Honda Power Unit Hardware & Software

[NL_Fer]

Power from ES to MGU-K is limited to 4MJ, Power from MGU-H to MGU-K is unlimited. Brake energy from MGU-K to ES is limited to 2MJ.

They way i see it, once the car exits the corner, driver applies WOT. First the ES is powering MGU-H to spin up the turbo and also powering MGU-K 120kw to accelerate as fast as possible. Once turbo starts boosting, the ICE delivers max torque and enough exhaust gasses to propel the turbine. Now the MGU-H needs to operate as a brake, limiting the turbo's pressure and rpm. Braking the turbine, the MGU-H is generating electrical energy to power MGU-K, so less energy is taken from the ES. It is believed that Mercedes' MGU-H is generating so much energy, it can run WOT with MGU-K enabled for a whole lap.

Now with the PU in quali mode, the ES can be depleted at the end of the lap, using all available energy for just one fast lap. If in normal racing mode, MGU-H can harvest enough energy to deploy MGU-K during all the WOT parts of a track (approx 60s for most tracks) there is no other way to deploy MGU-K more.

So they open the wastegate, which lowers the energy harvested bij MGU-H and lowers exhaust backpressure for the ICE, delivering more power at the crank, increasing accelaration.

[NL_Fer]

Anyone with some educated guesses what is the output power of MGU-H?

Most tracks they run 60-70% WOT, average laptime is 1m30s, so MGU-K is deployed for approx. 60s. Assuming At 120kw, it needs 60 x 120kw = 7,2Mj. If MGU-k can harvest 2Mj, MGU-h needs to harvest at least 5Mj during 60s of WOT, actually it will be less than 60s, since no energy is harvested when the turbo spins up. Roughly i would think of 100kw brake energy for MGU-H?

Now you understand, why these engines need a really big turbine and Alonso is screamng gp2 engine halfway the straight.

[Tommy Cookers]

.... Let me give you an example. The car is exiting a corner on a wet track. The driver applies the throttle. The boost starts to climb and the energy store is full. Traction is limited. Are you telling me that the MGUH will attempt to control boost by sending over 100hp to the rear axle when it is not needed?
That's just one situation where you simply tie up the whole system with a cumbersome and finicky way of boost control.
Put a waste gate in the situation and you have independent and precise control over you power unit and traction.

wastegate pressure is fixed so will be wrong for relieving most wheelspin situations ?

actually the driver applies some accelerator pedal displacement and the 2d mapping determines a throttle displacement
that displacement will change with rpm variously to reduce the torque increase with rpm ie help the driver
and the mgu-k slaved to the crankshaft is in continuous 2 quadrant operation, so resisting wheelspin and wheel locking

and the throttle displacements must surely 'know' whether eg the ES is full etc
and would also be used to prevent overboost ?
(and overboost is self-limiting at 100 kg/hr fuel rate)

wastegating to so-called zero (or partial reduction of) so-called backpressure is interesting
the turbine is at max rpm and will try to evacuate the exhaust manifold ??
the mu-h motoring power lost by driving the turbine will be how much ? ..... and how/why ??
we might want to throttle the turbine but cannot ?

[FW17]

Question on the wastegate

I thought only a mechanical waste-gate was allowed (as a safety feature); if in case it is to be open based on certain map on the ecu, how will it work? Is it actuator controlled? Is that permitted?

[PlatinumZealot]

Log.

Honda must be kicking itself after seeing that Mercedes ditched the log exhaust this season.

No, they have testing rigs to see what spec is better, too

Better in the case of packaging and reliability but tubular is always going to give more power. Let us see if Honda sticks to it for next year.

[Vortex37]

I believe the engine techs in f1 are programming the mgu-h strategy for every race, just to create the strategy which gets thr car around the circuit as quick as possable. Running the mgu at less than 120kw is useless, the faster a car can accelerate in the beginning of a straight, the quicker it can reach the end of it, it takes the momentum over the straights full lenght.

This was confirmed by Remi Taffin of Renault, just before the Monaco GP. The article was on the Renault F1 site.

@Ringo,Trini,Platinum Zealot, TC, GG. and the regulars.

Like I said months ago, these engines need to be considered as a complete system with multiple feedback loops, together with active/predictive software control. Turbo lag - why would you ever let it happen? As one of you alluded to above, multiple injection/ignition events per cycle are permitted. I still hold to the idea that they have a BOV, and GG mentioned it. I still think I would modulate both the waste gate and the BOV for multiple reasons. But these control events are going to vary depending on what is and what has just happened(WOT, lift off, braking, lift and cruise etc) and the state of the ERS. So adaptive control is almost certain, unless teams are using onboard GPS to decide track position, and therefore what control option is running. Maybe I missed it, but I couldn't see anything in the rules to ban it. Fluttering was mentioned, and I would add that others have mentioned a sound like misfiring during initial acceleration. With the sophisticated ignition and fuel injection control systems, it would be simple to do a running sequence across the cylinders. I don't discount the possibility of a few injection/ignition events during exhaust valve open, to maintain exhaust pressure/temp, during transition to acceleration.

@bergie88 Absolutely the Limebeer paper is essential reading.

[Blackout]

A lotus member in 2014 said:

For those interested. Here are some of the changes Renault have made since Jerez.
Change of battery cell provider. The individual cells that make up the Energy Store have individual over-charge and over-discharge protection. These were proving unreliable due to thermal/vibration issues. Whilst the energy store is sealed in our fitment, I am informed the cells are now supplied by Panasonic.
Change to MGU-K to Crank drive gearing. The original torque multiplication factor was calculated to give a wider spread of torque on acceleration. Track testing found that this was causing traction dificulties and overloading the gearset and causing failure of the crank casing.
Change to turbocharger wastegate function. Renault had originally intended for the H to regulate Turbine speed in 95%+ of normal running. They facilitated this by allowing the H to pull charge (when the energy store was at capacity) to an air cooled heat sink. This strategy proved ineffective in certain environments and a more coventional wastegate is now being used to supplement the H.
Due to both the change in K gear ratio, boost control strategy and the energy store, most of the software relating to the charge and discharge cycling has been modified daily and continues to be refined. There are still issues relating to turbine speed control via H but these are mostly to do with fine tuning of the control software and the syncronisation between H control and wastegate control.
There has been swift progress and software related driveability now appears to be the main issue.

Re the H to K transfer.
There are times when you can't (or wouldn't want to) transfer power from the H to the K but still need to limit turbo speed.
Think along the lines of a short burst of accel then slight decel then accel (such as feathering throttle for traction or in a switch back). The last thing the driver would want is for the K to feed power into driveline.
During those transitional situations Renault wanted to control the speed of the turbo by using the H to maintain shaft speed at or close to max rpm. It turned out (partly because of the driveability issues) that H was dumping to the heatsink far more than had been predicted or modeled. This was causing severe heat related issues in some cars at Jerez and the Renault 'patch' was to disable the H entirely and rely on the mechanical wastegate for the remainder of the test.
Obviously this resulted in dramatically increased lag and reduced performance but allowed some mileage to be put on the ICE.
In the interests of transparency it should be disclsclosed that we have had no first hand experience with the original spec PU used in Jerez.
Our running experience started with a version that had basic H functioning and limited K output.
enotsne

[wuzak]

Let me give you an example. The car is exiting a corner on a wet track. The driver applies the throttle. The boost starts to climb and the energy store is full. Traction is limited. Are you telling me that the MGUH will attempt to control boost by sending over 100hp to the rear axle when it is not needed?
That's just one situation where you simply tie up the whole system with a cumbersome and finicky way of boost control.
Put a waste gate in the situation and you have independent and precise control over you power unit and traction.

You are suggesting that the wastegate is used a s atraction control device - lowering the power in response to a loss of traction. That would be illegal, and i is also not how wastegates generally work in conventional turbos.

The boost requirement is dictated by the driver's foot. In detecting wheel spin he will reduce throttle input, reducing the fuel in the ICE and reducing turbine power.

With the MGUh->MGUK there are other options too. Like reducing ICE power such that the MGUH->MGUK fills the gap to the demanded power.

Using some numbers as an example only, let's say that the ICE at max power produces 500kW. And when the ICE is producing that power, 60kW is generated by the MGUH. Therefore the combined, self sustaining, power is 560kW.

But what if the driver demands 500kW only? There are a few options:

• Assuming ICE and MGUH vary in direct proportion, you could reduce the ICE power to 446.4kW, which would generate 53.6kW at the MGUH, which would then be directed to the MGUK to give the required combined output of 500kW. You are using less fuel to make the same power. And it doesn't matter the state of charge of the ES.
• Or you could reduce ICE power by less, and send the extra MGUH power not required at that time to the ES.
  eg
  ICE: 450kW MGUH: 54kW MGUK: 50kW ->ES: 4kW
  ICE: 460kW MGUH: 55.2kW MGUK: 40kW ->ES: 15.2kW
  ICE: 470kW MGUH: 56.4kW MGUK: 30kW ->ES: 26.4kW
  ICE: 480kW MGUH: 57.6kW MGUK: 20kW ->ES: 37.6kW
  ICE: 490kW MGUH: 58.8kW MGUK: 10kW ->ES: 48.4kW
  
  Where ICE is the power produced by the ICE, MGUH is the power recovered by the MGUH, MGUK is the power needed to be deployed by the MGUK to make the total target power and ->ES is the power not needed by the driver which can be sent to the ES.

I'd also ask how often will the ES actually be full? There will be the start of a qualifying lap, which wouldn't last long, as the ES will be deploying power as the car crosses the line. There is the start of the race, where the MGUK cannot be used until 100km/h.

The only real way the store is going to be filled to 4MJ during the race is if it is being used tactically, that is, not deploying from teh ES until you have the full charge to go for an attack. Or if you can't use the MGUh->MGUK connection.

[PlatinumZealot]

A lotus member in 2014 said:

For those interested. Here are some of the changes Renault have made since Jerez.
Change of battery cell provider. The individual cells that make up the Energy Store have individual over-charge and over-discharge protection. These were proving unreliable due to thermal/vibration issues. Whilst the energy store is sealed in our fitment, I am informed the cells are now supplied by Panasonic.
Change to MGU-K to Crank drive gearing. The original torque multiplication factor was calculated to give a wider spread of torque on acceleration. Track testing found that this was causing traction dificulties and overloading the gearset and causing failure of the crank casing.
Change to turbocharger wastegate function. Renault had originally intended for the H to regulate Turbine speed in 95%+ of normal running. They facilitated this by allowing the H to pull charge (when the energy store was at capacity) to an air cooled heat sink. This strategy proved ineffective in certain environments and a more coventional wastegate is now being used to supplement the H.
Due to both the change in K gear ratio, boost control strategy and the energy store, most of the software relating to the charge and discharge cycling has been modified daily and continues to be refined. There are still issues relating to turbine speed control via H but these are mostly to do with fine tuning of the control software and the syncronisation between H control and wastegate control.
There has been swift progress and software related driveability now appears to be the main issue.

Re the H to K transfer.
There are times when you can't (or wouldn't want to) transfer power from the H to the K but still need to limit turbo speed.
Think along the lines of a short burst of accel then slight decel then accel (such as feathering throttle for traction or in a switch back). The last thing the driver would want is for the K to feed power into driveline.
During those transitional situations Renault wanted to control the speed of the turbo by using the H to maintain shaft speed at or close to max rpm. It turned out (partly because of the driveability issues) that H was dumping to the heatsink far more than had been predicted or modeled. This was causing severe heat related issues in some cars at Jerez and the Renault 'patch' was to disable the H entirely and rely on the mechanical wastegate for the remainder of the test.
Obviously this resulted in dramatically increased lag and reduced performance but allowed some mileage to be put on the ICE.
In the interests of transparency it should be disclsclosed that we have had no first hand experience with the original spec PU used in Jerez.
Our running experience started with a version that had basic H functioning and limited K output.
enotsne

Ah. From the horde mouth. Good post!
This pretty much answers a lot of the questions and is as i suspected. As if video evidence wasn't enough for some people. Haha.

[Wazari]

These new PU's give me a headache and it sure appears now that I retired from race engineering at the right time. It baffles me that the FIA is concerned about "keeping costs down" while they made it necessary to spend enormous amounts of money in R and D, and production to make a competitive PU. After spending too much time analyzing these PU's, it really boils down to the relationship between the MGU-H,K, turbine and ES to maximize output for an entire race lap. It has become IMO, ridiculously complex to keep "costs down".

If it were up to me, I would just keep it simple. Max 2.0 liter displacement, number of cylinders, bore and stroke unrestricted, single turbo, max RPM 16,000 and everyone must run on the same 104 octane "pump fuel" supplied by one vendor only. 8 speed gearbox, no traction control and no ABS. Keep it simple, allow creativity in the ICE design and allow much fuel to be consumed with mid-race re-fueling. This is racing for goodness sakes....sorry for the rant....I feel better now.

[damager21]

These new PU's give me a headache and it sure appears now that I retired from race engineering at the right time. It baffles me that the FIA is concerned about "keeping costs down" while they made it necessary to spend enormous amounts of money in R and D, and production to make a competitive PU. After spending too much time analyzing these PU's, it really boils down to the relationship between the MGU-H,K, turbine and ES to maximize output for an entire race lap. It has become IMO, ridiculously complex to keep "costs down".

If it were up to me, I would just keep it simple. Max 2.0 liter displacement, number of cylinders, bore and stroke unrestricted, single turbo, max RPM 16,000 and everyone must run on the same 104 octane "pump fuel" supplied by one vendor only. 8 speed gearbox, no traction control and no ABS. Keep it simple, allow creativity in the ICE design and allow much fuel to be consumed with mid-race re-fueling. This is racing for goodness sakes....sorry for the rant....I feel better now.

I agree. I think over the years FIA has got it completely wrong. They have been changing rules in the name of cost cutting without thinking about implications. In season development has been clamped to cut costs, but this has led to one team dominating the sport with about 1 sec lead over everyone making the entire event boring because you know for sure that Hamilton / Rosberg will dominate every weekend.

This has led to drop in viewer ship of Formula 1 and as a result sponsors are moving away from F1. This is leading to increased pressure on smaller teams because they dont make enough revenues from sponsors and hence start making losses making their operations unviable.

FIA steps in again thinking that more cost cutting is required and this is nothing but a virtuous circle where cost-cutting steps is leading to F1 becoming boring and in the process sponsors are moving out leading to losses and hence more cost cutting.

You need faster cars, you need re-fuelling, you need access to more fresh tyres to make it exciting.

[j.yank]

Anyone with some educated guesses what is the output power of MGU-H?

Roughly i would think of 100kw brake energy for MGU-H?

Taking the rpm of MGU-H and the physical dimensions possible in this space I will be surprised if it exceeds 60 kW.

[Tommy Cookers]

2 years ago MM showed at Monza their mgu-h stated as 70 kW (whatever that means relative to peak power)
(it was big news in the big thread then)
and (everyone says) Ferrari have greatly increased their mgu-h generated power since 2014 (whatever that means)

[NL_Fer]

Anyone with some educated guesses what is the output power of MGU-H?

Roughly i would think of 100kw brake energy for MGU-H?

Taking the rpm of MGU-H and the physical dimensions possible in this space I will be surprised if it exceeds 60 kW.

I am thinking about Mercedes most of the time. It is obvious that Honda cannot deploy 60s of MGU-K per lap, so they probably harvest less than 100kw.

But more important, i personaly think Honda's design is comprimised by McLarens packaging demands. Poor Honda, made a fool of by Alonso's rant.

[wuzak]

These new PU's give me a headache and it sure appears now that I retired from race engineering at the right time. It baffles me that the FIA is concerned about "keeping costs down" while they made it necessary to spend enormous amounts of money in R and D, and production to make a competitive PU. After spending too much time analyzing these PU's, it really boils down to the relationship between the MGU-H,K, turbine and ES to maximize output for an entire race lap. It has become IMO, ridiculously complex to keep "costs down".

If it were up to me, I would just keep it simple. Max 2.0 liter displacement, number of cylinders, bore and stroke unrestricted, single turbo, max RPM 16,000 and everyone must run on the same 104 octane "pump fuel" supplied by one vendor only. 8 speed gearbox, no traction control and no ABS. Keep it simple, allow creativity in the ICE design and allow much fuel to be consumed with mid-race re-fueling. This is racing for goodness sakes....sorry for the rant....I feel better now.

I agree. I think over the years FIA has got it completely wrong. They have been changing rules in the name of cost cutting without thinking about implications. In season development has been clamped to cut costs, but this has led to one team dominating the sport with about 1 sec lead over everyone making the entire event boring because you know for sure that Hamilton / Rosberg will dominate every weekend.

This has led to drop in viewer ship of Formula 1 and as a result sponsors are moving away from F1. This is leading to increased pressure on smaller teams because they dont make enough revenues from sponsors and hence start making losses making their operations unviable.

FIA steps in again thinking that more cost cutting is required and this is nothing but a virtuous circle where cost-cutting steps is leading to F1 becoming boring and in the process sponsors are moving out leading to losses and hence more cost cutting.

You need faster cars, you need re-fuelling, you need access to more fresh tyres to make it exciting.

I disagree. Well, in some ways.

The FIA has been trying to cut costs for teams, not manufacturers. So while the PUs may cost a lot to develop, that shouldn't affect the teams. But currently it does.

The V8s also cost a lot, even though they were frozen for 7 years. But they were cost capped - while the PUs haven't been, as yet. But likely will be soon.

TV viewership has been declining for many years. to blame it o the current rules is wrong. And it is as much to do with commercial decisions by the FOM as it has to do with the on track action.

Take, for instance, FOM's attitude to social media and streaming content on the Internet - they think it is a passing fad. Yet younger generations are as likely to watch content streamed to their PC, tablet or phone as to watch a TV.

Also, the vast majority of the viewers of F1 over the years have been casual fans. Those that watch F1 because they could access it for free on TV, at a convenient time (in Europe). Now, the trend has been to move to pay tv providers, who may pick up the dedicated viewers, but it would be doubtful that any casual fans would buy into pay tv for F1. The casual fans that watch pay tv are the ones that already have a package.

And not to mention the move away from Europe, the home of F1, and still the strongest supporter base.

Refuelling won't help anything. But they could drop the race fuel allowance. Better tyres would help too. This woudl allow faster lap times, and less fuel saving and car saving.

Better yet, if the rules were to allow 6-8 ICE per season, instead of the current 4, they wouldn't have to be babied as much.