A test mule with virtual pivot steering

BalorNG

10 kW
Joined
Nov 10, 2019
Messages
516
I've tried creating a very interesting contraption - a virtual pivot steering bike with no had tube (the forks are something to attach bars to), virtually zero 'trail' in 0 deg handlebar position but greatly increasing positive trail when bars are deflected. Zero flop... basically XYZ cargo bike reversed.
It does feel kinda strange at slow speed for some reason (likely due to large tiller), but still feels very stable at high-ish speed (it is just bolted together, no brakes except for regen braking so it is scary to go fast). Preliminary tests are encouraging!
This is just a test mule, no pedals just e-drive.
My ultimate goal is a stable and compact fully faired recument that has no 'wind steering input' but self-corrects to the wind by means of CG/CP disribution (like a well-designed aircraft) and pneumatic trail effects, yet still stable at speed.

gRBFiUyh.jpg

GkuUjXvh.jpg

x6s7fb3h.jpg



Dual reduction with possibility of 80nm of torque, but frankly less than half is already a bit scary due to no brakes other than regen and overall 'shabbiness' of it all.
Did I mention wireless brake and throttle? :) Yea, not truly 'streetworthy' but my current goal is just to test the concept...
 
Warren said:
I would love it if you did a video of the steering mechanism. So does the caster actually vary with displacement from center, by the upper and lower arms following different arcs?

Exactly!
I may do a video later...
 
MorbidlyObeseKoala said:
Are all of your pipe clamps bike stems? That is creative haha

No, not all. Some are 3d printed/laser cut sandwhich pillow blocks :)
I don't weld and using composites for a 'test prototype' is luxury I cannot afford...
 
Virtual pivot steering kinematics. It does not show half of it though - real steering axis is 'instantaneous' and is located at intersection of diagonal steering members.
https://www.youtube.com/watch?v=Hhk93m9bWuM
 
Thanks for the video. I now realized there are not two links on each side...an upper, and a low...but a single link on each side. You are moving the tire patch relative to the vehicle center line without leaning the wheel vertically, relative to the rear wheel, like with normal forks. Interesting.
 
BalorNG said:
Virtual pivot steering kinematics. It does not show half of it though - real steering axis is 'instantaneous' and is located at intersection of diagonal steering members.
https://www.youtube.com/watch?v=Hhk93m9bWuM

Now that is cool. I don't know if I could get used to the steering axis location and reach changing that much with steering angle.

Seems like it might work favorably on a trike.
 
Chalo said:
BalorNG said:
Virtual pivot steering kinematics. It does not show half of it though - real steering axis is 'instantaneous' and is located at intersection of diagonal steering members.
https://www.youtube.com/watch?v=Hhk93m9bWuM

Now that is cool. I don't know if I could get used to the steering axis location and reach changing that much with steering angle.

Seems like it might work favorably on a trike.

That's pretty much how Velayo RWS velomobile works AFAIK, just even more extreme I think.
With VPS of this kind you can uncouple or actually reverse 'side force steering input torque' while preserving return to centre force inherent in positive trail and which is essential for stable handling.

Again, my goal is zero side force steering input but still plenty of 'trail' in latter sense. I've fully expected it not to really work, but it seems it does - maybe by merit of pneumatic trail and gyroscopic force alone... I'll need to do more experiments, but weather is particularly nasty of late, and I don't want to risk a short circuit.
 
Warren said:
Thanks for the video. I now realized there are not two links on each side...an upper, and a low...but a single link on each side. You are moving the tire patch relative to the vehicle center line without leaning the wheel vertically, relative to the rear wheel, like with normal forks. Interesting.

No, you get the same effect with positive trail actually, just not readily apparent. I've also thought how that affects things - indeed, it directly affects balance before there is any change in direction even, and must play a role in singletrack dynamics.
I've not seen this effect mentioned anywhere, in fact... and it should further interact with pneumatic trail effects if you think about it. Did I mention that singletrack dynamics is damn complex phenomena? :)
 
BalorNG said:
No, you get the same effect with positive trail actually, just not readily apparent.

With a traditionally fork with positive caster, the wheel leans, in addition to moving the contact patch off the center line.

How is this design functionally different than a vertical headtube bike?

https://motochassis.com/articles/experiments-with-steering-geometry/
 
Warren said:
BalorNG said:
No, you get the same effect with positive trail actually, just not readily apparent.

With a traditionally fork with positive caster, the wheel leans, in addition to moving the contact patch off the center line.

How is this design functionally different than a vertical headtube bike?

https://motochassis.com/articles/experiments-with-steering-geometry/

Because the steering axis itself ALSO moves (virtually), and hence lever arm from rolling resistance, braking and lateral forces.
On a typical head tube, you cannot have zero lateral force intput at zero deg, but *increasing* with angles of wheel defection.
Also you cannot have lateral force input and tons of return to centre force from RR.

And yes, I have a vertical head tube bike with negative offset :)
SbYPkxZh.jpg
 
BalorNG said:
Because the steering axis itself ALSO moves (virtually), and hence lever arm from rolling resistance, braking and lateral forces.
On a typical head tube, you cannot have zero lateral force intput at zero deg, but *increasing* with angles of wheel defection.
Also you cannot have lateral force input and tons of return to centre force from RR.

And yes, I have a vertical head tube bike with negative offset :)

How bad is the pedal steer on that vertical head tube, moving bottom bracket, FWD bike? I love the idea of a moving bottom bracket, FWD, but I find the pedal steer very annoying. Putting the cranks at the center of the axle reduces it, and some bikes put the cranks behind the wheel, which reverses the pedal steer, maybe? What I want is NO pedal steer. Would that be possible with a virtual pivot design?
 
Warren said:
BalorNG said:
Because the steering axis itself ALSO moves (virtually), and hence lever arm from rolling resistance, braking and lateral forces.
On a typical head tube, you cannot have zero lateral force intput at zero deg, but *increasing* with angles of wheel defection.
Also you cannot have lateral force input and tons of return to centre force from RR.

And yes, I have a vertical head tube bike with negative offset :)

How bad is the pedal steer on that vertical head tube, moving bottom bracket, FWD bike? I love the idea of a moving bottom bracket, FWD, but I find the pedal steer very annoying. Putting the cranks at the center of the axle reduces it, and some bikes put the cranks behind the wheel, which reverses the pedal steer, maybe? What I want is NO pedal steer. Would that be possible with a virtual pivot design?

Levand built virtual pivot steering MBB, but no, unfortunately not. In the end, he did not like it too much and still prefers LWBs (to be frank, so am I now, but mostly because you cannot build a streetable MBB streamliner unless you are Tom Traylor...).

You may want to read this tread:

https://www.bentrideronline.com/messageboard/showthread.php?t=159941

"the pedal feedback is there but unless you are pushing hard its minimal and 1 handed is very easy. I cannot go no hands yet but can just use my finder tips on the bars and ride comfortably."

'Pedal steer' is not the greatest problem you can adapt to this (though it DOES tire out my hands critically on ultra-distances, but that's 8+ hours of non-stop pedalling), the greatest problem is 'double flop' and steering inertia. Vertical steering axis fixes that, and Torkjunky actually had a 'real Cruzbike', but sold it because he was not comfortable enough with the handling.

You want to pay extra attention towards my 'leg push vs arm pull' musings from that thread.
 
BalorNG said:
because you cannot build a streetable MBB streamliner unless you are Tom Traylor...).

You may want to read this tread:

Flashback! I met Tom at an IHPVA meet at the fairgrounds in Milwaukee. He made MBB look easy. I also meet Bryan Allen, Paul MacCready's pilot. Paul was there, but I didn't actually approach him. He was my hero. He actually bought a high aspect ratio Soaring Wing kite my wife and I were making in the 1980s, from a kite shop in California. The shop owner sent us a letter telling us about it. I am still pleased about that, all these years later.

I once built a FWD, RWS recumbent, because of my desire for a short chainline. Rode it about fifty miles. It was ridable, and of course has zero pedal steer, but never felt relaxed at speed.

27FDRWS.jpg

After dozens of bikes, I have decided my dream bike is a dual 559-50, 60" WB, with a coaxial 250 watt generator bottom bracket, directly over the front axle. Grin single-sided DD hubs, front and rear (like Mike Burrows designs, for ease of flat and tire change), progressive hall sensor brake lever regen on both wheels, V-brakes for backup, throttle w/cruise control, plus torque sensing option from generator signal, two kWh battery under the seat. Chromo steel, Tig welded, about 80 pounds all up. Super simple and reliable, silent, all electric.
 
Warren said:
After dozens of bikes, I have decided my dream bike is a dual 559-50, 60" WB, with a coaxial 250 watt generator bottom bracket, directly over the front axle. Grin single-sided DD hubs, front and rear (like Mike Burrows designs, for ease of flat and tire change), progressive hall sensor brake lever regen on both wheels, V-brakes for backup, throttle w/cruise control, plus torque sensing option from generator signal, two kWh battery under the seat. Chromo steel, Tig welded, about 80 pounds all up. Super simple and reliable, silent, all electric.

Hah! Great minds think alike. ;)
I also like concept of coaxial 'MWB' with series hybrid drivetrain - search 'Gagarin recumbent' on youtube, just replace the twist chain drivetrain with jackshaft IGH to a generator, and indeed you now have possibility of AWD and even AWS - and without moving bottom bracket!

And indeed, while cantilereved wheel is not that important on the rear, a coaxial drivetrain is going to be a total PITA in case of a front tire flat, and there is actually a predencent of cantilevered drivetrain with front IGH (one of kervelos).

However, the main point here is solution of 'recumbent packaging problem' as per Craig Cornelius (cranks 'wanting' to be where front wheel is), and there are other ways to 'skin a cat'. For instance, by staggering two bottom brackets on space frame design like one above and connecting those with a common jackshaft with dual chains (and than using it as first step of reduction to small, lightweight generator like I did with my series hybrid project) should allow placing cranks with 'virtual axle' intersecing the wheel anywhere, allowing full suspension and easy wheel changes while solving the 'recumbent packaging problem'.
 
Yeah. It would be great to be able to have cranks where the wheel is, but as you discuss, you end up with a bunch of horrible compromises. Thankfully, my body is fine with the feet 6-10 inches above my butt, as long as I am not lying on my back. A 45 degree seatback angle is laid back enough for me, and I don't end up with massive tiller steer.

By coaxial generator, I was referring to this setup,

https://electrek.co/2021/08/31/radical-new-electric-bike-drive-systems-requires-no-chains-or-belts-entirely-ride-by-wire/

wound for a "50 volt" battery, rather than the wimpy "36 volt" they use. Just put it in a short boom, from the headtube, like a conventional recumbent, but no chains. With the generator/cranks directly above the axle, and pedals just clearing the tire in a turn, my heels clear.

The reason for one-sided rear is, I never want to have to deal with motor connectors. Fixing a flat on the side of the road is a PITA already. Add in torque arms, and electrical connectors, and it goes to another level.

Fixing flats on my mid-drive, with normal wheels was fine. In 33K miles on the DD hub bike, I did three flat repairs on the side of the road. If I was 20, it would be OK. At my age, it is barely doable. The Grin motors are much lighter than the massive Crown motor on my bike, and they use a single plug, rather than the phase, hall, and thermistor plugs on mine. But you still need to deal with torque arms.
 
Warren said:
Yeah. It would be great to be able to have cranks where the wheel is, but as you discuss, you end up with a bunch of horrible compromises. Thankfully, my body is fine with the feet 6-10 inches above my butt, as long as I am not lying on my back. A 45 degree seatback angle is laid back enough for me, and I don't end up with massive tiller steer.

By coaxial generator, I was referring to this setup,

https://electrek.co/2021/08/31/radical-new-electric-bike-drive-systems-requires-no-chains-or-belts-entirely-ride-by-wire/

wound for a "50 volt" battery, rather than the wimpy "36 volt" they use. Just put it in a short boom, from the headtube, like a conventional recumbent, but no chains. With the generator/cranks directly above the axle, and pedals just clearing the tire in a turn, my heels clear.

The reason for one-sided rear is, I never want to have to deal with motor connectors. Fixing a flat on the side of the road is a PITA already. Add in torque arms, and electrical connectors, and it goes to another level.

Fixing flats on my mid-drive, with normal wheels was fine. In 33K miles on the DD hub bike, I did three flat repairs on the side of the road. If I was 20, it would be OK. At my age, it is barely doable. The Grin motors are much lighter than the massive Crown motor on my bike, and they use a single plug, rather than the phase, hall, and thermistor plugs on mine. But you still need to deal with torque arms.

Oh, I see. Yea, that's neat setup - I don't trust their numbers at all, but it should still contribute meaningful range while certainly being cool and compact.
I DO have serious problem with high BBs, that's why I love the concept of moving BBS, LWBS and coaxial cranks - both allows low BB and decent rolling resistance.

Good point about hub motor - that's why I personally prefer middrives.
 
I am done with mid-drives through the pedal drivetrain. Absolutely loved the way it worked, but there is no redundancy, when a freehub fails, or a derailleur is destroyed. Had a chain link fail...no problem. Happened on our tandem too. That is why I carry extra links, a chain breaker, and quick links. Never had a problem in decades of just my effort. Double that, or more, and things break. Had the drivetrain fail twice in 33K miles on the mid-drive.

A left-side mid-drive is great, but again requires a chain, or belt. And I really like the idea of regen braking on both wheels. And having two motors and controllers makes for redundancy to get home.

Justin of Grin Tech did a presentation on this idea, minus being recumbent and pedal by wire. He thought it would make for the perfect bike. He said he would like to offer the progressive regen brake levers, but his volume is too small for suppliers to be interested.
 
Warren said:
I am done with mid-drives through the pedal drivetrain. Absolutely loved the way it worked, but there is no redundancy, when a freehub fails, or a derailleur is destroyed. Had a chain link fail...no problem. Happened on our tandem too. That is why I carry extra links, a chain breaker, and quick links. Never had a problem in decades of just my effort. Double that, or more, and things break. Had the drivetrain fail twice in 33K miles on the mid-drive.

A left-side mid-drive is great, but again requires a chain, or belt. And I really like the idea of regen braking on both wheels. And having two motors and controllers makes for redundancy to get home.

Justin of Grin Tech did a presentation on this idea, minus being recumbent and pedal by wire. He thought it would make for the perfect bike. He said he would like to offer the progressive regen brake levers, but his volume is too small for suppliers to be interested.

Well... my personal experience with left-sided middrive suggests that once tightened, it is possible to remove the wheel, reinstall the wheel and force the belt back on no problem. It is even easier with a chain, because it is it not as finicky about tension.

Belt is almost silent, and if you don't go overboard with too small 'pinions' it should act a long time and need no maintenance. Chains are fine too if you wax them.

So, you have a one step reduction belt or chain, no tensioners, regen capable, no clutch to break. Even better performance with 20" wheel.

You might have seen some of my musings about using multiple *middrive* motors, including fixed gear, overrunning clutch and even centrifugal clutch.
The problem with coupling e-drive with human power is that their torques sum up and result in huge chain tensions, and human power is low RPM/high torque anyway, which is actually pretty hard on chains.
By going with high RPM left-sided middrive with large rear cog, you get relatively low chain tension but high 'thrust' due to torque multiplication, while with human power you actually have torque reduction (front chaining is larger than the rear).

Personally, I don't think that you need a really powerful regen brake, powerful enough to serve as 'panic brake', and on both wheels!

On my Lwb with overall weight (cause I'm frankly fat) approaching someone light on a decent-ish motorcycle, I can still get away with measly 20nm of braking torque (rather underpowered setup) that gets activated by brake handle to scrub speed before using front disk brake to come to a complete stop - I'm harvesting like 90% of energy possible (and with high efficiency!) and with similarly little brake pad wear.
And I can simply take an other belt with me in case something happends (a cheap HDT 5M belt!)

If my overall weight was less or I'd have a bit more braking power - I'd be able to harvest like 99% except 'panic braking' - and than unless you have a HUGE battery panic braking from high speed using regen alone can damage BMS/Battery, and hopefully that happends very rarely.

Now, if I was to travel with full fairings somewhere in steep hills but still maintain high efficiency - than yea, you'll need VERY considerable propulsion AND braking torques. I'll be experimenting with centrifugal clutch for that...

DD motors, however, must be very heavy to provide enough torque with decent efficiency in a bicycle-sized wheel.

Btw, I've checked the simulator and my current setup is pretty much functionally equivalent to Grin All Axle run with 20A controller, however it is 'underpowered' on purpose so I maintain peak efficiency, but I can absolutely maintain same regimes with less heat losses... yet it weights a 4x less more (700 gr motor and about a pound of hardware) and costs *literally* an order of magnitude less - again, including belts and laser cut/3d printed hardware that is dirt cheap nowadays.

By switching to only a bit more powerful motor and gaining about half a pound, I'm getting more torque AND more efficiency and that motor costs like 20$ more - and you don't have to deal with motor connectors and torque arms ever as well!

But if you want 'zero maintenance' and absolutely dead silent and don't mind some (ok, lots) extra weight - than yea, a couple of those Grin motors sounds about right.

You *will* get lower efficiency at cruise due to double cogging losses though, which is pretty significant (about 50w at 50 kmh as I understand, PER motor. Basically, at given generator losses, you'll need to pedal quite vigorously just to compensate for cogging :))

In my case that's 30w at 50kmh.
I think you really want a much lighter and much more torque-dense front hub geared motor, overall torque is going to be pretty damn huge for bicycle (already approaching moped territory), but zero cogging when off, with disk brakes on the front.
 
Two motors is definitely less efficient than one, as Justin told me too. As to weight, two Grin DD hubmotors are 8 kg total, and good for 3 kW peak total, and probably still as overall efficient as one, less efficient, 9 kg, 3kW Crown motor.

What I will probably settle for is putting one fast wind Grin Tech DD hubmotor below the seat, running a fat 1/8" BMX chain on a 16 tooth VeloSolo on the left side, to a 22 tooth VeloSolo on the rear wheel, on my old Linear. Their fastest wind motor, on a 14s battery, should get me to ~30 mph hot off the charger. I would move the battery to ahead of the motor, and switch from my saggy old 12s LFP pack to a 14s NMC pack. The LFP was 25# for 1.1 kWh. The newer NMC could be twice the kWh for the same weight and size.
 
Warren said:
Two motors is definitely less efficient than one, as Justin told me too. As to weight, two Grin DD hubmotors are 8 kg total, and good for 3 kW peak total, and probably still as overall efficient as one, less efficient, 9 kg, 3kW Crown motor.

What I will probably settle for is putting one fast wind Grin Tech DD hubmotor below the seat, running a fat 1/8" BMX chain on a 16 tooth VeloSolo on the left side, to a 22 tooth VeloSolo on the rear wheel, on my old Linear. Their fastest wind motor, on a 14s battery, should get me to ~30 mph hot off the charger. I would move the battery to ahead of the motor, and switch from my saggy old 12s LFP pack to a 14s NMC pack. The LFP was 25# for 1.1 kWh. The newer NMC could be twice the kWh for the same weight and size.

Yup, that sounds about right! In fact, you can use more reduction and a stand-up scooter motor with disk brakes 6hole pattern to mount the cog to... I think there are off the shelf "brake mounts to 104 BCD spiders" adapters already? If not, just order one laser cut, again - that's almost nothing given what batteries and motors cost in general.
 
BalorNG said:
Yea, it does look cool, but frankly is an overpriced toy. Not suspension at all? Eh.

Yeah. If they are planning to run 50 mph on that bike, I would definitely want suspension, and full MC racing kit. My cargo bike is much longer, and I have had the rear wheel hop to the outside a few inches at 35 mph on a bump, on a downhill curve. Not scary, but definitely made me see the benefit of suspension at speed.

I don't see any point to the self-balancing setup, unless you are paralyzed. I suppose it might help on a streamliner, if it reacts quick enough.

I couldn't tell if they had a generator crankset, or were just wasting the pedal effort. I figured with two wheel drive, it would be easy for a computer to do ABS. Sounds like they have done it.

I love the single sided rear stay....obviously. I don't get why they skipped the front.
 
Back
Top