DrJB

Rather easy, look at the cab interior /operator space. It should give you a good idea of the relative scales.

I saw the topic/thread few days back and was debating ... where do you find the time ... to build MOCs almost 'full-time' ?

Pneumatics are 'closer' to real life, granted they are not precise, but very simple to install/connect/operate. With LA's, each one needs its own set of driving axles/gears ...

I beg to differ (one more time). Sometimes the geometry is used to 'alter' the effective stiffness of the suspension. In the past, it used to be that the balance between handling and comfort was achieved only by one factor: The aspect ratio of the tire. Nowadays, and with advances in suspension geometry/components, SUV vehicles use Run-Flat tires with very stiff sidewalls (poor comfort). The comfort is then provided by the suspension. There are so many performance measure/attributes in the tire suspension system that it is often difficult (from my own experience) to say simply 'this affects this performance and not that' ...

This discussion has suddenly become very interesting ... I did spend few years of my life working on exactly that, and this is rather refreshing. Back then, such 'knowledge' was not truly public as wheel/tire settings was kept 'secret' by the various auto-makers. I worked on balancing various tire performances including handling, noise, and irregular wear. While most European vehicles are 'tuned' for handling, many US cars are often tuned for comfort and max wear life of the tires ... Keep the discussion going, very interesting.

That's exactly it, and I said that in post #32. The kinematics is really easy here, and the intersection point MOVES when you steer ... @ nicjasno You cannot really build this and look at it. The clearances in the small ball joints will make it a bit difficult. It is much easier to understand it on paper ... If you've taken a kinematics course, it should be very easy.

How do you decide the scale of the 8043? It's got to be based only on the operator/cab size. Such excavators do come in various sizes and the only 'reference' is the operator cabin ...

Good to know ... would love to get your assessment on this.

For argument sake, picture this: a very simple suspension with a single A-arm on the top, and two links at the bottom. The top pivot does not change and we can both agree on that. Now if you extend the lower 2 links and find their intersection point, it will define a second point for that virtual king-pin. Now, if you steer the tires totally to the left, then totally to the right, that second point (virtual) will absolutely/positively move. I can sketch it quickly and post it, but it's a rather easy exercise. It is obvious to me that as the tires steers left/right, the virtual point (intersection of the two links) will move in the fore-aft direction. What are we missing here? I am fully aware of the many EU manufacturers using such geometry to 'offset' the virtual king-pin. I even wrote a paper published in SAE about this very same topic ...

well ... I need to pull out my vehicle dynamics books/notes and refresh my memory a bit ... will get back to you. :)

In a conventional suspension, with upper and lower A-arms, the king-pin is fixed. In a multi-link suspension however, the king pin does move when you steer the tires.

That is true, I only gave one explanation. The tire has a self-aligning moment (SAM), and that, together wit the scrub radius, determines how much overall return-to-center you have. If you really want to add other mechanisms that control the return to center, you must include the tire's inherent PRAT (Plysteer Residual Aligning Torque). This is very important in the US, where most roads are straight and vehicles sometimes tend to drift, because the drivers only hold the steering wheel lightly. In Europe however, because roads are typically very sinuous, the drivers hold on tightly onto the steering wheel, and typically cannot sense the pull/drift caused by PRAT ... Well, two more things: The Crown of the road (how non-flat it is), and the front-end alignment BOTH affect Return-To-Center ... True, Kingpin is for old cars. But the terminology stayed and is still in use today, at least with NA auto-makers. True, very few modern vehicles have an actual kingpin. What is referred to in this case, is the line joining the articulation with the upper and lower control arms. Very often also, such kingpin 'axis' is not fixed and tends to move, depending on load on axle ... etc, especially in multi-link suspensions.

Well, here it goes. For improved handling (how fast a vehicle responds to steering inputs, without overshooting), you want as small as possible a scrub radius. The scrub radius is the distance, measured on the ground, between the intersection of the king pin line and the center of the tire's footprint. That distance needs to be as small as possible. Since typically you cannot put the pivot (king pin) line inside the wheel, you want a king-pin that is inclined. Some upper-scale European vehicles achieve such small scrub radius by using multi-link suspensions. Also, while 'vehicle handling' is not something you can really 'experience' in a lego vehicle, all this discussion it to essentially mimic realistic suspensions/kinematics. My explanation above is very simplistic and there are more in-depth interactions between the various tire angles and vehicle handling ... I may have to wear my 'vehicle dynamics' hat, but it'll become very mathematical in no time :)

Nice, how do you attach a wheel+tire?

Thank you. From my recollection (and graduate courses in control theory) a classical PID loop is not possible to stabilize an inverted pendulum as some of the 'states' are not visible. Hence, the idea to use full-state feedback as the inverted pendulum is a true non-minimum phase system. I was hoping for a simple mathematical model, then an 'easy' translation into lego code ... but that is not what I saw ... I may need to spend more time and 'decipher' the lego solution. My thinking is, I can easily draw a block diagram to control the gyrobot, but translating that block diagram into lego code is not 'trivial'.

True, though gluing is not an absolute must. The main 'weakness' is securing the half pin to the spindle. I'm thinking a friction half-pin with some paper ... might be able to do away with the glue. I should try it on an actual car. Of course it'll fall apart as is ... What you should really do first is attach it to a vehicle and see what happens ...

Thank You.

Here is an LXF to illustrate the non-zero king-pin. The part in red should be replaced with the LONG A-arm. King-Pin.lxf

It seems you answered quickly before reading the entire post. Think about it again, a non-zero king-pin angle IS possible, but because of the limitation of tire spindles (x873c01 or 32186) this will also force a negative camber and that is why I hinted you'd need a balloon tire. But, all is not lost as THERE IS a solution (and that is why I started this thread). You can use the new spindle 11949 and offset the top ball joint so it is 'closer' to the vehicle on the top. The problem here is that friction alone might not be sufficient, so you'll need to glue together few parts. You'll then need to add the short A-arm on the top, and the long one on the bottom, and Voila! You have a tire/wheel WITH non-zero king-pin AND zero camber. I only have few 11949 and not ready to glue 2 of them 'permanently' ... You're correct if you're limited to flat crown tires (eg Unimog or 8421). Balloon tires should be no problem though.

Of course the challenge with cables is that they're very good at pulling, but cannot push. Thus, some 'smarts' are needed for bi-directional motion. Though, the nice thing is that cables can be pre-loaded, resulting in almost zero backlash ... Awaiting to see the final product of course, promises to be full of neat ideas.

With either the current 57515 or old 32195b control arms, the King-Pin angle is forced to be zero, as seen on many vehicles (8880, 8488, 8644, ...). But how about mixing the control arms? Granted, the offset is 1M and thus might be too large, but nonetheless still feasible. Of course this would have all sorts of implications on the various tire angles and the balloon tires might be the best fitment. Has anyone tried this? If so, what's your experience?

Good Progress. What does the chain do? I can't help but remember a course in aerodynamics I took years ago ... that 2-blade propellers are inherently 'unstable' ...

Not sure we'll ever see that ... but I agree that the current 'fake' engine has more than lived its life ... How about configurations such as W or 3-Cyl as seen on some German cars?

Looks very nice. Can't wait to see it as part of a larger setup.

While PF is more modern, has more control options ... Etc, it will cost you in terms of replacing/recharging batteries. Keep in mind that you are running motors, and those deplete batteries in no time. Yes, for a fixed amount of money, you get a lot more with PF in terms of parts than u do 9V, but in terms of operating costs, 9v is a sure winner. If you're doing only loops, ie no train switches, then there is a way to take the new track (cheap) and convert into 9v with some copper tape. What I would do is buy some of each 9V/PF, play with both, then decide. What I really would like to see is the PF to be compatible with EV3, the same way 9V is compatible with the old Mindstorms RCX. That, 9V/RCX compatibility enables you to do way more today than you can with PF.