Lightweight motorcycle build. The HillFighter project

[Chalo]

The issue I see with ebikes and high cg is the tendency for the bike to be unstable and very wobbly.

Try riding a tallbike and then render your assessment on how wobbly it is. As long as the frame is structurally sound, it's surprising how much steadier it is than a normal bike.

Now in terms of maneuverability, a tallbike suffers. But in terms of stability, it's better than a normal bike.

Here's the last one I worked on. The lady for whom I helped build it rode to Mexico on it.


I'm not arguing that there's any advantage in raising the system center of mass of a bike. But condensing it-- placing the bike's mass closer to the rider's-- has no disadvantage once you're moving faster than walking speed.

Look at it this way-- folks pretty uniformly prefer the handling of a lighter bicycle to that of a heavier one. The lighter the bike, the more concentrated the system's mass is in the rider, high up.

Also consider this: lowracer recumbent bikes are widely acknowledged to be very difficult to handle. By your reckoning, these should be the best handling bikes around, but instead they're among the worst. What's your explanation?

 

[DingusMcGee]

WoodlandHills,

In the face of such self-assured and determined ignorance all I can do is withdraw

It seems some here haven't a clue as to what physical variables come into play and are pertinent when making a numerical based design choice. But lucky for them they can do real world experiments with the variables which for this discussion is maneuverability in terms of the positioning of the CoG. All numerical based design is confirmed as being somewhat supported/correct when confirmed by experiments. So in the above section you quote of mine, I have suggest [ if you cannot see the physics] then this is a way to experiment with CoG positioning. Granted attaching a pack adds inertia specifically to the bike's controlling agent, the bike rider, but it does meet the criteria of being able to easily alter the system's CoG's.

You say, "...determined ignorance.." Maybe we have very different objectives? I do know I want ease of initiating a turn -- call that trait maneuverability -- as opposed to stability for the trail riding and the trials like attempts I do. Stability is cherished for the high speed riders. It seems with the likes of evidence you present -- the Honda Race world story -- you ride in different circles than recumpense and I do and you have little experience with the likes of trail and trials riding. For the purpose of this discussion it seems you have some ignorance of our particular needs in bike handling and how to design them to do what we want. Your talk is all on the other end of design choices -- How to get more stability.

And yes I can show the physics and do the math for the effects of the higher CoG in terms of maneuverability. Raising the CoG gives a non-linearly increase in the second moment of inertia and that physical variable varies with the square of the distance form the point of rotation. Therefore this particular physical variable when moved short distances changes maneuverability radically.

For a different wording of this effect see:

https://en.wikipedia.org/wiki/Bicycle_and_motorcycle_dynamics

... (with a high center of mass) can be easier to balance when ridden than a low one because the tall bike's lean rate (rate at which its angle of lean increases as it begins to fall over) will be slower.

The idea of lean rate is the same measure as what the second moment of inertia concurs about the CoG's rotation. A higher CoG means you need more time or force to get a lean angle established but you have more stability.





Here is an 11 minute video with a bike design similar to what recumpense designs and makes.

http://plugbike.com/2011/11/17/electric-motion-em-5-7/


click down low on:

Electric Motion EM 5.7 trial électrique

No high CoG's here. The trail ebikes of mine have very short wheelbases, 41" and 43", steep head angles and of course low CoG's. They are not the greatest for high speed downhill. But they are nimble. And that Honda racing bike you talk of would utterly fail in these maneuvers--ha ha!

 

[recumpence]

The issue I see with ebikes and high cg is the tendency for the bike to be unstable and very wobbly.

Try riding a tallbike and then render your assessment on how wobbly it is. As long as the frame is structurally sound, it's surprising how much steadier it is than a normal bike.

Now in terms of maneuverability, a tallbike suffers. But in terms of stability, it's better than a normal bike.

Here's the last one I worked on. The lady for whom I helped build it rode to Mexico on it.
View attachment 1

I'm not arguing that there's any advantage in raising the system center of mass of a bike. But condensing it-- placing the bike's mass closer to the rider's-- has no disadvantage once you're moving faster than walking speed.

Look at it this way-- folks pretty uniformly prefer the handling of a lighter bicycle to that of a heavier one. The lighter the bike, the more concentrated the system's mass is in the rider, high up.

Also consider this: lowracer recumbent bikes are widely acknowledged to be very difficult to handle. By your reckoning, these should be the best handling bikes around, but instead they're among the worst. What's your explanation?

Allow me to add one more comment that may bring this to more of a conclusion (and this may be the reason Calo and I disagree, but are Calo in agreement);

My experience with high cg is also rearward cg such as weight added to a rear rack. I have seen umpteen people on this forum add weight (battery) to the rear rack over the rear wheel with nearly disastrous results. The high cg, rearward weight bias, and inherent instability of the rear rack and rear wheel flexibility can make a bike, quite literally, unrideable above a modest speed. This is a HUGE problem with many bikes on this forum.

There are, also, other issues to consider such as wheelie tendency (a massive concern on my builds) and ease of manoeuvring the bike manually that make a lower cg as well as making sure the battery is placed centrally or slightly forward preferable. My Motoped was heavy enough that it was difficult to push around and hold up without concentrating greatly.

To sum up, I guess I am saying a low CG has many benefits in all of the builds (15 or so) that I have done (as well as all of the builds I have read about on this forum) and no perceived downsides.

Matt

Matt

 

[gogo]

…low Center of gravity is always a good thing on any wheeled vehicle.

This is the part you got wrong. Its true for '3 or more'-wheeled vehicles, but 2-wheelers are different animals. I'm not sure about unicycles.

Wobbles probably have more to do with overstress and improper use than weight location. I'd agree that keeping weight low on weak bicycles will help avoid wobbles, but a suitably robust system is a better solution.

 

[ArtemT]

Started to design the battery. The aim is the following: engineer 2 batteries of 120 Tesla batteries each, 72V in total. All the batteries will be held together by means of custom holders, printed on a 3D printer (the volume of each holder is approximately 106 cc). Each battery will have a formula 6P20S. Parallel connection is implemented by means of laser cut copper plates, thickness 2mm. At the narrowest point the cross section of the plate will be about 6 square millimeters. The plates will be fastened to the holder on the glue. Every battery will be soldered to the copper with thin copper wire.

 

[Rube]

Interesting design, thanks for posting. Do you plan on coating or insulating the copper and links to prevent corrosion? Or is that not an issues?

 

[Lurkin]

Its a damn shame madm3chanic hasn't been by with updates. His more recent frame is very similar to yours and the battery issue was investigated at length and tested by him...

It would be a good idea to read these threads. He made frames and actually tested these concepts rather than debated various physics and engineering concepts.

https://endless-sphere.com/forums/viewtopic.php?f=28&t=80032

https://endless-sphere.com/forums/viewtopic.php?f=3&t=62345

 

[Wheazel]

The issue I see with ebikes and high cg is the tendency for the bike to be unstable and very wobbly.

Try riding a tallbike and then render your assessment on how wobbly it is. As long as the frame is structurally sound, it's surprising how much steadier it is than a normal bike.

Now in terms of maneuverability, a tallbike suffers. But in terms of stability, it's better than a normal bike.

Here's the last one I worked on. The lady for whom I helped build it rode to Mexico on it.
View attachment 1

I'm not arguing that there's any advantage in raising the system center of mass of a bike. But condensing it-- placing the bike's mass closer to the rider's-- has no disadvantage once you're moving faster than walking speed.

Look at it this way-- folks pretty uniformly prefer the handling of a lighter bicycle to that of a heavier one. The lighter the bike, the more concentrated the system's mass is in the rider, high up.

Also consider this: lowracer recumbent bikes are widely acknowledged to be very difficult to handle. By your reckoning, these should be the best handling bikes around, but instead they're among the worst. What's your explanation?

I would like to add to this discussion aswell with some real world experience.
The comparision I am going to make is between my lowracer recumbent, a challenge jester, and regular upright riding position bikes.

The first time I tried riding the Jester I was questioning myself what I had gotten myself into. The very low CG(of the complete vehicle) makes it VERY twitchy. It feels like it can fall over in a millisecond.
The best way I can describe it is that I had to learn to ride a bike again. Like I had to tighten up my senses' play and tolerances for a more stable ride.
Even after alot of riding it is still twitchy and not mentally relaxing eventhough the seating position is. (This applies to riding at speed aswell, eventhough the gyroscopic effect from the wheels help quite a bit)
On a high seating position bike, I can "fall" to one side for a comparably long time and still be able to recover.

My judgement says that the best compromise lies somewhere between the extremes discussed.
There are many things to factor in under stability and nimbleness that comes from alot of different riding scenarios.

 

[recumpence]

That is a great observation. I would tend to agree. To illustrate this, I do not have any extremely low CG bikes. I have been comparing very high CG bikes (battery on a rear rack over the rear wheel) to my bikes with the battery integrated into the frame (very centralized CG). The centralized CG bikes handle far better than high CG bikes. So, I felt like low CG was the cause. I bet it is centralized CG that is beneficial in my case.

Thoughts?

Matt

 

[Wheazel]

My thought is that different cog positions are good for different things if you "isolate" them, and then you gotta decide the best compromise depending on what the bike should be used for.
For braking performance you might want the bike to affect the vehicles cog downwards.
For general handling and nimbleness you might want it more centralized so that less mass is flung around the rotational center (which will be affected/moved depending on the mass properties of the vehicle)
For a slow relaxing ride, a higher seatingposition and therefor a higher cog seem to be beneficial.

I dont have certain facts on this matter, I am just going with my gut feel based on experience and a general interest in physics.

Also I think this will be a sexy build. Looking forward to see the progress further!

 

[ArtemT]

Interesting design, thanks for posting. Do you plan on coating or insulating the copper and links to prevent corrosion? Or is that not an issues?

Of course, there'll be a layer of insulation - I just have not yet fully designed all the components

 

[madin88]

awesome work. Thumbs up!
I'm going to follow your progress

 

[ArtemT]

awesome work. Thumbs up!
I'm going to follow your progress

Thanks! That's very encouraging!

 

[Ohbse]

Started to design the battery. The aim is the following: engineer 2 batteries of 120 Tesla batteries each, 72V in total. All the batteries will be held together by means of custom holders, printed on a 3D printer (the volume of each holder is approximately 106 cc). Each battery will have a formula 6P20S. Parallel connection is implemented by means of laser cut copper plates, thickness 2mm. At the narrowest point the cross section of the plate will be about 6 square millimeters. The plates will be fastened to the holder on the glue. Every battery will be soldered to the copper with thin copper wire.

I really like the new frame design and your high level pack criteria (20s12p) are good, however I think you've got few critical things wrong

First, Split your pack into two 10s12p packs, connect in series. This ensures you don't have any potential cell level imbalance (or do not require cell level interconnects, more wiring = more chance for issues). It also lets you run the shortest possible battery leads. You could have cell 1 on the bottom of the left side and cell 20 bottom of the right, ensuring the leads to your controller are as short as possible. This is desirable from a controller perspective.

Second, Don't solder 18650's. This has been covered LOTS of times, it's simply not the correct way to terminate the cells. Soldering introduces substantial amounts of heat in an inconsistent fashion. This leads to inconsistent cell performance which will lead to inconsistent cell voltages, a big headache. Spot welded pure nickel is easy, well understood, substantially cheaper/faster than what you're proposing, will keep balance better and can be configured to have the same low resistance as the copper solution. It suffers from less corrosion related problems as well.

Third, I would suggest using subtractive machining rather than additive for your housing. Depending on the material, it's likely that machined solid will be substantially stronger than 3d printed parts, especially over the relatively large distances your pack covers. Here's a simple test example I made the other day:


This was cut on a CNC router in only a couple of minutes and the fit is flawless. This plastic is very easily machined, dimensionally stable, temperature resistant, an excellent electrical insulator. You can easily fix it using course thread fasteners and pilot holes to other pieces to create very rigid structures. Don't bother with the fan cooling. If your cells are getting hot, you either chose the wrong cells or didn't use enough of them. What sort of power are you intending on pulling?

Fourth, your cell layout should be focused on creating A) maximum conductive area between cell groups and B) ensuring that that conductive area is as even as possible across all the cell groups. Your current layout has groups with effectively 1 cell touching and it has groups with all 6 cells touching. This would again over time lead to cell imbalance.

 

[sn0wchyld]

Started to design the battery. The aim is the following: engineer 2 batteries of 120 Tesla batteries each, 72V in total. All the batteries will be held together by means of custom holders, printed on a 3D printer (the volume of each holder is approximately 106 cc). Each battery will have a formula 6P20S. Parallel connection is implemented by means of laser cut copper plates, thickness 2mm. At the narrowest point the cross section of the plate will be about 6 square millimeters. The plates will be fastened to the holder on the glue. Every battery will be soldered to the copper with thin copper wire.

I really like the new frame design and your high level pack criteria (20s12p) are good, however I think you've got few critical things wrong

First, Split your pack into two 10s12p packs, connect in series. This ensures you don't have any potential cell level imbalance (or do not require cell level interconnects, more wiring = more chance for issues). It also lets you run the shortest possible battery leads. You could have cell 1 on the bottom of the left side and cell 20 bottom of the right, ensuring the leads to your controller are as short as possible. This is desirable from a controller perspective.

Second, Don't solder 18650's. This has been covered LOTS of times, it's simply not the correct way to terminate the cells. Soldering introduces substantial amounts of heat in an inconsistent fashion. This leads to inconsistent cell performance which will lead to inconsistent cell voltages, a big headache. Spot welded pure nickel is easy, well understood, substantially cheaper/faster than what you're proposing, will keep balance better and can be configured to have the same low resistance as the copper solution. It suffers from less corrosion related problems as well.

Third, I would suggest using subtractive machining rather than additive for your housing. Depending on the material, it's likely that machined solid will be substantially stronger than 3d printed parts, especially over the relatively large distances your pack covers. Here's a simple test example I made the other day:


This was cut on a CNC router in only a couple of minutes and the fit is flawless. This plastic is very easily machined, dimensionally stable, temperature resistant, an excellent electrical insulator. You can easily fix it using course thread fasteners and pilot holes to other pieces to create very rigid structures. Don't bother with the fan cooling. If your cells are getting hot, you either chose the wrong cells or didn't use enough of them. What sort of power are you intending on pulling?

Fourth, your cell layout should be focused on creating A) maximum conductive area between cell groups and B) ensuring that that conductive area is as even as possible across all the cell groups. Your current layout has groups with effectively 1 cell touching and it has groups with all 6 cells touching. This would again over time lead to cell imbalance.

nice looking holder... whats the material?

 

[Ohbse]

nice looking holder... whats the material?

HDPE - most commonly seen as plastic cutting boards

 

[ArtemT]

After countless changes and improvements the final version looks like this:

The battery consists of 240 elements with a total capacity of about 2,8 kW. The battery will be organized with these holders https://ru.aliexpress.com/item/18650-battery-holder-Cylindrical-cell-2-10-plastic-holder-18650-lithium-ion-battery-bracket-plastic-case/32729220653.html?spm=2114.10010208.1000014.1.HRRPOG&scm=1007.13338.50051.0&pvid=ebc7791d-752e-465e-a6cb-91e04dfb3fb1&tpp=1

 

[skeetab5780]

I like the look of it before you designed it around the battery, this looks awkward to ride. I like the swingarm before it was adapted to the whole bike in the last picture, shock and angle seems off.

 

[ArtemT]

I like the look of it before you designed it around the battery, this looks awkward to ride. I like the swingarm before it was adapted to the whole bike in the last picture, shock and angle seems off.

well, the inner space available for the battery was too narrow to hold all the elements - I had to split the battery and its design became too complicated and expensive...

 

[ArtemT]

Finally, the frame is almost ready. I still have to add a steering column, rear shock mount and battery mount:

 

[macribs]

I thought we were discussing theory as illustrated by real world experiments: very low cg as tested by Honda during the '84 GP season? The theory should still apply....... If it's correct.

Is this the honda?

 

[skeetab5780]

Frame is looking pretty sweet! Interested to see its geometry when the headtube and swingarm are put into the mix. I like the motor/pedal mounts

Keep us posted on updates!

 

[Overclocker]

looking good!

is that a BHT motor?

 

[ArtemT]

looking good!

is that a BHT motor?

Well, I actually don't know. It's a nameless motor - there's not a single letter on it. I bought it from lmxbikes.com. Adam Mercier, the owner, says that it has 35kv and 72V

 

[Overclocker]

looking good!

is that a BHT motor?

Well, I actually don't know. It's a nameless motor - there's not a single letter on it. I bought it from lmxbikes.com. Adam Mercier, the owner, says that it has 35kv and 72V

in that case we have the same motor. got it from BHT ebike. 36kv. i'll be running 4:1 ratio on 17" moped tire