I can safely say that since I took up cycling again in 1990, I’ve not experienced such a dreadful start to a season as this year. Following a horrid cold (and subsequent post-viral fatigue) that effectively took me out of training for nearly a month to late January, by the end of February I had dragged my form back to where it had been in December. Then, I ricked my lower back again. This had the effect to making climbing on a bike sufficiently painful that I was unable to train for at least three weeks.

During that period, I tried a race (the NBRC club event in early March) only to find myself housebound with backache for a couple of days), and failed to start the Port Talbot Wheelers 2-up 25 in mid-March. Following that, the NBRC club event at Astwood was thankfully abandoned due to snow – thankfully because I might well have been tempted to try riding it.
The final club event of March was the so-called ‘Hardriders 22’ – this was held on a cold morning with a heavy frost. I rode out to see the start, but declined to race as I was a bit worried I’d aggravate my back injury which by this time seemed to be on the mend. Indeed, in the last week of March I’ve been able to resume training, albeit restricted to the joys of the turbo trainer.

Talking of turbo training, for some time now I’ve been using the Polar chain tension power meter to keep an eye on my training progress (see the review in six parts). Unfortunately this has been going through something of a hiatus after I unshipped my chain at speed, and it has proven rather difficult to coax the unit into working again. The explanation is that debugging problems with the device is rather complicated as the power unit is complicated to both set up and keep working. Complicated because there are three separate components:

1. The main sensor/transmitter. This picks up vibration in the chain, so needs to be the correct distance from the chain, but also positioned correctly on the chainstay. This requires judicious positioning in 3D for it to work consistently. The second function is to collect cadence data from the magnet on the crank arm. So the position of the sensor on the chainstay needs to take proximity to the crank arm magnet into consideration. If either of these don’t work, no data is sent to the head unit, with no indication where the problem lies.

2. The chain speed sensor. This is mounted on the rear derailleur, and the instructions aren’t terribly clear on its exact positioning. As I discovered yesterday, if it isn’t just right, no chain speed data are obtained. If it’s incorrectly positioned, the system may work in some gears only. The chain speed sensor is connected to the main sensor by wire – this connection can fail. If this doesn’t work, no data is sent to the head unit, with no indication where the problem lies.

3. The battery pack. Batteries can wear down, and the connection with the main sensor can fail. If this doesn’t work, no data is sent to the head unit, with no indication where the problem lies.

So, you can see that there are several points of failure, with no real diagnostics in place. If any point fails, the main symptom is that no power or cadence is displayed. This is the main reason I’m dissatisfied with the system. The most recent problem related to the position of the derailleur mounted chain speed sensor, which isn’t something that I’d suffered before, and frankly it hadn’t occurred to me! I’d think about reliability issues with any power meter system that I were to consider buying in the future. For the time being, I seem to be able to get along with the Polar system and I’m not inclined to change just now – though the cleat based system from Brim Brothers looks interesting (but may ultimately never be released).

So, to end in an optimistic frame, I’m hopeful that I’ve turned the corner and training can resume. I may even unwrap the CTT Handbook.