Tossing and Tortured 'Till Dawn

I come back to you now, at the turn of the tide.

Friday, December 08, 2006

iBike initial review --

You've heard the hate and the love for the new Velocomp iBike Pro. Depending upon who you ask, it's either an easy-to-use powermeter that costs less than half what everything else does, or a neat gadget that's of no use at all for serious training.

So, which is it?

INITIAL SETUP -- As you probably know, the iBike does not actually MEASURE power in the strictest sense -- it's not a dynamometer, but instead the iBike is equipped with an accelerometer, a barometric altimeter, an inclinometer, and a thermometer. It's got a wind gauge, too, which I'm sure has an -ometer in it somewhere, but I don't know what it is. (ANEMOMETER. Thanks Zwane.) You add in the weight of you, your bike, and equipment from your bathroom scale and do a coast-down to approximate the sum of rolling resistance, and off you go into the world of power measuring.

If only it were that easy. The iBike comes with a handlebar mount that includes a wired speed sensor, but mine did not include the bottom half of the clamp. John at Velocomp answered my email within 15 minutes, however, and sent out a replacement that same day, so I was up and running in two more days., or so I thought. Getting the handlebar mount to seat properly was a difficult process. The manual includes several warnings that the mount must be secured very snugly to the handlebar; the phrase "rock solid," is included, in bold, in the instructions. Included were both rubber shims to make the fit snug, and also double-stick tape. At first I tried securing it with the rubber shims, so that it would be fairly easy to swap the iBike from one bike to another, but I found it impossible to make the mount "rock solid" this way. I was afraid of stripping out the two tiny, tiny, screws in the mount if I torqued it down any harder, and then I would be completely out of luck. The plastic clip appears fairly fragile, but so far it has held up fine and Velocomp assure me that it's durable. Velocomp has clearly put a lot of hours and money of engineering work into their product, but the mounting device seems like a bit of an afterthought. If you're familiar with auto mechanics, those screw-down metal-banded hose clamps might work well, for a start.

After a frustrating time with the rubber shims, I resorted to the double-stick tape. 3M, I love you. I don't know how you invent this stuff, but the sticky pads made things secure in a hurry. Velocomp adds that using the sticky pads is the preferred method of attachment. The problem is, now the iBike mount is firmly affixed to the handlebar of my winter training bike. I'd have to plunk down another $50 to put it on my main road bike, or deal with ungluing the thing and sourcing some more sticky pads. The next part of the setup is a "tilt" test, whereby you mark two spots on the ground, hold your bike upright, hit "calibrate," rotate the bike 180 degrees, and hit calibrate again. The idea is it uses the two data points to figure out how far off level you've mounted it so that it can properly calculate the hill slope, and, thus, power. They say that the tilt test should be exactly accurate, or at least within 0.2%, but the first iBike unit supplied me was 1.5% out of calibration. In other words, no matter how many times I did the tilt test, the iBike was convinced I was on a 1.5% hill. If that doesn't foul up your power numbers, I don't know what will. Another email to John, another 15-minute response, and a new iBike unit was in the mail to me. Thankfully, they did not wait to receive the defective unit I returned to them before sending me a new one, so I was back in business two days later. The new unit was accurate within less than 0.1%, so it appears that my initial unit just had a glitch.

EXTERIOR, CONTROLS -- The controls of the iBike were clearly inspired by the Apple iPod, and I'm sure they're trying to borrow a bit of that ubiquitous device's fame with their "i-" name. Each of four sides of a circular perimeter changes what you're looking at on the 3-line display, and the center button functions either as a "back" button, or to cycle between the two main screens. It's pretty intuitive to anyone used to most modern gadgets. The LCD quality itself is disappointing, though, especially coming from Garmin's excellent Edge. The Edge feels smooth and refined, and feels like a tiny laptop computer. With its cheap white plastic, the iBike feels decidedly like a pocket calculator. A nice plus is that Velocomp opted for a replaceable CR2302 battery, rather than a rechargeable proprietary lithium like the Garmin. If you've ever used an Edge, the battery was probably one of your biggest complaints; you could MAYBE get 6 hours out of a full charge, and if you forgot to charge your Garmin before you hit the road in the morning, or left the thing on, you were out of luck for that ride. I can't tell you how many times that happened to me. You can also ask the ibike what the current voltage of the battery is, as in 2.88 volts, rather than just a little icon of a battery filled with a bar. Velocomp says the battery will last 25 to 50 hours, but they are cheap and available at any grocery store.

IN USE -- After that ordeal, I finally got a chance to RIDE the iBike. In operation, the unit is surprisingly simple. The coastdown test is straightforward and needs to be repeated only if you change something dramatic about your bike. It automatically starts recording, automatically stores trip files, and works fairly simply. Power is recorded at 1-second intervals -- ride harder, watts go up. Ride easy, watts go down. Coast, and the watts drift down to zero, though not instantaneously. From a 300 watt effort, it takes a few seconds for the ibike read 0. My iBike is not equipped with a cadence sensor, and I really, really hope that ones that are so equipped understand that when cadence equals zero, watts must equal zero. It's especially important for reasons I'll detail below. EDIT -- Velocomp has confirmed that, with the cadence sensor, wattage will be forced to zero when you're not pedalling. If you're going to buy an iBike, just don't buy it without cadence. You'll save yourself a lot of frustration.

Early complaints I heard about the iBike were that it did not measure sudden accelerations well, and that rough pavement disrupted its reporting considerably. The firmware has received a lot of attention since then, and I've had no problems in this department. Moderate chip-seal with occasional bad patches did not cause the iBike to falter, and while a section of gravel road obviously underreported the wattage, due to resistance changes, it continued to read steadily. Clearly this isn't a device for mountain biking or Paris-Roubaix type terrain, but beyond that I think you'll be okay. Even a short spurt of acceleration causes the wattage to spike as you'd expect; stomp on the pedals to get over a short rise and you'll quickly see 600 watts.

This is my first daily-use powermeter, and I'm happy to have it along for my rides. It's both rewarding and punishing to train with power; there are sections of my daily loop that often make me feel like a chump for going so slowly, and so it's nice to see that it's actually a brief section of 8% grade, and that I'm churning plenty of watts over it. The punishing part comes from the fact that there is no hiding place from power. Oh, sure, you THINK you're going quick, you're ticking along at 22 miles an hour, but the iBike will tell you, so sorry, charlie, but there is a tailwind and a 1/2% downgrade, you're only pushing 125 watts.

There's a "sub-trip" feature (they don't call it "interval") that's really nice to have; you can flag any number of "sub-trips" within a given ride, whether it's your favorite road or riding to and from a group ride. The instruction manual is a little confusing about this feature, and mostly seems to say "trip" about 50 times, but once you try it it makes a lot of sense. Whoever built this thing was a tech geek, and once you think like a geek everything is easy. I wish you could customize the display, though. You get 3 lines, like a Powertap, but nothing as cool as the Garmin Edge's 2 fully cusomizable screens, each with 1 to 8 automatically resized lines. In the two main modes, the top line displays either hill slope or speed, the middle line either watts or miles, and the bottom either calories (or cadence if so equipped) or travel time. There is a nifty feature where the iBike automatically changes the top line to show hill grade if the slope exceeds 2%, but, guy's, there's plenty of room. Can't I have both, so I can know I'm going 15 mph up a 10% grade? (Okay, okay, I lied.)

POST USE -- The iBike's included software is pretty basic. It works, but doesn't provide a lot of information. Velocomp is straightforward about this, since it bundles a demo version of Cyclingpeaks's WKO+ software with your iBike. Cyclingpeaks is really, really good software. It's the example all other power-training software should live by. A slightly annoying feature of setting up the iBike on the computer is that, despite taking a USB PORT, it's really using your PC's COM ports, and I got frustrated that I couldn't connect to my unit for a moment until I went into my computer's properties and found that the sucker had actually installed onto COM4. For the computer-unsavvy, this could be confusing; it would be nice if the software could autodetect. On 1-second recording, the iBike's memory can store about 10 hours of data, so you can get in a couple of rides without uploading if you like.

GLITCHES, IMPROVEMENTS, PROBLEMS -- It's not fair to the iBike to leave off with its problems, I know, but that's what I'm about to do. It seems to work really well in most situations, as far as training goes. For racing, the jury is still out. Velocomp says the wattage remains accurate within a peloton, and while I confirm that wattage does go down at a given speed when in the draft of another rider, I don't know how accurate it would be in a full peloton. Furthermore, the aerodynamic drag variations in the differences between the "hoods" and "drops" positions might skew the power data down too much. In training, you can stay in the hoods 95% of the time, or you can know which position to stay in for an interval, but in a race the conditions sort of dictate what you’ll need to do.

Two things that hurt the iBike in daily use: First of all, watts are overreported when you go up a steep hill out of the saddle. The motion of swaying the bike side to side is picked up by the iBike as part of the rise of the hill, I think, because it overestimates the hill’s slope. The error seems to be about 1-3% of slope, depending upon how severely you are rocking your bike. One particular area climb, that I know is about 12%, was reading 15% as I labored up it at full steam, and up the iBike reported I was pushing a Lance-caliber 475 watts. Talk about artificial ego inflation! I’m not sure what Velocomp could DO about this, so I think that, like the non-function of the iBike on indoor trainers, we’ll just have to live with it.

The other issue is correctable, and seriously annoying. When coasting on a descent, the iBike often reports wattage. Sometimes, this can be a LOT of wattage. The problem, again, comes from the coastdown. I do the coastdown in the hoods, and of course resistance is somewhat less in the drops, and somewhat less than THAT in the “downhill skier” descending position. Since resistance increases exponentially with respect to speed, the difference in resistance between the positions becomes more and more important. When I would power over the top of a hill, then tuck into a descending position, the watts sometimes don’t go down at all! Obviously, when analyzing the data, you can see that you’re going downhill and know that’s not accurate, but the average wattage and total kilojoules of the ride will be thrown off. That’s just not okay. Could they at least build something into it to where if the road’s slope was less than -6%, that wattage was forced to 0? It might not be entirely accurate but it would be better than this. EDIT: I’m told that, indeed, the cadence unit fixes this problem, and forces watts to zero when not pedalling. They just shouldn’t sell it any other way.

CONCLUSION – The iBike has significant limitations with respect to true power meters, but it costs less than half what they do, and its operation is simple and straightforward. It is a first-generation device, and clearly will have its teething issues, but with downloadable firmware updates and attentive customer support from Velocomp, it’s willing to work with you, if you’re willing to work with it.

Its price is fairly similar to a top-end Polar heartrate monitor that can download to your computer. Ibike might have some variation in different positions, etc, but attempting to use heart rate-based training zones is far worse. Save yourself a lot of frustration: buy the unit with cadence, and attach it to your handlebar using the adhesive pads. Now, if Velocomp could come up with a better mounting system, I'd be happier, but I'll let you know if mine stands the test of time.

Is it a "useful training tool" or just a "nifty toy?" I'm really not sure. I'm going to have to use it side-by-side with a power tap, take it in some different situations, and really get some use of it before I decide.

My conclusion on the iBike, a followup to the initial review, can be found here.


  • At 2:12 PM , Blogger allons-y said...

    very informative report, just like the one on sram you did. thanks alot, I found it very helpful.

  • At 4:58 PM , Blogger Zwane said...


  • At 2:51 AM , Anonymous Anonymous said...

    The concept of the ibike is good but if you are looking for a valid and reliable power meter look elsewhere. WHY?
    1. Power readings are inaccurate by up 10% or more. That is huge for power base training
    2. Menus are often confusing and difficult to navigate
    3. Eats through batteries like no other computer I know. They last no londer than 2 to 4hours maximum. Very frustrating and pointless. Why have a computer that dies mid way through a ride.
    4. Power reading only good for a trainer but you can get the same data with a $50 bike computer

    It is not all bad. The software is good, the data is extensive BUT you can get the a Garmin Edge 500 for $200 to $300 and get the same features PLUS downloadable maps, gps, a reliable battery and valid data. Yes you will not receive a power reading but what is the point of a power meter if it is often inaccurate. Save your $600 plus dollars and put it towards a real power meter.


Post a Comment

Subscribe to Post Comments [Atom]

Links to this post:

Create a Link

<< Home