PowerTap SL Cadence

MY02_STi said:

I've just received my PowerTap SL and was wondering if there's any real advantage to installing the cadence sensor 😕 as from what I understand, the SL will pick up 'virtual cadence' from the hub so is there any real point in running an extra cable etc just for 'real' cadence 😕

Any advice would be appreciated 🙂

Depends on how important cadence data is to you. Hub cadence is not overly reliable and over 130-140 rpm won't work. Some find it works OK, others find it problematic. It never worked properly for me. I use the crank sensor on my road bikes and don't bother with cadence on the track (I just know what gear I'm riding).

If using the crank sensor, get a better magnet (one of those small but powerful rare earth magnets) as the cadence sensor is pretty picky about the standard magnet being perfectly positioned in order to pick up the reading.

Alex Simmons said:

Depends on how important cadence data is to you. Hub cadence is not overly reliable and over 130-140 rpm won't work. Some find it works OK, others find it problematic. It never worked properly for me. I use the crank sensor on my road bikes and don't bother with cadence on the track (I just know what gear I'm riding).

If using the crank sensor, get a better magnet (one of those small but powerful rare earth magnets) as the cadence sensor is pretty picky about the standard magnet being perfectly positioned in order to pick up the reading.

the 2.4 reads from 20-250 (or so it syas in my manual). I have found that it jumps a lot though and readings could go from e.g 65-95-78 etc every second so it's not as "constant" as a cyclecomputer cadence.

I found the hub cadence to be extremely jumpy. If you're trying to do specific cadence work, I wouldn't rely on it at all.

Cadence from the cadence sensor also seems to suffer from a time delay: from the time you start to pedal, you may not see cadence values in the file until 1-2 seconds afterward. You don't have this problem using cadence from the hub. This may affect you during sprint workouts.

Steve_B said:

Cadence from the cadence sensor also seems to suffer from a time delay: from the time you start to pedal, you may not see cadence values in the file until 1-2 seconds afterward.

Do you have the cadence set for the shortest rolling average value (1)? I set my SL for pedal only and rolling average value = 1. Cadence changes appear to be reflected virtually immediately.

RapDaddyo said:

Do you have the cadence set for the shortest rolling average value (1)? I set my SL for pedal only and rolling average value = 1. Cadence changes appear to be reflected virtually immediately.

I was talking about what you actually see in the downloaded file. The rolling average you wrote about above will only change the displayed value on the PT screen and it has no bearing on the values in the file. To change what goes into the file, you have to change the data point collection period and mine is always set to 1 second, the minimum value allowed.

Steve_B said:

I was talking about what you actually see in the downloaded file. The rolling average you wrote about above will only change the displayed value on the PT screen and it has no bearing on the values in the file. To change what goes into the file, you have to change the data point collection period and mine is always set to 1 second, the minimum value allowed.

Yep - this is what I see too - RD remember the parsed Quadrant Analaysis discussion where I "discovered" this "feature". It takes a couple of pedal turns for the crank cadence option to register in the data, so while you have a power reading (since that is all generated by the hub and doesn't rely on crank cadence data) but the cadence is zero for 1-2 data points before it kicks in.

Also I find crank cadence doesn't register under about 30 rpm.

Alex Simmons said:

RD remember the parsed Quadrant Analaysis discussion where I "discovered" this "feature". It takes a couple of pedal turns for the crank cadence option to register in the data, so while you have a power reading (since that is all generated by the hub and doesn't rely on crank cadence data) but the cadence is zero for 1-2 data points before it kicks in.

Also I find crank cadence doesn't register under about 30 rpm.

Yes, I recall that thread. I think NomadVW had some cadence discrepancies with his file(s) as well. I can't recall which was more responsive and which was more consistent -- hub or pedal cadence? Not that it would matter for me. I like having the real-time data and am willing to give up some download file accuracy or responsiveness (cadence) in exchange.

RapDaddyo said:

Yes, I recall that thread. I think NomadVW had some cadence discrepancies with his file(s) as well.

I gave up using the hub because when you examine the files, the actual numbers it records for those first couple seconds while waiting to pick up the pedal/crank sensor are typically grossly inaccurate. I'd rather not have a number there, than have one that is wrong - and not just a little wrong.

NomadVW said:

I gave up using the hub because when you examine the files, the actual numbers it records for those first couple seconds while waiting to pick up the pedal/crank sensor are typically grossly inaccurate. I'd rather not have a number there, than have one that is wrong - and not just a little wrong.

Correct me if I'm wrong but I thought that it didn't switch from the hub to the pedal while in hub-then-pedal mode unless it didn't get a hub signal. Even if it did, it would mean that it would be receiving a hub signal before the pedal sensor and thus stay with that rather than switching. This would mean that the entire data stream is coming from the hub.

Steve_B said:

Correct me if I'm wrong but I thought that it didn't switch from the hub to the pedal while in hub-then-pedal mode unless it didn't get a hub signal. Even if it did, it would mean that it would be receiving a hub signal before the pedal sensor and thus stay with that rather than switching. This would mean that the entire data stream is coming from the hub.

I thought it was the other way round when that option is chosen: pedal, then hub if no pedal signal registers

Pedal, then hub. If no pedal, hub signal. Once it picks up a signal from the pedal, it uses that.

NomadVW said:

Pedal, then hub. If no pedal, hub signal. Once it picks up a signal from the pedal, it uses that.

I set mine up this way Pedal then hub, moved the cadence sensor out of range and spun for a couple of minutes without it ever picking up the signal from the hub. Does anyone know how long it is supposed to take before it picks up cadence from the hub if it loses it from the sensor?

Alex Simmons said:

I thought it was the other way round when that option is chosen: pedal, then hub if no pedal signal registers

Somewhere along the line, Saris changed things.

My old manual (2003 Pro) clearly indicates "hub then pedal". The latest manual (saris.com232.pdfOpen ↗) says "pedal then hub" (page 21). The latter makes more sense to me from an operational standpoint anyway.

The manual that came with my SL says "pedal the hub". All wired up and working well 🙂

MY02_STi said:

I've just received my PowerTap SL and was wondering if there's any real advantage to installing the cadence sensor

Beyond your question, I am curious as to how many people with the wireless PT SL 2.4 hub have gone with the wireless cadence sensor as well. I can't see it listed on the cycleops.com website (or reseller websites, e.g. competitivecyclist.com) and don't even know how much it costs and if it is available yet.

Beyond that, are there issues with installing with unusual tubing e.g. putting the wireless cadence sensor on a Cervelo R3.

Hub cadence has been wildly inconsistent for me. I have seen it jump from 60 to 126 and back to a number such as 80 on consecutive readouts. I ignore it entirely and view it as useless. The average cadence seems to be okay so I look at it. I have had PowerTaps (six different ones) for several years and all hub units have shown this cadence inconsistency.

Van Hilliard said:

Hub cadence has been wildly inconsistent for me. I have seen it jump from 60 to 126 and back to a number such as 80 on consecutive readouts. I ignore it entirely and view it as useless. The average cadence seems to be okay so I look at it. I have had PowerTaps (six different ones) for several years and all hub units have shown this cadence inconsistency.

Why don't you use the crank cadence instead? That works fine (better if you use a high quality magnet).