1.6l V6 Turbo engine exhaust temperature

[Steven]

I was wondering if anyone can offer an educated guess on the exhaust gas temperature of modern 1.6l V6 Turbo engines used in F1. I know of ballpark figures of around 1000°C floating around for the old naturally aspirated V8's, but haven't seen any figure for the turbo's.

For reference, we had a very nice thread on this topic in 2009, so relevant for the V8's
http://www.f1technical.net/forum/viewto ... 5&start=15

[gruntguru]

Comparing the NA V8 to the current engine. There would be a substantial reduction in CR - probably about 3 ratios eg 14:1 down to 11:1. This, along with the elevated exhaust pressure (about 3 bar) would increase exhaust temp by about 100 to 1100*C. OTOH the AFR is much leaner - between 1.3 and 1.4 compared to 0.9 for the NA V8. This will reduce EGT by perhaps 200*C (see chart below) so we are back at 900*C at the turbine inlet. If the exhaust gas is then expanded at 80% isentropic efficiency to atmospheric, the turbine outlet temperature will be about 680*C

[Wayne DR]

I would have guessed between 580 and 640 deg C varying with turbine efficiency. (The lower the temperature, the higher the MGU-H output is).

This is somewhat supported by the heat tint (temper) colour of the Ferrari turbo dump pipe of blue/dark blue (shown in photos from this weekend).

Colour - Temp (deg C)
pale yellow - 290
straw yellow - 340
dark yellow - 370
brown - 390
purple brown - 420
dark purple - 450
blue - 540
dark blue - 600

[gruntguru]

Sounds about right. My analysis above ignores heat losses from pipework.

[alexx_88]

@gruntguru: I get why the compression ration needs to be lowered when running a turbo with a port injection system, but why does this need to happen in a direction injection engine as well? I'm studying on GDI right now and din't find an actual reason so far. My thinking is that the ability to time the fuel injection would significantly reduce any chance of the mixture pre-detonating.

Thanks!

[ian_s]

@gruntguru: I get why the compression ration needs to be lowered when running a turbo with a port injection system, but why does this need to happen in a direction injection engine as well? I'm studying on GDI right now and din't find an actual reason so far. My thinking is that the ability to time the fuel injection would significantly reduce any chance of the mixture pre-detonating.

Thanks!

this is something that has intrigued me about the new engines, just when in a cycle do they inject the fuel? there are so many different options here.
if you inject during the intake stroke, you should get a better and more even mixture, but you will still have the problems of pre-ignition so the compression ratio will still need to be lowered.
if you inject during the compression stroke you run the risk of spontaneous ignition if the temperatures in the cylinder are high enough.
if you inject at the point of ignition, you run the risk of an uneven mixture, though if you get it right you can use a stratified charge to try and achieve better efficiency.

[godlameroso]

I would guess the benefit of having such high fuel pressures are precisely so you can inject during the compression stroke.

[RicME85]

@hondapu22 tweeted "My exhaust gases are over 1000 degrees. Hot enough to melt silver & gold."

[gruntguru]

She is just seeking sympathy.

[gruntguru]

Regarding DI and CR. DI enables the use of higher CR - about 1 or 2 ratios. The limit for CR in SI engines is almost always detonation. With very high octane fuel, DI and supercharging, the limit might possibly be imposed by peak cylinder pressure or rate of pressure rise.

DI has a number of modes/strategies.

Homogeneous mode (even mixing of all air and fuel) involves injection during the intake stroke and has similarities to port injection.
Stratified charge mode requires later injection during the compression stroke. This mode is likely used at all outputs in the F1 PUs due to its higher thermal efficiency and increased detonation resistance.

[itix]

I would guess the benefit of having such high fuel pressures are precisely so you can inject during the compression stroke.

I'd guess based on no facts what so ever that they were utilizing injection in several pluses over a 4-stroke cycle. That would makes sense to me as you'd create additional turbulence and spread the fuel more evenly.

I have no basis what so ever for this though so don't take it as any form of facts. Like gruntguru I'd also assume it was during the compression stroke that fuel is being injected.

[gruntguru]

I doubt that an "even spread" of fuel is optimal.

IMO the optimum level of turbulence is applied to the air by port and chamber geometry and injection is a single event late in the compression stroke. The ideal would be a "cloud" of air-fuel mixture at lambda 1.0 to 1.2, surrounded by air (but still in contact with the spark gap). The overall AFR (including the "air blanket") - approximately 1.3 to 1.4.

[itix]

I doubt that an "even spread" of fuel is optimal.

IMO the optimum level of turbulence is applied to the air by port and chamber geometry and injection is a single event late in the compression stroke. The ideal would be a "cloud" of air-fuel mixture at lambda 1.0 to 1.2, surrounded by air (but still in contact with the spark gap). The overall AFR (including the "air blanket") - approximately 1.3 to 1.4.

I am not questioning you here, I'm just wondering why you think that is the best?
In my mind, the more evenly you heat up the gas volume inside the cylinder, the higher MIP you achieve and therefor also a higher end power. Now if the fuel that is to be combusted is more evenly spread around the cylinder the heat should distribute better to all of the gas volume in the cylinder.

The fuel injectors I work with are diesel injectors so I can't say much about a petrol direct injection engine but they (the diesel ones) have extremely small holes in them drilled evenly in a circle around the injector in various angles and height, leading to as even distribution as is possible... but since the injection pump is mechanical and not a common rail system with a solenoid valve (or some other form of controllable valve) there is only one injection pulse.

[Edax]

I would have guessed between 580 and 640 deg C varying with turbine efficiency. (The lower the temperature, the higher the MGU-H output is).

This is somewhat supported by the heat tint (temper) colour of the Ferrari turbo dump pipe of blue/dark blue (shown in photos from this weekend).

Colour - Temp (deg C)
pale yellow - 290
straw yellow - 340
dark yellow - 370
brown - 390
purple brown - 420
dark purple - 450
blue - 540
dark blue - 600

You have to watch out with these colors. These are for steel. The color is originating from thin film interference on the oxide scale that forms on the metal. Your basically looking at the thickness of the oxide scale. The exhaust systems likely are made from a superalloy like inconel which is much more oxidation resistant and thus give a wrong indication.

I don't know for these ICE's but for the old V8 engines I would guess that the temperature is 900C-1000 at the manifold. If you see them on the dyno you can see the exhaust appears translucent (it is not really translucent but you get these weird cold shadowing effects). That happens in this temperature range.



The temperature then drops off. It is clear from the Mercedes picures that they go to greath lengths to insulate their exhaust so they will experience very little drop-off.


Last year Renault and Ferrari had their exhaust unshielded so there the temperature loss will be much greater. Don't know if that is still the case.


If you look at the arrangement then I think you can say that the Mercedes was running their turbo hotter than others (I still think that is part of their advantage). Probably Honda copied that strategy, so 1000 degrees seems possible, although because of material limitations I don't see how they can get much hotter than that.

For the other engine builders I don't know. Last year they seemed to run their turbos colder but I don't know what they have done for this year.

[R_Redding]

Entropy and Enthalpy ...

Gordon McCabe explains ...

http://mccabism.blogspot.co.uk/2014/12/ ... halpy.html