Auto/Manual Gear Shifter

Kenny17

Hello everyone,

I'm currently working on my major project in which we have to come up with an idea to work with over the length of the course. I was thinking of making an automatic gear shifter which can be overridden and enter manual mode. The basic overview is that it has three modes, road, trail and manual and they will have different settings and algorithms to find the optimal time to change gear. These would be used in conjunction with a little machine which will analyse the gradient in front and change gear accordingly. It would also have different settings for different age groups. It uses pressure and cadence to change gear in addition to the gradient machine and the different modes.

There's a lot more to it, but this is the basic overview to save you a read.

What I would like to know is:

If you were to buy this product, why would you buy it?

If the price was low enough, would it affect whether you would buy it or not?

How much would you be willing to pay for it?

Would you buy this product like this?

Please feel free to ask any questions.

Thanks in advance

The Rookie

Survey monkey is a wonderful tool.......

1/ Would I buy it, probably not as the cost could probably be spend elsewhere for a better benefit
2/ Would pricing affect my decision, yes of course
3/ How much would I pay, about the cost of a higher end shifter and rear derailleur
4/ Thats the same as question 1!

It wouldn't be hard to add a wheel and crank speed sensor and use that to trigger an existing electronic shift system such as Shimano Di2, replacing the button inputs with an input from a controller.

02gf74

Interesting but it does smell of a solution to a problem that does not exist, certainly not now we have Di2.

In the olden days with multiple chain rings, it was a faff when changing gear that required changing the front ring along with rear sprocket. The move to 1x systems (11 and 12 speed) has done away with that. Also Shimano electronic system will deal with the changing front and rear at the same time and is done by a single button press.

For the sake of simplicity, cost and being in control, a single front shifter works fine for me. I am not sure I would like to have someone do the thinking for me - e.g. when on a steep hill, I don't like shifting when the chain is under maximum tension - I tend to speed up then slow down so putting less force through the drive train before shifting - I am not sure how an automatic system can do that.

I have not ridden electronic gears but having a battery puts me off - with manual gears I can not use a bike for months or years knowing that it will work, can the same be said with one that relies on a battery?

figbat

Would I buy it? No.
Am I interested to see someone figure out how to do it, for the sake of "because I can"? Yes.

I've not even got to current electronic shift options yet, let alone an automated version.

figbat

Also, I think this would work better on a road bike than an MTB.

w00dster

I've also thought about this - but always came to the conclusion that the benefit didn't warrant the amount of R&D needed. Bit of a mind dump from me......only my thoughts not in anyway a criticism.

1. Does this take into account the rider's fatigue levels? (Road use for example, for certain climbs I may use different gears at the start of a 100 mile ride than nearer the end)
2. How does it know what my preferred riding style is? (Road wise I may want to take a hill at 70rpm stood up, but my team mate may want to ride the same hill at 90rpm sat in the saddle - how would a computer know what our preference is?)
3. Whilst you have stated there is an option for rider age, does that mean that if using the system there is a category for age 40-45 for example? This will then say a 40-45 year old will want to climb at 90rpm using gear 36/25? Within the cycling world (and especially road cycling), we tend to use watts per kilo as a measure of fitness not age. This is also a more effective measurement for climbing than what age is. (Watts itself isn't ideal as a heavier rider may use different cadence / gears and have a higher power output than a lighter rider - I'm 67kgs, I may ride uphill at a different speed/cadence and in a different gear than what a 85kg person the same age as me may use)
3. Does this take into account wind speed or whether I am drafting? For example, cruising at 25mph in a peloton is not the same as riding at 25mph into a headwind. Do you need an algorithm that will work out wind speed/cadence/heart rate as well as any coming hills? When it is my turn to go to the front of the peloton/group I will drop down a gear or two, what is the delay in this? (I may only be at the front of the group for 30 seconds so would need to be instant?)
4. On a Mountain Bike, will it work out not only the gradient of a hill but also the length of the hill? For example, a short hill we may be fine to "mash" out in a larger gear as we know that in 15 seconds the hill will be over and we want to be in the same gear for going down the other side?
If we are on a longer hill will it know to gradually drop gears over time? For example, we may go into a hill at a strong pace and want to allow the hill to come to us (slowly changing gear as the hill / speed / effort demand) - this is very individual, but I wouldn't want to drop a handful of gears all at once because the computer has seen the oncoming gradient.
5. How does the computer know when is the most efficient time to change gear? If putting in a big effort on a climb we can't just change gear mid pedal stroke as this can lead to chain snapping or chain jumping off the front ring.
6. What happens when the battery is dead? Is this back to manual changing as normal?

I may not be the target demographic, but it may be worth trying to differentiate between road and mountain bike riding. Unless of course you are performing a theoretical uni study, then obviously you don't need to find a real problem to solve, only a theoretical one.

Generalising now, but cyclists tend to want to go faster for a longer period of time. Any device that gets added to the bike needs to help that criteria. I may add extra weight to my bike if I thought it genuinely helped me to achieve both of those criteria, but I wouldn't add something if all it did was reduce the need for me to move a finger - which to most of us is pretty much a natural instinct, almost just as natural as braking or pedalling. Because of how individual climbing is and to some extent, how technical, I think you will find this an incredibly tough project with no financial reward at the end. If you are serious about this, be interesting to know how often you ride, whether that is technical terrain, serious road riding or commuting only?

02gf74

as ^^^

You need to tell us what the system will do and what benefits it gives before anyone can answer the questions.

another issue is if you are racing and want to stand up on the pedals for a sprint finish - what will it do then - will it change gears down because more power is available all of a sudden to maintain the same cadence or what?

There's a lot more to it, but this is the basic overview to save you a read.

ok, I'll bite - where is the rest of it so I can have a read, may answer some of the above questions.

Kenny17

So essentially there are a lot of variables and considerations involved if it were to be made for an experienced rider?

cooldad

I can't see how it would ever work without some sort of psychic link.

Sometimes for a short sharp hill, you'll just stand and mash in a high gear - if it suddenly decides to shift down, you'll end up with trashed shins or crushed nuts.

And going down, especially if there's something technical at the bottom, I might stick in a low gear, reducing chainslap, and being prepared for whatever is ahead.

Or, to put it another way, gearing is often used in anticipation of something ahead, or not necessarily using optimum ratios for that specific bit of trail.

w00dster

The variables are considerable yes. Lots been listed here.
Not saying it is impossible. And if this is a Uni research project it would be interesting. But in terms of do-ability, I can't see it. (Hope we don't sound overly negative!)
I do think there may be something that can tie in to eBikes or possibly for older riders. But in terms of a product for the mass contingent of regular cyclists then I would say that the variables are indeed large and complex.

steve_sordy

This sounds exactly like the sort of task that an engineering student would be set at uni. It feels like you want us to do your homework for you. My time at uni didn't even have calculators, let alone the internet to assist, just slide rules, but good luck to you.

After skimming the other posts on this thread, I will add only one comment. Forget the age groupings! Do you even ride a bike?! It's all about fitness, strength and the quality of the joints. In other words, nothing to do with age, or sex for that matter.

02gf74

steve_sordy wrote:

It feels like you want us to do your homework for you. My time at uni didn't even have calculators, let alone the internet to assist, just slide rules

Slide rule :shock:

You were lucky, we had to queue up to use the logarithm book.