I’ve jumped back into the tri world after a brief 10 year hiatus from taking it seriously.
I remember when di2 first came out some folks thought the eventual path was leading to an erg mode scenario where the tour de France or kona was won by an athlete that put 100% of their effort on the physical aspect of racing and due to this new electronic shifting fad there were zero missed shifts. Surely it was a scandal waiting to happen gasp!
Personally I thought it’d be a cool idea and hoped it would. Fast forward to now and we’re so close but there seems to be a missing link and I’m wondering why it hasn’t been done yet?
It seems to me that the calculation to power erg mode is happening from the headset, be either a phone tablet or computer so it must not need too much computing power.
di2 has synchro shifting so were close there. Pair it with a power meter and cadence sensor which is standard on almost any electronic shifting bike and the data to make the calculations should be enough.
When I go into the app store I see some non shimano written apps that control shimano e bikes so I can’t imagine reverse engineering their tech is that complicated.
Just curious any thoughts why this hasn’t been implemented either through official channels or hacked together through a diyer yet. Maybe I’m off in how tough simulating erg mode is. Maybe it’d take too many shifts that the battery would die quickly, but what if we gave a range of say 10-20 watts to stay within it might work better…
Are you talking about a system that keeps your cadence steady? That would be trivial and could be done on an arduino if you could reverse engineer the Shimano signals. Personally I see why this is not desirable, imagine you’re 2000w sprinting for the line and your bike decides you should be in a different gear. That’s a one way ticket to crashville.
Actual Erg Mode though keeps power steady no matter the cadence. That’s a completely different problem. I don’t see any way this could be implemented in an outdoor cycling situation without the addition of a resistance hub or something of the sort.
Whatever you’re talking about, ‘erg mode’ probably isn’t the right term. I can’t envision any real-world-riding mechanical system that forces constant power that makes any sense while racing.
If you’re talking about shifting to keep a constant torque-or-cadence, that could conceivably make some sense in some scenarios, but that is not ‘erg mode’ but something different.
Many people in triathlon aim to hit a constant power for the entire bike so not sure why you think it would make no sense. Maybe road racing it’s not ideal but for triathlon it seems like an ideal situation.
Many people in triathlon aim to hit a constant power for the entire bike so not sure why you think it would make no sense.
A smart trainer is essentially an automated braking system. Which is exactly what you don’t want in the real world. You never want the brakes on except when you want to slow down for a corner, etc.
Merely shifting gears in the real world to achieve a desired torque-or-cadence does not achieve the same thing that the braking system in a smart trainer achieves. Erg mode on a smart trainer sets a power level, but lets you self-select your preferred cadence and torque. You can do 300W at high torque/low cadence. Or if you’re a spinner you can do it at high cadence/low torque. Erg mode doesn’t care.
I don’t understand how to replicate that in the real world without using the bike’s brakes, which is, of course, disaster to real world performance.
As I said, I see possible value having the shifting system keep you at your preferred cadence/torque regardless of what power you’re outputting. But that is not the same thing as erg mode.
a few years ago someone was selling a gadget that would auto-shift to keep you at target power (as much as possible), this didn’t take off
technically it would be very easy to implement this on the head unit if you have the di2 component to connect wirelessly
definitely does not make any sense in peloton situation and if you look at the VI of pro triathletes, even on hilly course, its very close to 1. so really don’t see the point of this
Many people in triathlon aim to hit a constant power for the entire bike so not sure why you think it would make no sense.
A smart trainer is essentially an automated braking system. Which is exactly what you don’t want in the real world. You never want the brakes on except when you want to slow down for a corner, etc.
Merely shifting gears in the real world to achieve a desired torque-or-cadence does not achieve the same thing that the braking system in a smart trainer achieves. Erg mode on a smart trainer sets a power level, but lets you self-select your preferred cadence and torque. You can do 300W at high torque/low cadence. Or if you’re a spinner you can do it at high cadence/low torque. Erg mode doesn’t care.
I don’t understand how to replicate that in the real world without using the bike’s brakes, which is, of course, disaster to real world performance.
As I said, I see possible value having the shifting system keep you at your preferred cadence/torque regardless of what power you’re outputting. But that is not the same thing as erg mode.
Exactly what I was looking for. Our 2.5 year old has been sick so I blame my brain for not clicking that there’s some sort of braking system in a trainer too! So there’s basically 2 scenarios for this to ‘work’ 1) integrate the brakes into it which would be pointless, or 2) have the ability for the user to accidentally spin up too high and send their power over the goal
the idea of keeping at a preferred cadence/torque is interesting though, I wonder if there’s any thought to a system that will let you say, don’t let me spin below a certain RPM before auto shifting to a lower gear, but that could have unexpected results when slowing or stopping your cadence around a corner.
Anyways, this was always my big question around di2 and electronic shifting systems was how far can they go in aiding the rider, so thank you for your help talking this out!
It does exist, but only for ebikes. I think it’s more cadence based like the ideas described later in the thread. There also are a few internal hub shifters that support automatic shifting - one from Shimano that’s geared and another that’s a CVT from Enviolo. These are also usually on an ebike and are more common on the cargo bike models.
Baron Biosystems had some interest in doing that. I never saw it take off.
I tested this and loved it (though I needed cell phone reception for it to work, I found, and that was problematic!).
It had several modes, two of them were: Cadence and Power.
I loved Cadence because I’d simply slow rpm’s for easier gears and if I stood up to sprint, it *quickly *shifted to harder gears which I thought was just awesome.
The Power mode was pretty much erg mode out on the road. Like erg mode, I had to keep the cadence up because if I had it set at 300w and slowed to 50rpm, it would just shift to harder gears! I liked it, but preferred the Cadence mode and would *love *to see this brought back. I mean, setting my cadenice at 90-95rpm and never worrying about shifting was awesome.
OK, but you’re more likely to encounter a dangerous situation on a training ride.
I mean, I (sort of) get the desire for a system like this, but to me it’s just bonkers to introduce a gearing system that would try to keep the rider to a number. Have folks never had to jump out of the saddle and stomp on it to avoid a wayward car or animal?
OK, but you’re more likely to encounter a dangerous situation on a training ride.
I mean, I (sort of) get the desire for a system like this, but to me it’s just bonkers to introduce a gearing system that would try to keep the rider to a number. Have folks never had to jump out of the saddle and stomp on it to avoid a wayward car or animal?
The problem is you can’t present a constant load to a human without some internal braking mechanism inside the hub when the road does not present enough load (wind resistance plus rolling resistance). This is basically what you have on a trainer but that is external to the bike/hub. On the trainer there is no external load from wind and rolling resistance, just the trainer load. If you wanted to do it on the road, the sensor would have to meaure your power output and if it is lower than the target, just apply enough internal hub braking force to top it up. It should be doable, because there should be enough airflow cooling to dissipate any heat built up in the hub.
The gear we are in is irrelevant. Let’s say your target is 250W, and the road is only presenting 190W of load the internal braking system adds another 60W of load instantaneously. It would be tricky to make it smooth because if you soft pedal, the internal hub would suddenly apply a lot of load to bring it back to the target and then you’d have to apply a lot of torque to get things back to stable.
On the computrainer, I remember doing intervals in erg mode at 270W. I would do 55 seconds “on” and take 5 seconds at zero watts, but I would have to stay standing to apply enough crank torque from zero to get the flywheel speed up so the software did not apply infinite force from the load unit when my initial output was detected at zero.
Power = torque x rpms. So if you set your rpms and desired torque or power (which given the formula gets you torque) the controls should change your gears to change torque if you start losing rpms or gaining rpms.
There are some limitations. You run out of gears to get the torque needed (going up hill or down hill). You need continuously variable transmissions or you have to accept that you will have varying gaps because there are gaps in gear ratios because of the discrete gearing ratios on a derailleur bike.