F1technical.net

Mclaren and Honda officially denying the claims that that they will be using 2015 spec power units is a tight slap on all those outlets claiming BS and using sensationalizing headlines just to invite more hits.

Wazari wrote:McLaren Honda will not be taking any 2015 ICE or PU's to Barcelona, only new 2016 PU's.

Oh sorry, I didn't see the above article when I posted this originally.

Thanks @Wazari.

@Joseki

You said you know the guys from F1Analitica.....please kick them in the butt for us.

Thanks.

Postmoe wrote:

taperoo2k wrote:

Del Boy wrote:
I may be wrong in my interpretation, but I thought software has to FIA approved and be used in the standard FIA designated (supplied by McLaren electronics) ECU.
2016 technical regulations 8.2.1 Control electronics

I believe teams can run their own code on the ECU but it has to be FIA approved and is restricted so a team can't develop software that controls traction as an example. Everything else is to do with the ECU is locked down. McLaren do have expertise in software, so that's not really an issue for them. No idea if Honda do the PU maps themselves or get help from McLaren via data mining the data collected during races.

I'm thinking about the software they need to develop the car holistically, not the ECU.

They cannot use an excel sheet, they need some kind o ERP-like tool to manage the different layers of information from the different teams, only for the engine.

If combustion is key to harvesting, they need to manage several equilibriums. The way I see it, they need a solid sofware only for tuning the engine for a specific GP, with what if scenarios, analytics, etc. I mean... if not, they would need to gather info from the ICE guys, then the MGU guys, then the fuel guys... nightmarish.

Technically, there is nothing avoiding you to put middleware to link the ECU system to this development tool, as long as you don't touch the code.

The teams already have their own "what if" analytical and predictive PU software. On the actual implementation of the software though, Arai admitted that the way the PU was mapped was a big weakness at the start of the 2015 season. They duly sorted it out, but of course software is not the real differentiator here. The good old hardware and how you use the software to manage it is the limitation. If you don't have the hardware and application knowledge, all the best software can't help and vice versa. See Mercedes customer teams; same software, different programming, different PU performance. And as for hardware side, I think if you ran and tuned the Honda PU on the Mercerdes programs you won't get much more out of it than what the Honda tuners got on their programs.

OT: IIRC customers can not play with SW as much as Mercedes. They can play with mapping and such but not too deep. Also did last season give Mercedes the last spec to the customers teams?

PlatinumZealot wrote:

Postmoe wrote:

taperoo2k wrote:
I believe teams can run their own code on the ECU but it has to be FIA approved and is restricted so a team can't develop software that controls traction as an example. Everything else is to do with the ECU is locked down. McLaren do have expertise in software, so that's not really an issue for them. No idea if Honda do the PU maps themselves or get help from McLaren via data mining the data collected during races.

I'm thinking about the software they need to develop the car holistically, not the ECU.

They cannot use an excel sheet, they need some kind o ERP-like tool to manage the different layers of information from the different teams, only for the engine.

If combustion is key to harvesting, they need to manage several equilibriums. The way I see it, they need a solid sofware only for tuning the engine for a specific GP, with what if scenarios, analytics, etc. I mean... if not, they would need to gather info from the ICE guys, then the MGU guys, then the fuel guys... nightmarish.

Technically, there is nothing avoiding you to put middleware to link the ECU system to this development tool, as long as you don't touch the code.

The teams already have their own "what if" analytical and predictive PU software. On the actual implementation of the software though, Arai admitted that the way the PU was mapped was a big weakness at the start of the 2015 season. They duly sorted it out, but of course software is not the real differentiator here. The good old hardware and how you use the software to manage it is the limitation. If you don't have the hardware and application knowledge, all the best software can't help and vice versa. See Mercedes customer teams; same software, different programming, different PU performance. And as for hardware side, I think if you ran and tuned the Honda PU on the Mercerdes programs you won't get much more out of it than what the Honda tuners got on their programs.

Yes, but I'm not talking so much about mapping. This would be the most basic layer and that is easy to develop, since it's only an evolution from the classical approach. I'm thinking more about organizational software dedicated to the development of the car, specially for the engine part. It's more about workflows, coordination on the go and so on. And that is extremely important, because those engines are way more complex than in the V8 era.

I have the feeling that Mercedes is specially good at this, btw. It's the way they adapt, the way they manage fridays, how they bluff since 2014... this isn't only old school experience since this formula is new. They have a good backbone.

The problem may be electronic hardware related. It's much more efficient to send high voltage than high current, and that may be a limiting factor. The batteries are probably limited to about 5-600 volts, and that's not really the issue. The issue is being able to send high voltage to the MGU-K from the H at the 1,000 volt limit while maintaining 5-600 volts at the battery would be very very difficult, probably impossible. The increase in efficiency would be tremendous because you can lower resistance by increasing voltage and reducing current.

Rest assured the efficiencies are already very high - especially voltage drops in cabling. There is very little left on the table.

Sure the cabling is not leaking much but I'm talking about the controllers, non of them are even capable of handling 700 volts let alone 1,000.

godlameroso wrote:The problem may be electronic hardware related. It's much more efficient to send high voltage than high current, and that may be a limiting factor. The batteries are probably limited to about 5-600 volts, and that's not really the issue. The issue is being able to send high voltage to the MGU-K from the H at the 1,000 volt limit while maintaining 5-600 volts at the battery would be very very difficult, probably impossible. The increase in efficiency would be tremendous because you can lower resistance by increasing voltage and reducing current.

Why do you believe it's impossible? There has to be an inverter and it would be between both motor/generators and the battery.

godlameroso wrote:Sure the cabling is not leaking much but I'm talking about the controllers, non of them are even capable of handling 700 volts let alone 1,000.

There is little to be gained by running higher voltages when the efficiencies are already close to 100%. The losses avoided by increasing the voltage are mostly I squared R (resistive heating) losses.

Del Boy wrote: Why do you believe it's impossible? There has to be an inverter and it would be between both motor/generators and the battery.

Only if they are running an AC motor in the MGU's - if it's all DC then you dont need and inverter (merely voltage regulation).

djos wrote:

Del Boy wrote: Why do you believe it's impossible? There has to be an inverter and it would be between both motor/generators and the battery.

Only if they are running an AC motor in the MGU's

They are. All the photos show 3 (heavy gauge) cables to each MGU.

BTW "Brushless DC" is also actually an AC motor.

Great example. Note the yellow cables - especially from 1:55 onwards.

https://www.youtube.com/watch?v=HJ3EIbhwx2w

gruntguru wrote: They are. All the photos show 3 (heavy gauge) cables to each MGU.

BTW "Brushless DC" is also actually an AC motor.

Cheers, I wasnt sure.

PS, 3 cables is no guarantee of AC power.

Sure, but a good indication - especially if all three are the same gauge. Check the Andy Cowell video if you haven't already. (I added a link to the post above.)

gruntguru wrote:Sure, but a good indication - especially if all three are the same gauge. Check the Andy Cowell video if you haven't already. (I added a link to the post above.)

For sure but in this case the 3rd cable will be a 2nd phase, not earth as in your home AC wiring (Active, Neutral & Earth).

if it was a true DC motor, the 3rd cable could be for a 2nd winding.