The Fool on the Hill: Why not a bicycle?

This is a new post in a series of essays which, up to now, has been about tricycles. At the start I ruled out a bicycle because

1. I'm designing a vehicle for my old age;
2. my balance is deteriorating;
3. I live at the top of a considerable hill;
4. after Covid, I had considerable difficulty getting an unassisted bicycle up the hill;
5. I've always been wary of recumbent bicycles because of their alleged poor climbing ability.

Those are, together, fair reasons. So I've been looking at tricycles because they have good static stability. But they don't have excellent dynamic stability, and, looking at the geometry, I'm worried that the dynamic stability of my proposed design will be unacceptably poor.

It's poor because the proposed design has the single wheel at the front, and consequently both descending a hill and braking reduce stability; and that, combined with adverse camber, could cause the vehicle to flip over suddenly. You also can't move the centre of mass very far back, because the rider's feet have to be in front of the front wheel.

If you reverse the design and put the two wheels at the front, then you increase your effective frontal area, undermining the attempt to get good aerodynamics; and you carry that increased frontal area everywhere you go, slowing you up, whether or not you're descending.

So let's reconsider the bicycle solution. A bicycle has lousy static stability — it topples over very easily at very low speeds — but it has excellent dynamic stability. In general, with bicycles, you can solve the low speed toppling problem by putting your foot down, and in any case low speed falls tend not to cause injury.

However, in the special case of a fully faired bicycle, you can't put your foot down. Several fully faired bicycles — again I'm thinking of Beano, Snoopy and Woodstock — use retractable landing gear to keep the vehicle upright at speeds lower than that at which dynamic stability is established.

They are not, however, typically being used on bloody big hills. If, climbing a bloody big hill, you had to deploy retractable landing gear, then unless the landing gear wheels roll really well, you're increasing drag at a point where you are by definition on the limit anyway. And getting off and pushing any recumbent — let alone a fully faired one — up a hill is going to be a fair bit more awkward than pushing an upright bike up one.

However...

However, I'm going to use electric assist in this vehicle anyway. So let's run the numbers. The formula for watts consumed for elevation gain is

(mass (Kg) x gravity (m/s²) x elevation gain (metres)) / time (seconds) = watts

So let's simplify that by considering a single second:

(mass (Kg) x gravity (m/s²) x rate of climb (metres/second)) = watts

Let's punch in some numbers:

Assume that the bike has dynamic stability above 8 Km/h, that the bike has a total weight including motor and battery of 30 Kg, that I weigh 70 Kg, my clothing weighs 5 Kg, and that the maximum cargo I want to carry is that damned gas bottle, which weighs 35 Kg, and that the maximum gradient is 10% (I think our track may be steeper than this, need to check, but the main road isn't):

That comes out at 305 watts (ignoring aerodynamic and mechanical drag for the moment, but on this bike at this speed they're small). The maximum I can sustain over any significant time these days is probably 200 watts, the maximum legal motor output is 250 watts, so neither I nor the motor, alone, can do this. But together, I and the motor could output 450 watts, which should be plenty.

I could, in fact, on these numbers, almost do a 15% gradient; also, very encouragingly, with no cargo, I could manage an 8.7% gradient unassisted.

So this looks moderately doable, provided I can get dynamic stability by 8 Km/h and keep the bike weight under 30 Kg.

General arrangement

So, suppose, instead of a tricycle, we go for a bicycle, what does it look like? Well, I was trying to avoid having to combine front wheel drive with front wheel steer, but the alternative to that, on a bicycle, is to run the drive chain all the way through the belly of the beast to the rear wheel, and I really don't want to do that, because maintenance problems more than anything else (although, yes, it would also add weight).

So I would have to bite the bullet and combine front wheel drive with front wheel steer. I think I would go for a Beano-style arrangement with the primary chain on the right hand side driving a gear system on a cross-shaft above the steering pivot, which would in turn drive a secondary chain on the left hand side which would drive the front wheel. To avoid adding further complexity, the motor would be in the rear hub.

So this ends up being a vehicle which looks very like Woodstock, but with (at least) the following differences

1. Cross shaft and left hand secondary chain;
2. Battery and motor driving the rear wheel, battery probably positioned under the slope of the seat back;
3. Hydraulic disk brakes on both wheels;
4. At least the rear of the upper shell permanently bonded to the lower shell, with a permanent roll-over hoop or bulkhead immediately behind the seat;
5. A four or five point harness for the occupant, to prevent being thrown out in a crash;
6. About 300mm extra length behind the seat for cargo bay;
7. Larger diameter landing gear wheels;
8. Front and rear lights and also turn signal lights, for use on the public road;
9. Raspberry Pi or similar to do data logging, motor control and data display;
10. Rear view camera mounted in the tail and displaying to the Raspberry Pi's display screen;
11. Thought given to better weatherproofing around the front wheel aperture.

Issues

Turning circle

On a vehicle like this, having more than a few degrees of steering lock greatly harms aerodynamics. At speed, Beano, Snoopy and Woodstock can all manage remarkably tight turns on the tracks they race on — certainly tight enough to manage a 'T' junction turn from a two lane road onto another two lane road. But whether they can turn that tightly from a standing start I don't know, and I would say that I'm skeptical.

Access

Both Stephen Slade (Beano) and Russell Bridge (Woodstock) access their vehicles by lifting off the whole top half of the shell, sitting in, and lifting the top half of the shell back on. There are two reasons this doesn't work for me:

1. It obviously compromises structural integrity in roll-over, which has to be a risk — these vehicles are fundamentally egg-shaped and in a crash they're likely to tumble or roll;
2. Managing the quite large upper shell in a public place, particularly in wind, strikes me as being potentially hazardous.

I would prefer the covering for the access aperture — the 'door' for want of a better word — did not need to fully detach from the bike. I would prefer to access just through the windscreen aperture, so that both the front part of the upper shell, and the rear part of it, could be permanently bonded to the lower shell.

But this may not be practical.

My guess — I have not yet seen it done — that the way people get into these vehicles is to straddle the lower shell, sit down, and then tuck their legs in. Getting in from one side may not be practical for balance reasons. Also, any increase in height between the seat and the 'sill' may make things still more awkward, especially as I get older and less flexible. Some very sketchy prototype work can probably test this.

It would mean that the landing gear, when deployed, would have to be able to take the weight of the user getting in and out.

Tagging this 'Tricycle', since it relates to other posts under that tag.

Share on social media

I'm no longer offering links to kleptocrat owned social media. If you're still using it, please think about whether you should be.