Racing - Sprinting By The (Power) Numbers

This is related to my jump/sprint numbers from the other night.

Me getting led out at the now defunct SUNY Purchase Tuesday Night Sprints, probably 1991 or so.

My good friend and former teammate sent me a link to this interesting sprinting article: "The Profile Of A Sprint" from The Science of Sport website. It's a great article, opening with one of those bike cams mounted on a couple pros' bikes (on the Giant team). Go ahead, watch it, I'll wait.

Clip from above linked article.

Okay, so that was exciting, right? Really chaotic, and, frankly, it's really difficult for me to tell when the rider should jump. I guess that's why they have markers every 100 meters for the last kilometer (or more), because if I were in such a sprint I'd have one eyeball on them and another on the riders around me.

I'll sort of follow along their article and inject some of my own thoughts.

Sprint Length

First, they state that a pro sprint typically lasts about 10-15 seconds at the end of typically flatter (and therefore less selective) stages, allowing for a large number of riders to arrive at the finish together. I'd want to add here that for at least the Tour de France, the Champs Elysee stage seems to end up being a much longer sprint. One winter I studied a number of stage sprints while on the trainer and found that most sprints were short, like 10-13 pedal revs in the wind. This is well under 10 seconds in the wind.

A notable exception? The Champs consistently seemed to be 20 revs or so, or almost twice that of a "regular" stage. Not having every sprinted on the Champs I can't tell you exactly why this is the case, but I'm guessing it has to do with the wind (tailwind?), the slight down hill nature of the finish, and the fact that it's a super prestigious stage for a sprinter to win so they'll go a bit earlier.

Good Sprinters are Good Sprinters

What's really interesting is that in the Grand Tours from 2008 to 2011 there were 79 field sprints. 54 of those sprints were won by 5 riders. 19 others accounted for the remaining 25 sprints. This means that there were really just 5 really effective sprinters for those years. That's a very, very, very small number.

Power Numbers

Now for some nitty gritty - wattage and speed. What's incredible is that the power numbers aren't out of this world. For the six sprinters studied the average peak power was about 1250w, a number that even I can hit.

Sustained power, for the average 13 second sprint, was about 1020w. Again, this is a totally attainable number by a number of riders, me included.

My "good" sprints have been about 1250w peak with the best sprint being a sustained 1100w effort for 18 or 19 seconds. Typically my peak will be about that, like 1250w, but depending on the sprint my sustained effort is usually shorter and usually is in the 1050w range.

Top speed averaged about 41 mph. That's fast but not outlandishly fast.

Basically even the best pros are sprinting at numbers you might see in a Cat 3 race.

Of course there's a huge difference between the pros and a Cat 3. Cat 3s aren't racing for 3-6 hours before the sprint, they don't have a massively fast (and tough) final hour, and they typically don't have to hold 30+ mph before they get into the sprint.

Thoughts

I think that the numbers only tell part of the story. Obviously hitting the numbers counts. But reaching the end of the race in good position, with as much in reserve as possible, is critical. I'm sure there are much stronger sprinters out there that don't get the job done regularly, and there are weaker sprinters out there doing super well. Race craft counts for a lot, but it's good to know what sort of numbers the big boys are hitting. At least it's a point of reference.

How To - SRM PCV Battery Replacement

One of the "drawbacks" of the SRM powermeter system is the lack of "user serviceable batteries". To be technically correct it should say "easily user serviceable batteries". The computer head ("power control") and the cranks ("power meter") are both designed to be serviceable, just not with a nickel to twist off a battery cover. You need a bit more than that to service them.

The Power Control 5 (or PC5 or PCV in Roman numbers, and we'll go by PCV since that's what SRM calls it) falls under this "not so easily serviceable battery" bit.

I have two of them, one that came with the Cannondale's SRM system and one that I came with a used SRM system I bought off of an eBay seller.

When I bought the Cannondale the crank battery died almost immediately. SRM very kindly replaced it under warranty - it was maybe a year old at that point and I hadn't used it anywhere near the 15,000 km that it might last.

Since then I've replaced one crank battery. I went to replace the other but when I checked the battery it was fine - apparently the wiring harness was the culprit.

The PCV (remember that's the computer head) battery worked fine though. It happened to fade just about when I bought the used SRM system so I've been using the second PCV all this time.

Now, with two bikes rolling, each with an SRM power meter on them (the cranks), and with each SRM power meter holding different offset values (some variable you have to program into the PCV), it's easiest to use two separate PCV heads.

(To program the offset you have to do something akin to reprogramming the clock on a cyclometer. You can do it each time you switch bikes but it's sort of a pain. This is why I looked for a used SRM system, not just the cranks.)

Now I need to replace the PCV battery in that original, came-on-the-Cannondale PCV. I put it off until I had a rolling bike, meaning I put it off for a couple years. With the now-red Tsunami finally on the road I ran out of time.

Here's a short pictorial on how I did it. It took about 30 minutes total and if I was better it'd have taken 10 minutes or less. I note how much time it actually took to do the stuff in each picture - the extra time came from me being super careful or checking and rechecking too much or overthinking things.

PCV head, 4 screws removed.

Green "drum" is a stack of batteries.

Time: 1 minutes to remove screws, 1 minute to remove battery from silicone glue stuff.

The PCV has clear glue that was holding the battery in place. I popped the battery out with a small flat blade screwdriver.

The new battery to the right.

I wanted to reuse the wires so I saved them off the old battery, cutting them at the solder points.

Time: 1 minute to cut leads clear of battery.

You'll notice the new battery is shorter than the old one. It doesn't hold as much of a charge but unfortunately the taller original battery isn't readily available. I read of one person that got them custom made for him. He was selling them too but I never got a response to my request for a battery or two. I bought some lower capacity batteries from Mouser Electronics. They'll be good for the 6 or 8 hours data capacity of the PCV.

I bought a few batteries while I was at it. One for this one, one for the other, and one spare. 

New battery soldered.

Time: 2 minutes to solder each end.

I soldered the two wires on, red to positive, black to negative. I did check the battery first to make sure it was putting out 3.7 volts and it was.

The black stuff is brush-on electrical tape. It's great stuff, just brush it on. It's sort of like black tubular glue, consistency-wise, and it dries to a semi-hard state. It doesn't transmit electricity so it insulates the connection. In my case I "painted" the whole end of the battery just to be safe. I'm not sure but I think I got the stuff at Radio Shack.

In the picture above I have not brushed the black/negative wire/connection.

New battery soldered, both ends covered with brush-on electrical tape.

Time: 1 minute to brush both ends with brush-on electrical tape.

I got specks of the brush-on electrical tape on the surface below the PCV (and a bit on my fingertips too).

Battery in place.

Time: 1 minute to reassemble.

I decided to use the brush-on electrical tape as the glue to hold the battery in place. The black gunk was a bit wet when I put the battery into place - the battery didn't move so I figured it was okay.

All set!

I'm holding at an angle because the flash makes the LCD screen look like it's not working right.

It's on the charger, as evidenced by the "+" symbol on the right side of the screen. We'll see how it goes.

The second PCV, still going strong.

Note the different color buttons.

If you look closely at both PCVs you'll see there's some "loop" of "hook and loop" on the sides. This was to hold the PCV on the Cane Creek Speed Bars when I used them. I haven't been strong enough to use them recently so they're sitting on a bookshelf waiting for me to get stronger. I hope to have a better solution when I put the Speed Bars back on the bike.

The download cable, fixed with brush-on electrical tape.

The download cable for the SRMs tend to crack and break at the plug end. This is just like any other power plug that connects to a device - think of a DVD player or a toaster or whatever power cord that has to bend a bit.

With the SRM download cable the black bit is pretty rigid and the white cable is pretty flexible. This leads to a conflict of sorts where the two meet. Inevitably the rigid black plastic cracks and breaks, and shortly after the flexible white cable separates from the black plastic, exposing the wires. Eventually the download cable fails.

This download cable, that I got used with the eBay SRM power system, arrived already cracked. Given a bit of time the cable would have failed. I used the brush-on electrical tape, liberally applying it to the area where I could see the wire innards amongst the cracked black plastic. The download cable works fine and it appears to be plenty durable.