring bicycle being sold with road gearing. Given how cheap it is to replace this, I will definitely be doing so.

I would also like to replace the Handlebars. If they aint broke, why fix em? Well, they are very narrow. It was the fashion in the 80s to give touring bikes very narrow and very stiff handlebars. I'm quite tall and have wide shoulders, so 36cm wide bars are not going to cut it. While I'm at it, I'd like to replace everything above the stem. Replace the bars and the rotting tape will have to go. The brake levers are small and have hoods which are oddly shapes and not very comfortable. They also have the wires coming out of the tops, preventing me from using certain hand positions. The brake cable linings are cracked (dangerous!), and the bar ends are in, uhm, interesting shape :) This should be a pretty easy and pretty cheap fix up which will yield large improvements in comfort.

The Pedals that came with the bike are are very toe clip friendly. However, I ride neither clipped in nor clipless, but feet on pedals wearing converse all stars. Since I'm not racing, this setup works really well for me. With some good grippy pedals, I can get a lot of power before I start to feel in danger of slipping. I've been thinking of getting clipless for the road bike since I am able to put out more power at the top and bottom of the rotation than traction allows, but not for very long, so it's mostly a moot point. Anyway, I'll be looking for a good grippy set of platform pedals. Something that's nice and wide to allow a lot of different leg positions and prevent any hot spots.

Lastly, I'd like to talk about Fenders and Racks. The problem here is simple: the bike didn't come with any! I have two unused sets of SKS fenders lying around in sizes 35 and 45mm. With the large clearance provided by the caliper brakes, I'm sure one of these sets will fit great on this bike. I have a crappy aluminum rear rack on the road bike that doesn't really need to be there anymore. For the time being, I'll move it to the Heavy Bike. Ideally, I'd like to get some nicer racks, but one can easily spent $250 on good racks after you get a front and a rear and I don't have that sort of money. The verdict is out on how I can solve this. It was another long article today, but I think it's going to take a lot of thought and planning to build a $400 touring bicycle. Hopefully some readers can use this blog to help them do the same!

In the last post I tried to narrow down exactly what to buy to build my Heavy Bike. I ended up feeling somewhat frustrated because the best options were to get a hybrid bike, which is terribly boring, or to hope beyond hope that I could get a used touring bicycle within my budget. The hope was not in vain!

I just bought a 1986 Schwinn Passage touring bicycle! This is a great base to build upon for the Heavy Bike. It has most of what I am looking for, and I think I can fix up the rest.

As good as it is, there are a number of changes which will be required. Thankfully, I still have the majority of my budget left to spend on fixing it up. The first step will simply be to clean up the bicycle. It looks like the last owner left it outside for the better part of its 24 years. The paint has faded to gray (which is quite fetching) instead of the original green. There are many nicks and light rust on many pieces of the bicycle. Everything is bone dry and in need of oil and grease. There's dirt everywhere, the handlebar tape is rotting off, and the brake pads look like they are cement instead of rubber. Cables are frayed and cable housings are cracked. It is actually dangerous to ride a bicycle in this condition as the brake lines or pads could stop functioning. I'm going to be very careful until I remedy that.

The next step, however, will be to put it into operational order and take it for a little test ride to make sure that the frame and wheels are sound. The tires look terrible (at least one seems to be original from 1986) and I will definitely be replacing them. On initial inspection, they seem to hold air, so I'm willing to the bike around the block on them and hope they don't fail. The wheels are actually 27”, which is a terrible nuisance since the whole world seems to have decided upon 700c. The difference comes out to only 4 milimeters of radius, but that may as well be a mile. It means that I can't put on any of the 700c tires I have for the test ride and that I won't be able to swap out my nice kevlar tires from the road bike onto this.

The 27” wheel size actually refers to diameter of the original wheel plus tire. This is a lousy way to specify wheel sizes, because most of the time you are more concerned with the size of the rim on the wheel, not the outer diameter of the tire. A 27” wheel can take a 27”, 27 1/8”, or 27 1/4” tire. These tires are all the same size at the rim, but some of them are wider, which will increase the effective diameter of the wheel and tire as a whole. Thus, you can have a 27” wheel with a 27 1/4” tire that will have a diameter that's larger than 27”! This is a terrible way to do things. Better is to talk about the diamater at the rim. For a 27” wheel, the diameter is exactly 630mm, for a radius of 365mm.

The 27” wheel is pretty much dead. It's hard to find tires for them and I don't think anyone is making new bicycles that use this wheel size. Everyone has switched to 700c. Here, the same mistake was made: 700 refers to the number of millimeters of the wheel plus tire when the size was originally created. However, with different tire widths, the size changes. Of primary concern to us in the rim diameter, which is exactly 622mm, for a radius of 361mm. This is exactly 4mm smaller than on the 27” wheel, meaning that you can't mix and match. This means that I'll have to do a lot of research to get good tires for the Schwinn. But enough of this talk, I should be test riding the bike!

... Out for a test ride ...

It rides great! Everything is very smooth, stable, and quiet. It needs a lot of work, but this is the right base to start on. Therefore, I propose a project:

I will continue blogging about the project and progress. The mission will be, to build the ideal Heavy Bike!

In Part I of the definition, the requirements were determined to be:

I would like to add a few extra requirements and then start figuring out which type of bike to buy!

First, I want to look at whether or not a Road Bike will meet my needs. A new road bicycle is out of my budget (they mostly start at $600), so I would have to go to the used market. I can certainly find a used road bike that has a steel frame and drop handlebars. Finding one with all of the eyelets for fenders and racks will be difficult, though. At best, road bikes tend to lack front fender eyelets and at worse, make no provisions for anything. Tire clearance can also be an issue since a lot of frame builders like the look of a tight frame around skinny wheels. It will almost certainly have caliper brakes, which is another problem. The best I can hope for from the gearing would be a racing triple, which doesn't quite meet my gearing needs. This approach doesn't get me a bike that's very different from what I already have.

“What about a Mountain Bike? Doesn't it address all of the problems you just quoted?”, you ask. Yes, yes it does. Unfortunately, by doing so, it also made itself unsuitable in other ways. First, a good thing about mountain bikes is that they are cheaper than road bikes. They are mass produced and can be had for half the price, allowing a new one to fit within my budget. However, these are not great bikes on the road. The big knobby tires have a lot of rolling resistance. I could replace the handlebars with drops and end up with something of a Frankenstein (Actually, the Salsa Fargo has been looking intruiging) that might work. The frame will be short which can cause problems with panniers: if the rear bags are too far foreward, you hit them with your heel. If the front bags are too far back, you can hit them with your toe. This might be a viable approach by replacing some parts.

If the road and mountain bicycles are at the ends of a spectrum, the Hybrid Bicycle attempts to be in the middle. It is often a mountain bike that has been made more road friendly. It is equipped with larger diameter wheels that are smaller and smoother to work much better on the road. They are mass produced and cheap, and they are certainly not afraid of fenders or racks. The handlebars would have to be replaced, but that's easily done. The only downside to this type of bicycle is that I'm not excited about it. It meets most of the requirements pretty well, but it's not going to be a very fun bicycle to ride. So far, it's in the lead, though!

The Cyclo-Cross Bicycle attempts to mix road and mountain bikes in a different way. The hybrid started with a mountain bike while the cyclo-cross starts with a road bike and then makes it more capable off road. Alas, this means that it's expensive like a road bike and I will not be able to afford a new one. Even a used one is quite an intriguing prospect. It's a bike which is good on roads that is also meant to go off road and take some abuse. It has the right capabilities until we notice one thing: they are built light. In a cyclo-cross race, you end up carrying your bicycle a good bit (as counter-interuitive as it is to have a race which involves carrying a bicycle, there is some reasoning behind this madness), so they aren't very stout. Eyelets for racks seem to be hard to come by and the bike probably will not handle well under load.

Next up will be the City Cruiser bicycle. Think Dutch: heavy, single speed, built like a tank. You can ride these bikes in a business suit while carrying groceries. They are almost impossible to hurt. However, actually importing a new Dutch bicycle is well out of my budget, so I looked at a couple of local options. The first was a new bike from Schwinn: the Willy. At first blush, this looks to be almost perfect. Upon further inspection, it seems that the Schwinn name was sold in the 90s and is now being badged onto very low quality bicycles. I got worried when the bicycle only came in two frame sizes. I'm 6'5”, two frame sizes is not going to be enough! Oh well, onto the next possibility...

Public Bikes recently launched with a lot of fan-fare. They produce city bikes and are selling them at prices that suggest that they aren't crap. They have a NYC test ride location (in a shoe store in SOHO, of all places), so I went to go for a test ride... Alas, this was not the bike for me. Even on their largest frame size, it felt like I was riding a child's bicycle. The paint is even brighter in person than on the website, which contracted the rule about looking cool. The bike seemed solid and reasonably well built. I think they need to learn from the Dutch and offer them in black. Different handlebars would go a long way, too. The only option has you sitting straight upright, which is not for everyone.

A really good fit seems to be the Touring Bicycle. These are similar to a road bicyle, but meant to carry a lot of weight. The frame is heavy, there are eyelets to mount racks and fenders. They have a lot of clearance for reasonable sized wheels and they come with drop handlebars. Oddly, they usually come with racing triple gearing, which is not what you want on an actual tour. Most of these bicycles are sold to people who don't ever want to take them on a tour, so the racing triple is a better set up for them. Alas, even the cheapest touring bicycle is more than three times my budget. Even a used one can easily exceed it and I know that I'll probably have to swap out some of the drivetrain to get new gears.

I looked at the idea of building a bicycle from scratch. Unfortunately, the manufacturers only sell nice frames as discrete parts. I'd love to spend $450 on a Surly LHT frame, but that's more than my budget for the whole bicycle! So far, my two best ideas are to buy a cheap hybrid and modify it, or to keep looking on ebay and craigslist for a cheap touring bicycle to fix up.

In the Frank Discussion of Gearing, Part I, we discussed five different units of measure that are used to discuss the gearing of a bicycle. Today I want to look at some common drive train set-ups and see how applicable they might be for the Heavy Bike. All of the charts will be in gear inches, so a few useful facts about that:

For low gears, it can be useful to think of a clown riding a unicycle of the same wheel size as the number of gear inches. What happens when a clown rides around a really small unicycle? They go reallllly slow. When the wheel size is around 25 inches, they can move comfortably at a walking pace and somewhat faster. Gearing this low is useful when you are mountain biking, carrying a large load up a hill, going up a really steep hill, or are trying to do something that requires a lot of control of the bicycle at slow speeds. The clown at the left is riding a unicycle that looks to be geared at about 6GI (you can calculate this from his RPM and speed. If anyone does a more accurate calculation, please email me). Because a 6 inch wheel on a unicycle would be too small to pedal, he's constructed it with a tiny wheel and a small drivetrain to put the pedals higher.

The first setup I want to contemplate for the Heavy Bike is the simplest: Single Speed. This has a number of advantages: it's cheap, it's very maintenance friendly, it's relaxing to ride (you never need to think about gearing!), and it is reported to make you more attractive to the opposite sex. The downside is the lack of flexibility. Because you only have one gear, you have to choose one which is low enough for the steepest hill or heaviest load you will be riding, otherwise you will be walking the bicycle. Most single speeds have gearing of around 60-65GI, which is a pretty good comprimise. When you aren't carrying anything, you can get over almost every hill (with some extra effort) and it's comfortable on flat ground. As long as you don't go fixed gear, you can always coast if you go down a really big hill. Unfortunately, one of the requirements for the Heavy Bike is the ability to haul things. That means that my lowest gear (and thus my only gear) has to be low, at least below 30GI. This would require that I pedal at more than 150RPM to cruise around town.

A lot of road bikes (skinny tires and drop handlebars) come with gearing called a Racing Double. The "double" part refers to the number of chainrings, and it's "racing" because this gearing lets you go fast. A good setup for racing is to have lots of high gears, a very high largest gear, and small spacing between them. By having small spacing, you can always get into the most efficient gear and that let's you go faster. Since the number of gears on a bicycle is limited, you have to give something up to get lots of high gears, and you give up low gears. In the example gear chart, the lowest gear is barely 45GI. For racing, this makes a lot of sense. Unless you are in the Alps, using any gear lower than this means that you are going too slow for the race. However, the Heavy Bike is not a racing machine, and I will need low gears.

If you are racing, and you are in the Alps, what you need is a Racing Triple. This is very similar the racing double, but it adds a smaller chainring and will often have some larger sprockets in the back. This setup gives you lower gears at the expense of extra weight and complexity along with somewhat wider spacing between gears. An interestering development lately has been the Compact Double. If a typical racing triple has chain rings of 52, 39, and 30 teeth, then the compact double will have two chainrings of 50 and 34 teeth. This means that your high and low gears are almost as good as a triple. This is a popular comprimise if you want the lighter weight and simpler gearing, but would still like to be able to climb big hills. Both of these are popular choices for your average road bike that needs to do more than just race. The chart for the racing triple is on the left, the compact double on the right.

A popular choice for city cruiser and comfort bicycles is the Internal Hub. All of the gears and whatnot are hidden away in the rear hub and the visible part of the drivetrain is a single chainring running to a single cog. Because all of the complexity is hidden from the elements, this is a very durable and low maintenance solution. The better hubs usually have 7 or 8 different gears and can have up to a 300% change in spacing. This means that you can have a low gear of 30GI and a high gear of 90GI in nice easy steps. This is a great solution for an around-the-town-bicycle. These hubs are expensive, though, and usually run above $200. There is also the Rohloff Speedhub with 14 gears that cover a 500% range. This is the epitome of the internal hub, and it will cost you more than a thousand dollars to have one.

A similar, but far cheaper, setup is the Single Chainring. This can also be known as the 1x7, 1x8, or 1x9 depending on how many sprockets you have in the back. This differs from the internal hub by using a wide range cassette with a derailer instead of an internal hub. This can be done for less than $30 and can provide just as good of a range and just as much simplicity as an internal hub. Since everything's exposed, it's more difficult to maintain and you can't use a chain guard (aren't those cool? Save your pants they will!). Here's the chart for an example 1x8. The high gear is high enough for everything except racing and the low gear is low enough for everything except hauling a large load up something flatness challenged. The gear spacing is a little wide (steps which are colored yellow), but except for the the transition to the lowest gear, it should work well. Since you will likely be toiling away madly up a steep hill when going to that lowest gear, I doubt the large jump will be much of a concern. I like this setup a lot.

The last setup to look at today is Mountain Bike Gearing. This is meant for bikes that go off roads, that navigate obstacles, that need to be agile. You might hit a section of trail that's at a 45% angle. You might need to climb over an obstacle. This adds up to mean that you get lots of low gears. In the example chart on the right, a whopping 16 of the gears are lower than the racing double's lowest. Huzzah! The trade-off is a widely spaced top end. You don't have a lot of gears which are useful for normal road riding or for that rare cherished ride when you have a tailwind. The top gear of 104GI should be sufficient for pretty much anything, however.

For the Heavy Bike, I'm torn between a 1x8 setup and mountain bike gearing. I would actually select a smaller chainring than the example provided above to provide lower gearing with a range between 25GI and 78GI. I tend to spin the pedals relatively fast (80-90RPM), so I almost never use any gear above 80GI. The low gear of 25GI would be low enough to carry a good load up a pretty steep hill, but isn't quite low enough for all situations. Going with mountain bike gearing would give me a lot more low gears (and a few high ones) at the cost of a lot of extra complexity. The racing double and triple are not suited for the purposes of this bike. If I want to go out on a fast light bike, I can ride the Salsa, which has a compact double. I like the internal hub, but they will be too expensive for this project. I may be able to find one used and hope it's laced into a good wheel. I think the bicycle I select will help me determine which gearing to use.

Bicycle gearing is something you either want to never think about, or it's something you want to think about deeply. A shallow contemplation of this topic is sure to leave one dissatisfied. There are many trade offs and no perfect solution. There are always things which you'll want to change about a drivetrain, but with the proper forethought, it should be possible to pick the trade-offs so that you aren't much bothered by the negatives of your selection.

The first problem you have when thinking about gearing is how to do it. Count chainring teeth in front and sprocket teeth in back? That works as long as everyone has the same size wheel and tire. That doesn't work so well in practice, since bicycles have widely different wheel sizes. Riding with a 42 tooth chainring and a 17 tooth sprocket on a road bike with 700c tires is very different than having the same gears on a folding bicycle with 16” wheels. Even if the wheel size (which is effectively the size of the rim) is the same, different tires can increase the diameter of that which rolls. Clearly, just counting teeth is not going to be enough. And so, I'd like to discuss five different units of measure that are frequently used to measure gearing. All of them are useful in certain situations.

One potentially useful measure is to say that a certain gear will go a certain speed at a certain RPM. I've constructed an imaginary bicycle that has chainrings with 52 and 39 teeth. It has 9 sprockets in the back that range from 12 to 27 teeth. This is a very standard setup for certain types of bikes. Below, you'll find charts that display the MPH and KPH for this setup at 80 RPM.

Specifying a gear in terms of speed at a particular RPM has a few advantages. It is very concrete. Everyone knows what you are talking about when you exclaim, "I can ride 19.2 miles per hour at 80 revolutions per minute!" If you aren't careful about where you say it, they might look at you funny, but in general, you should be well understood. This system is not very generalizable, however. The numbers only make sense at a particular RPM. It will also cause a great deal of confusion around the US-Canadian border. It's useful to do this calculation for yourself to see what speeds you'd be riding at in certain gears at your preferred RPM. However, it's not as useful to communicate with someone else who has a different preferred RPM.

The system of Gear Inches is very old. It's so old that "inches" refers to wheel size of a penny-farthing bicycle (specifically, the diameter). During the 19th century, the pedals were connected directly to the wheel. There were no gears, no chain, and you couldn't coast. This was partly because it was the simplest way to do things and partly because the technology didn't yet exist to make good quality gears and chains. If you have no gearing, there are only two ways to go faster: either pedal faster or get a bigger wheel. Since you can only pedal so fast, people quickly made bicycles with ever bigger wheels. The limit was how length of the rider's leg. The taller the rider, and thus the longer their legs, the larger wheel they could spin, and so the faster they could go for a certain RPM.

What does all of this have to do with a modern drive train? Well, when gears were first introduced, it allowed smaller wheels to be used. However, the bicycle maker needed to specify to the customer what the gearing was. Since this new style of bicycle (at the time, called a "safety bicycle" because it was a lot safer than the penny farthings, which were called "ordinaries". We are still riding these "safety bicycles", except for you recumboids) didn't have a large wheel, the effective wheel size was specified. If the diameter of the wheel was 25” and it had a 2:1 gear ratio, it was sold as being equivalent to a penny farthing having a 50” wheel size. The term "gear inches" was later formalized around this ad hoc measure. In some ways, this is a good way to talk about gearing. It's somewhat concrete and it doesn't care how fast you are spinning the cranks. It tells you what you need to know about the gearing, taking into account the wheel, tire, sprocket, and chainring.

The metric world (Everyone whose country's name doesn't begin with "Unite" and end with "tates") uses a similar term called "Meters Development". This will tell you how many meters you (and your bicycle, if you are lucky) will be moved forward for each and every revolution you pedal. This is a very concrete number and has all of the same advantages of gear inches with the further advantage of being intelligible to the majority of the world. It's downside is that the numbers are small. "So what?" you are probably asking. Most people, The Doty included, tend to round off numbers they see with a lot of decimals. When someone tells you that "the fizzledoo is 3.24!" you probably remembered that as being "about 3". Unfortunately, meters development suffers from a lot of these rounding errors. The numbers are all small with a lot of decimals. If you round in your head, then most of your drivetrain is "about 5, right?". Here are the charts for gear inches and meters development for our imaginary bicycle.

The last way to talk about gearing that we will discuss today is Gain Ratio as suggested by Sheldon Brown. If you don't yet know about Sheldon Brown, I highly suggest you spend the next 17 weeks exploring his site. It is the largest wealth of information about bicycles that I have ever seen. Sheldon noticed that all of these units of gearing ignored crank length and that all of them were tied to a particular system of units. By adding the crank length, he was able to get the units to cancel out, providing a pure ratio of distance moved by the pedals compared to distance moved by the bicycle. This is pretty cool. It's a number that is meaningful whether you speak furlongs or nanometers. It reports consistent results whether you are riding a 8” child's tricycle or a big-wheeled 29er mountain bicycle. Furthermore, since it's unit free, it's the same number, no matter what type of bicycle you have, or what country you happen to live in. It suffers from similar problems as meters development because the numbers are small. It's also new, and people don't like new things, so they might not want to listen to your new-fangled-gain-ratio-talk. There's also the discussion of whether or not crank length actually matters. In some ways, it certainly does. If your legs are really short, long crank-arms would be a drag. At the end of the day, I think it boils down to whether or not RPM or distance you've pushed the pedals is what is important. I fall on the RPM side of this debate. I know that a longer crank arm provides more mechanical advantage (ie, it makes the gearing "steeper", making you go faster for a certain RPM), but it seems like the more important thing is how many times you go around. I consider crank length to be largely a comfort thing: go with what works for you. Here's our example bicycle drive train in gain ratios:

This ended up being a long post, but fret not! The next entry will discuss the types of drive trains you might encounter. I'm producing these charts with the Bicycle Gear Head app I made for the iPhone. I wanted something that would allow me to really look at gearing and to produce beautiful charts that I could post online or email to people. This didn't exist, so I made it. At the end of the day, I prefer to use gear inches most of the time. This is partly because it was the first thing I learned, and partly because it works for me. I live in the US where inches are understood and I believe that RPM is more important than how far I've pushed the pedals. All subsequent discussions will be in GI unless public outcry demands otherwise.

In the last post, I listed what I wanted to use the new bicycle for. That still didn't tell me what the new one should be. Do I get a hybrid? Another road bike? A touring bike? A Dutch style city cruiser? Do I buy new? Used? Build from scratch?

To start, I will list out the specifications that are important to me and will allow me to use the new bicycle in the ways I want:

Because of all of the limitations discussed in my last post, it was clear that I need a new bicycle (yay!). Because I don't have a lot of room or money to use on bikes, I have decided to sell the single speed to help fund the project. This decision was reached with some amount of teeth gnashing and sniffling: this was my first bicycle love, after all. In the end, the practical needs won the day and I headed for craigslist. The sale should (hopefully) give me about $300 to spend on the bike and all upgrades. It is now time to list all of the ways I want to use the new bike:

After I made this list, it has become clear that I need a heavy bike. My new friend is not going to be light and fast. It won't jump out of a stoplight or nimbly dodge potholes and tourists. It is going to haul, it is going to lumber, and if I am lucky, it will do it all in comfort and style.

Most of this summer I've been plagued with the idea of getting a new bicycle. I already had two, so this didn't seem to make much sense until I thought more about what my current bikes were not good for. Knowing this will help me determine what the new bike should be good for.

I have a Salsa Casseroll road bike that's built up with a lot of custom racing components. If I had known more when I bought it, I would have selected components that were more durable instead of lighter, as the shop had done. It has a steel frame, and the manufacturer suggests that it be used for randonneuring, century rides, and anything that isn't loaded touring. It has a rack on the back and braze ons for fenders, but there's no clearance between the racey caliper brakes and my 32mm tires for them. Everything on this bike is light and expensive: you don't want to leave it locked up somewhere and you don't want to abuse it. All told, this is an awesome bike for a long ride when the weather is nice.

My other bike is my first bike, a Raleigh Rush Hour. It has a single speed drivetrain with a fixed/free flip-flop hub in the back built, and it is built on a track geometry cromoly frame. This bike is fun to ride! This is the bicycle that caused my terrible addiction about which I'm blogging and you, dear reader, likely also suffer. Fun as it is, this bicycle is very limited. Anything you want to carry, you'd better carry it on your person, because the bicycle isn't going to help. The aggressive geometry is very tiring to ride longer than 10 miles. I found that I wasn't riding this much anymore.

I also have an interest in bicycle touring and recently completed my first partially loaded tour. I road the Salsa with two panniers in the back and the tent between them on the rack. This equated to camping gear and some food and clothes, but it was certainly not fully loaded touring. At maximum, with two water bottles, two extra quarts of Gatorade, and all of the food for three days, I was probably carrying an extra 50 pounds. The bike was able to carry this, but not without some problems. The handling got squirelly and it developed the dreaded shimmy: a low frequency side-to-side oscillation. It was doable for a 100 mile tour, but I wouldn't want to cross the country this way.