[WIP] 40+ km/h racer - need your help!

[TechnicSummse]

22 minutes ago, nicjasno said:

This is not very good for high speed runs on very uneven surfaces.

Hmm what do you mean?

The soft suspension, because of the low ground clearance?

I think on my testtrack (pretty much like regular road asphalt) it should work pretty good. I had the best results until now with my crashed car, using the same suspension-system... well it was a bit harder... but i can adjust it, to be a bit harder ;)

[nicjasno]

There is no suspension at all. What you have here is a vehicle that is going to skip uncontrollably over bumps, small stones etc...

The higher the speed, the more important a good chasis becomes. While you have figured out how to get the most out of the motors, it's the chassis that is holding you back.

The front needs a bionicle ball joint front suspension with caster and camber, springs and shock absorbers. This is what you need at the very least.

The scale of the car is what makes this necessary. A 1:1 car would not really need any of that, because the road surface would be smooth for it, but for a lego car, the road is a pothole ridden obstacle course.

[TechnicSummse]

 

1 hour ago, nicjasno said:

There is no suspension at all. What you have here is a vehicle that is going to skip uncontrollably over bumps, small stones etc...

The higher the speed, the more important a good chasis becomes. While you have figured out how to get the most out of the motors, it's the chassis that is holding you back.

The front needs a bionicle ball joint front suspension with caster and camber, springs and shock absorbers. This is what you need at the very least.

The scale of the car is what makes this necessary. A 1:1 car would not really need any of that, because the road surface would be smooth for it, but for a lego car, the road is a pothole ridden obstacle course.

Did you look at the video... maybe i did not show it well enough there?

70-80% of the cars weight (meaning basicly the whole body) is connected via flexible beams to the front and rear axle. It's like a leafspring-suspension. This is the best what i could do.. it is more then i thaught i could do with this project.

To reach the highest possible speed, the motors need to be connected as direct as possible to the rear axle. In my case this means just 1 pair of gears between motor and axle. Thats the reason, why i can not put a suspension between axle and motors, and the motors have to be part of the rear axle and like that, part of the unsprung mass.

But at the end... thats it.. only the axles/wheels (front+rear) are part of the unsprung mass... everything else (wich is just the battery-boxes -> ~55% of the whole weight) is sprung-suspendet. But like this is built... a big part of the motors weight is also sprung-suspendet, because the motors rotate arround the rear axle.

(Well i have to admit... the PF-box is not 100% suspendet, because it is on the motors end of the flexible beam, i could change this, but it was easier to build like that)

 

To the front wheel...

...camber on just 1 single wheel doesnt make sense ;) ... but with caster... you are right... and thats what i got here . The whole steering is based on caster.

 

I hope the following video, wich i made especially for you  could show it in a better way:

___________________________________________________________________________________________________________________

 

To todays progress:

I made a short testdrive, wich was really fast. The best i could see on the speedo was a 16kph (*2,25 gearing=36kph ), but it was far away from maximal possible speed. I made 2 runs where i wanted to go for full speed, but i came across the following problem at both runs... seems i need to reinforce the steering:

But except this problem...the steering works much better then i thaught... maybe a bit to good... it steers pretty much, wich makes it hard to make just small corections... but makes it really fun, to drive arround :D

The suspension also works really good, i can drive over small bumps without problems...it drives over them like there would'nt be any bump. I will reinforce the steering, and give it another try tomorrow.

 

[nicjasno]

The problem is, that you need bigger suspension travels and a softer suspension.

What i see in your model is just some chasis flexing (i assume this is what you interpret as "suspension") . This causes just lots of vibrations, which in term cause the wheels to loose traction, the steering to fall apart, etc...

Your setup is good for a flat surface like a gym hall, not for a road outside.

I hope this makes sense.

[TechnicSummse]

@nicjasno Last week i made another video especially for you (and for me, to see what really happens with my car )

Please watch it at 0.25 speed setting on youtube.

 

_____________________________________________________________________________________________________________________

Last week i also wanted to test my latest setup at full speed.

I thaught i found a nice place for my testruns:

It looks really nice ... if you are far far away...

... but if you take a closer look....

.... you can see... 

.... it is not the best place for a Lego high speed car to run...

 

So i had to go back to my old testtrack on the "penny-market" parking.

Performance was really nice... i had some nice results at the speed computer... but... i simply can not run it at full speed. I had a lot of problems with oversteering at 30+kph with the new frontwheel.

That's why i decided, to rebuild my steering-system to something similar to the ones @Marxpek and @mocbuild101 used on their fast cars. Below you can see some pictures and a video of the new steering:

The way i designed the steering now... i am also able, to use the speedcomputer at the frontwheel, instead of the rearwheels... for a more accurate measuring (some claimed about possible slipping at the rear axle  ).

 

Well... but sadly the new steering did not bring any better results. It is way more accurate, drives more straight... but still... at higher speeds it oversteers exactly as my previous steering-system. Every little steering results in an instant complete turn of the car.

But i think i allready found the problem... i think the supporting wheels are mounted to close to the front of the car... and if they hit the ground, this results in a verry small turning radius like this.

I will make some changes again, and hope for good weather tomorrow.

[nicjasno]

I could help you design a steering that would actually work at high speeds if you want. I am really interested how fast this thing can really go.

Also, it's not hard to see why the car would oversteer. The wheels loose traction, because there is no suspension that would ensure they keep in contact with the road. The flex in your chasis is not suspension. It's just a flexing chasis.

Edited September 2, 2017 by nicjasno

[PorkyMonster]

6 hours ago, TechnicSummse said:

That's why i decided, to rebuild my steering-system to something similar to the ones @Marxpek and @mocbuild101 used on their fast cars. Below you can see some pictures and a video of the new steering:

  I've tried to bring across the point that a "caster wheel" is NOT a wheel with "caster angle" to @Marxpek, in his thread before, but failed ... 

Let me try again here .

First, what we want, in a fast vehicle, are wheels that can self-align to the vehicle, NOT to the direction of movement - and this behavior should be more apparent the higher your speed is. Second, we want more steering efforts the higher the speed - so we'll need to send more power to the servo, or we'll need bigger bumps, in order to change the direction of the vehicle.

Caster wheels, the same kind used in shopping trolleys, and as implemented in your models, DO NOT exhibit the above properties. You need wheels that have their steering axle properly angled in order to achieve high speed stability.

 

Regarding suspension - I agree with @nicjasno that you need proper suspension (yes, there might not be significant effect on speed due to lost traction (possible reasons mentioned in my earlier post), but suspension will affect steering and that will prevent you from staying the course and reaching the highest possible speed). For proper suspension to work, the unsprung weight must be much less than the sprung weight (target less than 10%, even lower if possible). If the ratio is not significant enough, the wheels (being part of the unsprung weight) will send vibration up onto the rest of the vehicle and cause the entire racer to rock and you'll lose traction as a result.

As for hardness/travel of suspension, that really depends on your speed and track surface - something to be experimented (in general, harder+shorter for flat surface, softer+longer for rough surface...).

 

Edited September 3, 2017 by PorkyMonster

[mocbuild101]

7 hours ago, TechnicSummse said:

Well... but sadly the new steering did not bring any better results. It is way more accurate, drives more straight... but still... at higher speeds it oversteers exactly as my previous steering-system. Every little steering results in an instant complete turn of the car.

Your steering lever is too long (2 studs point-to-point), my steering - and I think @Marxpek's too - used a 2L beam, which resulted in only 1 stud distance between the turning points, meant that the wheel didn't move as much.

[nicjasno]

What @PorkyMonster said. I think i speak for us all here, we want to see this thing go as fast as possible. But for this, you need to completly redesign the chasis.

The suspension type that would best fit this is that of a baja truck. With the narrow motorcycle wheels of course. But you need a suspesnion travel of at least 2-3 studs, and the suspension must be soft enough that it settles about 1 stud when you put the weight of the car onto it. This ensures that the wheels remain in contact with the ground even if they encounter a hole/dip in the road surface.

Edited September 3, 2017 by nicjasno

[TechnicSummse]

8 hours ago, PorkyMonster said:

  I've tried to bring across the point that a "caster wheel" is NOT a wheel with "caster angle" to @Marxpek, in his thread before, but failed ... 

Let me try again here .

First, what we want, in a fast vehicle, are wheels that can self-align to the vehicle, NOT to the direction of movement - and this behavior should be more apparent the higher your speed is. Second, we want more steering efforts the higher the speed - so we'll need to send more power to the servo, or we'll need bigger bumps, in order to change the direction of the vehicle.

Caster wheels, the same kind used in shopping trolleys, and as implemented in your models, DO NOT exhibit the above properties. You need wheels that have their steering axle properly angled in order to achieve high speed stability.

I know about that... i am a car mechanic and i lead a garage ;)

The design is not meant to be like the one on a shoppingcart... but it has to have the turning point infront of the wheel...maybe you did not see it.. but there is also a caster at all... the whole body is tilted forwardly 3 studs, as you can see in the following pictures:

 

All together this should make a realy stable frontwheel.

8 hours ago, PorkyMonster said:

 

Regarding suspension - I agree with @nicjasno that you need proper suspension (yes, there might not be significant effect on speed due to lost traction (possible reasons mentioned in my earlier post), but suspension will affect steering and that will prevent you from staying the course and reaching the highest possible speed). For proper suspension to work, the unsprung weight must be much less than the sprung weight (target less than 10%, even lower if possible).

 

Any idea, how i could achieve this with a 2-gear drivetrain? As mentioned earlier... i am happy to have a ~70% sprung weight now... and just ~30% unsprung (well i know this is still a lot...).

I would need to undock the motors from the rear-axle... but this would mean a way more complicated drivetrain, meaning heavier construction AND especially producing alot more friction, because of more axle/pinhole bearing points and more gears required...

8 hours ago, PorkyMonster said:

If the ratio is not significant enough, the wheels (being part of the unsprung weight) will send vibration up onto the rest of the vehicle and cause the entire racer to rock and you'll lose traction as a result.

As for hardness/travel of suspension, that really depends on your speed and track surface - something to be experimented (in general, harder+shorter for flat surface, softer+longer for rough surface...).

2 hours ago, nicjasno said:

What @PorkyMonster said. I think i speak for us all here, we want to see this thing go as fast as possible. But for this, you need to completly redesign the chasis.

The suspension type that would best fit this is that of a baja truck. With the narrow motorcycle wheels of course. But you need a suspesnion travel of at least 2-3 studs, and the suspension must be soft enough that it settles about 1 stud when you put the weight of the car onto it. This ensures that the wheels remain in contact with the ground even if they encounter a hole/dip in the road surface.

 

I would create such a kind of suspension for a fun model. But 2-3 studs of travel means 3-4 studs of ground clearance... this will lift the center of gravity way to high... resulting in a unstable driving... and especially resulting in tilting to the sides with just 1 front wheel.

This could be realised with 4 wheels... but then again there will be 33% more rolling resistance, wich i assume as one of the biggest torque eater at all.

14 hours ago, nicjasno said:

I could help you design a steering that would actually work at high speeds if you want. I am really interested how fast this thing can really go.

Also, it's not hard to see why the car would oversteer. The wheels loose traction, because there is no suspension that would ensure they keep in contact with the road. The flex in your chasis is not suspension. It's just a flexing chasis.

I would like to hear/see any suggestions here. If you like... feel free to pm me. 

6 hours ago, mocbuild101 said:

Your steering lever is too long (2 studs point-to-point), my steering - and I think @Marxpek's too - used a 2L beam, which resulted in only 1 stud distance between the turning points, meant that the wheel didn't move as much.

 

I know... it can be changed easily... i just wanted to start testing with a bigger turning radius... since there is a 3 step servo at the rc-unit, i should be able, to do just small turns, also with a 3 stud long beam, if i move the steering-stick just a litlle bit to the first position.

 

 

14 hours ago, TechnicSummse said:

But i think i allready found the problem... i think the supporting wheels are mounted to close to the front of the car... and if they hit the ground, this results in a verry small turning radius like this.

Not sure if you can see it properly in the video below... but it demonstrates exactly what i assumed yesterday:

When steering the car tilts a bit to the side... this causes one of the rear wheels to lift from ground, and the supporting wheel to contact with the ground. 

Furthermore the supporting-wheel acts as lever, and also lifts a bit the second rear-wheel, to also loose grip.

At first you see a steering to the right, with dismounted supporting wheel... followed from a steering to the left with attached supporting wheel on this side... i think you can see the difference...

 

[nicjasno]

Your caster is tilted in the wrong direction. That's the first problem.

I do have a very good and concrete idea how to do the suspension :)

The motors and gears will be part of the rear live axle. The front will be an independent suspension with proper caster and kingpin angles, to ensure the wheels self center. The front also needs to be made with bionicle ball joints, to eliminate slack.

I can design it all if you like. Don't have any of the r/c motors or the battery pack, but i can design the basics and let you adapt it to what you have.

[TechnicSummse]

46 minutes ago, nicjasno said:

Your caster is tilted in the wrong direction. That's the first problem.

Hmm... i dont see, why you all think my caster is wrong... its just crazy :D

I made a video again...i removed the steering lever, to let the front wheel do, what it would do while running.

You can see what happens in wich direction... i think i made everything right here 

 

Quote

I do have a very good and concrete idea how to do the suspension :)

With the new steering setup now, i also got some ideas, how to suspend the frontwheel.

But i still would see different ideas :)

Quote

The motors and gears will be part of the rear live axle.

But then again the motors are part of the unsprung mass...

Quote

The front will be an independent suspension with proper caster and kingpin angles, to ensure the wheels self center. The front also needs to be made with bionicle ball joints, to eliminate slack.

The (relativly small) slack will be romoved with a hockey spring... when my order arrives :( This should be here allready 2 days. To the self centering... take a look at the video :D

Wheels? Sounds like you want to mount 2 wheels at the front also? As i sayd earlier... this will result in 33% more rolling resistance.

Quote

I can design it all if you like. Don't have any of the r/c motors or the battery pack, but i can design the basics and let you adapt it to what you have.

I would like to see a sketch of your idea

Edited September 3, 2017 by TechnicSummse

[nicjasno]

Just now, TechnicSummse said:

Wheels? Sounds like you want to mount 2 wheels at the front also? As i sayd earlier... this will result in 33% more rolling resistance.

 

Not in combination with some camber ;)

[TechnicSummse]

Just now, nicjasno said:

Not in combination with some camber ;)

sure... the wheels have a round profile... camber does not change anything at rolling resistance here ;)

Did you see the video in my latest post related to my caster?

[nicjasno]

Yes, i did. But the problem is, you're doing testing on a smooth surface at low speeds, giving you false results.

[PorkyMonster]

2 hours ago, TechnicSummse said:

The design is not meant to be like the one on a shoppingcart... but it has to have the turning point infront of the wheel...maybe you did not see it.. but there is also a caster at all... the whole body is tilted forwardly 3 studs, as you can see in the following pictures:

2 hours ago, nicjasno said:

Your caster is tilted in the wrong direction.

1 hour ago, TechnicSummse said:

why you all think my caster is wrong... its just crazy :D

LOL you're absolutely right that its crazy!!! why do you think your caster is right when there are tonnes of literature all over the web on that topic? 

For a car, or a bike, with proper caster, there is tendency for the front wheel(s) to center even while stationary... this property does not exist in your "caster".

What you see as centering behavior is caused by the rear wheels being fixed in forward direction - so when your front wheel hits a bump, you're relying on your rear wheels to prevent your racer from deviating from straight line, and this may place more stress on the rear axle and slow down the racer (and of course, caster should not be tested on flat/smooth surface).

In contrast, proper caster will self-align, relieving the rear axles to focus on its main purpose - to propel the racer forward.

[nicjasno]

The current setup is exactly what a shopping cart has. And just look at the wobble that happens on those wheels everytime they hit something, even a slight surface imperfection.

[TechnicSummse]

2 hours ago, PorkyMonster said:

LOL you're absolutely right that its crazy!!! why do you think your caster is right when there are tonnes of literature all over the web on that topic? 

Well... phew.. its allways hard to tell people things wich they think they know allready... Again... i am a car mechanic master craftsman (think this should be the best possible translation to english for my job). I have to do with such things pretty often ;)

Quote

For a car, or a bike, with proper caster, there is tendency for the front wheel(s) to center even while stationary... this property does not exist in your "caster".

This happens in cars only... did you see any parking bike with a straight frontwheel? 

I guess you will say now... all the bikes are inclined to the left side... and yes you are right... but this shows another point, i will explain.

Like the caster on bikes is done, the wheel lifts the bike in straight position... meaning you have to push the bikes weight into the straight position. If you pushed a motorcycle you will know what i mean... the more you steer while pushing, the harder it is to go to the straight position again.

I think what you mean on cars is because you lift the car while steering to the edges of the wheels, and it has its deepest point in straight position. The wheels want to go to the ground again... as you can see in the following picture (perfect paint skills  )

Quote

What you see as centering behavior is caused by the rear wheels being fixed in forward direction

If so... why do i have a fluttering wheel in reverse direction, as you can see in the video?

Quote

- so when your front wheel hits a bump, you're relying on your rear wheels to prevent your racer from deviating from straight line, and this may place more stress on the rear axle and slow down the racer (and of course, caster should not be tested on flat/smooth surface).

In contrast, proper caster will self-align, relieving the rear axles to focus on its main purpose - to propel the racer forward.

Caster has nothing to do with the surface you are running on. Caster is by definition the difference between turningpoint and the point the wheel hits the ground. This does not change with changed surfaces. At the same time, if the wheel is lifted because of hitting a bump... caster does not matter at all... because the wheel has no contact to the road.

In the image below, you can see my angles at the front:

As you can see in the image... i have a huge caster with the turningpoint infront of the wheel. At the same time the caster is reduced because of the body-tilting... (maybe this is what @nicjasno meant).

But like this.... i can realise the exact opposite of the "bike-problem". Like this, i lift the car, when turning the frontwheel. The weight additionally to the caster pushs the wheel to the straight position again.

Edited September 3, 2017 by TechnicSummse

[nicjasno]

The bikes, when at speed, are stable in part, because of that very same caster angle. In cars, you also have kingpin inclination, which in combination with the caster provides the proper self centering.
 

 

Edited September 3, 2017 by nicjasno
Adding video

[TechnicSummse]

3 minutes ago, nicjasno said:

In cars, you also have kingpin inclination, which in combination with the caster provides the proper self centering.

Thats right... i did not want to talk about kingpin inclination, since it has nothing to do with just 1 wheel ;)

Why did you draw my turning-axle completeley wrong? The turning-axle at my car is exactly as i did draw it... its angled backwards at the same amount, as my body is tilted to the front. And like this it hits the ground exactyl at the point i did draw in the image ;)

 

By the way.. the turning axle mostly does not hit the center of the wheel like you did draw it... it mostly looks like this:

 

 

By definition, caster is the difference between the point where the wheel hits the ground, and the point where the turning axle (imaginary) would hit the ground.

For a positive (wanted) caster, the point where the turning-axle hits the ground, has to be infront of the point where the wheel hits the ground.

And this IS given at my car

 

[TechnicSummse]

2 hours ago, nicjasno said:

The current setup is exactly what a shopping cart has. And just look at the wobble that happens on those wheels everytime they hit something, even a slight surface imperfection.

I did not see your post earlier, thats why i did not answer to it.

... also i just found your video in your last post.

Did you look at the right wheel in you video? It runs perfectly smooth ;)

The left wheel is fluttering, because it does not really hit the ground. There are a few reasons why shopping carts start fluttering even if they have a positive caster.

The wheels are pretty small, and also the positive caster is pretty small, it only helps to align the wheels in your whished direction. Another point is the 100% vertical turning axle, wich is right above the wheel...

Imagine something like this:

Do you think the wheel will flutter? The wheelfluttering is allways a question of the lever act. This counts for all points of the wheel-geometry including kingpin.

In my image above the force to bring the wheel out of its current position must be huge, because there is a big lever pulling it in the front direction.

Edited September 3, 2017 by TechnicSummse

[nicjasno]

Pictures:
https://www.dropbox.com/sh/31ai04x1sck22uy/AAC0YLEAdiwdN5taF2khQ0G6a?dl=0

[PorkyMonster]

9 hours ago, TechnicSummse said:

This happens in cars only... did you see any parking bike with a straight frontwheel? 

 My bad  I forgot that tyre width and a squarish cross-section matters a lot when it comes to wheel self-straightening while stationary. Without width and a squarish cross-section, KPI will be necessary (as @nicjasno already mentioned) and therefore 2 wheels are needed. Once in motion, the trail (i.e. the distance between the point where the steering axis touches the ground and the tyre's contact point with the ground - hope we have the same definition) contributes to the straight-line behavior.

9 hours ago, TechnicSummse said:

Like the caster on bikes is done, the wheel lifts the bike in straight position... meaning you have to push the bikes weight into the straight position. If you pushed a motorcycle you will know what i mean... the more you steer while pushing, the harder it is to go to the straight position again.

I think what you mean on cars is because you lift the car while steering to the edges of the wheels, and it has its deepest point in straight position. The wheels want to go to the ground again... as you can see in the following picture (perfect paint skills  )

Great, so we're at the same point  so far.

9 hours ago, TechnicSummse said:

If so... why do i have a fluttering wheel in reverse direction, as you can see in the video?

That's due to your trail - there is reason why the steering axis hits the ground in front of the tyre's contact point with the ground. However, this fluttering effect is much reduced for cars with wider and more squarish cross-section tyres.

9 hours ago, TechnicSummse said:

In the image below, you can see my angles at the front:

9 hours ago, TechnicSummse said:

As you can see in the image... i have a huge caster with the turningpoint infront of the wheel. At the same time the caster is reduced because of the body-tilting... (maybe this is what @nicjasno meant).

But like this.... i can realise the exact opposite of the "bike-problem". Like this, i lift the car, when turning the frontwheel. The weight additionally to the caster pushs the wheel to the straight position again.

I believe you have huge trail, but almost negligible caster angle (although contributed by the body tilt, but front weight tends to go down as speed goes up). Coupled with thin tyre, you have near zero wheel self-centering desire. What your huge trail allows is for the front wheel to follow wherever the movement takes the vehicle.

Much like a trolley with 4 caster wheels - you can push it in any direction and the wheels will all follow happily without any desire to revert back to any particular direction - that's what trails contribute. Whereas with caster angle and wider and squarish cross-section tyre, the wheel itself will resist any attempt to change its direction, thereby contribute to greater stability. So imagine a 3-wheel vehicle like yours - when any wheel hits a bump, you really need the remaining two to stay on-course... but implementing the front this way, you've relieved your front wheel from this duty.

8 hours ago, TechnicSummse said:

In my image above the force to bring the wheel out of its current position must be huge, because there is a big lever pulling it in the front direction.

You do have a point here, but it depends on where the force originates from - if its a bump sustained by the front wheel, yes, you're right that it won't dis-orient the wheel easily due to the big lever effect... but if its a bump to one of the rear wheels, the entire vehicle will dis-orient pretty easily (again, due to lever effect from the other direction  which renders the self-centering ability of the front wheel useless).

 

[TechnicSummse]

@PorkyMonster

Hmm... i have to check something here again. In german there is just one word for caster and trail.

The german translation for cater is: "Nachlauf" and to trail means "nachlaufen".

So basicly the word to "trail" is the verb of the noun "caster"

Quote

You do have a point here, but it depends on where the force originates from - if its a bump sustained by the front wheel, yes, you're right that it won't dis-orient the wheel easily due to the big lever effect... but if its a bump to one of the rear wheels, the entire vehicle will dis-orient pretty easily (again, due to lever effect from the other direction  which renders the self-centering ability of the front wheel useless).

Well you are right... it depends on the direction of the force... as allways :D

The bigger the caster is, the more stable the car/bike will be in driving direction. But at the same time... the bigger the caster is, the more unstable it will become vs side-forces.

[PorkyMonster]

25 minutes ago, TechnicSummse said:

Hmm... i have to check something here again. In german there is just one word for caster and trail.

The german translation for cater is: "Nachlauf" and to trail means "nachlaufen".

So basicly the word to "trail" is the verb of the noun "caster"

That's rather surprising ... my understanding is that both caster and trails mean very different and independent things. Caster defines the angle between the (1) steering axis and the (2) vertical line that cuts through the rotating axle of the wheel, and trail defines the distance between the point where the lines cut the ground.

Huge caster does not necessarily imply huge trail, and vice versa, (all depending on where the two lines intersect, which need not be the center of the wheel) and varying each of them separately can lead to different straight-line performance.

32 minutes ago, TechnicSummse said:

The bigger the caster is, the more stable the car/bike will be in driving direction. But at the same time... the bigger the caster is, the more unstable it will become vs side-forces.

Note that this is only true for single, thin, wheels... when you have a wider, more squarish cross-sectioned wheel, or even better - two wheels with KPI, you can avoid such problem. So... time to consider having two front wheels?