To be fair though apioqm, someone at uni has a stumpjumper evo that I got to ride round the car park before getting to demo the normal stumpy properly. I don't think I'd want the evo as the standard is low and slack enough; the evo takes it too far and loses some of all round attraction that the normal stumpy has!

Through axle on the rear and qr on the front? Sounds like specialized's spec department got a bit confused... I had the chance to ride a stumpy recently at a demo day, the rear suspension definitely felt nice and active and ready for some thrashing although the slightly longer stem they had on the bike was a little off putting. Shove a short stem on the bike and it'd be more than ready to let loose on some downhill tracks; saying that the bike is XC oriented definitely doesn't do justice for what is a very good and very fun bike all round.

http://www.pinkbike.com/video/253561/ I wouldn't be so sure man. Those trees are planning something...

The trick was simply so sick that it broke the walls of time and space.

Fair play. Hope it works out for you, good luck with it all!

Nice effort... looks like spud's been building again!

The only reason I was suggesting using the speed of the shock and forces to calculate the actual damping forces is that you could then get some real time heat correction going on which might allow for weight savings elsewhere if you came around to designing a bespoke shock unit, as you might not need to worry about heat dissipation as much. I think I see what you mean about the system never thinking the damping is anything - I can't work out if you might get errors accumulating if the actuator overshoots the position it is meant to be in ever so slightly though? I could imagine trying to get the brake dive stuff sorted would take an awful lot of time and programming - however I think it's one of the places you're likely to see the biggest difference in terms of improving rider feel, or maybe the most noticeable - just because all the other stuff has to be so subtle!

I can definitely see that a fully closed loop would be impractical, are you going to keep track of the real time damping though? I could imagine that in relation to the rest of the system it would not be too hard to implement - an accelerometer integrated once can give you shaft velocity and it would be possible to measure the force on the shock through a strain gauge. I would imagine it would be pretty important to know what the damping settings are so as to be able to make accurate changes as there are a lot of variables that may change the damping from what the computer will think it is... Also, would it not be possible to have a seperate set of computations for the brake dive? I would imagine that if you could get the preload on the fork to track the extra force required from the weight transfer quickly and accurately enough, the fork would still feel the same to the rider and so wouldn't feel unreliable like if the damping changed too quickly or too far. I could imagine the actuation of this being as much of a problem as anything though.

Going off what you're saying about keeping it deployable across multiple platforms - I'm fairly sure you'd have to make bespoke damping units for the forks and shocks, more likely entire bespoke units? So I'd imagine that next to that, adding your own brake lever unit and cranks would practically be just a drop in the ocean!

Ah, I see what you mean; I hadn't began to think about differences in frequencies! I think you'll have to be very careful though; I could imagine that in some circumstances brake dive may look similar to a nose heavy landing. Both will have high amplitude and a high initial shock but low frequency compared to general trail buzz, and both will have the rear of the bike unloaded. During brake dive you'd want the preload of the fork to be increased and compression damping reduced; the exact opposite of what you want during a nose heavy landing! Using the frequency of the pedaling motion to isolate it is almost definitely the right way to go about it, although again I think it would be difficult to identify and isolate pumping forces. Thinking about it, I'm not sure the strain gauge would make it any easier to identify when a rider is pumping however it might be worth a think about if it would help identification and if its something you'd want to program in anyway. I'm curious as to why you're using an open loop controller? We've started off control theory (I'm doing an integrated mech & elec course so it's pretty heavy on that), and whilst its all pretty new, my lecturers definitely seemed convinced that closed loop control was superior. I had a quick look at the sigmoid function stuff, its a shame that wikipedia doesn't explain it at all. My first guess that by finding where the gradient on the curve is a maximum you can assume that's where error is zero and program your system to go to that point - with the curve ending up looking like a damped oscillation? Might be way off though...