News stories from June 2017

Life on the Limit

15 June 2017

It all used to be so simple. Once upon a time, racing engines could be neatly divided into those that used some form of forced induction and those that did not. Mechanically, of course, there were myriad complexities, but as far as the combustion was concerned that really was about the size of it. Except now there’s a third way.

The last few years have seen the pinnacles of open wheel racing and sports prototypes both switch to a fuel flow limit as the key constraint in their engine regulations. While the motives behind this may be different for Formula 1 and LMP1, the result is that engine designers in both categories have had to adopt similar strategies to get the best out of their powerplants.

“There is a huge step from a conventional turbocharged or naturally aspirated engine to a fuel-restricted engine,” comments Ian Whiteside, chief engineer for advanced projects at Ilmor. “You have to unlearn a lot of what you’ve done before. Fuel flow restricted engines tend to be a lot more complicated to develop, with many more compromises required. That means you need to carry out a lot more simulation and testing upfront.”

For companies like Ilmor, which has worked behind the scenes on one of the recent turbocharged V6 Formula 1 engines, it means adopting a whole new philosophy. But it also raises a question: With the fuel flow limit having such a decisive impact on the combustion design, and the engine configurations used in F1 and sports cars looking increasingly similar, would it be possible to share elements of a common design between the two?

It’s not a new idea, of course. Cosworth, Matra and Ferrari have all adapted Formula 1 engines for use in sports cars in the past. But these days the regulations actively mandate a lot of common ground. The 100 kg/h fuel limit imposed in Formula 1 is tantalisingly close to the non-hybrid LMP1 Lite (LMP1-L) limit, which currently stands at 101.4 kg/h. Turbocharging (which is mandatory in F1) is virtually a prerequisite in LMP1 too, while there’s also a very sound argument for using the same six-cylinder configuration.

In the hybrid (LMP1-H) category, things get more complicated. The cars are given a different fuel allocation per lap based on their energy storage capacity and the type of fuel. This opens up a multitude of different options right from the start, unlike the F1 rules that dictate the basic specifications of the IC engine and the hybrid system. Nonetheless, the use of a fuel flow limit and an electric hybrid system still opens up the potential for quite a lot of crossover – conceptually if not physically.

This isn’t just an academic exercise, either. “We have looked into the possibility of developing an LMP1 Lite engine,” Whiteside confirms. “A number of teams have approached us about it, which has led us to carry out quite a lot of simulation to assess the feasibility.”



So how much commonality could there be between the two applications in principle? The fundamental design of a Formula 1 is laid down in the rules: It has to be a 90-degree V6, displacing 1.6 litres, with four valves per cylinder and one single-stage turbocharger. The bore is also mandated at 80 mm, which implies a stroke of around 53 mm.

In comparison, the LMP1 regulations are wide open. The first dilemma you’d encounter if you were to apply F1 design principles to a sports car engine would be the displacement. Although comparatively small turbocharged engines are favoured, you can in theory go right up to 5.5 litres in LMP1-L, or as far as you like in LMP1-H.

This is exactly the sort of thing that Whiteside and his colleagues pondered during their investigation. “Generally you don’t want the cylinder pressures to be too high, so that might push you towards a slightly larger capacity,” he says. “On the other hand, you need to think about mass and frictional losses when you start increasing the size of the engine.”

Another factor is engine speed. In theory, this is unlimited, although the fuel flow rate effectively imposes an upper boundary. “You wouldn’t want to rev it as high as a Formula 1 engine – the 15,000 rpm limit in those regulations is partly written in to increase the spectacle. Speed equals friction, so I think you’d be looking at 8,000 to 9,000 rpm at the most. Maybe 7,500 rpm. Under the Lite regulations, that would probably steer you towards a (turbocharged) engine with a capacity of 2.5 to 3 litres.”

Current F1 engines have a bore to stroke ratio of just over 1.5:1. This heavily oversquare configuration works well at high rpm, where it allows for larger valves and lower peak piston acceleration than a long-stroke engine. For a somewhat slower revving sports car engine, however, it’s likely you’d want to run a ‘squarer’ ratio (somewhere in the order of 1.2, Whiteside suggests). This tends to increase thermal efficiency, because less heat is lost to the swept portion of the bore than to the cylinder head.

From there, the next question is cylinder configuration. This takes in a range of factors, such as heat loss, friction and the individual masses of the reciprocating components. For a capacity of around 2.5 litres this all points towards a six cylinder configuration. Packaging considerations and the potential issues with crankshaft flex on inline sixes, mean that a vee layout is essentially a given.


Forced Induction

“This tallies with what we see in LMP1-H,” Whiteside points out. “The Toyota is a 2.4-litre V6, while the Porsche is a 2-litre V4. I’d imagine four cylinders was starting to become a marginal decision at 2 litres, but it has the benefit of reducing the engine’s dimensions.”

In theory, you could build a naturally aspirated LMP1 engine – until recently several manufacturers did – but the general consensus now is that the fuel flow restriction favours forced induction. The reason is that the engines now run comparatively lean – around 1.2 to 1.3λ – to make the most of the limited fuel allocation. As a result, they need more air than a traditional engine, so without some form of turbocharging they would either have to rev significantly higher or use a much larger displacement, both of which have their drawbacks.

So, do you run one turbocharger or two? For hybrid applications with heat energy recovery – particularly if you were looking to share some of the hardware with an F1 unit – it’s likely you’d go for a single turbo. Running two motor generator units would add weight, cost and complexity, plus it would deviate further from the F1 recipe.

 On an LMP1-L engine, however, it’s likely you’d go for two. “Twin turbos often allow for better packaging,” comments Whiteside. “You can have one either side of the engine rather than a single larger unit in the centre of the vee, so you can lower the centre of gravity. Depending on the design of turbine housing used, there is also a secondary consideration of potentially inferior exhaust tuning on the single turbo due to pulse interference between the two banks.”

Traditionally, race engines have run slightly rich to maximise power and reduce the risk of detonation. That’s no longer practical in a fuel-limited formula, so it’s become even more important to ensure the in-cylinder conditions prevent knock. From a design perspective, one of the most critical aspects is charge motion.

“We’re trying to create a tumbling motion inside the cylinder,” comments Whiteside. “That’s another reason it pays to use a comparatively long stroke – it gives you more space to set up that motion, whereas a heavily oversquare engine tends to result in a wide, flat swept volume.”

This is where much of the testing – and, in particular, in-cylinder CFD work – comes in on a fuel limited engine, he explains: “One of the fundamental departures from traditional engines is that you’re less concerned about the total amount of air you can squeeze into the cylinder. Instead, it’s about getting the right charge motion and the right lambda value and to suit the available fuel supply. It’s a trade off, but generally it’s worth compromising the peak airflow value to improve the mixing, because it increases combustion efficiency and reduces the risk of detonation.”

The downside to prioritising charge motion is that more boost is required to force sufficient air into the cylinders. This increases the engine’s pumping losses, but it’s generally worth doing, Whiteside notes.

There’s a similar shift in the philosophy behind the design of the inlet and exhaust systems. Traditionally, these were designed purely to maximise volumetric efficiency, but now they’re tuned to reduce the risk of detonation, he explains: “On the air-restricted engines previously used in LMP1, you had to be very careful to avoid blowing any of the inlet air straight through the engine. With the restrictor it wasn’t possible to make up for the lost air. That’s no longer an issue on the current engines, which have no air restrictor. Instead, you can focus on scavenging the exhaust gases from the chamber during valve overlap. That way you’re removing heat, which minimises the chances of knock.”

This new breed of engines is pushing far more reliance on in-cylinder CFD. Ilmor has recently invested in a new computing cluster, plus software licences for Converge CFD. Gamma Technolgies’ 1D package GT Power also continues to be used extensively for design and optimisation.

“In-cylinder CFD is relatively new for us,” comments Whiteside. “We’ve used 1D simulation for many years, but that’s now far more intensive. It’s so important to understand the intricacies of what’s going on inside the combustion chamber.”


Shared technology

It’s clear that there’s a degree of synergy between Formula 1 and LMP1 engines, but could we actually see shared hardware?

“Physically, I wouldn’t expect to see that much crossover on the base engine,” says Whiteside. “You can’t just scale up the architecture of an F1 engine into an LMP1 unit – it’s likely to be running at two thirds the speed, for a start, which means the mechanical loads are quite different. But the benefit that comes from sharing a philosophy between the two categories is that you can carry over the knowledge of how to optimise and develop a fuel-limited engine. There is quite a lot of learning required the first time you do it, and potentially quite a lot of investment in the simulation tools, so any company that has worked on a current Formula 1 engine would certainly have a head start if it decided to tackle an LMP1 project.”

The major castings are likely to require a full redesign. Tempting as it might be to machine off an unused boss here and there, Whiteside reckons it would take more effort to adapt a common design than it would be to develop each application from scratch. Nonetheless, the basic concepts, such as the casting process, the cylinder layout and the design of the water jacket could be shared.

Not surprisingly, a hybrid LMP1 engine could have more potential for crossover. With comparatively similar fuel allocations and a common layout, it’s likely the fundamentals – and perhaps even some of the physical components – of a Formula 1 heat energy recovery system (MGU-H) could be shared with an LMP1-H sports car engine. Part of the challenge, Whiteside explains, would be pairing the inertia of an electric motor to a high speed turbocharger (“that knowledge is very definitely something that would transfer,” he notes).


Commercial case

One of the biggest differences between modern Formula 1 and sports car racing – certainly where privateer teams are concerned – is the budget. So would the technology be prohibitively expensive to share? “You could share the same core principles without carrying over the more expensive aspects of an F1 engine, so it could be quite a cost-effective way of producing an LMP1 unit,” comments Whiteside. “It’s hard to say, though. A lot of the cost comes from developing the engine, which either has to be paid upfront or amortised over a number of units. That’s a difficult business decision for an engine supplier until you have an idea of how many you would sell.” He estimates the lead time would range from around one to two years, depending on whether there was a suitable race engine design which could be used as a base. This could also be tricky for privateer teams, where there isn’t always a guarantee of on-going funding.

One option would be manufacturer sponsorship of the engines within the privateer class – similar to the Ford-badged DFVs found in the back of numerous privateer chassis in 1970s F1. Whether the FIA and the ACO would allow that – when they have indicated that manufacturers belong in LMP1-H – remains to be seen. Similarly, the manufacturer would have to be able to get enough value out of the sponsorship deal to make it preferable to a full works effort.

Of course, the other option is that a Formula 1 manufacturer might look at entering LMP1-H. There are no obvious candidates at the moment (even the persistent rumours of Ferrari re-entering sports prototypes seem to have gone quiet) but that’s not to say it couldn’t work the other way. An LMP1-derived Formula 1 engine from Porsche or Toyota? It’s not impossible. July 2017

Texas Motor Speedway

12 June 2017

FORT WORTH, Texas – By the stat book, Will Power’s 180 laps led and win in the Rainguard Water Sealers 600 may seem mundane.

But to anyone who watches it was anything but, as Power claimed his 31st career Indy car win and second at Texas Motor Speedway. Power (No. 12 Verizon Team Penske Chevrolet) led Chip Ganassi Racing’s Tony Kanaan (No. 10 NTT Data Honda) and teammate Simon Pagenaud (No. 1 DXC Technology Chevrolet) to the checkered flag in the ninth race of the 2017 Verizon IndyCar Series season.

But not without lots of drama leading to the race finish under caution. With nine cautions on the evening, the race became one of attrition as multiple on-track incidents left just nine cars running at the conclusion of the 248-lap event on the high-banked, 1.5-mile oval. The final incident of the evening on Lap 244 came when championship contenders Takuma Sato (No. 26 Andretti Autosport Honda) and Scott Dixon (No. 9 NTT Data Chip Ganassi Racing Honda) made contact on the front straightaway, spinning into the SAFER Barrier in Turn 1.

Also collected were Chip Ganassi Racing’s Max Chilton (No. 8 Gallagher Honda) and AJ Foyt Racing’s Conor Daly (No. 4 ABC Supply Chevrolet), forcing the remainder of the race to be run under caution. The race’s most notable incident occurred on lap 152, when Kanaan made contact with James Hinchcliffe (No. 5 Arrow Schmidt Peterson Motorsports Honda) entering Turn 3, triggering a massive pileup involving nine cars. No one was injured but six cars – for Hinchcliffe, Mikhail Aleshin, Ryan Hunter-Reay, Carlos Munoz, Tristan Vautier and Ed Jones – were eliminated.

The race was red-flagged after 154 laps to clean up debris from the incident. On the ensuing Lap 159 restart, Kanaan was assessed a 20-second stop-and-hold penalty for avoidable contact and blocking. Despite losing two laps, the Brazilian used subsequent cautions to regain lead-lap status and charged forward in the final laps to claim his first runner-up finish since Road America last year.

Following the red flag, INDYCAR announced it would require competition cautions after every 30 green-flag laps and mandate teams to change four tires under caution, as race conditions affected tire wear in a manner different than previous on-track sessions on the recently-reprofiled and repaved 1.5-mile oval.

Despite his late race exit and ninth-place finish, Dixon remains in the championship lead. The four-time series champion is 13 points ahead of reigning champion Pagenaud and 14 ahead of Indianapolis 500 winner Sato.

Power’s win vaults him into fifth in the championship, just 40 points out of the lead.

The Verizon IndyCar Series next turns to popular Road America for the Kohler Grand Prix, June 23-25. By Mitch Robinson | Published: Jun 11, 2017


05 June 2017

Graham Rahal continued his impressive form to dominate in Detroit, becoming the first man to win both races of IndyCar's Belle Isle double-header.

Starting third, Rahal showed no signs of having lost any pace from Saturday and made short work of Ryan Hunter-Reay, passing him on lap nine of 70 with an audacious move around the outside of Turn 7 for second. After Newgarden pitted again, Rahal claimed the lead and started to build a large gap over the rest of the field, holding a 16-second advantage at the end of the final pitstop phase. Coming up to lap Hunter-Reay - who dropped through the field after contact with Helio Castroneves in the early stages - Rahal was unable to close up to the Andretti driver, allowing Newgarden to whittle away at his advantage. Newgarden - who had made his three-stop strategy work with some blistering pace - closed the gap to just under six seconds, before Hunter-Reay eventually allowed Rahal past.

Newgarden was under threat from Penske team-mate Will Power directly after the restart, but managed to hold onto second place across the line, over a second ahead of the Australian.

Sato was unable to convert his pole into anything greater than fourth place, narrowly beating Simon Pagenaud and Scott Dixon to the line. Alexander Rossi made steady progress through the field to finish seventh as Charlie Kimball just managed to secure eighth, finishing only a tenth ahead of Castroneves. Tony Kanaan rounded out the top 10, after AJ Foyt's Conor Daly dropped down the order following contact at the restart.

Autosport 4 June 2017