Help needed with virtual pivot steering please.

[ukbajadave]

I'm working on a MOC with virtual pivot steering and I'm struggling to grasp the theory behind it.

This is the build so far, the PF servo drives the 24t gears which move the steering arms. Small turntables are used instead of hubs because I have turntables...

I am a physical builder who mashes bricks together until they work so perhaps someone can explain to me (with small words and pictures) what happens when variables change. In the following pics assume the chassis is grey and the yellow beam is the hub.

Firstly what is the effect of arm length? Longer arms widen the track but how do they affect the angle and movement of the hub?

Secondly what happens when you change the spacing between arms?

As always any help gratefully received 

[MajklSpajkl]

Hi, virtual pivot point is where both longitudinal axis of the black liftarms cross. This point should be in the centre of your wheel, so the wheel doesn't circle forward  and backward around the pivot poimt. So when altering the lengths of the liftarms and distance between their attachment points you are moving virtual pivot point closer or away from the actual build.

https://images.app.goo.gl/P7XDN6CV3kEgnRvr9

On the left, pivot point is on the inner side of the wheel, and on the right, it is far to oitside of the wheel.

Edited March 31, 2019 by MajklSpajkl

[ukbajadave]

Thanks @MajklSpajkl ! I can't believe the explanation is so simple. Does this mean that the yellow liftarm follows an circular path around the virtual pivot as badly drawn here?

 

[MajklSpajkl]

I think that is correct.

[Jurss]

In this moc

this solution worked very well

https://drive.google.com/file/d/0B3hhR5y2V2lARl9MeUpwZ3dwMXc/view?usp=sharing

[Lucullus]

One thing that has to be kept in mind is, that while the virtual pivot is always located at the intersection of the steering arms it doesn't mean that it will always be at the same place during the steering motion. The wheels won't turn on a circular path like they would do with a nonvirtual pivot. In the above picture this is especially evident if the yellow liftarm is turned so far, that it lies in one line with one of the black ones. In this moment the pivot point is on the 2l axle at the opposite end of the yellow liftarm.

[ukbajadave]

1 hour ago, Lucullus said:

...while the virtual pivot is always located at the intersection of the steering arms it doesn't mean that it will always be at the same place during the steering motion...

...the yellow liftarm lies in one line with one of the black ones...

Just when I thought I'd sussed it!

Of course it is obvious when you think about it that the virtual pivot moves with the steering arms. Does the yellow liftarm still follow a circular path around the virtual pivot though? That would suggest shorter arms give a larger steering angle for a given rotation at the expense of greater fore/aft movement of the pivot point? And yes, when the liftarm lies inline with the steering arm the whole thing locks up 

I'd love to get a handle on this because as @Jurss shows this seems like a great way to steer smaller models without needing huge wheel clearances.

After reading this thread I've rebuilt the steering on my MOC and managed to make it slightly worse