34th America's Cup: multihulls!

[philSweet]

"

The mechanical feedback was purely based on the position of the board trunk, and there was no measurement of the motion of the boat. So no stability augmentation system. It was a rules-compliant hydro-mechanical servoactuator that compensated for varying hydraulic supply pressure to give a consistent response to the controls. All it did was control the position of the valve - the position of the valve in space.

Yup, all day, everyday, with or without manual inputs. That is what I think the committee missed. It must have been a nightmare to get it to work properly given the space constraints.

I have this image of some guy banging on the board box with a 4 pound rubber mallet all night long trying to loosen the bearings up a bit, or hitting some titanium pin with a die grinder after it was machined to micron tolerances. Prybars and portapowers littering the floor. Different neoprene bushings and bungs being tried every day

Anyway, the result was spectacular. It is just a wand operating on the effect of the water, rather than the water itself. It just goes to show how powerful a computer you can build with only one servo. Like I said before - old school.

It's ironic that a gadget that would barely rate at a high-school science fair appears to me to have been the deciding factor. Of course, the expertise and modelling needed to make it practical is all still in a computer somewhere.

I don't like it, which should not surprise anyone. But it is impressive. Apparently ME's still have some relevance in today's world.

tech question - Oracle's lower board box pivot is -

A. A gimbal axis mounted to part of the cage?

B. A spherical bearing mounted to the hull?

C. What about on ETNZ?

 

[tspeer]

...This is a stability augmentation device that happens to also be the prime mover for foil rake. ...
This is a data acquisition system. And it is a control system. It senses ride height excursions in the form of board case deflections and initiates the appropriate corrective action. ...

It was not a stability augmentation device. Ride height did not feed back to board rake. First of all, there's a lot of friction in the board trunk bearings, and this tended to fix the position of the board. Second, the vertical lift, and the drag associated with it, stayed at the the same distance from the hull regardless of ride height. Even if the board did move with ride height (and it didn't), the effect of the hydromechanical servo system was to keep the board at the desired position regardless of ride height. In aircraft, this is known as stick fixed stability, vs stick free stability in which the elevator is allowed to float.

In order for it to have operated as a stability augmentation system, there would have had to have been a change in wing incidence (board rake) based on a measurement of flying height. By definition, stability augmentation requires a feedback sensor in a closed loop system to provide artificial stability that isn't obtained through the forces and moments alone acting on the surfaces. There wasn't any measurement and there wasn't any change in board incidence with flying height. So there wasn't any stability augmentation.

Instead, there's a natural coupling of sideslip to vertical lift due to the dihedral angle of the wing on the L foil. As the ride height is increased, vertical board area is decreased, leeway increases and this reduces the angle of attack of the wing and reduces the lift. This coupling meets the Design Rule requirement for direct action of the hydrodynamic forces.

Board rake acted more like a trim device, determining the equilibrium lift about which the leeway coupled stability acted. The actuation of board rake was a hydromechanical servo system, responding to a manual position command, and powered by human pumped hydraulics.

From the AC72 Class Rule,
"19.2 The use of stored energy and non-manual power is prohibited, except:
...
(e) for electrical operation of
(i) hydraulic valves. These operations shall only provide the input for the position of the valve;..."

The system on 17 did exactly what the rule said. It used a small electromechanical actuator to provide input for the position of the valve in space.

 

[philSweet]

It was not a stability augmentation device.

I was under the impression that Oracle had referred to it as such. I wasn't trying to be controversial at that point. I'll gladly call it by what ever name it goes by.

By definition, stability augmentation requires a feedback sensor in a closed loop system to provide artificial stability that isn't obtained through the forces and moments alone acting on the surfaces.

That sort of begs the question. What do you call a system that does the same thing, but doesn't have a closed loop (besides legal under rules interpretation).

My thinking is that the system uses an open loop to increase the leeway-stability sensitivity by coordinating small rake changes. For instance, if the claimed 0.5 degree increment is repeatable to 5% accuracy, the switch is sensitive to 0.025 degrees of rotation (granted, there will some additional hysteresis to trigger the thing). On a 1 meter throw, that equates to around 0.5mm linear motion. Are we wrong to wonder if there isn't the chance that the board box might respond to a change in height by flexing a bit, and that there might be a 0.5mm registration error caused by that flex? If someone wanted it to behave that way, couldn't the lower bearing carrier have just a little squish so that the ram head became the pivot point when there wasn't an active control input?

At any rate, I had missed a large chunk of the discussion up till now, and could have saved everyone the trouble of wading though it all over again. I'm firmly in the camp of Earl and Remmlinger and a couple others. If Oracle had wanted to avoid a controversy, they could have strapped the gadget to the ram. It is specifically the potential difference between the ram extension and the servo registration that is the provocation. As long as the board box moved exactly about the same center with no squishy bits that caused variances in the instantaneous center of rotation, it is nothing more than what it appears. If the board box's instantaneous center of rotation can be influenced by outside loads, then there is the possibility that the switch on the servo is measuring a variance that can be related to ride height. I haven't seen any post that questions how the system works when responding to a control input. It's what it is doing in the absence of a control signal that I find controversial.

Also, I find absolutely no fault in Oracle running the proposal up the flag pole. I would have loved to be a fly on the wall when Oracle got the news it passed muster.

 

[tspeer]

...That sort of begs the question. What do you call a system that does the same thing, but doesn't have a closed loop (besides legal under rules interpretation).

An open loop control system.

...Are we wrong to wonder if there isn't the chance that the board box might respond to a change in height by flexing a bit, and that there might be a 0.5mm registration error caused by that flex?...

Yes. You are wrong to think that. There are a great many things that affect board flex besides height. Side force on the sail rig being the principal one. There's simply no way to deduce the flying height from the loads at the board trunk.

There were things that the Measurement Committee did not allow. One was regenerative hydraulics that would have bled pressure from the ram that lowered the board to drive other functions. The MC had members that were quite knowledgeable about hydraulics, so the notion that OTUSA slipped one past an ignorant MC is just not the case.

 

[philSweet]

It all just seems so Rube-Goldberg compared to slapping one of these on the boat and calling it a day.

http://www.google.com/imgres?imgurl...a=X&ei=uVqVUvuRFOmdyQGT64HABA&ved=0CEAQ9QEwAw

 

[oceancruiser]

tech question - Oracle's lower board box pivot is -

A. A gimbal axis mounted to part of the cage?

B. A spherical bearing mounted to the hull?

C. What about on ETNZ?

What about on ETNZ

Any takers.

OC

 

[tspeer]

It all just seems so Rube-Goldberg compared to slapping one of these on the boat and calling it a day.

http://www.google.com/imgres?imgurl...a=X&ei=uVqVUvuRFOmdyQGT64HABA&ved=0CEAQ9QEwAw

There are two problems with that valve. First, production spool valves are rarely rated for 10,000 psi supply pressure, and their leakage rates result in power losses that can't be tolerated with a manually cranked system. That's why poppet valves were used instead of spool valves.

The second is that it doesn't provide the mechanical feedback to compensate for changes in supply pressure.

There are many ways to achieve precise control with hydraulics. Achieving precise control with manually powered hydraulics, using no stored energy and within the weight constraints of the AC72, is a completely different problem.

 

[Doug Lord]

First Foiling America's Cup

Oracle CFD simulation video:
http://www.youtube.com/watch?v=AJgADA9_3GM

 

[petereng]

Hi All - Do we have any news on the size of the AC35 boats? I've seen 65ft mentioned and I've seen that the AC72 platform may be used. Anyone have leads yet on this? Regards Peter S

 

[Doug Lord]

Hi All - Do we have any news on the size of the AC35 boats? I've seen 65ft mentioned and I've seen that the AC72 platform may be used. Anyone have leads yet on this? Regards Peter S

===============
Peter, as soon as I hear anything you can rest assured it will be posted in the 35th America's Cup thread. I'd like to see them stick with the 72's so it's not starting from scratch again.
I imagine we'll hear something shortly after the first.
http://www.boatdesign.net/forums/multihulls/35th-americas-cup-foiling-multihulls-48428-5.html

 

[powerabout]

Tornado's on foils?
I think the 72 will be binned as the operational costs are too high even if they gave you one.

 

[Doug Lord]

First AC on Foils!

Interesting article-there is trickle down-and not just uptip foils:
http://www.sailingworld.com/racing/americas-cup-34-trickle-down-effect

 

[Baltic Bandit]

Pretty hollow fluff article "as cat sailors dream of".... not really.

 

[Blackburn]

Hi All - Do we have any news on the size of the AC35 boats? I've seen 65ft mentioned and I've seen that the AC72 platform may be used. Anyone have leads yet on this? Regards Peter S

Andy Green has gossiped lately on the rumor mill also known as Twitter:

"sounds like a 60ft foiling cat 8crew one-ish design wing 45 series 15&16"

(that last bit meaning the 45's will be sailing in '15 and '16)

"AC35 boats not launched before 1st Jan 2017 (my guess). RC said def a crew nat rule of some amount"

 

[Blackburn]

There are two problems with that valve. First, production spool valves are rarely rated for 10,000 psi supply pressure, and their leakage rates result in power losses that can't be tolerated with a manually cranked system. That's why poppet valves were used instead of spool valves.

The second is that it doesn't provide the mechanical feedback to compensate for changes in supply pressure.

There are many ways to achieve precise control with hydraulics. Achieving precise control with manually powered hydraulics, using no stored energy and within the weight constraints of the AC72, is a completely different problem.

Tom is all too modest, and makes no mention of the two very interesting articles you can find here, and here.