Manifold pressure sucks?

[PRB]

So, in a normally aspirated engine, the manifold pressure gauge will simply indicate whatever the outside air pressure is, if the throttle is wide open (or sitting on the ground, not running...) In FS, with a supercharged engine, the MP will continue to read a steady max MP value until critical altitude is reached. If you continue to climb, MP will start to decrease from that point, but at what rate? I would have guessed that MP would begin to change to reflect the change in atmospheric pressure from critical altitude upwards. In the SOH F7F, with a CA of 22000 feet, MP reads 55.7 inches all the way up to 22000 feet. Beyond 22,000, MP begins to reduce at approximately 2 inches every thousand feet. However, that is not the rate of change in pressure up there. It's more like half an inch per thousand feet.

Is “two inches per 1000 ft” about correct, and is that behavior built into the FS “flight engine” or is it written into the plane's air file?

This is a great resource, by the way:
http://www.advancedpilot.com/downloads/prep.pdf

- Paul

 

[Henry]

once upon a time!

I USED TObe able to HELP,I HAD Abrain then:costum:ive lost it somewheresorry cannot even place an atachmentH

 

[stansdds]

I don't know nuthin' 'bout codin' or programmin', but I do know a little bit about engines.

On the ground, regardless of throttle position, when an engine is not running its manifold pressure will be the same as the barometric pressure.

Naturally aspirated engines (no turbocharger, no supercharger), will create a vacuum within the intake manifold, so manifold pressure will be less than barometric pressure when the engine is running.

Turbocharged and supercharged engines are just the opposite. Turbochargers and superchargers force air into the intake manifold, so manifold pressure will be greater than barometric pressure when the engine is running.

As altitude increases, barometric pressure decreases. To compensate for decreased barometric pressure at altitude and to maintain engine power, some engines were fitted with multi-stage superchargers. Each stage increased the manifold pressure, thus the engine still produced near sea level power at altitude. MS FlightSims do not model multi-stage superchargers, thus the "critical altitude" and that is sometimes set to a higher altitude than normal just to compensate for the lack of correct supercharging.


I found you a nice temperature and pressure vs. altitude chart. Pressure change vs. altitude is not a constant, but diminishes with increasing altitude.

http://www.sablesys.com/baro-altitude.html

Hope this helps.

 

[PRB]

What I'm trying to do may be a little ambitious. I want to predict, using only the aircraft.cfg and air file, a plane's maximum speed and range performance in FS, at any altitude. Every time we have a race event, either the annual Round The World race, or one of our “home grown” events, we spend a huge amount of time testing planes to see if they are “hot”, or faster in FS than the real world version. I hate the endless testing of planes for "hotness" in a race event! I would like to be able to run the air file and aircraft.cfg through a program that simply tells us how fast it will go, and how far.

It may be beyond my mathematical skills, but I just started... I have the atmospherics done. Density, density ratio, pressure, from sea level on up. Now I'm working on power curves. Apparently power drops off at an almost linear rate from sea level, or from critical altitude if you have a supercharged engine. I think I'll be able to nail that one without relying on manifold pressure at all. The hard part will be predicting speed at a given altitude and power setting, because that will involve equations of drag, lift, and thrust. The problem with calculating speed using thrust is that the thrust equation uses speed as an input, so it's an “iterative” process, or one requiring calculus... Not to mention that thrust, in a prop driven plane, is complicated by issues of propeller efficiency, and other things. That may be where I hit the wall. We'll see...

 

[fliger747]

Unfortunately there are Mach drag effects that add to the curve to make it non linear. Somehow the sim engine takes all of the plane and atmospheric parameters and comes up with a result, on might be able to do this?

T

 

[Milton Shupe]

I think you are over-complicating the issue.

The only rule is that the aircraft.cfg Reference speeds must be accurate and reflect the RW aircraft.

The reason that works is that you cannot "overspeed" against the stated numbers in the cfg reference at any altitude.

[Reference Speeds]
cruise_speed = 269
max_indicated_speed = 304
max_mach = .56

So, if these are correct, would that not solve the question?

 

[PRB]

I think you are over-complicating the issue.

The only rule is that the aircraft.cfg Reference speeds must be accurate and reflect the RW aircraft.

The reason that works is that you cannot "overspeed" against the stated numbers in the cfg reference at any altitude.

[Reference Speeds]
cruise_speed = 269
max_indicated_speed = 304
max_mach = .56

So, if these are correct, would that not solve the question?

You may be right. After reading your comment, I'm trying to figure out why we spend so much time testing planes! Of course, I've always wondered that... It's an interesting project in any case, and it's helping me understand what's inside those air files. When I posted that "where do I begin?" thread, I really was at a loss as to how to even begin building an FM. This is helping me answer that question, if nothing else, so it's not a complete waste of time.

 

[Milton Shupe]

You may be right. After reading your comment, I'm trying to figure out why we spend so much time testing planes! Of course, I've always wondered that... It's an interesting project in any case, and it's helping me understand what's inside those air files. When I posted that "where do I begin?" thread, I really was at a loss as to how to even begin building an FM. This is helping me answer that question, if nothing else, so it's not a complete waste of time.

And it may still be worthwhile regarding ability to reach limit speeds quickly, especially in aircraft that need to FL300 or better to get to top speeds. How important is that if 20-30% of the time is climbing to best altitude is the question. It would be nice to have a simplistic answer to the specified need however.

And, I still think that Jerry's FDWB is your best "view" inside the air file and its contents.

 

[fliger747]

It is possible to build a plane with the correct top and SL speeds but has extraordinary climb and acceleration. In did this once with a twin engine Corsair (both co located) in an effort to get a two stage supercharger. Didn't exactly work, but boy the climb and acceleration was spectacular. Just hump up the drag to Mach the extra thrust and the max speeds remain correct, but the plane is not....

T

 

[Ivan]

Perhaps this was already covered sufficiently, but here goes anyway:

For a naturally aspirated (non supercharged) engine, the maximum manifold pressure will be the ambient pressure. Look for the "Standard Atmosphere" to figure out what it is spec'ed at for various altitudes. Local conditions may differ, but this is the benchmark.

For an unsupercharged engine, if it is not running, the pressure behind the throttle plate of your carburetor will be the same as the outside air pressure (roughly 29.8-something inches of Mercury IIRC). When you turn on the engine, the cylinders will try to draw in fuel-air. The Throttle plate in your carburetor will restrict the amount of air coming in, thus at low throttle settings, there will be a vacuum relative to outside pressure. 8 inches of Mercury isn't terribly unusual IIRC.

As you open the throttle, you reduce the restriction and the pressure differential decreases. At max (Wide Open Throttle) there will be no difference between pressure behind your throttle plate and outside pressure: 29.8-something inches Hg (Mercury).

With a supercharger, the measured pressuer behind your throttle plate may be well over ambient pressure because there is a pump which is increasing the air pressure in front of your throttle. 45 inches manifold pressure (roughly +15 inches Hg) isn't unusual. The numbers may go much higher. 72 inches manifold pressure wasn't unusual for a late WW2 fighter.

At low altitude, your supercharger may have much more capacity than your engine can take. Thus you may have several speeds for different altitude ranges. At some point, your supercharger cannot compress the thin air sufficiently to maintain maximum allowable manifold pressure. That would be your critical altitude above which your engine power falls off. Your actual critical altitude for speed is generally slightly above this because of ram effects because your aircraft is moving.

For testing, I do a quick engine power reading (using Jerry Beckwith's gauges) at 500 ft, 2500 ft, 5000 ft, and 2500 ft intervals above that to about 35,000 to 40,000 ft. You can tell when manifold pressure and engine power fall off. Maximum speed will be somewhere around there. The power check for all altitudes can be done in a couple minutes (about 10-15 seconds per altitude) because you aren't waiting for things to stabilise as you would for a maximum speed run.

Just to make it really interesting, just about all the countries in WW2 used their own system for measuring manifold pressure. I wrote a spreadsheet a couple years ago to do this conversion because at least in CFS, all the numbers are specified using the US system and it wasn't instantly obvious what German 1.42 Ata, Japanese +300 mm should be translated to in the AIR file.

Hope this helps.
- Ivan.

 

[srgalahad]

I think you are over-complicating the issue.

The only rule is that the aircraft.cfg Reference speeds must be accurate and reflect the RW aircraft.

The reason that works is that you cannot "overspeed" against the stated numbers in the cfg reference at any altitude.

[Reference Speeds]
cruise_speed = 269
max_indicated_speed = 304
max_mach = .56

So, if these are correct, would that not solve the question?

Part of the issue is how knowledgeable the model designer is about the meaning of those speeds when translating from sometimes very inexact historical data. Innumerable sources quote simply one 'max speed' (not even specifying KIAS, KTAS, MPH) and certainly not recording whether this was "max. attainable speed in level flight" or design Vne. Often a modeler will not question or cross-check this data and simply "makes it so". In fact some of the historical data from the WW II era was deliberately vague in this regard for propaganda purposes. However, we have found examples where Vne was used in the .cfg file, but then the model was tweaked to allow it to reach toward/near this in powered level flight, not just as a limit in a dive. Alternately, aircraft models that can reach real-world Vne (interpreted by the modeler as "max. speed" - level) in level flight then have a .cfg limiting speed much higher to enable speed to increase in a dive. These anomalies don't always show until the model is flight tested against r/w data. Others have more-or-less accurate numbers but the model has been crafted to reach that number at the worst altitude and thus is far too fast at Sea Level.

For things like the events of which Paul speaks then would require corrections to be made (and somehow enforced) or the model rendered unacceptable since there is no current way to monitor changes or accuracy of an individual's flight model.

While I agree that it would be useful to be able to decompile the data from the tables, building a tool to do so might be more hours of work ( P likes to play with data ) than the flight testing if done on an ongoing basis.

 

[fliger747]

Generally it is possible to get speeds fairly close to RW ones, probably as close as the rw data justifies, at least from WWII era data which was read off of a gauge by a busy test pilot onto a knee board. If the detailed engine data is available, for altitudes, MP and RPM, it is a big help. As FS does not model supercharging per se, but turbocharging, various misunderstandings of how the engines work exist. First it must be assumed that any racer will run the engine at full throttle as long as he can unless otherwise restricted. Several payware planes I can think of are capable of Hp greatly increased over the real world examples at SL because a critical altitude is not set, but the MP is allowed to fall off from a very high value to where it is correct at say 20,000 ft. One popular carrier aircraft with an R2800 I tested developed something like 4000 hp at SL!

The turbo model can be used moderately well with supercharged aircraft, however several stipulations apply, especially for multi stage supercharging. More or less we attempt to best fit the jagged line of the supercharger output with a smooth curve. Another item is that the high blower versions use up more shaft HP, so total output is reduced, which may require a critical altitude slightly lower than the book value to get the lines to more or less intersect at this point. Get the HP right, then drag can be set to get correct speeds fro the output. This includes Mach drag which can be used to bend the drag curve vrs alt to fine tune things.

T

 

[Skyhawk_310R]

You may be right. After reading your comment, I'm trying to figure out why we spend so much time testing planes! Of course, I've always wondered that... It's an interesting project in any case, and it's helping me understand what's inside those air files. When I posted that "where do I begin?" thread, I really was at a loss as to how to even begin building an FM. This is helping me answer that question, if nothing else, so it's not a complete waste of time.

I think perhaps the best answer I can give to that insightful question is that it pits pilots versus FDE engineeers, and they bang their heads together to see what lies, deceits, and innuendos can be devised to wring out the best overall match to reality, while both the engineers and pilots walk away bruised and frustrated by the compromises that have no work around because of the myriad of artificial limits placed on the effort by FSX's baseline code!

Ken

 

[Skyhawk_310R]

Part of the issue is how knowledgeable the model designer is about the meaning of those speeds when translating from sometimes very inexact historical data. Innumerable sources quote simply one 'max speed' (not even specifying KIAS, KTAS, MPH) and certainly not recording whether this was "max. attainable speed in level flight" or design Vne. Often a modeler will not question or cross-check this data and simply "makes it so". In fact some of the historical data from the WW II era was deliberately vague in this regard for propaganda purposes. However, we have found examples where Vne was used in the .cfg file, but then the model was tweaked to allow it to reach toward/near this in powered level flight, not just as a limit in a dive. Alternately, aircraft models that can reach real-world Vne (interpreted by the modeler as "max. speed" - level) in level flight then have a .cfg limiting speed much higher to enable speed to increase in a dive. These anomalies don't always show until the model is flight tested against r/w data. Others have more-or-less accurate numbers but the model has been crafted to reach that number at the worst altitude and thus is far too fast at Sea Level.

For things like the events of which Paul speaks then would require corrections to be made (and somehow enforced) or the model rendered unacceptable since there is no current way to monitor changes or accuracy of an individual's flight model.

While I agree that it would be useful to be able to decompile the data from the tables, building a tool to do so might be more hours of work ( P likes to play with data ) than the flight testing if done on an ongoing basis.

Astute observations about propaganda issues. But, also add in that captured aircraft used in testing were a limited sample size, and sometimes flown using aircraft that failed to achieve the design spec's for a lot of reasons. The same airplane can have its performance vary by a significant margin due entirely to the differences in density altitude from one day to the next.

Ken

 

[Skyhawk_310R]

Perhaps this was already covered sufficiently, but here goes anyway:

For a naturally aspirated (non supercharged) engine, the maximum manifold pressure will be the ambient pressure. Look for the "Standard Atmosphere" to figure out what it is spec'ed at for various altitudes. Local conditions may differ, but this is the benchmark.

For an unsupercharged engine, if it is not running, the pressure behind the throttle plate of your carburetor will be the same as the outside air pressure (roughly 29.8-something inches of Mercury IIRC). When you turn on the engine, the cylinders will try to draw in fuel-air. The Throttle plate in your carburetor will restrict the amount of air coming in, thus at low throttle settings, there will be a vacuum relative to outside pressure. 8 inches of Mercury isn't terribly unusual IIRC.

As you open the throttle, you reduce the restriction and the pressure differential decreases. At max (Wide Open Throttle) there will be no difference between pressure behind your throttle plate and outside pressure: 29.8-something inches Hg (Mercury).

With a supercharger, the measured pressuer behind your throttle plate may be well over ambient pressure because there is a pump which is increasing the air pressure in front of your throttle. 45 inches manifold pressure (roughly +15 inches Hg) isn't unusual. The numbers may go much higher. 72 inches manifold pressure wasn't unusual for a late WW2 fighter.

At low altitude, your supercharger may have much more capacity than your engine can take. Thus you may have several speeds for different altitude ranges. At some point, your supercharger cannot compress the thin air sufficiently to maintain maximum allowable manifold pressure. That would be your critical altitude above which your engine power falls off. Your actual critical altitude for speed is generally slightly above this because of ram effects because your aircraft is moving.

For testing, I do a quick engine power reading (using Jerry Beckwith's gauges) at 500 ft, 2500 ft, 5000 ft, and 2500 ft intervals above that to about 35,000 to 40,000 ft. You can tell when manifold pressure and engine power fall off. Maximum speed will be somewhere around there. The power check for all altitudes can be done in a couple minutes (about 10-15 seconds per altitude) because you aren't waiting for things to stabilise as you would for a maximum speed run.

Just to make it really interesting, just about all the countries in WW2 used their own system for measuring manifold pressure. I wrote a spreadsheet a couple years ago to do this conversion because at least in CFS, all the numbers are specified using the US system and it wasn't instantly obvious what German 1.42 Ata, Japanese +300 mm should be translated to in the AIR file.

Hope this helps.
- Ivan.

And this is exactly why in such an engine type, an engine failure will lie to you if you reference manifold pressure. You can be above 7,000 feet and when the engine fails it shows the exact same MP as it does when running with the throttle wide open! This is why pilots are taught to use EGT. It doesn't lie -- it just takes its time to tell you the truth!

Ken

 

[Ivan]

I don't believe any of the stuff I fly on the simulators has a EGT gauge. The A6M Type Zero is sposta have one for fine tuning mixture for best economy. Would a drop in oil pressure or RPM be a quicker indication of an engine failure?

FWIW, I posted an Engine Tuning Tutorial for Combat Flight Simulator a couple months back that I believe was quite successful in being able to duplicate performance statistics and to a lesser extent, engine power. When given a choice between engine output and performance, I chose to replicate performance statistics.

http://www.sim-outhouse.com/sohforums/showthread.php?77148-Engine-Performance-Tuning-Tutorial

Regarding the testing of captured aircraft, I am currently in a couple discussions on that subject in other forums.
Modern documentation on Japanese aircraft performance is especially bad. Each type seemed to have its own peculiarities that make the test results less than representative of what the aircraft could actually do.

- Ivan.

 

[Skyhawk_310R]

Real world? No. FSX world, it would I suppose depend entirely upon how the virtual aircraft was coded and I'd be the wrong person to ask for the details on that one.

But, in real world, oil pressure would take more time than reduction of EGT unless the engine failed as a result of low oil pressure. However, RPM is a very bad means to determine engine failure, and here is why: RPM is simply the measure of engine rotation and since there is the aerodynamic forces on the prop to cause it to rotate due to forward velocity, the airstream can be sufficient to windmill the prop at an RPM that would be normal for cruise flight. This is particularly true for a constant speed propeller since the governor will rotate the blades to achieve and maintain the RPM the pilot selected using the prop control lever.

In flying multi-engine piston aircraft, you have to actually shut down your engine in flight and a reason is so the MEI can demonstrate how MP and RPM can deceive a pilot into being lulled into a false understanding about engine failure (especially partial failure which is really just a significant drop in output power). It was eye-opening the first time I did this and flew single engine at cruise and let the prop windmill long enough to see how misleading RPM and MP can really be.

Whether virtual aircraft feature working EGT gauges in FSX is entirely dependent upon how the plane was coded. But, to fly an actual aircraft without any EGT gauge is a bit unusual though anything is possible, especially older aircraft. My Skyhawk and 310R both have EGT gauges, and both actually have graphic engine monitors that provide me with CHT and EGT for each cylinder. Though that is add-on for older aircraft, any I have flown have at least had what's called a totalizer single EGT that measures the combined EGT at the point where the exhaust manifolds come together.

Cheers,

Ken

 

[Skyhawk_310R]

BTW: Relative the other powers fighting World War II, the Japanese had a particular problem with their engine technology. Compared to the Russians, British, Americans, and Germans the Japanese never did develop engines that were capable of the same horsepower output that these other nations achieved. Even Italy did a better overall job with their engines.

The Japanese overcame this problem by simply making their aircraft very light. So, by this method, they achieved excellent power-to-weight ratios for their aircraft, bombers and fighters both. That gave them excellent range plus for their fighters an incredible combination of range and maneuverability. In fact, during 1941 and most of 1942, Allied intelligence actually believed the Japanese had far more airbases than they really did simply because they rejected the possibility that IJNAF had the kind of range needed to attack targets based on the actual bases! But, they did!

Ken

 

[Ivan]

I can't agree with you regarding your summary of WW2 Engine technology.

The Japanese had no problems designing the engines. They just had a problem in manufacturing them as a quality control issue, particularly in regards to the Ha-45 Homare series and the Ha-40 / Ha-140 series of inlines. With the Ha-40 engines, they actually had a lighter engine than the original German DB 601 they licensed and the engine even had slightly greater MP (1.47 ATA = 330 mm) and output.... When it was working correctly.

With the Homare series, they managed to design an engine that could put out around 2000 HP (actually 1990 HP) on only 2000 cubic inches. The prototypes could put out a bit over 1850 HP and as US tests of a Ki-84 showed, if properly maintained, a service example with all the bugs worked out was actually capable of putting out its claimed performance.

We have a pretty good discussion going at WW2Aircraft.net on this subject under Aviation / Best Japanese Fighter.

The J2M Raiden had a quite reliable 1850 HP radial: The Kasei. The same 1850 HP engine drove the H8K flying boat which was one of the fastest around.

The Mitsubishi Kinsei was a very reliable 1500 HP engine that powered the Ki-100 fighter and others.

The Russians during the Great Patriotic War never managed to get their inlines up to the level of the other countries. Their Klimov VK-107 didn't have its issues worked out until the end of the war.
The Shvetsov ASh-82 which powers the modern FW 190A Replicas was and still is limited to around 1800 HP.

The Italians never managed to get their domestic designs right. Their licensed versions of the Daimler Benz inlines never got past the 1500 HP range which was about equivalent to an early German DB 605.

As for lightweight Japanese aircraft, I believe this was more from the requirements side than the ability side. Even their late war fighters such as the N1K2-J Shiden were awfully light for their size and it was due to not carrying any armour at all. Their structural strength was quite good though. The aircraft is the size of a Hellcat but is about a ton lighter.
The J2M Raiden which did well when evaluated by the US forces wasn't well liked because it didn't fit into the Japanese philosophy of how fighters should be built. It was an energy fighter and handled very well by US standards though poorly by Japanese standards.
The Ki-61-Id Hien was actually a fairly well armoured fighter. It wasn't a lightweight though it was underpowered and it evolved into the excellent though slow Ki-100 series. If you check the weight specifications of the Ki-61-Id, you will find that it is very similar to a late Merlin Spitfire.

- Ivan.

 

[Ivan]

.....
Modern documentation on Japanese aircraft performance is especially bad. Each type seemed to have its own peculiarities that make the test results less than representative of what the aircraft could actually do.

- Ivan.

As an example of these US Test results:

The J2M3 Raiden is typically listed as having a maximum level speed of 371 mph. The actual US test put this aircraft at 417 mph which sounds more reasonable for a light fighter with a very small laminar flow wing and 1850 HP. The 371 mph number IS also a maximum level speed, but that is while carrying a drop tank.

The N1K1-J is typically listed with a maximum speed under 370 mph. Its US test put it at 407 mph and it wasn't the fastest version of the N1K?-J series.

The A6M2 Aleutian Zero was tested at 332 mph even though its automatic mixture control didn't work. The later test of the A6M5 put its maximum speed at 335 mph which is astonishing because it would have put the maximum speed BELOW that of the earlier aeroplane even though it had greater engine power.

The Ki-61-Ia Gloucester Tony was cobbled together with an aircraft that was apparently sidelined by engine trouble. Its tests were concluded by a main bearing failure that wasn't repaired until the aircraft was transfered to the USA. If the common 348 mph maximum speed is from this aircraft, it probably isn't indicative of the type. Even the Japanese claimed 590 kph for this aircraft which is 366.6 mph.

The maximum speed numbers for Japanese aircraft are also misleading because all of their testing was done under Military or Rated Power, not Take-Off or Emergency Power.

Just a few details about what I was describing.
- Ivan.