Engine Performance Tuning Tutorial
Page 1 of 2 12 LastLast
Results 1 to 25 of 37

Thread: Engine Performance Tuning Tutorial

  1. #1

    Engine Performance Tuning Tutorial

    Here by popular demand is a short Tutorial on how to tune the Engine Output of a CFS Aircraft.

    The Straight Line Performance of an Aeroplane is very dependent on the amount of Thrust (and therefore Engine Power) that is available at each altitude. Fortunately, this is not terribly difficult to do for Combat Flight Simulator using a simple AIR file editor. The only other necessary tools would be a means of measuring the effects of our changes to the AIR file variables.
    The best set of measuring tools that I have seen are the gauges included in Jerry Beckwith's Test Panel which can be found here:


    For those who might be wondering, I will not make any claims that this method is a perfect solution. There are limitations to what CFS can represent. I also do not know enough to be able to generate my own AIR file Propeller (511 & 512) Records, so they will be "borrowed" from other stock aircraft.

    One of the greatest limitations of CFS is that it cannot represent multi-speed superchargers. Aircraft can only have a single speed supercharger which means that we will be trying to make our power curve best fit the aircraft's actual power curve. Typically this means that the aircraft will have a smooth power curve plotted against altitude rather than a sawtooth curve that dips at the supercharger shift points.

    The Subject Aircraft will be a Focke-Wulf FW 190D-9. Since I haven't built such a model yet, My FW 190A has agreed to be the stand-in for now. Most of the specifications are quite similar between the two aircraft and I happen to like the look of it.

    - Ivan.

  2. #2

    Engine Data

    The FW 190D series used the Junkers JuMo (Junkers Motor) 213.
    The main production variant, the FW 190D-9 used the JuMo 213A-1.
    Later variants used the JuMo 213E and JuMo 213F which had slightly greater power and significantly higher critical altitudes as well as the ability to mount a co-axial cannon.

    The following data is required to use this method of engine tuning:

    Cylinder Count
    Displacement per Cylinder in Cubic Inches
    Compression Ratio
    Crankshaft to Propeller Reduction Gear Ratio
    Maximum RPM
    Propeller Diameter

    Take-Off or Sea Level Power Rating
    Manifold Pressure used to achieve this power level

    War Emergency Power Rating
    Manifold Pressure used to achieve this power level
    WEP Type

    Critical Altitude
    Additional Power @ altitude data points are also useful.

    Propeller Pitch Range is not absolutely necessary but very helpful.

  3. #3
    SOH-CM-2019 hubbabubba's Avatar
    Join Date
    Sep 2005
    Montréal, Québec, Canada
    Quote Originally Posted by Ivan View Post
    (...) I also do not know enough to be able to generate my own AIR file Propeller (511 & 512) Records, so they will be "borrowed" from other stock aircraft.(...)
    - Ivan.
    Sorry to interrupt...
    You can change Records 511 and 512 using AirUpdate Utility. Also from Jerry Beckwith (HERE)

    It will "dump" the AIR file data in a TXT format that you can edit, tables included, and recompile.

    The download is HERE.

    Back to you Ivan.

    P.S.- And it is FREE!!!
    Torture numbers and they'll say anything.

    Hubbabubba, Touche à tout.

  4. #4
    Hello Hubbabubba,

    I have also written my own programs to pull out specific records and I can translate them to anything I want or even into a CSV so I can load into a spreadsheet. Changing every value in the table is no problem either.

    The issue is that I don't know what the graphs should look like for a given engine power curve.

    Thanks for the suggestion.
    - Ivan.

  5. #5

    JuMo 213A-1 Specifications

    Specifications for the JuMo 213A-1 are the following:

    Cylinders: 12
    Bore: 150 mm
    Stroke: 165 mm
    Compression: 6.5
    Reduction: 0.417 <---- This was hard to find. I found it in Janes Fighting Aircraft of WW II.
    Maximum RPM: 3250
    Propeller Diameter: 3.300 meters

    Performance (varies a bit depending on the source)
    1750 PS (1726 HP) or
    1755 PS or
    1776 HP at Take-Off (3250 RPM)
    Throttle setting should be 1.8 ATA

    WEP at sea level produces 2050 HP to 2240 HP depending on the source you believe
    Throttle setting should be 2.02 ATA

    Critical altitude is around 6500 meters altitude.
    Aircraft critical altitude (presumably with ram-effect) is around 7000 meters.

    Power at altitude varies all over the place depending on your source but appears to be at least 1600 HP at 18,000 feet.

    If anyone has more reliable information, please respond along a listing of the source.

    - Ivan.

  6. #6

    Propeller Power Coefficient - Part 1

    My understanding of the Propeller Power Coefficient is that it is a number representing the power absorbing ability of the propeller. It represents how hard the propeller is to spin. In the case of a constant speed or variable pitch propeller as typically found in a WW2 fighter, it is a representation of how hard the propeller is to spin at each pitch setting as plotted against its advance ratio.

    This is the means by which the game selects the correct pitch setting to use based on engine power and how fast the aeroplane is moving. Strictly speaking, it isn't necessary to get this right to get the proper engine power but it does influence how the power is used.

    The Power Coefficient will vary depending on air density, but I chose to use an altitude of 500 feet to gather my data.

    Since I don't know how to generate my own graphs, I will copy the closest match from one of the stock aircraft. The values for the stock aircraft are shown in the attached spreadsheet.

  7. #7

    FW 190D Propeller Power Coefficient

    Note that the specifications for the JuMo 213A did not include the propeller pitch settings.
    I could not find that data so will be using the values for the stock FW 190A which seem fairly reasonable:
    (23-65 Degrees). In any case, although the attached spreadsheet lists them, they are grayed out because they are not actually used.

    Although the prior post doesn't show it, the power output of the engines of the FW 190A-8 and FW 190D-9 were very similar at low altitudes. The better engine power at altitude, slightly better streamlining and lighter weight is what I believe made the difference.

    Note that the Power Coefficient from the spreadsheet attached is 0.24. This is a measurement of how hard the engine with its reduction gearing can turn a propeller. Note that this calculation completely ignores the number of blades and the profile (activity factor) of the blades.

    None of the stock aircraft go quite that high. The two closest are the Me 109G (0.1721) and the FW 190A (0.1684). I will probably start with Records 511 and 512 from the Me 109G and do a final test with the FW 190A. Perhaps it is a good idea to start with the FW 190A propeller because of its similarity to the one on the FW 190D.

    Although the choice influences flight performance to some extent, it does not affect engine power.

    The AIR File records may be copied from one file to another using AirEd.

    Next comes the first "Flight Test"..... (which will not be tonight).

    Good Night.
    - Ivan.

  8. #8

    ATA Values

    Finished with the bedtime reading to my son a bit earlier than I thought I would....

    So far, the calculations for values to plug into the AIR File have been pretty obvious.
    Volume of a single cylinder was simple geometry and metric conversion.
    (I got 177.93246 Cubic Inches per cylinder from the bore and stroke values listed.)

    RPM is obviously as listed.
    Critical altitude of somewhere around 20,000 - 23,000 feet will do. (I don't think this number makes any difference at all.)
    The Throttle settings for German Aircraft were listed in ATA (Atmospheres Absolute) and THAT conversion may not be obvious.
    Attached is a spreadsheet to help with that calculation.
    To use, just plug in your values in the Left Column and see what the results are in the other columns.
    I wrote up this spreadsheet to help with conversions almost 3 years ago.
    If anyone finds an error, please let me know.
    Also let me know if you know what the Russians used for Throttle settings.

    Initial values are for ambient pressure, so don't change them; They make a good reference and starting point.

    The values I got were the following:
    1.80 ATA = 52.315 inches Mercury
    2.02 ATA = 58.709 inches Mercury

    - Ivan.

    P.S. If anyone wants to comment but doesn't want to do it as a post here, please contact me at Ivan1GFP@yahoo.com or in a PM.

  9. #9

    A Basic Airframe....

    For those that are curious, I figure that the numbers might have a bit more validity if the engine is tested in an AIR file that is very similar to what a complete FW 190D-9 version would look like. Having weights and basic aerodynamic numbers correct will make a service ceiling test more repeatable in the finished AIR file.

    Here are the planned modifications to the stock P-51D:

    Wing Area
    Wing Span
    Wing Efficiency

    Zero Fuel Weight

    Fuel Tank Volumes and Locations

    Cockpit Viewpoint <---- Has no effect but makes testing not look as strange.

    DP File <---- Affects Flying Weight.

    CL Graph will not be changed ---- YET. Up to the stall, it looks fairly reasonable.

    Coefficient of Drag ---- Will be changed after first correct reading of Sea Level Engine power.

    Hope this makes sense.
    - Ivan.

  10. #10

    Updated Damage Profile

    For the purposes of working on Engine Tuning, it helps to get a fairly good estimate of the aircraft's weight.
    Attached is an updated Damage Profile that should reflect the weight of the munitions carried by the FW 190D.
    The source for the round counts is a Focke Wulf manual.
    The source for the weights of a round and one link of the disintegrating belt is from a Schiffer Book about Kurt Tank, Focke Wulf's Designer and Test Pilot.

    The Focke Wulf manual specifies that the cowl guns have no convergence. The wing root guns are set to converge at 600 meters.

    Damage values are the ones I worked out based as a compromise between CFS Stock values and 1% values. The differences are mostly due to my belief that a cannon shell that relies on explosives for effect have no significant additional damage due to extra velocity.
    The choice of using 3 x 250 Kg bombs is because although the FW 190 could carry a much larger bomb, a single 250 Kg bomb can also mimic the load of a 300 liter drop tank.

    - Ivan.

  11. #11

    FW 190D Basic Airframe

    Attached is a copy of the stock P51D Flight model with just the basic data changed to match the FW 190D-9.
    The engine has not been modified.

    Here are the basic changes and how they were calculated:
    Record 301
    Modified to match the visual model I am using.

    Modified to describe the purpose of the AIR File.

    Record 1204
    Wing Area: 196.98 square feet becomes 197 in the AIR file
    Wing Span: 10.51 meters according to the factory specifications.
    Wing Efficiency: Changed to 5250 which is a fairly average value.

    Fuel Tanks (according to Kurt Tank flight test contain 569 liters).
    Center 1: 232 liters = 61.3 Gallons (Same as stock FW 190A)
    Center 2: 337 liters = 89.0 Gallons (Doesn't match with any numbers I have seen)

    Zero Fuel Weight
    Calculated from 4293 Kg (9464.5 pounds) loaded "Clean".
    - 800 rounds 13 mm ammunition (150 pounds)
    - 500 rounds 20 mm ammunition (219 pounds)
    - 232 liters fuel (367.73 pounds)
    - 337 liters fuel (535.24 pounds)
    8192 pounds
    The weight of MW50 is specified as 125 Kg for 115 liters in a tank behind the cockpit.
    Perhaps we should deduct half of that for an average load condition in combat?
    If so, that would make the Zero Fuel weight 8054 pounds.
    Either value still seems a touch heavy to me, but it is close enough as an estimate for engine testing.

    For what it's worth, this weight is almost two TONS heavier than the Zero Fuel weight of the stock P-51D which is why I believe the P-51D has totally unreasonable values.

    - Ivan.

  12. #12

    FW 190D Basic Engine Installation

    I finally got around to installing the JuMo 213A in the FW 190D basic AIR File and took it for a test flight. Attached is the updated AIR file.

    Here is what was changed: (All unstated data has been specified in prior posts.)

    Record 500 - Reciprocating Engine Specifications
    Propeller Pitch Max
    Propeller Pitch Min
    Propeller Diameter

    Record 505 - CFS Reciprocating Engine
    Displacement per Cylinder
    Compression Ratio
    Cylinder Count <---- Unchanged from 12
    Maximum RPM
    Maximum Horsepower <---- Changed to 1600. I believe this only affects sound effects.
    Negative G Effects <---- Becomes "None". This is a fuel injected engine.
    Engine Type <---- Remains "Water Cooled".
    Supercharger <---- Remains "Yes".
    Max Manifold Pressure <---- 1.8 ATA equivalent in Inches of Mercury.
    Supercharger Boost Gain <---- THIS WILL NEED TUNED LATER!!!! Unchanged for Now.
    Critical Altitude <---- 6600 Meter equivalent in feet.
    WEP Type <---- Becomes MW.
    WEP Manifold Pressure Increase <---- Becomes difference between 1.80 ATA and 2.02 ATA.
    No Mixture Control <---- Copied from Stock FW 190A: "KommandoGerat" controlled everything.
    No Panel Magneto Switch <---- Copied from Stock FW 190A.

    Record 508 - Engine Torque versus RPM

    Maximum RPM changed from 3000 to 3250.

    Record 509 - Engine Per Cylinder Friction Loss
    Maximum RPM changed from 3000 to 3250

    Record 510 - Propeller Parameters
    Propeller Diameter
    Propeller Pitch Max
    Propeller Pitch Min
    Reduction Gear Ratio <---- Changed to 1/(0.417).

    Record 511 - Propeller Efficiency
    Copied from Stock Me 109G using AirEd.

    Record 512 - Propeller Power Coefficient
    Copied from Stock Me 109G using AirEd.

    Attached for explanation of parameters is a copy of the FDE Control File I use.

  13. #13

    First Flight Test of the Jumo 213A

    The First Flight Test was either very good or very bad depending on your point of view:

    Without changes to anything other than what was specified in prior posts, the engine power at an altitude of 500 feet was almost exactly what we would want. The maximum speed was also pretty much exactly what we were trying for.

    The Sea Level (Take-Off) Power is somewhere between 1726 HP and 1776 HP depending on which source you believe. Kurt Tank's Flight test of a standard D-9 showed 586 kph (360 mph) at 300 meters altitude.

    See Attached Screenshot of performance values from my version of Jerry Beckwith's excellent test panel. One thing to note from this screenshot is that the propeller pitch is a couple degrees higher than I would have wanted. I expect a bit of overspeeding at higher altitudes with less dense air.

  14. #14

    Good or Bad - Depending on Outlook

    Values that match exactly what you are looking for are good if the goal is something to be used in a project. For a tutorial, this is bad because there isn't an opportunity to demonstrate how to adjust parameters to get what you want.

    For the purposes of this tutorial, we will pretend that we are not trying to reproduce the mid-production Jumo 213A. Early versions of this engine typically produce about 100 HP to 150 HP LESS than advertised, so we will attempt to detune this fine running 1740 HP engine down to around 1620 HP without changing the manifold pressure limits. Depending on how quickly I get bored, we may also try to simulate a "Blueprinted" engine of 1800 HP or so.

    At this point, some folks probably have figured out what is coming next. Adjusting the engine power is actually quite easy:

    To INCREASE Engine Power,
    Increase Torque in Record 508 -OR-
    Decrease Friction in Record 509
    (or some combination of the two)

    To DECREASE Engine Power,
    Decrease Torque in Record 508 -OR-
    Increase Friction in Record 509.

    Note that the effects are not the same.
    Although the engine output at Sea Level (and up to critical altitude) will be the same with either method, the engine power will fall off faster if Torque and Friction are both higher for the same power output.

    Next comes a lot of Modify and Test cycles....
    (Lather, Rinse, Repeat....)

  15. #15

    The Different Ways of Tuning

    Attached are two screenshots showing the results of Reducing Torque and of Increasing Friction.

    Stock Torque multiplier is 0.56. It was reduced to 0.53 to get 1620 HP (Actually 1621 HP).
    It took two tries to get here.

    Stock Friction number is 68 I believe. To get power down to 1620 HP needed a Friction number of around 84.3. It took about 10 tries to get here because I wanted a power reading equal to the other method. The difference is only 1 HP so I believe that is sufficient. You can reach any level of precision as long as you have sufficient time to test.

    Note that we have dropped around 7 mph by losing 120 horsepower.

    - Ivan.

  16. #16

    High Altitude Comparision

    Attached are two screenshots showing the power output of these two engines at altitude.

    Note that the stock supercharger values are much too high and are providing too much power at this altitude. This isn't surprising because we haven't discussed tuning superchargers yet.

    Nothing is different between them other than Records 508 and 509, but the power output is noticeably different. Power difference isn't very much yet (only 33 HP), but then again, we do not appear to be anywhere near the ceiling of this aircraft even at 40,000 feet.

    It should also be obvious that we can continue to raise both the Torque and the Friction in a balanced manner to reduce high altitude power even more.

    - Ivan.

  17. #17

    Edit and Test Cycle

    I will leave the 1800 HP tuning as an exercise for whomever wants to try it. The starting AIR File was posted as Basic Engine Parameters.

    The Edit and Test Cycles can be done very quickly:

    The attempt to reduce power to 1620 HP was completed (at 1621 HP) in two cycles in around 5 minutes.

    The Jumo hits full RPM on the ground about 5 seconds after throttle is advanced. I noticed that the power reading on the ground (Sea Level +250 feet) was about 1617 HP, so I used that as the target reading when tuning the engine for increased Friction. This took perhaps 10 cycles in 15 minutes total time. The extra time was to complete a Maximum Speed run at 500 feet for a screenshot.

    You have all the tools. Don't hesitate to try it out!

    - Ivan.

  18. #18

    Critical Altitude

    The next step in Engine Tuning is to set the "Critical Altitude".

    "Critical Altitude" can mean at least two things:
    1. The maximum altitude at which the Engine can maintain Sea Level Boost pressures.
    2. The altitude at which the Aeroplane achieves its maximum speed. Generally this will be several thousand feet higher than the the Engine's critical altitude (1). On a real aeroplane, this happens because the forward motion adds a ram effect to the engine's intake.

    I make no claims to be an aeronautical engineer, so I may be playing a bit loosely with terminology here.
    The basic idea is to tune the aeroplane's engine so that its power output at each altitude will give appropriate performance.

    I believe this is not really possible with Combat Flight Simulator.
    Many aircraft of WW2 had Multi-Speed Superchargers.
    CFS can only simulate a Single-Speed Supercharger.
    What this means is that CFS cannot simulate the Blower "Shift Points" at which the engine power will be at its lowest.
    Instead, those middle altitudes are those at which the eingine power will be its HIGHEST in CFS.

    This is the reason that I believe the 1% ideas don't work well with CFS.
    They try to match the performance at two points: Sea Level and Critical Altitude.
    When that is done, typically somewhere between those two places, the aircraft will be much more powerful and faster.

    As an aside, please observe that in the attempt to adjust engine power downwards from 1740 HP to 1620 HP, the difference of 120 HP only resulted in 7 mph less speed.
    This is pretty much as expected: the ratio of airspeeds is the cube root of the ratio of engine power.
    What this means is that we be pretty far off with engine power without being too far off for maximum speed.

    Instead of trying to precisely match performance (speed) at two altitudes, I believe it is a better idea to shift the curve as necessary to not be too far off anywhere in the altitude range.
    While this idea works with maximum speed, it does NOT work well with climb rate.
    The climb rate will be quite a bit high at medium altitudes.


  19. #19

    Target Performance

    It helps to know how well the actual aeroplane performed in order to properly simulate it.
    My data comes from a flight test by Kurt Tank flying a FW 190D-9 in autumn of 1944.

    With the Take-Off power setting and three minutes for speed to stabilise:
    360 mph @ 300 meters altitude
    426 mph @ 6600 meters altitude (21,650 feet)
    Neither was done with MW50 injection (War Emergency Power).

    Climb Rate is noted as 16 meters per second AND 1000 meters per minute.
    Climb is noted as being higher than for the FW 190A, but the numbers do not agree.
    Perhaps something less than Take-Off power was used because it was a fairly long sustained climb.
    Eric Brown's test of a Dora after the war showed it to be a close match to a Spitfire Mk.XIV which would mean its climb rate is MUCH higher than Kurt Tank's test shows.
    Allied tests of Arnim Faber's FW 190A-3 and the Smithsonian's FW 190G show both aircraft to have a maximum climb rate of around 4000 feet per minute which is much faster than what Kurt Tank recorded.
    At some point, we need to decide which data is most believable.

    Other sources comment that although the power output of the Jumo 213A wasn't any greater than for the BMW 801D, it could maintain that power several thousand feet higher.
    The BMW 801D was typically capable of 1730 HP - 1750 HP at Sea Level and 1550 HP at 18,000 feet or so.


  20. #20

    Engine Power Graphs

    Pardon the poor quality, but these two graphs give an idea of how the actual engines performed.

    There are several things to note here:
    The BMW 801D test is from 1942. At this time, 1.42 ATA boost was not cleared for service use.
    Although tests of Arnim Faber's FW 190A-3 used it, the intended limit was only 1.32 ATA.
    The graph of 1.32 ATA here also is unfair because the RPM limit is too low.
    The reality is somewhere in between.

    Note however that at 1.42 ATA and 2700 RPM, the HP peaks at around 1800 PS at sea level and 1800 HP a touch higher.
    Note that the altitude scale on the bottom on this graph is NOT LINEAR.

    The Jumo 213E graph is even more interesting.
    First of all, remember that the 213A that we are trying to simulate only had a two speed supercharger.
    The 213E has a three speed supercharger.
    Note that with Take-Off power (labeled "Start" on the graph), only about 1725 HP is being produced.
    Note that with WEP (labeled "SonderNotLeistung" on the graph), only about 2050 HP is being produced.
    This isn't even close to 2200-something HP that is usually quoted but quite in agreement with the WEP increase I am getting in CFS.

    To get from 1740 HP to 1722 HP at SL, the Torque multiplier goes from 0.56 to 0.555.


  21. #21
    Thank you Ivan for this thread, although some is above my Intel level, it is very interesting to follow.

    Thumbs up dude,

    "Laissez les bon temps rouler"

  22. #22
    Thanks, No Dice.

    I had thought this thread would be much easier than this.
    I thought I would be done after posting the next installment about tuning supercharger settings, but there is still:
    Testing with AIR files
    Idle speed
    Propeller animation which I will probably take back to the thread I started a couple months ago.

    The reference data is also quite a lot more difficult to find than I first thought it would be.

    - Ivan.

  23. #23

    More Engine Power Graphs

    Here are a couple more Engine Power Graphs.
    Please observe that the forms are quite similar.
    The number of peaks corresponds to the number of speeds in the supercharger.

    The Merlin graph is fairly self explanatory but the Jumo 213A graph might need some translation:
    Sonder-Notleistung ==> Special Emergency Power (WEP)
    Start und Notleistung ==> Take-Off and Emergency Power
    Steig und Kampfleistung ==> Climb and Combat Power
    Mehrleistung durch GM1 ==> Increased Power via Nitrous Oxide Injection
    Reiseleistung ==> Cruise Power
    Krafftstoff verbrauch ==> Fuel Consumption

    The bottom of the graph shows exhaust thrust.

    Note that 1 PS is slightly less than 1 HP but for our level of precision you can pretend they are the same.
    (1750 PS = 1726 HP)


  24. #24

    Target Engine Power

    Attached is an annotated graph of Jumo 213A Engine Power.
    The Blue shows the specific power curve we are trying to approximate.
    The Red Dots show the three points at which we will be trying to best match the power curve.

    They represent:
    1. Power at Sea Level.
    2. Power at Critical Altitude where Aircraft's maximum level speed is achieved (6600 Meters).
    3. Power at Service Ceiling (11400 Meters).

    We have already seen how easily the Engine Power can be adjusted at Sea Level.
    We have demonstrated how Engine Power at high altitude (Service Ceiling) can be adjusted to some extent.


  25. #25

    Tuning Medium Altitude Performance

    The maximum speed for the FW 190D was achieved at 6600 meters or 21650 feet.
    On the graph in the last post, engine output looks to be 1440 PS.
    The conversion factor is 1.01387 so this is equivalent to 1420 HP.

    In the AIR file, the variable to adjust for tuning critical altitude can be found here
    Record 505: Supercharge Boost Gain.

    Since we started with the stock P51D, it is still 5.36.
    A quick flight test gives the following results:
    00500 feet ==> 52.3 inch MP 1722 HP
    17500 feet ==> 52.3 inch MP 2020 HP
    20000 feet ==> 52.3 inch MP 2071 HP
    22500 feet ==> 52.3 inch MP 2123 HP

    25000 feet ==> 52.3 inch MP 2177 HP
    27500 feet ==> 52.0 inch MP 2219 HP

    Only the two rows in BOLD are really relevant.
    In fact, a single test at 21650 feet would have told us quickly that we needed to adjust.

    Changing the Supercharger Boost Gain to
    2.68 ==> 1163 HP which is obviously way too low.
    3.30 ==> 1549 HP which is closer
    3.20 ==> 1487 HP
    3.10 ==> 1424 HP which is pretty close.
    (There is no point in getting any closer at this stage.)

    A quick speed run at this altitude gives us 421 mph which is way too low.....

Members who have read this thread: 5


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts