As I have been somewhat a critic of Taxi645’s ideas around narrow/shorter tires and flexible/passive aero. I figured I would give him and others a chance to be just as critical with an idea that I think does have legs. I am also not particularly much of an F1 historian, so apologies if I got some of the history wrong.
A recent history of active and passive control in F1
F1 has a history of active control solutions. The heyday of this in F1 was in the mid 1990’s in which you could see anti-lock brakes, traction control, semi/full automatic gearboxes, some drive by wire aspects and lastly fully active suspensions. While most of those have since been banned, you can see that more recently F1 has brought back DRS and brake by wire for rear brakes (which use a combination of kinetic recovery energy recovery along with traditional caliper/rotor setup to slow the car.)
With many active systems banned, F1 has also used and subsequently banned some passive systems. These include chassis tuned mass dampers, front and rear interconnected suspensions (FRIC) and suspension inerters (aka J Dampers). Mass dampers were banned in 2007, FRIC in 2014 and suspension inerters for 2022.
What was active suspension about and why was it banned?
Initially Lotus brought active suspension to F1 to solve underbody wing porposing problems (like we are seeing today). Given that the issue was a right height control issue, things such as stiffer springs, etc. was not a good solution. The idea was to create an “active” spring that could dynamically adjust ride height. And this was done via a computer-controlled system. While this was initially about solving porposing, it was clear this was just one of many things the concept could be used for.
Williams took this to the next step. Their system was more about obtaining optimal aero benefits (such as adjusting ride height to reduce drag on long straights) than trying to focus purely on optimal ride height. Additionally, the system could learn the circuit and know where it was so it could adjust the suspension in advance of bumpy parts of a circuit or adjust ride height to increase speed on straights.
This all sounds fantastic, so why was it banned? If you are able to control the attitude of the car so tightly over an entire circuit, you can then optimize the aero solution to work within that new narrower window of expected situations. The argument for the ban was that this more narrow performance window created cars that could be very unstable in non-optimal scenarios. Additionally the cars were able to create higher cornering speeds and the regulators have always had to periodically dial back the performance of the cars (for safety reasons) as technology reduces lap times. F1 banned active suspension in 1993.
Why has active suspension not returned?
With the rework of the regulations for the 2022 season, active suspension was a serious contender for returning as an allowable option. It is nearly three decades since it was banned and given the technology is now not particularly bleeding edge, it could be much easier to implement today than it was back then. With ease of following leading cars being a large design goal of the 2022 regulations, there was concern that bringing back active suspension might once again create particularly peaky solutions that would be contrary to the 2022 goals. So it was not included.
https://www.racefans.net/2019/07/19/...nsion-in-2021/
I hypothesize that another reason for this is that given the cost cap introduction that the teams had a large amount of rework on their hands. They had to not only switch from a generally flat bottom car to one with an underbody wing, they also had to adapt to the wheels and tires. You could say that active suspension might make both challenges much easier, but in the end, I suspect it was just too much for the teams to tackle.
Should we bring back active suspension now?
I don’t think active suspension should be introduced now. It would need to wait until the next set of significant technical regulation changes. And that will be in 2026. While details are not final, it is generally expected that the basic internal combustion setup of today will carry forward (turbo v6), but there will be the removal of the complex MGU-H system and the MGU-K will be expanded in size. This includes a larger battery as well. With the potential for the cars to be slightly less efficient (loss of MGU-H, I suspect there may be an increase in fuel capacity as well as a larger batter pack. I fully expect these core changes will trigger all new designs in 2026 from the chassis perspective. Core things such as fuel cell size, battery size, and potentially overall PU dimensions will drive core design aspects of the car. This could be the time to introduce active suspension.
What problems does active suspension solve?
It solves the same problems it tried to solve back in the mid-1990s. That is to achieve the level of suspension control that designer want, you either need some type of active system, or you have to implement complex mechanical and pneumatic systems to get around active system bans. Therefore, things such as mass dampers, FRIC and suspension inerters came about. All of which were banned due to their complex and esoteric nature.
I also say there is another reason to do this. The analogy I use is that F1 suspension is very cool, but also very esoteric and expensive Swiss mechanical clocks in a world of cheap commodity quartz clocks that tell better time. The complexity and intricacy of classic race car suspension (especially in the bespoke world of F1) is impressive. But it is also totally unnecessary. They are entirely gratuitous in their implementations. But they are this way only because more modern solutions have been banned.
I think an active system could be a huge cost savings for teams, perform better and allow F1 to stand out from other series. There have also been major advances in computing, sensors and actuators and the general miniaturization and cost reduction of all the above. Building an F1 active suspension should be relatively cheap and trivial. Money could be spent elsewhere.
How do we avoid the issue of “peaky” cars?
The active suspension of the 1990’s was impressive for it’s time. And I understand the concerns of it creating highly optimized solutions. If you can control a ton of variables that previously was difficult or impossible to control (car attitude at all locations on the track) then you are able to build a much more optimized solution. And this optimized solution is likely to have much worse than optimal performance when outside of that window. I guess this is the definition of “peaky”.
I think the solution is to allow for a closed loop system, but to restrict it from things like “learning the track” dynamically or even knowing where it is on a specific track that has been pre-programmed into it. Both of which can allow for it to have predictive actions.
If the car has no knowledge of what is about to happen, but rather only what is happening “right now”, then this can provide a solution that is better than today, but maybe not quite as good as what was done in the 1990’s. Teams could not create such highly optimized solutions. I am suggesting that it would control more, but not all of the variable. With the further extension being that the cars maybe don’t have to be as peaky as the regulators fear.
What are the downsides?
The concerns about peakyness might be real. The concerns about failure of active systems might be real. People are concerned the cars are too easy to drive. This will make it easier. The initial cost of transitioning from legacy to active suspension might be a bit of a pain. Teams have a vested interest in how things work today. They have staff who focus on the design and tuning of the suspension. And generally speaking, they have a specific and unique set of knowledge with that being about the legacy style of suspension. While some of their knowledge will transfer over to active suspension, it would be a revolution in that those departments with winners and losers.
What would the regulations look like?
As I mentioned above, I think this would be part of the 2026 regulation change. If the desire to move to active suspension is generally settled upon in advance, teams can devote time and budget toward that goal in advance of the switch over.
I think the regulations should have a few goals
• Cheap to implement and develop
• Simple mechanically
• Controls in place to move innovation into software vs mechanical solutions.
• Limits to prevent a costly arms race.
• Capability limits to prevent overly optimized solutions
My solutions we be as follows
1. Implement a standard ECU just like is used for the Power Unit
2. Hard exclusions/requirements
_a. Track prediction and learning.
__i. The cars should not be able to figure out they are on X track and then act accordingly.
__ii. The cars should not be able to learn the track
__iii. No technology such as GPS, geolocation, dead reckoning navigation, etc. to try to determine their location.
_b. No car to pit or pit to car communication with respect to active suspension.
__i. The car is only governed by the local ECU. No AI elsewhere telling it what to do
_c. Driver controls
__i. Driver can set some limited modes. (such as via rotary knob on wheel). Otherwise it is just the ECU doing it’s thing
_d. Fault detection mode that triggers some type of “limp mode” that provides a safe operation in the event of a failure. In short, don’t crash the car if some sensors fail. This will allow a driver to safely retire the car in the event of a serious failure.
3. Set a fixed number of input channels for the ECU.
_a. Each input is tied to a single sensor.
_b. You can’t combine multiple sensor inputs into one channel.
_c. This means input channels count equals sensor count
4. Set a fixed number of output (control) channels for the ECU
_a. Each output is tied to a single actuator
_b. This means output channel count equals actuator count
5. Provide a standard set of homologated sensors and actuators
_a. This prevents tricky multipurpose sensors and actuators.
_b. It puts everyone on the same playing field
6. Provide some additional language on implementation
_a. Allowable suspension elements (such as basic double wishbone or limited multi-link setups)
_b. Prevent actuators from driving downstream complex mechanical systems. This may be as simple as saying that a linear relationship should exist from an actuator movement and movement of the suspension. Non-linear actions could be controlled in software.
No doubt I am missing something. I think the last item will be the area that would need the long-term adjustments as team look to find loopholes. With software updates being the primary method of R&D and bringing updates to the track, I am on the fence as to allowing “per race” updates. Which would create the situation of teams creating packages that could be optimized (including mode operation) to an individual track. You might be able to resolve this by limiting teams to a fixed number of updates per season, or require them to only update every so often (with and exception to address safety or reliability issues, but with them having to provide evidence that the changes are for those reasons only)
In short, I think it would set F1 apart, be cheaper, can be regulated if done correctly and also be done in a way that avoids the issues from prior implementation by creating something that is better than what we have today, but not to the full extent of what could be technically capable.
Lastly, while I don’t mention it. I think active aero (beyond DRS) could be implemented using pretty much the same formula of standard ECU, standard homologated components, limited on implementation capabilities, etc. You could even combine the two concepts into one or have them work with each other.
Thoughts?
Richard