Golden Rule

One final reflection about recent events in the A Class, before resuming normal programming.
Future posts will focus on hopefully interesting technical commentary, and updates on our projects...

Flourishing

The last couple of years gave us extraordinary advances in the capabilities of inshore racing sailboats.
Arguably the root cause is development in the America's Cup, where big foiling multihull 'daysailers' benefited from the focused study, and substantial R&D budgets, of professional teams in a high-stakes arena.

This work furthered our understanding of how to make such boats faster and safer.
It also helped popularise a kind of fast sailing that previously existed on the fringes of a sport not necessarily known by the layman as particularly dynamic.

We have seen multihulls 'go mainstream'. Preconceptions about poor maneuverability, and unsuitability for close racing were dispelled.
Once full foiling became possible, a new dimension of speed, technical challenge, and spectacle became reality.
Basic insights, tools, and technology were quickly applied to 'accessible' craft, including one-design classes, and even fast cruisers.

The top tier of the sport was rejuvenated, and the appeal of sailing broadened.
This did not detract from 'mature' disciplines (the monohull scene is still thriving. Only now there is more on offer).
Keelboats, foiling Moths, and foiling multihulls, all benefit from a broadening of the base. Lessons learned in one specialty advance the others.

Origin

Reflecting with some fellow sailors, an important realisation dawned on us:
ALL of the above flowed from the literal interpretation of a rule specifically designed to prevent foiling.
Think about it: The AC72 Rules had provisions intended to limit foil volume (and hence foil area) to a value insufficient for full foiling.
Clever designers in Emirates Team New Zealand spotted that the wording of the rule allowed a literal interpretation completely contrary to intent.
Yet once written, the rules had to be applied as worded.
Consistency dictated that the plain meaning must be upheld.
Though this gave an initial advantage to the discoverers of the 'loophole', it also opened the door for others to follow. Even to improve on the original idea.

Sports, architecture, and law, are littered with examples of 'rule cheats' that either opened the door to progress, or exposed the absurdity of outdated regulations.
Normally regulators react by permitting such solutions until rules can be properly changed, usually after a season or cycle has concluded, then everyone can start again afresh.

Some interesting examples for the curious (links to external articles):
- Due to sloppy rule drafting the class rules could be literally interpreted to read...
- Committee had exceeded its jurisdiction in issuing an interpretation that changed a Class Rule...
- The F-duct was so smart that the FIA used it in creating what's known as DRS today...
- Brawn's double diffuser in 2009...
- One of many unintended consequences of tax policy...
- Slicing part of the rocker overcomes the prohibition against hollows in the hull with an 'appendage' not as it was intended...
- The judicial is separate from the legislative precisely to objectively interpret law...
- 'Contrary to the spirit of the law', exposed as a ludicrous question...

Lesson

Rule administrators have a duty to be impartial when issuing interpretations.
They must keep intent at arm length, look at the wording with fresh eyes, and read the rule as written. They must ask themselves: "What would this mean to someone reading it with no preconceptions, coming from a different background, unburdened by the baggage of what was in the inscrutable mind of the writer?"

Saying that we should "not [be] spending our time looking for loopholes" is disingenuous.
It is doing a disservice to the Class and the sport.

Those charged with issuing interpretations have a duty to objectively apply the rules.
Attempting to enforce an arbitrary view of preferred outcome is ineffective and unfair.
Ineffective because it invariably fails to achieve the intended outcome, rather fostering elaborate workarounds that divert energy from genuine development. Unfair because it makes it impossible for participants to reliably know what will be deemed acceptable by the whim of the arbiter.

Once loopholes have been found, acknowledged, and exploited, then the discussion can begin to either tighten or get rid of defeated rules.
This is finally happening in the A Class, thanks to the persistence of those who take to heart the history and ethos of a unique and exhilarating development class.

"The golden rule is that the words of a statute must prima facie be given their ordinary meaning."
Viscount Simon, in Nokes v. Doncaster Amalgamated Collieries, [1940] A.C. 1014, at p. 1022.

Growing

We now have an opening for a composite fabricator/shipwright.

Our workload is growing so we would like to take on another team member.
Projects include UAV airframes, customer parts, RC yachts, and our A Class catamaran.
A good mix of one-off prototypes for R&D, and production items.

If you possess skills in carbon fiber pre-peg construction, with an aerospace or marine background, and would like to join a growing team working on interesting products, contact us by email: info@carbonicboats.com

Paradox 2015

First Public Introduction of our All New A Class Catamaran

Concept

-          Speed, stability, easy to tune for different conditions, value, elegant engineering.

-          Benefiting from three years of structured testing, data collection and validation.

-          Developed in close collaboration with Glenn Ashby.

-          Built in Australia to aerospace standards.

Hull Shape

-          High volume combined with narrow waterline beam through U shaped sections.

-          Flat bottoms for maximum planing lift and minimum dynamic wetted area.

-          Rocker shaped for responsive trimming: Easy transition from bow-down (lowriding) to bow-up (step back/takeoff).

-          Bows have generous volume underneath and peaked low-freeboard tops for wave piercing and water shedding.

Platform

-          Low windage and high stiffness.

-          High modulus beams, Nomex cored hulls.

-          Integrated construction, sealed low-stretch trampoline, streamlined rear beam.

-          Future-proof foil case design, able to take any shape foil. Go from Z to L with no mods.

Foils

-          Optimised Z foils with variable section (camber changes along the span).

-          Full use of permitted lifting span.

-          Precise rake adjustment through worm-gear, as used by proven foiling classes.

-          Good foil support (no slop, no jamming) with precise toe-in through rotating bearings.
-          Optional L/V foils, plug and play. 

Steering

-          All new: A leap forward from the existing dagger/cassette concept that we pioneered and has since been widely adopted.

-          The new system allows much better refinement of rudder planform as well as easy rake adjustment on the water and greater safety.

-          Superior grip at low speed, low drag and precise control when foiling.

Availability

-          Customer deliveries expected to start in August 2015.

-          Ongoing deliveries after September including containers to the Americas and Europe.

-          Contact us now to lock in a hull number with a conditional holding deposit.

Contact

Email: info@carbonicboats.com

Phone: +61 412 127 388

Farewell 2014

Some images looking back on a year of regrouping, transition and growth:

Production L rudders, proven at the A Class Catamaran Worlds 

Mould for experimental T rudder elevator with junction bulb.

 Part of extensive R&D work on appendages

Rudder gudgeon assembly with 'between races' rake adjustment

Experimental gudgeons with 'on the fly' rake adjustment
via tiller extension twist-grip

Billet rudder cassette. Our concept of 'dagger' rudders with offset axis has been widely adopted

First A Class 'V' foil concept. 'Inspiration' for current Z foils

Retrofit foil case kit with rotating bearings

Moth bow swivel fitting developed with Scott Babbage. Production version available here: http://www.sailingbits.com/class-specific/moth-bow-mechanism/

Moth bellcrank developed with Scott Babbage. 
Production version available here: http://www.sailingbits.com/class-specific/moth/moth-adjustable-bellcrank/

Bolts with streamlined heads. Used on UAVs and various sailboat classes

Experimental Finn mast chocks for NB Sailsports

18' Skiff rig spanners for Allmarine. Available here:
http://www.allmarine.com.au/shop/boat-specific-products/18-foot-skiffs/all-marine-rig-spanner/

Tasar fittings for NB Sailsports.
Available here: http://www.nbsailsports.com.au/store/product-info.php?pid1365.html

Fairleads

Trophies for A Class Catamaran Nationals

Choices

We have received many questions regarding the differences between ‘active’ and ‘passive’ foil systems for full foiling.
So here is a look at the principles with respect to performance.

Active

An active system consists of a foil with variable camber or variable angle of incidence controlled by a sensor that measures heave position (ride height).
The input can be via a mechanical device such as a wand/float or an electronic sensor.

Usually the main lifting foil is fully submerged. In order to minimise the total lift necessary, the submerged foil should be angled to provide both vertical and horizontal force components.
The vertical component holds the boat up and the horizontal component resists leeway.
Moths achieve this vectoring by heeling to windward.
By vectoring the lift from the submerged/active span, the vertical struts are not significantly loaded so surface-piercing effects are minimised.

It is interesting to note that where active T foils have been tried on catamarans the results have been less than promising because vectoring was difficult to achieve. Sideforce was provided by the surface piercing vertical struts. These got smaller with increasing ride height. Also their pressure field interfered with the main lifting foil degrading efficiency.

With twin Ts it may be possible to set the hulls up for differential ride height (set the neutral point on the respective sensors differently for windward and leeward foil) thereby encouraging the platform to stabilise at a heeled ride height. However the downside is that the windward foil will have a long span of submerged strut (since the foils are far away from the centreline, the difference in immersion from upright to heeled is large).

One concept we tested, designed by Dave Lister, showed promise by combining active heave control and lift vectoring via angled fully submerged lifting spans for minimum wetted area.

On an active control setup, lifting foil area does not change with heave. The submerged portions of the vertical struts get shorter but this has little effect on total lift. Instead lift is controlled by changing the lift coefficient of the main foil, either through altering angle of attack or, most effectively, through adding camber by deflecting a flap.

A flap alters camber and changes the angle between chord line (light blue)
and oncoming flow (dark blue)

Passive

This solution comes in different forms. Variations on V configurations rely on a decrease in immersed foil area with heave.
Other solutions such as the acute L/V rely on a coupling between heave and leeway such that increasing leeway reduces the effective angle of attack of the main lifting surface.
Where leeway values are very small, an L/V foil can also use a reduction in lifting area (inboard tip breaching the surface) as a last-resort means of limiting ride height.

Tradeoffs

Mechanically it can be argued that the overall complexity is similar: Active systems have swivels, pushrods and bellcranks that require significant refinement and must be looked after correctly. Passive foils require hull and deck bearings and means of adjusting depth and rake.
So ultimately the cost differences are minimal.
Active foils need some form of articulation built in (a shaft or flap) so they are more complex to produce. But they tend to be made from straight segments whereas passive foils tend to have curved spans so their tooling is more expensive.
Again, on balance cost is not a deciding factor.

Active foils with mechanical sensors tend to be at a disadvantage in light winds and marginal foiling conditions because there is a drag penalty associated with the control system.
In non-foiling conditions the sensors can be disconnected and retracted. But then no lift is available so any puffs would see the passive boat move ahead in foil-assisted mode.
Arguably the active setup is also heavier depending on where the sensors are located and how they connect to the foils.

So on a small cat the passive foil would have the competitive edge in very light winds.
The exact crossover remains a subject of investigation and will be found to depend on variables such as displacement/length ratio, sail area/wetted area ratio and the exact design of the foils...

Once foiling the active system requires less deliberate correction by the skipper.
This favours the less advanced sailor but probably makes little difference to the nuanced expert who is constantly making adjustments by muscle memory.

The crucial difference is this: An active foil can be smaller for a given takeoff speed because lift coefficient can be maximized when needed and dialed out when not required.
You can have an aggressively cambered foil on takeoff and a flat low-drag one at high speeds.

This is not impossible with passive foils. For example, the section used in the upper portion can have more camber than the one used near the tips.
But the compromise is more critical.
It is more difficult to have early takeoff and low drag at high speeds.

If the rules are tested and the A Class decides that active controls are not desirable, then passive systems will evolve rapidly and the problems will be solved.
Hopefully the decision will be an informed one based on a good understanding of the options rather than on prejudice and fear of the unknown.
In either eventuality the development process will continue to be fascinating.

Graph from UNSW Team 1: Sam Paterson, David Kirkby, Byrce Edmonds, 

Ashley Thornton, Felicity Kelleher, Nick Tenison, Syafiq Nazarudin 

And Team 2: Jarred Grimmond, Nay Myo Lwin, Stephen Narunsky,
Julia Shields, Tyler Steer, Hu Su

Bling

New Moth bow mechanism developed with Scott Babbage now available from SailingBits.
CarbonicBoats worked with Scott to develop a system with less friction, almost zero play, and built-in adjustment of gearing and wand length.

Four prototypes incorporating different shapes and bearing materials were created and tested before the production version could be signed off.
A great project for learning about the complex tradeoffs between mechanical efficiency, weight, reliability, repeatability of tolerances and cost effectiveness.

Milled from a single block of aluminium the new design takes out the wobble and play from other systems. With a larger diameter axle, and bronze bearings, the bow mech takes away the opportunity for unwanted movement that develops in other designs.

Incorporating adjustable gearing, adjustable wand and a large fast-point variation, you have a large range of adjustment to get you through the full range of conditions.

And for all those aero junkies, it incorporates and aerofoil maystick to reduce drag.

Available in Black, Silver, Red, Blue, Purple & Yellow.

Order yours now by going to: http://www.sailingbits.com/class-specific/moth-bow-mechanism/

Checking In

Click here to read a Q&A Session with Martin Vanzulli who is doing a great job of keeping the A Class website up to date as well as running the Catsailingnews blog.

The interview covers our V3 Paradox A Class design (nearing production) as well as our ongoing foil R&D work.

The final questions are about how some of our recent experiments with control system foils fit with the A Class rule. We reiterate that our design decisions for production are informed by proactive consultation with the Technical Committee to make sure we are always within current rule interpretations when introducing innovations to the market.
Experimental work goes on in parallel. It is aimed at demonstrating what is possible and, increasingly, at satisfying market demand for 'pure' full foiling solutions.
Personally my hope is that fair, objective, literal and consistent rule interpretations will allow further development within a knowable and predictable design space.  

http://www.catsailingnews.com/2014/11/a-class-development-dario-valenza-will.html
http://www.a-cat.org/?q=node/474

Inevitable

Here is an extended mix showing some early runs with experimental control system foil configurations.
Though there is still vast untapped potential, these sequences give a flavour of what is surely to come.
Everyone who tried it commented, through a persistent grin, that it is easy and feels secure.

As often repeated on this site, the goal is performance, not foiling at all costs.
Passive systems such as L/V foils give some measure of heave stability at the cost of some additional lift-induced drag. If properly designed they are competitive and manageable. The key is to design the system to work with the hull so that the highly foil assisted mode remains fast. In the right conditions and with the right technique the skipper can then push beyond a 99% lift share and transition to full foiling.
So far this has only been proven to pay downwind in flat water when fully powered up. But undoubtedly the profitable flight envelope will steadily grow, expanding to lighter winds and upwind.
By all accounts engaging this mode is hard work and requires judgement to give net gains in VMG. But since gains are definitely available, it is a challenge to be relished.
V, comma, and now Z foils have improved performance, added a challenge and made the A Class safer to push hard, without taking away from the delicious responsiveness of this lightweight boat.

The difference between a passive system and a control system is that the latter is simply relentless. The boat will remain foilborne essentially until it stops, allowing for the skipper to look around, sit in, change gears and ride out lulls... All while the ride height is directly reacting to changing inputs.
So enjoy this first glimpse into just what is possible under this great class rule!

Make no mistake: there is a concrete measurable difference between current passive systems and truly stable foiling. As long as speed gains are out there, people will experiment and discover ways of realising them. Whether this is made easy and safe or expensive and dangerous is determined by how the rule is administered.

Paving the Way

Lots to report as we continue to test over the winter... 

We are working on the next-generation Paradox A Class Cat design for the 2015 season, resolving the details for all new tooling to be created in-house at a new facility. 

Now that our new expanded production facility is operational we can tackle such jobs with confidence. This gives us more control than before when we relied on contractors for certain aspects of production.

The path we are taking is, as always, very empirical. Every idea is assessed for potential merit, tested objectively, evaluated, then either discarded or developed for the next round of testing.

The focus is on perfecting a foil package that will be a significant improvement on current designs. ‘Improvement’ in this case is strictly defined as the ability to generate better performance around the racetrack in most conditions. So ease of handling, maneuverability and acceleration play a role as well as outright straight-line speed.

We began this phase of R&D by prototyping a series of ‘acute L’ (AKA 'L/V') foils. These all shared a common vertical strut but had incrementally different horizontal chord, span, tip-up angle, and section characteristics. For testing they were inserted from below into simple straight (parallel-sided) cases. These cases are installed in one of our test platform (the orange boat nicknamed Glamorous Glennis) in the exact same position as the production ‘comma’ foils we used at the NZ Worlds.

Following are some thoughts on the testing process and the state of play in the Class:

Rudders

Two candidate revised rudder designs were tested. More on the selection of rudder design in future posts. For those of you who missed the previous related post, the 2014 version of the cassettes is pictured below. 
You will notice that the rake adjustment system has been simplified and construction beefed up to maximise stiffness.

Robust cassette assembly machined from billet. Available now.
Rod end/spherical bearings have been deleted and rake adjustment can now be easily done on the water

Continuing Foil R&D

Imposing the constraint of a straight vertical strut simplifies progress by reducing the number of variables. It also reduces production cost, allows us to use the full horizontal span permitted by the rule and makes fitting of the structural foil case very simple.

Relatively quickly we came to some definite conclusions regarding ideal tip-up angle, shape, and area for reliable stable foiling using the leeward foil only. Needless to say this configuration is extremely promising with upwind foiling and foiling jibes being a given. The key is the ability to use all the beam of the boat to generate righting moment. A marked difference can definitely be felt when the windward foil is out of the water and no longer pushing the windward hull up.

As an aside, the market has proved very hungry for this type of foil. Many customers want to retrofit their boat with the simplest, most cost effective package to just get out on the water and enjoy foiling.

Since racing in the A Class was always integral to our design brief, we have also devised a way to legally fit the final selected L/V foil in compliance with Rule 8. Perfecting this aspect of the concept will be the next step and hopefully the result will be relatively elegant. I say relatively because any solution other than inserting from below will be more complex than strictly necessary. But our challenge is to minimise the rule-mandated unnecessary complexity.

No Stone Unturned

Part of the test series is a radical concept that could potentially achieve two goals simultaneously: Firstly it could be inserted from above through a very modest slot/case with no complex cassettes. Secondly it could displace the horizontal lifting surface forward, increasing separation from the rudders, without affecting helm balance. 

A side-benefit is that the full horizontal span could be used without needing to put the vertical extremely outboard. 

Stability would still come from a tip-up angle (leeway coupling) exactly as for an L/V foil. 
This concept does involve a wetted area penalty (in the form of the area of the horizontal tube). 
It poses some structural challenges (flex in the tube and twist in the vertical foil) and it has a higher induced drag because it has more free tips exposed to the flow. 
Preliminary calculations showed that it had enough potential to warrant building a prototype for testing. We will know soon how it does in the real world…

In Parting

That sums up our status along the fascinating journey of performance development. 

Now to explain the title of this post: Observing competition in Europe we have been happy to note that the approach we took for the production V2 Paradox is now finding acceptance by other manufacturers.

Our 'bent' foils (as opposed to curved) that exit the hull vertically then transition quickly to a span with pronounced dihedral, have been emulated and refined to different extents (functionally the working portion of the foils in this concept is not dissimilar to that used successfully by Hydroptere).
Interestingly some newer designs place the ‘elbow’ further down so that the hulls effectively sit higher when the foils are working in equilibrium. It looks more spectacular and arguably gives a bit more wave clearance, but the penalty is extra foil area - a compromise with respect to performance in displacement mode. This can be alleviated by raising the windward foil such that the lower bend passes above the hull floor when sailing upwind and in light airs. Getting the foil to locate properly when partially retracted requires engineered bearings rather than a simple slot. Our bearing technology remains unsurpassed. The effectiveness of our self-aligning bearing design is such that our ‘bent’ foils ‘autotack’.

Our V2 production foils pictured at the NZ Worlds.

This concept of transitioning from a vertical exit to a Hydroptere style dihedral setup was a first in the A Class and has now adopted by others.

The upper bend in our design allowed the windward foil to adjust automatically to optimum dihedral when sailing upwind.

It is certainly great to see a move away from unstable J foils toward more stable (less unstable) arrangements. The guys at the Europeans are to be congratulated for some great performances with well set up ‘four point’ arrangements. It is also great to see validated our findings that loaded surface-piercing foils require careful treatment of camber and entry angle to delay ventilation. Mischa Heemskirk using sections designed by Gonzalo Redondo of D3 seems to have nailed that aspect of foil setup.

Interestingly the foils and beams on other designs have moved forward to closely match the positions seen on our production V2 boats. 

We were happy with the performance of our equipment at the NZ Worlds. But the next steps are already in testing. So that is where we are concentrating our energy now. 

There is yet another avenue we are exploring that has shown great potential in terms of safe, easy, reliable, fast foiling. More news on this and on our new testing centre in the coming weeks... 

Soon we will have to decide which way to go for the production boat. It may be that the market will continue to demand ‘unadulterated’ equipment in parallel with a competitive rule-legal version. So we will continue to offer both options.

The flattery of imitation is a great confidence booster, but pushing forward into the unknown is an even greater thrill.

Setup

As already mentioned, we have spent lots of time on the water recently, with different foil concepts, testing, evaluating and tuning. Some tests with Paradox sailing alone and some in the company of other known fast A Cats. It is safe to say that we are getting a handle on key issues and how they will be addressed on the production boat.

Two boat testing session

As a general observation, we are in unexplored territory for this class and arguably for this type of boat at this scale simply because the relative effect of foil setup on overall performance is much greater when the foils are working hard enough to support most of the mass of boat and skipper most of the time (and all of it some of the time).
Put simply, when the foils are doing most of the work, getting the settings right is much more influential than if they were only helping out a little.
In hindsight this should be no surprise. Ask any 'Mothista' about the effect of a small fraction of a degree of foil angle and they will say it is like night and day. When the foils are the only part of the boat actively interacting with the water (in the case of Paradox the hull may still be 'skimming' but our measurements tell us it is supported by the foils, not by the water) their effect is dominant.

The saving grace is that the correct setup is mostly related to crew weight and remains constant for different conditions. Once the correlation is understood, it should be easily duplicated.
Now that we understand the (far reaching) effects of main foil shape, toe-in and rake, the key is getting the right amount of 'lift share' so that the sterns are supported by the rudder winglets, but a step back can still raise the bows up sufficiently to 'pop' the boat up onto the foils.
This is a function of some combination of winglet Angle of Attack (AoA) and winglet area.

Lets say we want X amount of lift from the rudders such that they will support enough weight to keep the sterns 'flying' but not so much that the stern cannot be made to sink somewhat when the skipper takes a step aft.
We could obtain the desired lift with small winglets at a big AoA or with big winglets at a smaller AoA.
Assuming aspect ratio can be optimised in both cases, the lowest drag solution will come down to the chosen foil section - and the lift coefficient (Cl) it is happiest at.
However the choice will also have an effect on stability: If the AoA is larger, then the boat will trim down further before the winglet goes through a neutral AoA and begins pulling down to restore the desired pitch attitude.
In reality having the winglets actually pull down will only happen in rare 'extreme' situations. However it is a helpful way to visualise the dynamics at play.
Simply reducing the AoA with bow-down trim is enough to introduce a stern-down restoring moment.
The vital part is the rate at which this moment increases since its rate of change is key to stability in pitch.

Optimum ride height with skipper not all the way aft and correct lift sharing by the rudder winglets.
As the bow pitches down, rudder winglet AoA decreases.
In fact we found that the most stable setups tend to takeoff 'stern first' and stay level or slightly bow-down in flight.
The boat happily sits in this attitude when set up correctly. Notice the absence of wake other than spray.

We are cristallising a useful map of how this foil system works and how it can be exploited.
It appears that performance is good when it is set up correctly.
The most impressive aspect has been the utter predictability and controllability of some foils (more than others) when pushing hard downwind. That is definitely an aspect of the brief that was met successfully.

What remains to be proven is whether the gains are exploitable around the course.
One finding for example has been that with some foil types there is quite a significant benefit in raising the windward foil when sailing upwind.
In a close racing situation this can only be exploited if the system to raise and lower the foils is extremely easy and fast to use with minimal distraction.
We have therefore experimented with a series of mechanical solutions and it seems the last iteration meets the criteria.
In short it uses elastics to raise the boards automatically and a single line with significant mechanical advantage to lower them. More detail on the evolution of these mechanical systems will be released later.
As already mentioned, this is a problem that some of our competitors will also have to solve as they adopt 'S' foils with outward inflection at the top that alters dihedral angle as a function of foil vertical position.

Having explored this development path we will only adopt in production a system that is reliable and easy to use without distracting from 'keeping eyes out of the boat'.
If we are not satisfied that such a system can be engineered (meaning that, after friction is overcome and single line operation achieved, the burden on the skipper is still judged to be excessive) then we will change the foil system so that it does not need to be touched during racing.
This may involve changing foil shape and/or finding a compromise toe-in setting that is optimised for always having both foils down.

Referring to the design brief for Paradox, the final balance to be struck must be in favour of best achievable speed around the course.
If a certain setup cannot be sailed at a high percentage of its potential for a large percentage of the time around a course (by a 'mere mortal'), then a slightly compromised variation that is more exploitable will be more competitive.

With this in mind, simplifying the boat is vitally important and what we are learning now will inform the choices for the next prototypes and the production setup.