Sign in to follow this  
TopGun

Using Prop RPM to control Prop sound in turboprops...

Recommended Posts

In most turboprop sounds, the waves used to represent the sound of the props are controlled by N2 or "COMBUSTION" as MS would call it. Basically, we will remove this unrealistic reliance on N2 and link the prop sounds to the existing yet underutilized Prop RPM system. They will then respond to changes in Prop RPM instead of turbine speed.Check out this thread for some more info on the benefits of such a system:http://forums.avsim.net/dcboard.php?az=sho...76848&mode=fullSome familiarity with the SOUND.CFG would help in understanding the following instructions. I suggest you read and understand the sound section of the Aircraft Container SDK if you haven't yet.I have used the default King Air SOUND.CFG as an example:(FLTSIM)product_code=FSIM(SOUND_ENGINE)number_of_engines=2eng1_combustion=COMBUSTION.1.00 - Remove these two COMBUSTION lines since we won't be using them.eng2_combustion=COMBUSTION.2.00 - >eng1_prop=PROP.1.00eng1_starter=starterAeng2_starter=starterBeng1_shutdown=shutdownAeng2_shutdown=shutdownBeng1_combustion_start=combstartAeng2_combustion_start=combstartBeng2_prop=PROP.2.00eng2_jet_whine=JET_WHINE.2.00eng1_jet_whine=JET_WHINE.1.00...Now comes the tedious part. If we merely linked the N2 or "COMBUSTION" sounds to the prop sounds, at lower power settings the props would sound much higher pitched and louder than they should because the original parameters were not designed for the relatively limited operating range of the props. We need to adjust these parameters to use values that are more appropriate for propellers rather than turbines.But first, an explanation of the vparams line is necessary. Imagine a graph with X (Horizontal) and Y (Vertical) coordinates, both values starting from 0 and rising to the right and up, respectively. There are 8 points on the graph, and each one from left to right is represented as an X and Y coordinate by vparams as shown:vparams=X,Y,X,Y,X,Y,X,Y,...vparams=X,Y(1st Point),X,Y(2nd Point),X,Y(3rd Point),...Given X which is the % of maximum speed (N1, N2, or Prop RPM), the sound will play at Y loudness. FS uses only values equal to or between 0 and 1 for X, while there doesn't appear to be any relevant numerical limitations for Y. While we will not concern ourselves with Y, you may find tinkering with it useful in fine tuning the sound later. When you connect the dots, or rather points in this case (FS will do the interpolation), they form the "Amplitude Envelope" of the sound. We'll need to move all the points to the right so that they match the range of the Props, while maintaining the relative proportions between each of them. The formula for this is:New X = Original X ( 1 - ( Minimum Prop RPM / Maximum Prop RPM )) + ( Minimum Prop RPM / Maximum Prop RPM )Back to the CFG, the first COMBUSTION line reads like this:(COMBUSTION.1.00)filename=kan21flags=0viewpoint=1rparams=0.000000,0.989000,0.998000,1.761000vparams=0.000000,49.000000,0.002000,49.600000,0.169000,55.200000,0.241000,16.000000,0.380000,0.000000,1.000000,0.000000,1.000000,0.000000,1.000000,0.000000link=COMBUSTION.1.01For example, the King Air's minimum prop RPM is 1050 (Power at Idle, Condition at Low Idle) while it's maximum is 1700. The second point's (Since this is the first sound it's ok for the first point to start at 0), equation would then be :0.618411 = 0.002(1 - 0.617647) + 0.617647We should also set "flags=4" to tell FS that it's a prop sound. Flags=2 doesn't seem to work as well for some reason.After applying the modifications to the relevant remaining points (The points being repeated at the end, 1s and 0s in this case, in the default CFGs aren't crucial) it should look like this:...(COMBUSTION.1.00)filename=kan21flags=4viewpoint=1rparams=0.000000,0.989000,0.998000,1.761000vparams=0.000000,49.000000,0.618411,49.600000,0.682264,55.200000,0.709794,16.000000,0.762941,0.000000,1.000000,0.000000,1.000000,0.000000,1.000000,0.000000link=COMBUSTION.1.01...For this example I used all available digits, though in practice this level of precision won't be worth the effort. I used values rounded down to the second decimal place in my sound packages to simplify calculations. From what I've read, users of FSSoundStudio will find entering the values much easier.You should apply these modifcations to each relevant X value in all the COMBUSTION lines. For the default King Air CFG, there are 16 sections in total. 4 internal and 4 external sounds for the left engine under the COMBUSTION.1.0x header, while the COMBUSTION.2.0x header signifies the right engine's set of sounds. The default sounds always have the same parameters for both left and right engines, no doubt to make things easier. You may want to simply cut and paste the vparams for the left engine to the right engine.After all that, scroll all the way to these two sections at the end:...(PROP.1.13)filename=KAP1Dflags=4viewpoint=1rparams=0.060000,0.500000,0.354000,1.431000vparams=0.000000,0.000000,0.058000,0.000000,0.157000,39.200000,0.255000,41.600000,0.377000,11.200000,0.475000,0.000000,0.475000,0.000000,0.475000,0.000000...(PROP.2.13)filename=KBP1Dflags=4viewpoint=1rparams=0.060000,0.500000,0.354000,1.431000vparams=0.000000,0.000000,0.058000,0.000000,0.157000,39.200000,0.255000,41.600000,0.377000,11.200000,0.475000,0.000000,0.475000,0.000000,0.475000,0.000000This is where we link the COMBUSTION sounds to the existing Prop system, as shown (Remember vparams is only one line) :(PROP.1.13)filename=KAP1Dflags=4viewpoint=1rparams=0.060000,0.500000,0.354000,1.431000vparams=0.000000,0.000000,0.058000,0.000000,0.157000,39.200000,0.255000,41.600000,0.377000,11.200000,0.475000,0.000000,0.475000,0.000000,0.475000,0.000000link=COMBUSTION.1.00...(PROP.2.13)filename=KBP1Dflags=4viewpoint=1rparams=0.060000,0.500000,0.354000,1.431000vparams=0.000000,0.000000,0.058000,0.000000,0.157000,39.200000,0.255000,41.600000,0.377000,11.200000,0.475000,0.000000,0.475000,0.000000,0.475000,0.000000link=COMBUSTION.2.00That's pretty much it, although while the result may be technically sound (At max power, try moving the Prop levers up and down) it may not be aesthetically so. After these initial steps, I did a lot of additional tweaking to get the sound just right from startup to shutdown. I may add more tips depending on the response to this thread.BTW, for those looking to apply this to Flight1's Piper Meridian like I have, you'll need to add an entire Prop section manually, since the SOUND.CFG looks like it was taken from the 737.

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

Hi there, Is this technique applicable to all turboprops (e.g. PMDG 1900)? J.C. (MYNN)

Share this post


Link to post
Share on other sites

Listen to the sound files referred to by the COMBUSTION sections (In the King Air's case, the Kan2x.wav files.) If they sound like the props then yes, it's applicable.Some turboprops such as the Meridian, use Sound.cfgs that were meant for jets. In this case, instead of linking the COMBUSTION sections to the PROP sections at the end, you should add the prop sections at the start:This is part of the 737's sound.cfg:(SOUND_ENGINE)number_of_engines=2eng1_combustion=COMBUSTION.1.00eng2_combustion=COMBUSTION.2.00eng1_jet_whine=JET_WHINE.1.00eng2_jet_whine=JET_WHINE.2.00eng1_starter=starterAeng2_starter=starterBeng1_shutdown=shutdownAeng2_shutdown=shutdownBeng1_combustion_start=combstartAeng2_combustion_start=combstartBChange it to:(SOUND_ENGINE)number_of_engines=2eng1_prop=COMBUSTION.1.00eng2_prop=COMBUSTION.2.00eng1_jet_whine=JET_WHINE.1.00eng2_jet_whine=JET_WHINE.2.00eng1_starter=starterAeng2_starter=starterBeng1_shutdown=shutdownAeng2_shutdown=shutdownBeng1_combustion_start=combstartAeng2_combustion_start=combstartB

Share this post


Link to post
Share on other sites

Hi,Glad to read that I am not the only one to curse against all so called payware planes with so called " Realistic turoboprop sounds ".If you remember I started some threads about that some weeks ago.I was even "censured" on the PMDG forum and Flight One Forum just starting threads about that. With the help of FSSOUNDSTUDIO I could tweak myself some sounds and get the realistic low pitch and high pitch sound when setting the props levers. But being not an expert in sound tweaking I believe that there is a lot possible to come close to what a specific turboprop sound should be.DASH8 by Olek Frolov has the most realistic sound ever made for a turboprop, but Olek uses a gauges for that purpose and do not use the basic sound.cfg file for it. Roger

Share this post


Link to post
Share on other sites

I am afraid that this thread and all the others making similar claims are based on a misunderstanding. In many turboprop engines the airscrew is attached to the turbine shaft via a constant ratio step down gear box of one kind or another. When the shaft rpm changes, the airscrew rpm changes, so the pitch of the screw and the sound changes as the power lever is moved. Many aircraft have turboprop engines in which screw rpm is controlled using only the power lever. They have no rpm lever even though they have variable pitch airscrews. Other aircraft have turboprop engines in which neither turbine shaft rpm, nor airscrew rpm, ever change from take off to touch down. In these aircraft airscrew pitch is also controlled solely by movement of the power lever. These aircraft also have no rpm lever even though they have a constant speed airscrew. The sound files need to follow the screw pitch, not shaft, nor airscrew rpm, both of which are constant, if they are to work correctly. A third type of turboprop engine has a variable ratio gearbox that allows the pilot to control airscrew rpm independently from turbine rpm, using an airscrew rpm lever. The PT-6A is one such engine. In these engines the sound should alter when the pilot moves either the power lever or the airscrew rpm lever. The pitch of the screw changes whether the pilot varies screw rpm at constant power, or power at constant screw rpm. Even in this variety of turboprop engine, most of the changes in sound are the result of movement of the power lever and there are no sound files that should not respond to 'combustion', i.e. movement of the power lever. The sound needs to change when airscrew pitch changes, and when engine rpm varies, whether or not airscrew rpm changes. In real life, even when flying turboprop aircraft that have rpm levers, it is quite common for pilots to ignore them completely and operate the aircraft by reference to torque and ITT alone, as though they were flying a larger turboprop aircraft with no rpm levers. All the sound changes in flight are then due to movement of the power levers. The Cessna Caravan is commonly operated in this way, as are other aircraft with PT-6A engines, whatever the POH says.If you wish to operate a PT-6A powered aircraft 'by the book' you will still in most cases make only two movements of the rpm lever whilst airborne. From fully fine to cruise rpm at top of climb, and back to fully fine (or high rpm) for the approach. You will manage rate of climb and descent, as well as cruise speed, with the power lever, not the rpm lever. Whilst the changes proposed at the top of this thread 'work as advertised' they result in a less realistic engine sound environment. The sound needs to change with every movement of the power lever, more than it needs to change with the two potential movements of the rpm lever.Now take all the above into account and it should be easy to see why both freeware and payware producers choose to tie the engine sound environment to movement of the power lever just like Microsoft.From the point of view of writing any code for MSFS there is no such thing as 'a turboprop engine', but movement of the power lever always changes the pitch of the screw in all three main types of turboprop engine unless it has already hit one of the pitch stops. The FDE author codes the pitch stops and takes care of that aspect of sound management, but MSFS has only limited support for variable pitch stop technology that is an important feature of some turboprop airscrews.Turning to Roger's post in particular, I have no connection with the businesses he mentions, but I have no difficulty in understanding why neither of them wish to host or promulgate misinformation concerning the link between combustion and sound in their retail products. On the other hand no producer should claim levels of realism that their product does not deliver. These are however separate issues.Producers can improve upon the default method of writing sound files for 'turboprop engines', to match the characteristics of specific engines, but the default method is not 'wrong', and producers are under no obligation to do so. In common with all the Microsoft default code the sound.cfg code is less than fully accurate. It is correct as far as it goes, but the bells and whistles are missing.Payware has to be produced to be fully compliant with MSFS else consumers may have a claim against the retailer. Criticising payware producers for releasing fully compatible products may be counterproductive. If files are not fully compatible, then every time a new version of MSFS is released, the aircraft no longer works. Those who buy payware are the first to complain when they discover that there will be no update of their expensive purchase for the new version, or that they may have to wait months whilst non compliant code is rewritten into compliance. Consequently those who crave maximum authenticity will usually have to turn to freeware producers, since they can push the envelope into areas of incompatibility that would be unwise for a payware producer. Freeware is written to serve only the needs of the producer. It can be so complex and accurate that it will never appeal to the target payware consumer. To enjoy full turboprop sound authenticity in aircraft with rpm levers it is either necessary to create a plug in and interface it via FSUIPC, or else to create gauges containing potentially complex FDE extension code. Both methods risk lack of future compatibility since neither method produces a 'fully compatible product'. It does not matter how many consumers post to multiple forums the misinformation that turboprop sound should not be dependent on combustion, (i.e. movement of the power lever), because it is not going to change the commercial imperative for payware producers to produce compliant code, and it is false anyway.Fortunately all three main types of turboprop engine, in a variety of turboprop powered aircraft, with 'realistic' in flight sound output, are available as freeware for use with the FS8 flight model. At least one, Frolov's DHC-8, has already been recoded and released for use in FS9. The real problem is the number of turboprop aircraft whose engines have no rpm lever in real life, but which have been released with panels, or operating instructions, which lead consumers to believe that rpm levers should be present and should be used to manage rpm. In the real world relatively few types of turboprop aircraft are flown with airscrew rpm as a reference value, and in others where it is available and displayed it is never varied whilst airborne.FSAviator

Share this post


Link to post
Share on other sites

Hi PCAVIATOR,Lets talk a bit about practice and not theories. For having logged some hundreth of hours on complex aicrafts I can agree for 20% of what you explaining with talent and skill but maybe just with theoretical experience. Sorry being french I cant be as sharp as you are in your explaination and will probably make some language errors, but as I said you are wrong in what you try to explain.What are the most flown turboprop nowaday.Fokker 50 - I agree with you. This plane has a kind of FADEC called differently but two levers manage throttle Prop and fuel.DASH8 - 100/200/300/400. - You have four levers - The Prop Lever handles the props and handles also the fuel.SHORT 330 - 360 ( My COckpit) has six levers Throttle Prop and FuelATR 42/72 - has Four levers with separate control of turbine and Props. BEECH King air Family - Six levers>>>Even in this variety of turboprop engine, most of the changes in sound are the result of movement of the power lever >>>>Absolutely wrong!! It's NON SENSE!!>>> The sound needs to change with every movement of the power lever, more than it needs to change with the two potential movements of the rpm lever.>>>Hey PC AVIATOR. You probably have even never flown a piston constant speed aircraft then you would know that even here when you setting the manifold pressure with the throttle lever you have almost no change in sounds in the cockpit, but once on final and setting to high RPM, you will feel a earthquake under your a....Dont tell such thing, thats' simply wrong an stupid. And when tuning the torque in a turboprop, you also have no change in sound in the cockpit, believe me. >>It does not matter how many consumers post to multiple forums the misinformation >>>> Stupid and insulting>>>In the real world relatively few types of turboprop aircraft are flown with airscrew rpm as a reference value, and in others where it is available and displayed it is never varied whilst airborne.>>>>WRONG. Reference value is one thing, and resulting sound is an other thing. Anyway, all current available payware turboprop are turboprops which separate controls:FDS Cheyenne, PMDG Beech 1900, AEROSOFT ATR42/72 PSS DASH8 Q300 Anyway, the DASH8 by Frolov is far the best and most realistic plane ever produced for any desktop simulator.So and now please be so courageous and unveil your identity and let know for which company you are working. Roger

Share this post


Link to post
Share on other sites

>A third type of turboprop engine has a variable ratio gearbox that allows the pilot to control airscrew rpm independently from turbine rpm, using an airscrew rpm lever. The PT-6A is one such engine.Lets talk a bit about practice and not theories. For having logged some hundreth of hours on complex aicrafts I can agree for 20% of what you explaining with talent and skill but maybe just with theoretical experience. Sorry being french I cant be as sharp as you are in your explaination and will probably make some language errors, but as I said you are wrong in what you try to explain.What are the most flown turboprop nowaday.Fokker 50 - I agree with you. This plane has a kind of FADEC called differently but two levers manage throttle Prop and fuel.DASH8 - 100/200/300/400. - You have four levers - The Prop Lever handles the props and handles also the fuel.SHORT 330 - 360 ( My COckpit) has six levers Throttle Prop and FuelATR 42/72 - has Four levers with separate control of turbine and Props. BEECH King air Family - Six levers>>>Even in this variety of turboprop engine, most of the changes in sound are the result of movement of the power lever >>>>Absolutely wrong!! It's NON SENSE!!

Share this post


Link to post
Share on other sites

Douglas,It is very difficult to be balanced and smart when you use a language which is not your mother language.About game: Let's please some people try to make with FS and with the help of several other programs a tool and not a game. When I compare my cockpit with the simulator I used in Los Angeles when I made my IFR rating 1985, there are lightyears between both and the one was FAA approved. Here we were playing in the sandbox.Furthermore, I not a flight Enginer, I am a pilot and can in my poor english only tell what's happened in a cockpit with a given airplane and I have never heard any change in the cockpit sound level when adjusting the power lever neither with a twin prop nor with a twin turboprop. But when adjusting even for 50 Rpm, you can dramatically change the cockpit sound level and this as well in ATRS, DASHs, BEECHs FOKKER 27s, JETSTREAMS, DORNIERS, HERCULES, METROLINER, anbd so on.........Roger

Share this post


Link to post
Share on other sites

Firstly I don't mind at all if posters correct mistakes in my posts. That is a good thing. However FS9 is all about 'A Century of Flight'. It is not really about aircraft being manufactured in 2004. Turboprops have been in airline service for the last 55 years. The first two generations of turboprop engine had no rpm levers. They are typified by the Rolls Royce Dart and the Allison 501. They work in very different ways to one another, and to later generations of turboprop engine. Many aircraft were powered by these first two generations and many are still current. A problem arises whenever, based on a specific problem, posts appear in forums claiming that this or that is broken in MSFS and/or promulgate 'fixes' for things that are not broken. The problem is worse if posts imply that a proposed fix will improve all existing turboprop sound files, freeware or payware. I have explained that the fix works as advertised, but also why the proposed fix does not have generic application. The method of coding sound files described in the relevant SDK is not broken. The coded values may be incorrect in some cases. The problem is specific, not generic. << ..........when you setting the manifold pressure with the throttle lever you have almost no change in sounds in the cockpit, but once on final and setting to high RPM, you will feel a earthquake under your a....>>When creating flight dynamics for any aircraft the relevant engine and airscrew dynamics must be coded to deliver this. The sounds are driven by the flight dynamics in the same way that the gauges are driven by the flight dynamics. Even if an error in sound output is present it does not follow that the problem is in the sounds or the sound.cfg. The values being output from the engine dynamics into the sound module may simply be wrong. For instance, if the assorted 90 percentile turbine spool rpms are linked to the wrong input torque and screw rpm for the engine variant in the air file, then a perfect sound file cannot play the correct sounds for that input torque or ITT and screw rpm. Stretching or contracting an envelope defined in the sound.cfg is not going to reshape several curves defined in the air file. With enough effort it might be possible to distort the sound.cfg data so that the errors thus created within it exactly offset errors in the engine dynamics, but that is not the best design time solution. Contrary to popular belief there is no 'turboprop engine model' in MSFS. There is just a highly specific PT6A engine model. There is no reason to suppose that turboprop engines that work in completely different ways can be replicated solely by manipulation of the default PT6A code. Consequently it may be necessary to modify or add dynamic output using embedded gauge code or a plug in, but because the sound system in MSFS is not broken, everything may work without being 'fixed' if the air file delivers the input required by an SDK compliant sound file. Douglas, thank you for the detailed explanation of the inner workings of constant speed airscrews and some third and fourth generation turboprops. With hindsight I agree that the analogy I chose of single ratio and multi ratio gear boxes, in an attempt to simplify the concepts, was unwise since it was ultimately inaccurate and misleading. Having again cautioned against using the proposed method as a generic 'improvement' to all turboprop sound files I have nothing further to add. FSAviator

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this