Sunday, September 25, 2016

Shimano Dura-Ace 9000 vs Ultegra 6800: Road Shifters

During the latest upgrade of the Wheelsport Fantasy mini velo, I had the chance to take a closer look at the Dura-Ace road shifter ST-9000. This was also a good chance to compare it with the Ultegra ST-6800 road shifter.

As I have already done many times, I like to do a feature-by-feature comparison of various components of different grades. This makes it easy for us to make an informed comparison to decide which grade of component to use.

Today, I will compare the Dura-Ace ST-9000 road shifter with the Ultegra ST-6800 road shifter. They are both 2x11 speed road shifters that are cross compatible. So, how are they similar and different?

As a Dura-Ace shifter, the utmost importance is to ensure the smoothest and lightest shifting performance possible, while also keeping weight down to the minimum. The next grade down, Ultegra would aim to replicate most of the performance of the Dura-Ace shifters, while ensuring that the weight is kept low and at a more affordable price.

Dura-Ace ST-9000 on the left, Ultegra ST-6800 on the right. Note the black axle on the Dura-Ace brake lever for the slightly more integrated look.

The rubber hood on the Dura-Ace shifter has a two-tone colour that supposedly has a softer compound at the grey area, but I cannot really feel a difference. The grip size and feeling is the same across the two shifters.

To save weight, the Dura-Ace shifter on the right side has a titanium clamp band, as opposed to the plated steel clamp band used on the Ultegra and below grades of road shifters.

The carbon brake lever on the Ultegra shifter on top is chunkier on the inside and has a rather thick steel insert within the carbon fiber lever for strength, while the Dura-Ace brake lever at the bottom is slimmer on the inside, and has a very thin piece of reinforcement. Most of the Dura-Ace brake lever is carbon fibre, which is where most of the weight difference comes from.

The clear coat finishing on the brake levers look very different up close. The Dura-Ace finishing is super smooth and glossy, while the Ultegra finishing is not as smooth. Not obvious from this picture, but visible when viewing the actual part.

Dura-Ace right side shifter weighs 186 grams...

...while the Ultegra right side shifter weighs 212 grams, just 26 grams more.

These are the 5 main physical and visual differences for the road shifters:
1) Hood colour and material
2) Titanium vs steel clamp band
3) Brake Lever reinforcement
4) Quality of clear coat finishing
5) Weight

As for performance, the Dura-Ace shifters have a very refined, smooth and light clicking feeling, as it uses sealed bearings for some of the shifting mechanism inside. The Ultegra shifters work very well too, although it just lacks that extra bit of smoothness as compared to Dura-Ace. That is totally excusable, given that the Ultegra shifters cost half of the Dura-Ace shifters, and yet works 90% as well.

If you like to have the best, the Dura-Ace shifters are the best that you can buy. However, Ultegra and 105 shifters are much more commonly seen on bikes, as they are much more affordable and work almost as well.

Sunday, September 4, 2016

Avanti Inc 3: New Bottom Bracket SM-BB52

As you may already know, the Avanti Inc 3 is the ideal all weather commuting bike, according to my requirements for a commuting bike. Besides being able to handle all weather riding with little or no maintenance, it needs to work and ride well with good durability.

I have had this bike for almost 3 years, and the bottom bracket has also been used for almost 3 years, ever since I upgraded the crankset to the Hollowtech II type. The bottom bracket used is a SM-BB51 Hollowtech II type, which is considered the Deore grade of Hollowtech II bottom bracket in the MTB range.

SM-BB51 Hollowtech II BB on the Avanti Inc 3

During the past few rides, I have noticed a difference in the pedaling feel on this Avanti Inc 3 bike. It felt a bit rough and unstable. Since I ride and customize all my bikes personally, I am quite sensitive to changes to the feel of the bike. For example, if the seat post height has been changed slightly, or if the tire pressure is lower than usual, I will be able to feel and detect these differences.

Finally, after finding some time to take a closer look at the bike and do an inspection, I found that the crankset felt loose. There was quite a bit of play between the crankset and the bottom bracket. My initial suspicion was that the crankarm was loose, but after removing the left crankarm, I found that there was quite a big amount of play between the crankset spindle and the bottom bracket bearings.

This means that the bearings or ball races in the bottom bracket are probably worn out, causing the large amount of play and also rough rotation feeling. The bottom bracket is considered a consumable part (similar to the chain, cassette, brake pads), which means that it is prone to wear and needs to be replaced after running for a certain amount of mileage.

In order to change out the bottom bracket, the crankset needs to be removed. Due to the belt drive system, this bike uses an eccentric BB insert for adjusting the belt tension. As such, it will not be possible to remove the crankset without first removing the belt from the front pulley.

It is possible to remove the front pulley from the crankarm and just remove the crankarm from the bike, but due to the belt tension and also the rusted chainring bolts, I decided that it would be easier to just remove the entire crankarm than dismantle the front pulley from the crankarm.

This brings us to the next step. In order to remove the belt from the front pulley, it will be necessary to remove the rear wheel, so that the belt tension is released for the belt to disengage from the front pulley.

Removing the rear wheel. I took a picture to remember that the green non-turn washer is on the left side, while the blue non-turn washer is on the right side. Your bike may have different coloured washers, depending on the angle of the rear dropout.

With the rear wheel removed, there is no more tension on the belt and the belt can be removed from the front pulley easily.

Since the rear wheel is out, I weighed it (Alfine 11 hub rear wheel + sprocket + disc rotor + inner tube + tire), and it is 3.4kg! A typical road bike rear wheel would weigh about 1.5kg (rear wheel + inner tube + tire + cassette).

As mentioned earlier, the left crankarm is first removed to expose the crankset spindle.

Gritty bottom bracket, as it has never been serviced despite riding the bike in rain often. Also shown here is the eccentric BB insert that is used to adjust the belt tension.

The crankset removed. Black grease shows that it has been in use for quite some time. Also shown is the rusty chainring bolts, which I would rather not touch as it is likely to give me more problems.

Time to remove the old SM-BB51 bottom bracket!

The bottom bracket was very tightly fixed to the frame, and it was too tight to remove with a normal Hollowtech II tool. The solution is to slip on a Dahon seatpost, which is large enough to go over the handle of the original tool. This gives plenty of leverage which makes it easy to unscrew the bottom bracket.

Once the old bottom bracket is removed, I took the chance to clean up the whole bike, since the rear wheel is also already out of the frame. After that, it is time to put in a new bottom bracket and reinstall all the components!

Deore grade SM-BB52 Bottom Bracket, an updated version of the older SM-BB51. From what I heard, they improved the sealing to be equal to the higher grade SM-BB70.

Close up view of SM-BB52 bottom bracket. If the sealing has been improved, it will last longer than the SM-BB51 that was previously on the bike.

It came with a few spacers of assorted thicknesses, which is great if you need to fine tune your chain line. Will be useful on Dahon bikes too from my previous experience with modifying Dahon bikes.

Cleaned up eccentric BB and frame, ready to accept the new SM-BB52 bottom bracket!

SM-BB52 bottom bracket installed easily

Sufficient grease on the metal-to-metal contact surfaces to prevent binding and improve sealing against water.

It is recommended to use a torque wrench to achieve proper tightening torque at critical areas, such as crank arm bolts.

After the new bottom bracket has been installed, all the other components can be fixed back onto the bike.

Although changing out a bottom bracket may take quite a bit of effort and time, it is not a difficult job if you are using a Hollowtech II type of bottom bracket. Also, it is not often that you change a bottom bracket, as a good bottom bracket can last at least a few years.

With the new bottom bracket, the pedaling feel is smooth again! There is no more play in the crankset, and the bike is now running well again after this servicing.

Saturday, August 20, 2016

Avanti Inc 3: DIY Steering Stabilizer

Do you ride a bike that has a heavy front? By heavy front I mean that when the bicycle is parked, the handlebar tends to tilt to the side, and over-rotate. This causes the whole front wheel to flip to the side, and if the front of the bicycle is heavy, the whole bike will fall over, even if it is resting on a kickstand or resting against the wall.

For most road bikes, they can be supported easily against a wall or on the kerb using the pedals, as they are lightweight. Mountain bikes are usually placed with the bar end against the wall, or just laid down on its side on the ground. Other bikes may have kickstands to hold the bike up when parked.

In my case, I have some problem with parking the Avanti Inc 3 when outside. As there is no kickstand, I will rest the bar end of the Ergon GP3 grips against the wall when parked. However, the bike will sometimes roll forward or backwards due to a slight incline of the ground. This causes the handlebar to rotate and the whole bike will fall over.

When there is no wall to lean against, I will rest the bike against my body when stopping for a short while. Even then, the heavy front end will rotate and cause the bike to fall over easily. This is due to the wide handlebar and the many accessories on the handlebar.

To solve this issue, I decided to get a steering stabilizer to prevent over-rotation of the handlebar. It looks like a very simple device, with just a spring to prevent the fork from rotating too much to either side. After failing to get it from Taobao as it was out of stock, I decided to DIY my own steering stabilizer from some common parts.

The basic component of a steering stabilizer is the spring, together with the mounting brackets at each end that mount to the front fork and the downtube of the frame respectively. The spring has to be sufficiently strong to hold the front fork, yet have some play to allow normal steering.

The spring comes from an unused kickstand, which seems to have a suitable length and spring force.

The L-shaped mounting bracket is something that I found from my box of random hardware, and it fits nicely on the back of the front fork.

Basic construction of the DIY steering stabilizer. One end of the spring is held to the front fork by a bracket, while the other end is fixed to the cable holder.

One end of the spring is hooked around the bolt that fixes the cable holder to the frame. Since the force is transmitted through the metal bolt, there is no strength issue here.

The other end of the spring hooks onto the L-shaped bracket, and is fixed to the front fork using the fender mount.

After that, it is a matter of adjusting the spring force to ensure that there is little or no spring resistance during normal steering. On the other hand, the spring force will be activated when the handlebar is rotated beyond a certain angle. This is adjusted by stretching and deforming the spring.

After installing the spring, the handlebar will encounter resistance when rotated beyond 20 degrees to either side. This is more than the steering angle required for normal riding.

As shown here, the spring prevents further movement of the handlebar, which enables the bike to be parked this way without falling over.

When the bike is held by the saddle, the handlebar can only rotate naturally 20 degrees to either side. Does not affect steering at all.

DIY steering stabilizer fixed under the downtube of the bike, and is not obvious at all.

With this steering stabilizer installed, parking this bike is now easy, with little risk of the bike tipping over. Riding and steering of the bike is not affected at all because there is almost no spring resistance at normal steering angles.

If you have the same problem with your bike, this steering stabilizer will solve the issue!

Saturday, August 13, 2016

Merida Scultura 5000: Di2 Battery Draining Issue

The Ultegra Di2 upgrade was recently done on the Merida Scultura 5000 road bike, and I really like the effortless and easy shifting offered by the electronic shifting system. It was working well for a few weeks, but one day I realised that the battery was flat just as I was going out for a ride.

This is a very unusual situation, as I always charge the Di2 battery once the battery level drops to 50% or less. Also, the Di2 battery is supposed to last a long time, even more so if you are not using it often. As I have been using Di2 since I upgraded the Dahon Boardwalk 3 years ago, I am well aware of the battery life of the system.

My first suspicion was that one of the shifter buttons had been depressed continuously during storage, as there was a chance that another of my bike could be pressing on the shifter buttons on the Merida road bike. However, upon charging up the battery to 100% and leaving it for a couple of days, the battery level dropped to 60% even without touching the bike.

The next suspicion was a faulty battery. This was a possibility as the battery was already 3 years old, and there was a chance that the battery life has been compromised. I swapped the battery with the other one on the Dahon MuEX, and monitored the battery level on both the bikes for a few days. With both batteries at 100%, we will see if it is the battery issue or some other problem.

The result was that the battery (from the Merida) remained at 100% on the Dahon MuEX, which ruled out a faulty battery. On the other hand, the battery (from the MuEX) that was installed on the Merida went down to 80% after just one day, without any usage. Further monitoring saw the battery level drop by about 20% every day.

With the battery level dropping rapidly, by around 20% a day, this was definitely not normal. Something was wrong with the system, but it is going to be tricky to diagnose the source of the problem. Most of the components are ported over from the Wheelsport Fantasy mini velo, and it had worked perfectly. The only differences are the wiring in between the components, as I have used different lengths of wires. For some unknown reason, there is a problem with the Di2 system on the Merida road bike that is causing serious battery drain.

Current Di2 wiring layout on the Merida road bike

In order to troubleshoot the battery draining issue, I had to isolate the components one by one to see which is the component causing the battery drain. The obvious place to start would be from the battery. Since I have already confirmed that the battery is working fine, the next component to test would be the battery mount.

I plugged the battery into the battery mount, and disconnected the battery mount from the rest of the system. Wait one day, reconnect the wiring, and check the battery level with the Garmin Edge 510. The good thing of having the Garmin to check the battery level is that it has a digital readout of the battery level, at every 10% intervals. This is more precise than having to read the blinking lights on Junction A to determine the approximate battery level.

No battery drainage detected with just the battery mount connected, so I continued by connecting more components to the battery (through the battery mount).

Connect battery mount to Junction B, disconnect all RD, FD and Junction A from Junction B. No battery drainage, so I moved on to connect just the FD, followed by just the RD, and so on until the entire system has been tested individually for battery drain.

This is very time consuming, as it takes one day to detect any battery drain, so testing the entire system component by component took more than one week.

I eventually found that there is battery drainage when Junction A is connected to the system, but the RD, FD and shifters are not connected. It seems that Junction A might be the one causing the problem. In order to test the hypothesis, I borrowed a new Junction A to swap with my current Junction A.

The new Junction A is the three port type, unlike the older Y-junction EW67 type that has two integrated wires for the shifters. With this new EW90 Junction A, there is a need for two separate wires to connect Junction A to the two Di2 shifters.

New 3 port Junction A, SM-EW90-A

With the new Junction A wired up, I monitored the Di2 system for a few days, and was glad to see that the battery is no longer draining fast. Under normal circumstances, the Di2 battery level can stay at 100% for at least a couple of weeks if it is not used. I did not wait that long, but the battery level did stay at 100% after a few days, which is much better than the 20% drain per day.

I was glad to see that the Di2 components are working fine, and that the problem was actually due to a faulty Junction A. Not sure why this problem will surface suddenly after working fine for these few years, but I am glad that this can be easily solved with a new Junction A.

After waiting for a few more days to confirm the diagnosis, I was sure that the old Junction A was the problematic component. With that, it is time to formally swap in the new Junction A onto the bike!


New 3 port Junction A with the mounting bracket and the rubber strap. This will normally be installed on the stem. 3 more wires are needed to connect Junction A to the two shifters.

Original Ultegra EW67 Junction A, with the two wires for the shifters built into it

3 port Junction A installed onto the Merida bike for testing and monitoring

Rubber strap with bracket mount installed on the stem. The rubber strap needs to be cut to the correct length.

Junction A clipped in under the stem. There is a charging port on this Junction A, but it only works if the new internal Di2 seat post battery is used.

Two wires lead to the two shifters, while the third wire leads to Junction B.

The other side of Junction A has the LED indicator lights, for indicating the derailleur adjustment mode or to indicate battery level.

Similar wire routing with the new Junction A, using the cable wrap to keep things neat and tidy

With the new Junction A, there is no longer any battery draining issue. The Di2 battery can last a really long time with some light usage. Glad to be able to find the source of the issue and solve it with a change of Junction A.

Sunday, July 17, 2016

Merida Reacto 4000: Part 3 - Full Components Specifications

After taking the effort to disassemble and clean all the bike components on the Merida Reacto 4000, it is time to assemble back everything. Also, it is a good chance to take a look at the stock specifications for this bike.

Part 1: Introduction and Disassembly
Part 2: Disassembly (continued) and Restoration

Other than cleaning the disassembled components, the bike frame was also cleaned and waxed to get a nice glossy finish. Here are the pictures showing the results!

Large downtube with bottle cage bosses

After waxing the frame for a couple of times with Hi Glaze 88, it is now glossy, maybe even more than the original condition!

View of the rear triangle

View of the seat tube. All clean and glossy.

The D-shaped aerodynamic seat post, with the special rubber insert for added comfort

Hidden seat post clamp for a clean appearance. Works well with no slippage.

Now, let's take a look at the component specifications for this bike. Before the cleanup, it was not ideal for highlighting the various components as they are too dusty. With all the components cleaned and restored, this is the best time to show the component specifications.

Shimano 105 5800 2x11 speed road shifters. Reliable performance at a great price point.

The damaged silver name plate was changed to a new one

Controltech aluminium stem. This seems to be a 110mm length stem.

Internal cable routing, with the shifter cables running into the top of the top tube, through a plate specially designed for this purpose.

Tektro direct mount brakes, works OK but lacks power and smoothness compared to Shimano direct mount brakes.

One way to improve braking power is to change out the original Tektro brake pads, to the Shimano brake pads taken off from the Shimano Ultegra 6800 brake calipers.

The rear section of the rear brake caliper outer casing was changed out to a new section, as shown here. It exits from the side of the downtube and routes in a smooth curve to the rear brake caliper, which is mounted under the bottom bracket.

Another view of the hidden rear brake caliper. Maybe good for aerodynamics, but troublesome for adjustments, and braking feeling is not really good due to the longer and more complex cable routing.

FSA BB386 crankset with the Shimano 105 5800 front derailleur. All cleaned up and looking new.

View of the crankset from the left side

52/36T chainring combination, which is a popular mid-compact gearing nowadays. Shiny chain and clean chain rings.

Totally cleaned up drivetrain, front derailleur and bottom bracket area. Way different from the dusty condition when we first got the bike.

Shimano 105 5800 11 speed rear derailleur

Shimano 105 5800 11-28T 11 speed cassette, all clean and shiny again!

Even the tires and rims were cleaned by wiping with a wet rag. Looking new again! Continental Ultra Sport tires, 700x25C for a bit more comfort compared to 23C tires.

No more dusty spokes! The quick release axles were changed to Shimano QR for better clamping force. To see the difference between Good and Bad QR Axles, click here.

Overall weight of 9.0 kg without any accessories or pedals. Not lightweight, but acceptable for this aero setup, with a heavier frameset, heavier high profile wheels and also the not-so-lightweight Shimano groupset.

Front view of the restored Merida Reacto 4000

Looking fast with a small frontal area

Aero frame looks good with these high profile rims

With the bike fully restored, it looks like a brand new bike again!

All it needs now is to put on a pair of pedals, and it is ready to roll!

With this bike restoration, I have learnt more about this bike and the setup. For those who are considering an aero road bike with a good component specification, this is one bike you can consider. If you have a larger budget, you can consider the higher grade Merida Reacto 5000 which has Ultegra components.