Fuel system and in-flight re-starting

I decided to open this thread to discuss properly about the capability of inverted flight of the Bf-109.

At the moment in the DCS:Bf-109K4, when flying inverted for 5-8 seconds, fuel pressure go down quickly and the engine loses power, but no stop.

In 2013 there was an accident at the Roskilde airshow due to excessive time of inverted-flying of Bf-109G "red 7" as the aeronautical authorities of Denmark reporting.

This is de official report.

http://www.havarikommissionen.dk/index.php?option=com_contentbuilder&title=search-aviation-hclj510-2013-242-bulletin&controller=details&id=4&record_id=41&Itemid=225&limitstart=0&filter_order=&lang=en&contentbuilder_download_file=c43f2f96ca5292e717bcfe1f71c67b2ddaef0539

According to this report, I find the following differences with DCS: Bf-109K4

1/ The fuel pressure drop should occur at 34 seconds of inverted flight, (not 4-8 seconds)

2/ During those 34 seconds, the engine runs normally at full power

2/ 34 seconds after the inverted flight was initiated, the engine must be start to misfire.

4/ Then, the engine must stop completely, (Not gripped pistons, only stopped)

5/ The electric pump P2, is not necessary for normal flight, and the engine must run normally without it at cruising speed. It is only for Take-off, landing, combat or aerobatic conditions. (edited: this is wrong acording to Falke says)

6/ The manual fuel priming, and the in flight electric start must be possible.

Acording with the book of Jean-Claude MERMET "Messerschmitt Bf 109 G-1 to K-4 Engines & Fittings"-1999

in page 21 say:

""Incidentally, contrary to the accepted idea, the DB 605 was equipped

with an electric starter permitting in-flight restart."""

P.D.

The report from Denmark concerns a Bf-109G, but if Yo-Yo take the elevator trim data from a Bf-109G2/R6 taken by the VVS, ..these data are valid too.

/5 is wrong. Already discussed here, booster pump is to be started before engine start and shut off after landing.

http://forums.eagle.ru/showpost.php?p=2308443&postcount=13

Nowadays one tries to get off hours from the boost pump (it's not easy /cheap to get spare parts) and since in flight conditions where no hard maneuvers are flown in terms of moderate descend and climb (no neg G) fuel supply only through suction from engine fuel pumps will be sufficient.

That is why normal procedure of red7 say boost pump off (same as historical 109G). For 109 K boost pump on is obligatory.

In the Denmark incident of red7 an operating boost pump would most probably saved the pilot with a re-started engine.

The "MUST" is not the right word for this case. There are some reasons for it:

1. The fuel consumption in this accident is specified for 2300/1.3ata but you are asking for the same time for 2600/1.42 or even for 2800/1.8???

well, e.g. , applying simple cross-multiplication, it would be something like;

2300rpm = 24seg (data from report)

1150rpm = 48seg

575rpm (idle) = 96seg = 1 min. 36 sec.

2800rpm = 20sec

3000rpm = 18seg

in any case always more than the current 5 to 8 seconds

3. Windmilling engine behaviour depends on the IAS and prop pitch, so to see what happens in DCS you must create the same conditions. Just again - the windmilling prop was compared with real P-51 and matches it.

The airscrew governor in the real P-51 was driven by hydraulic pressure from oil system of the engine. When the engine and propeller is spinning at low speed or stop, there is no oil pressure, and governor no work properly.

The Bf-109 is equipped with, electric-motor driven airscrew governor, this device, can work with the propeller and engine spinnig at low speed or stopped, whenever the aircraft battery have some energy.

the windmilling characteristic of Bf-109 must be some different from P-51.

/5 is wrong. Already discussed here, booster pump is to be started before engine start and shut off after landing.

...

That is why normal procedure of red7 say boost pump off (same as historical 109G). For 109 K boost pump on is obligatory.....

In the Denmark incident of red7 an operating boost pump would most probably saved the pilot with a re-started engine.

I agree with you Falke, .. I Don't have the real 109K4 flight manual, and I don´t know all differences with the Gustav.

You really think that fuel flow is proportional to rpm??? CPSU always tought us that the fuel flow is proportional to the air mass flow regardin mix strength.

Yes.. but the air mass flow is proportional to RPM. Is there any piston-engine that consumes more fuel at idle, than at full power rpm?

it's an erratic statement, deeply erratic... the hydraulic pump in P-51 works fine as the prop is rotating

P-51D propeller governor need that propeller and engine, spinning at idle rpm at least, for working properly, Otherwise, oil pump does not produce enough oil pressure in the circuit. and blades pitch control is not possible.

The Bf-109 electric aiirscrew pitch control does not have these limitations, works in any condition while the aircraft battery has sufficient energy.

Yes.. but the air mass flow is proportional to RPM.

This assumption is plain false. I could have an engine revving 6000 with throttle closed. It surely has less air mass running through than revving 1500 with throttle full open. And there are tons of effects in the system 'engine' that lead to a total mismatch in terms of rpm being proportional to air mass flow. Then again revving with 6000 throttle full open with 600km level compared to 6000 rpm at airspeed 120km on top of a zoom climb tells a totally different story. Now take the boosting (supercharger) in account. You see: if it was that simple ...:smartass: ... Even I would be able to built a perfect injection system for an engine.

This assumption is plain false. I could have an engine revving 6000 with throttle closed. It surely has less air mass running through than revving 1500 with throttle full open. And there are tons of effects in the system 'engine' that lead to a total mismatch in terms of rpm being proportional to air mass flow. Then again revving with 6000 throttle full open with 600km level compared to 6000 rpm at airspeed 120km on top of a zoom climb tells a totally different story. Now take the boosting (supercharger) in account. You see: if it was that simple ...:smartass: ... Even I would be able to built a perfect injection system for an engine.

Obviously, I mean with the engine running normally not windmilling...and this condicion the assumption is plain true. and the air mass flow is proportional to RPM.

And .. obviously a direct fuel injection engine spinning 6000 rpm in windmilling, with throttle closed, consume a bit or nothing fuel.

I find it absolutely valuable that you try to contribute getting the 109k to a status where there is no wrong behavior. Don't be offended but your understanding of the whole concept seems to lack major knowledge.

An engine system with constant speed prop will have major differences in mass air flow at constant rpm of the engine linked to the flight path and coarse or fine state of the prop gear and temperature, air density and many more effects. That's the magic behind the system anyway. If you deny the existence of that fact by saying closed throttle means windmilling and stating that rpm and air mass flow is proportional in such a system you simply do not understand what's going on.

Again do not feel offended, I value your interest in the topic, after all I'm just a 109fanboy, our goals are the same.

:pilotfly:

Falke, taking in account that Otto is an A&P (TMA in spanish), and everytime of his work are aviation engines.... sure.

And btw fuel mix is automatic in 109, and that's the situation that Otto remarks, no windmill one.

greets

Falke, taking in account that Otto is an A&P (TMA in spanish), and everytime of his work are aviation engines.... sure.

If so, he has a great talent in phrasing his points in a way one either gets him wrong or does not understand at all what he wants to point out.

:music_whistling:

For example how could breathed air mass into the engine be the same at e.g. 6000rpm and boost 1,3 vs 1,4 ata?

Most of this is dynamic in flight, many parameters changing rapidly with speed, altitude, temperature, manifold pressure etc.

Most of the time the CSP system will not be able to keep the RPM exactly the same value.

And btw fuel mix is automatic in 109, and that's the situation that Otto remarks, no windmill one.

And that supports the assumption that air mass is proportional RPM?, I mean really???

Maybe thats not his point and I do not understand what he wants to point out?

PBKAM (Problem between Keyboard and Monitor) on my side? I dunno ...

But like I said, everyone who contributes to the goal of an as perfect as possible modelled and bug-free 109K, I give kudos for!

As far as I understood Otto the claimed bug is

that time passing by until missfire and shutdown of engine occurs

after extensive neg G or inverted flight

is in doubt.

It depends on the time until the rest of fuel in the fuel lines is used up

and if - in this specific case -

neither the motor mounted fuel suction pump

nor the intank boost pump

is able to re-establish sufficient fuel pressure as a positive G, not inverted flightpath is re-gained.

One could question if the underlying calculation of this consumption is correct.

But this needs to be answered by Yo-Yo I think.

As far as I understood Otto the claimed bug is

that time passing by until missfire and shutdown of engine occurs

after extensive neg G or inverted flight

is in doubt.

Exactly, that's what I try to explain.

One could question if the underlying calculation of this consumption is correct.

But this needs to be answered by Yo-Yo I think.

My calculation is not accurate, it is only a prorated based on data of time to stop the engine from the Danish report. It is only an example

Danish report data are not exact, (according to the report) they are taken from engine noise in a video of a spectator.

My calculation is a good approximation, because the mass of air and fuel consumption (in weight units, and applying the stoichiometric ratio) is more or less, proportional to engine RPM.

As an example, this chart from fuel injection supercharged V8 engine Cadillac:

For example how could breathed air mass into the engine be the same at e.g. 6000rpm and boost 1,3 vs 1,4 ata?

Falke, I think you guys are talking past each other. Both statements are correct depending on the type of engine being discussed.

Otto says:

the mass of air and fuel consumption (in weight units, and applying the stoichiometric ratio) is more or less, proportional to engine RPM.

That is correct in a fixed propeller aircraft, automobile engine, and a turbine engine.

PLANK/33000 tells us that.

However, in a constant speed propeller aircraft, that relationship of air mass flow being proportional to rpm does not hold up under all circumstances. It goes even farther from the mark with a supercharged engine. Just look at the power production for any supercharged constant speed propeller aircraft after critical altitude.

You can see that in Table 2, Hooker's Expression shows us that charge flow, air plus fuel is the main relationship.

http://www.enginehistory.org/members/articles/ACEnginePerfAnalysisR-R.shtml

As far as I understood Otto the claimed bug is

that time passing by until missfire and shutdown of engine occurs

after extensive neg G or inverted flight

is in doubt.

I think Otto has a point. Rolling upside down should not destroy the engine but neither should intentional inverted flight for more than a few seconds be allowed.

Falke, I think you guys are talking past each other. Both statements are correct depending on the type of engine being discussed.

That is correct in a fixed propeller aircraft, automobile engine, and a turbine engine.

PLANK/33000 tells us that.

You can see that in Table 2, Hooker's Expression shows us that charge flow, air plus fuel is the main relationship.

http://www.enginehistory.org/members/articles/ACEnginePerfAnalysisR-R.shtml

I'm talking about a very specific case. It is a DB605 engine in inverted flight, and running at normal power, or combat power.

The formula for IHP or Hooker's Expression, is not contradict that I am saying.

I think Otto has a point. Rolling upside down should not destroy the engine but neither should intentional inverted flight for more than a few seconds be allowed.

Not a few seconds!! ... Specifically 24 seconds at 2300 RPM!

or more, if you run the engine a less RPM.

I quote my previus post again

well, e.g. , applying simple cross-multiplication, it would be something like;

2300rpm = 24seg (data from report)

1150rpm = 48seg

575rpm (idle) = 96seg = 1 min. 36 sec.

2800rpm = 20sec

3000rpm = 18seg

in any case always more than the current 5 to 8 seconds

. I'm talking about a very specific case. It is a DB605 engine in inverted flight, and running at normal power, or combat power.

Well, then you are wrong on the point of air mass flow as others have explained. I am an A&P too, btw. :music_whistling:

As for the time...24 secons vs 8 seconds is a few seconds....

A few seconds is subjective but the fuel air flow vs volume of fuel in the metering system can be easily calculated to give a reasonable approximation.

Just need a good source on the volume of usable fuel in the system when inverted.

Well, then you are wrong on the point of air mass flow as others have explained. I am an A&P too, btw. :music_whistling:

 

 

As for the time...24 secons vs 8 seconds is a few seconds....

 

A few seconds is subjective but the fuel air flow vs volume of fuel in the metering system can be easily calculated to give a reasonable approximation.

 

Just need a good source on the volume of usable fuel in the system when inverted.

My dear colleague, .. I think I'm not mistaken, but if you're sure that the Bf-109 can fly inverted 8 seconds only, at any engine rpm .. you'll have to give me a reason for it.

I think my calculation is a "reasonable approximation".

but, You can also make a rough estimate, of how many seconds of flight you get ..

At the moment the only specific data that we have, is what documented in the report, of the Danish Accident Investigation Board:

and they say:

" The engine fuel consumption at that time with 2300 rpm was approximately 2.9 liters/min.

After approximately 10 seconds of inverted flight the fuel amount in the fuel supply system was decreased by approximately 0.5 liters and the remaining fuel amount in the fuel supply system was approximately 1.4 liter.

Approximately 24 seconds after the inverted flight was completed, the engine started to misfire.

In this period of time, the engine consumed approximately 1.2 liter of the above mentioned 1.4 liter of fuel. ""

It's a pleasure discuss this topics with a colleague,:beer:

The mistake is not the amount of time or definition of a few second inverted flight

Otto says:

the mass of air and fuel consumption (in weight units, and applying the stoichiometric ratio) is more or less, proportional to engine RPM.

Air/fuel consumption is not proportional to RPM in a CSP equipped aircraft as others have pointed out to you.

Do you have manifold pressure data for this accident?

During the display flight for a time period of approximately 10 seconds, the aircraft flew inverted.

 

For that period of time, the fuel stand pipes in the fuel tank were not emerged in fuel and air was

sucked into the fuel supply system.

The engine fuel consumption at that time with 2300 rpm was approximately 2.9 liters/min.

 

After approximately 10 seconds of inverted flight the fuel amount in the fuel supply system was

decreased by approximately 0.5 liters and the remaining fuel amount in the fuel supply system was

approximately 1.4 liter.

Approximately 24 seconds after the inverted flight was completed, the engine started to misfire.

In this period of time, the engine consumed approximately 1.2 liter of the above mentioned 1.4 liter of

fuel.

The test of the fuel tank in inverted position and with various amount of fuel verified that the fuel tank

and the fuel supply system were not able to deliver fuel to the engine during inverted flight.

You have read the report wrong. The aircraft was inverted for only 10 seconds.

24 seconds after that inverted flight, the engine began to misfire.

file:///C:/Users/Owner/Downloads/76d4dfdb8440aa5b39111a089cbcc567_HCLJ510-2013-242.pdf

You have read the report wrong. The aircraft was inverted for only 10 seconds.

 

24 seconds after that inverted flight, the engine began to misfire.

 

file:///C:/Users/Owner/Downloads/76d4dfdb8440aa5b39111a089cbcc567_HCLJ510-2013-242.pdf

NO my friend, I've read the report correctly .. the engine NO stop until 24 seconds, .. you can add 10 seconds more of inverted flight if you want.

I've read the report correctly

No, read the report again.

It took 24 seconds for the engine to start misfiring AFTER only 10 seconds of inverted flight.

In the last part of the display flight and for a period of approximately 10 seconds, the pilot demonstrated inverted flight.

 

 

Thereafter, the aircraft made a right turn (180°) in normal flight position and followed a traffic pattern

with a second right turn to the cross wind leg to runway 21.

While flying the cross wind leg approximately 24 seconds after the inverted flight was performed, the

engine started to misfire and afterwards lost all engine power.

According to the Aircraft Flight Manual the engine was not suitable for inverted flight. (See appendix 2)

During the display flight for a time period of approximately 10 seconds, the aircraft flew inverted.

air was sucked into the fuel supply system.

The accident was caused by fuel starvation as a consequence of inverted flight. The on board fuel system was not designed for inverted flight.

10 seconds of inverted flight resulted in air in the fuel lines. 24 seconds later...the engine quit in the resulting vapor lock.

The airplane ONLY flew inverted for a total period of 10 seconds.

file:///C:/Users/Owner/Downloads/76d4dfdb8440aa5b39111a089cbcc567_HCLJ510-2013-242%20(1).pdf

It is modeled correctly in DCS.

Otto English report says exactly what Crumpp told you. You misread it.

S!

No, read the report again.

It took 24 seconds for the engine to start misfiring AFTER only 10 seconds of inverted flight.

 

The airplane ONLY flew inverted for a total period of 10 seconds.

 

It is modeled correctly in DCS.

Ok, I forgot to add in my calculations, the 10 second inverted flight.

At the moment, the time of inverted flight or negative G´s in the Bf-109K4 of DCS remains of 4-8 seconds after thats the engine lost power and start windmilling.

Today as you've read in the report, the airplane in the second 9, 10,11, 12 and 33 after starting the inverted flight, .. it was flying at normal power.

In the second 34 (24 after complete the inverted flight) the engine start misfiring.

It's 34 seconds, it's not 4 or 8 seconds .

Crumpp, You can read the page 10 y 11 too, of the Danish.

..

Normally the electrical fuel pump was only “ON” during take-off and landing.

 

With the electrical fuel pump “OF” the engine driven fuel injection pump supplied fuel to the engine

by motive flow.

..

Both fuel filters were almost full of fuel.

according to the comments of several pilots who I have consulted, the engine would not have stopped if the electric pump had been in operation.

The electric fuel pump P2 of Bf-109 is a backup system to ensure the flow of fuel to the injection fuel pump. Unlike the mechanical fuel pump, the electric pump pressurizes the fuel lines rather quickly.

according to the comments of several pilots who I have consulted, the engine would not have stopped if the electric pump had been in operation.

 

The electric fuel pump P2 of Bf-109 is a backup system to ensure the flow of fuel to the injection fuel pump. Unlike the mechanical fuel pump, the electric pump pressurizes the fuel lines rather quickly.

I agree, it probably would not have at least restarted had the electric booster pump been running given time if it shut down at all. Clearing air in the lines and overcoming vapor lock is reason that pump exist's on low wing aircraft designs.