Chris C.

Still, both a nice-to-have for all but competitive driving. the 911 is very agile as it is. If you can afford it, and you'd regret it down the road, it would make my options list.

I like the look of SPASM but some days regret the rougher ride and yes I scrape - I live in a hilly area. When I was shopping GTSs I would have bought one new if they didn't all have SPASM!

OC 991

Next time get PDCC with your S-PASM, you'll be amazed how sophisticatedly comfortable it is.
If you're not apposed to that kind of thing in a sports car
Might as well get RAS while you're at it lol.

worf928

Racer20

I mean, once you have PASM, PDCC, PSM, etc., all the sensors and signals are there anyway. While I'm not positive about exactly what signals Porsche uses to control RAS, it would be silly of them not to use, at the very least, throttle, brake, and some kind of yaw sensor. The programing is trivial and the additional tunability is significant. Keep in mind that steering angle and wheel speed does not give you enough information to know what the car is doing, since grip and slip angle can vary with normal load, tire pressure, and surface friction.

And you'd be surprised how much chassis calibration is still done in-vehicle by highly skilled engineers. If anything, the tuning is harder than designing the hardware and control system. THAT can be done in simulation.

digits

Yes, it is speed dependent - inverse of front steering angle for tighter turning at low speeds and then aligned with front at high speeds. Warning: I haven't watched the video, or even read anything about RAS internals BUT I would highly doubt that the speed and steering inputs to RAS are mechanically coupled in any way at all - in other words I suspect it's software monitoring the inputs and driving the outputs to control rear wheel position. Assuming this is the case, it's possible that only Porsche knows which of the numerous available input signals are used in that software, but I couldn't imagine being a Porsche engineer fine tuning the system for real world performance and not considering any tweaks to the model that involve input signals other than speed and steering such as lateral acceleration, roll, etc. They'd almost certainly do it for their race cars, and (hopefully) it trickles down to their production cars. So I'll +1 the above question: has anyone seen any claims one way or the other?

worf928

It isn't a question of what's 'harder' it's a question of the development expense, the anticipated return on investment, and (remember, lot's of lawyers involved) the potential liability of a life-ending failure that's plausibly traceable back to PAG as a design fault.

Do we know for a fact that RAS is implemented on Porsche's (real) Race Cars?

I doubt it. A RAS system as sophisticated as you and Racer20 postulate would almost certainly be classified as an active suspension system. I think even the simplest RAS system (which is what I postulate *we* are driving around with) would also be classified as an active suspension system.

Active control of the suspension (along with the differential, brake bias, aero, and probably other things) has been banned, as far as I know, in all forms of organized motorsports for decades. This is certainly true of F1 as of 1995.

If true - that RAS is, or would be, banned, in motorsports - which I'm willing to bet lunch it is, Porsche has no motorsports-based reason to develop a highly-sophisticated RAS system. The current PTV/PTV+ system which mimics (arguably poorly) a computer-controlled dynamically adjustable differential is - as far as I know - also not legal in organized motorsports. This, though, serves as a counter-example: Porsche might spend money on 'active' technologies with no motorsports-based motivation. The follow-on question is: how sophisticated do they need to be?

My Porsche habit has been paid for by about 25 years of experience in developing, fielding, and supporting software-based control systems in mission- and life-critical environments. There's at least an order of magnitude difference in the expense of developing and fielding between a simple RAS system and one that can dynamically adjust toe through the entry and exit of a corner.

It is not, at all, impossible. Most (maybe even all) of the sensors you would need are - as you point out - already on the car. The processing power is also, I suspect, already on the car. An F1 team with experience from 'back in the day' could knock-it-out in a few months. But, in F1 expense is rarely a limitation if it will buy you several tenths and lawyers are further and fewer between. (And there are other simplifying factors.)

In the real world expense is a limitation. I'm skeptical that Porsche (via ZF) has gone to this expense since there are no motorsports reasons and the street-marketable benefits (that we are aware of) can be achieved without the sophistication you two speculate might exist.

On the subject of Porsche's acumen viz-a-viz embedded control software: have you two read the thread(s) on 991s going bonkers on the banking at Daytona? (https://rennlist.com/forums/991-turb...t-daytona.html) ... Doesn't leave me with the impression that Porsche's got its $h1+ together on the subject and strengthens my skepticism (perhaps fear) of a highly-sophisticated RAS implementation.

Without some actual technical data on what's going on inside the RAS code, further speculation is, while perhaps fun and interesting, not going to do anything but expend time.

If you've got it post it. For me, it would be more fun and enlightening (and possibly scary given the above "Daytona Fail") to be proven wrong.

worf928

On this I would bet an important body part: it is ALL in software.

Racer20

They don't need one. RAS is a road car thing that has been developed by road car suppliers and automakers. There are other benefits as well across many market segments: Improved emergency handling, significantly improved stability when towing, smaller turning circle, improved stability at high speeds, etc. Automakers are looking at putting these systems on everything from pick-up trucks to large sedans to sports cars.

I think you're overestimating the cost and complexity of this; not to mention that using only wheel speed and steering angle to control the RAS behavior wouldn't work because those signals alone don't give you enough information to know what the RAS system needs to do. All the sensors to know individual wheel speeds, yaw rates, each corner's suspension motion, the body's vertical motion, each tire's slip angle, and accelerations are already on the car. They are already being used to control numerous other systems, such as PASM, PSM, PDCC, etc. These control systems have been in continuous iterative development for years. Once the basic control scheme is there, adding additional dependancies can be done well within the budget and time frame of a new model development period or a .1 to .2 update. We are generally past developing the basic functionality of these systems now, and the industry is working on making them better, faster, more integrated, and cheaper.

BlueNorther

worf928

Yes. I was responding to 'digits' not you.

I've rarely been accused of over-estimating LOL. I may however, as you point out below, not be counting-up the cumulative investment that has occurred over the past decades.

You're so sure of this? For what Porsche tells us RAS does? Why?

We? Do you work in the industry? On these systems? Or with folks that are and do? Tell us.
Agreed. You have, however, missed my primary point: What non-race regime marketable or objective benefit (the two are not always the same) would derive from a RAS control system with a sophistication and response period fast enough to dynamically change toe through a corner? Certainly it would manifest on the Northloop, but what's the ROI versus areo, power, tires, etc. If it did what you say it does, wouldn't Porsche's Marketing Machine be all over it?

Let's cut the speculation. My knowledge of RAS is based upon what Porsche says it does and what my butt-o-meter says it does. The rest is somewhat-informed speculation based upon non-automotive (but physical) control systems. Regardless of how many sensor inputs are needed to make RAS do what Porsche Marketing says it does (and as far as I've seen they do not claim that RAS changes toe dynamically through a corner) do you have any knowledge of RAS's actual requirements or design? If you do then let us know and tell us what's really going on in there.

And tell us when Porsche's going to fix the control software for Daytona? (No comment on that?)

Otherwise it's just us speculating on what might or might not be true, necessary or expensive. I think it is certainly possible to dynamically control toe through a corner with the hardware that's on the car. I just don't think they actually did it.

Racer20

I'm not sure what you're envisioning here . . . it's not like the rear wheels are self-performing all the tiny corrections that a racing driver does at the steering wheel in real time. But braking, entering/exiting the corner, and acceleration are all different driving situations. Braking in a straightline is different than trailbraking, which is different than accelerating while unwinding the steering wheel. Ensuring stability during all those maneuvers would be very easy to program via different lookup tables and command priorities.

If brake pressure > 0
rear steer rate = x
else y

x and y would be different look up tables or gain ratios where steering angle is determined by vehicle speed and steering wheel angle. There would be a similar one for acceleration, and probably one for when stability control is activated.

Also, keep in mind that the system would have to respond quickly enough to deal with an emergency avoidance maneuver, otherwise the rear wheels would be left pointing the wrong way on the correction steer and create a yaw moment rather than prevent one. The system has to respond at least as fast as a driver can work the steering wheel. All the ZF literature mentions integration with stability control and the ability to adapt to "driving situations" besides just speed and steering angle.

I work in controlled chassis systems. My group doesn't do RAS specifically, but my company does something similar. I don't work on Porsche, but PASM, PDCC, RAS, and similar systems throughout the industry share similar architectures and functionality. I didn't come here to prove anything or to argue, just to share some interesting information with my fellow Porsche enthusiasts. Cheers . . .

Taffy66

I agree with everything you say..Porsche spend a fortune to fine tune the RAS on different iterations within each model line..As an example the 991S,GTS,R,GT3 and GT3.RS all have different RAS tuning..In fact the 991.1 GT3 and 991.2 GT3 have different RAS tuning.The 991R also had a different RAS tuning to the RS.
All of this extensive fine tuning between different 991 versions would be rather pointless IMO if the RAS merely steered opposite to the front wheels up to 30mph and in the same direction above 50mph..I partly base this assumption on owning both 991.2 GT3 and a 991.2 GTS with RAS..

BlueNorther

Hard to believe Porsche would not make more claims about RAS in the 991.2 if it did more than this... and charge more for it LOL.

BeepBeep88

Archimedes

One of the early track reviews had some comments from the reviewer who drove cars with/without RWS back to back on track, including through a coned chicane and he said the difference was very noticeable and that the RWS car was much more nimble.