New dura-ace R9100

[jeffy]

wonder how long it will take for the bar companies to bring out 9100 hidden JBox "ready" bars.

[strangely]

Guess I will have to try this out on my carbon 3T bars also (will update later with findings) since I suspect they'll also have this reinforcement.

Worst case scenario is that I guess I could run an e-tube wire from a spare brifter port down the outside of the handlebar (under the tape) and then maybe drill a hole very near the end of the drop for it to enter again. I already have my brifters wired with a straight wire from the BB with no junction box up front, so don't need a Y cable or the junction to be there as way of joining the brifters, since they are also wired directly and I only need to plug it in.

If this was done near the end of the tube, then hopefully this wont be too detrimental to the strength of the bars and therefore there shouldn't ever be much leverage there, given my hands are normally on the curved part when I'm in the drops?!

[glepore]

Yeah, unless you need both climbers and sprinters, I'd go shifter to shifter and eliminate the front junction entirely.

The reinforcement area behind the shifters never occurred to me...

[KWalker]

The best news for me is about the nature of the power meter.

Wired intra-crank communication and magnet cadence measurement is a feature, not a drawback.

Accurate cadence is critical to accurate power data and I don't think accelerometers are up to the task at least from my experience with them.

Elaborate on this please because I've never ever once seen evidence of this. Accelerometers are used in countless rotational measurement use cases in a number of industries and products with high sample rates with great success.

[kulivontot]

I do enjoy some baseless armchair engineering, especially when it comes to power meter accuracy discussions.

[strangely]

Guess I will have to try this out on my carbon 3T bars also (will update later with findings) since I suspect they'll also have this reinforcement.

So I checked my 3T Carbon aeronova bars, and they are clear of the obstruction

I should have remembered this really since I actually ran some string through it a year ago (to assist pulling through an e-tube cable) in view of doing some sort of custom connector for charging in the future.

Now I just need the new junction and I'm set... Common Shimano, hurry up and release these already... I've already seen somebody with some of the new 9150 stuff (he works in the bike industry), so it cant be long before it's released.

[TheDarkInstall]

The best news for me is about the nature of the power meter.

Wired intra-crank communication and magnet cadence measurement is a feature, not a drawback.

Accurate cadence is critical to accurate power data and I don't think accelerometers are up to the task at least from my experience with them.

Elaborate on this please because I've never ever once seen evidence of this. Accelerometers are used in countless rotational measurement use cases in a number of industries and products with high sample rates with great success.

Speaking from experience, with a Power2Max TypeS, the built in cadence system is extremely laggy, like over 3 seconds to update to changes, which makes it unusable. Compare this to the Garmin GSC10 which is way faster, and what I use in conjunction with the Power2Max.

[cyclespeed]

So I checked my 3T Carbon aeronova bars, and they are clear of the obstruction

That's good news as that's what I have. I was going to check them today. Although hope there's no difference between the LTD (what I have) and the Team model.

I had been mulling over some kind of Dremel style flexible drill bit shaft to get up there and drill the sucker out!

[ms6073]

Probably a silly question, but any chance these 'obstructions' are simply bits of an internal bladder levt over from molding tthe bars?

[hambini]

The best news for me is about the nature of the power meter.

Wired intra-crank communication and magnet cadence measurement is a feature, not a drawback.

Accurate cadence is critical to accurate power data and I don't think accelerometers are up to the task at least from my experience with them.

Elaborate on this please because I've never ever once seen evidence of this. Accelerometers are used in countless rotational measurement use cases in a number of industries and products with high sample rates with great success.

Speaking from experience, with a Power2Max TypeS, the built in cadence system is extremely laggy, like over 3 seconds to update to changes, which makes it unusable. Compare this to the Garmin GSC10 which is way faster, and what I use in conjunction with the Power2Max.

I'd agree, as an engineer in industry, the accelerometers for industrial applications are much better than the one's they use on bikes. The bike ones are laggy and have a tendency to move by about 1 or 2 rpm. The reed switch is much more accurate IMO. On high speed industrial machinery, accelerometers are not used for speed measurement, usually the shaft is encoded and a sensor picks it up - like turbo trainers.

[Calnago]

Somewhat related, I'll add my experience with the Garmin magnetless brick that straps to the hub, and cadence sensor that straps to the crank. At first I was never sure it was actually working because it took so long for the head unit to pick up. Contrast that with the simple magnet and a GSC-10, or other magnet type sensor, and the readings are picked up instantly with no question that it's working properly.
The strap on magnetless speed sensor is convenient for sure, but butt ugly imo when strapped to an otherwise nice looking hub. I've since gone back to a separate magnet/sensor arrangement on bikes without a power meter, and for my SRM equipped bikes I use the K-Edge chain catcher with the magnet attached to it.

[ms6073]

I had issues with the Garmin magnetless speed sensors in which the speed displayed by my Edge 520 would constantly vary every few seconds by as nucg as 2 mph. Took a chance on the Wahoo RPM speed sensor and it is much more stable, although one of the 3 that I purchased had some kind of issue that seemed to prevent the sensor from going to sleep and exhausted the battery within a 15-20 days.

[deek]

The best news for me is about the nature of the power meter.

Wired intra-crank communication and magnet cadence measurement is a feature, not a drawback.

Accurate cadence is critical to accurate power data and I don't think accelerometers are up to the task at least from my experience with them.

Elaborate on this please because I've never ever once seen evidence of this. Accelerometers are used in countless rotational measurement use cases in a number of industries and products with high sample rates with great success.

Talk to the Metrigear guys on this. Cadence from an accelerometer while riding indoors is really easy; it gets more difficult to separate acceleration caused by rotation when your BB is also accelerating in different directions outdoors.

[kulivontot]

Until somebody shows me a data plot showing power accuracy vs cadence type I remain unconvinced. The experience of an averaged measurement sent over ant+ at 1hz is really not related to what a power meter does internally with an internal accelerometer sampled at a higher rate. In short, unless you've built a power meter, please refrain from making blanket statements regarding design choices as you have no idea what is happening internally.

[hambini]

Until somebody shows me a data plot showing power accuracy vs cadence type I remain unconvinced. The experience of an averaged measurement sent over ant+ at 1hz is really not related to what a power meter does internally with an internal accelerometer sampled at a higher rate. In short, unless you've built a power meter, please refrain from making blanket statements regarding design choices as you have no idea what is happening internally.

I wouldn't say that anyone has really made a blanket statement but as this is an open discussion... I think you would be hard pushed to find an engineer who would pick an accelerometer over an encoded shaft (reed switch) if accuracy was a key parameter.