• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

FSXA Table 1548 Turboprop/helicopter torque scalar

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Torque should drop off with density as altitude is increased. For example at 10,000 ft torque should be 0.735 of the sea level value.

Table 1548 inputs ambient density (slugs/ft^3) and outputs a scalar on "corrected" shaft torque. For 10K feet, 3 up from the bottom, the value of the scalar is 0.735. That would be fine except it is a scalar on "corrected" torque, not actual torque. With that value, torque in a helicopter at 10K feet is around 0.54 of sea level.

Corrected torque factor should be sqrt(sigma) not sigma. If you replace the values in 1548 with their square root the drop in torque with altitude now exactly follows the density ratio.

This is another example of where the MS engineers did not understand what "corrected" values mean when constructing the stock turbine tables.

I have been working on what these values should be in all the turbine tables and will post on the subject in the near future.

Roy
 

jx_

Messages
555
Country
unitedstates
Looks to me like it's just a scalar. All altitude drop off calculated internally. Corrected tables are not used to substitute natural atmospheric drop off.

Basically, you are getting 0.54 because you are scaling 1.0 down to 0.73 which is giving 0.73 * 0.73 = 0.54.


If 1.0 at 10,000 is 0.73 then by setting 0.73 you are getting:

0.73 * 0.73 = 0.54


If you want 0.73 at 10,000, leave it at 1.0.

1.0 * 0.73 = 0.73


By trying to input the scalar directly you are squaring everything.



At 10,000 feet, the actual corrected value must be greater than sea level torque, so that it can be scaled back down to density altitude.... the scalar is just a tweak.
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Joe,
Turboprop torque is corrected by Sqrt(sigma). As such it is not the same correction factor applied to things like thrust. Got that fact from the USN Test Pilots document
The torque readout at 10K feet should be 73%.

Having made the change I now get 73% torque at 10 K feet which agrees with the data in the HH-3 manual. Previously I was getting 0.73 times 0.73 or 0.54 at 10 K and I now get that value at just below 20K altitude.

My readout is (A:TURB ENG MAX TORQUE PERCENT:1, percent) and it is not a corrected value.
The ESP/P3D description of the table is:
;Atmospheric density on shaft torque (max 10 entries))
;IN: Ambient density (slug/ft^3)
;OUT: Scalar on corrected shaft torque
TOKEN_BEGIN AIR_80_DENSITY_ON_TP_TORQUE

The original data would have worked as a scalar on torque, but not a scalar on corrected torque. That is what I fixed and it now works as intended.

Roy
 

jx_

Messages
555
Country
unitedstates
Yes sigma for torque... I know all that, but altitude scaling is not done in the table. Set it to 1.0 and you should get 73%; the sim already scales thrust/torque/etc correctly with altitude.

Previously I was getting 0.73 times 0.73 or 0.54 at 10 K

By changing it to 0.86 you should be getting 0.86 * 0.73

If you changed it to 0.86 and it is working then something else is wrong somewhere in your process (most likely another table is compensating somewhere), because it's not mathematically possible to get 0.73 * 0.73 changed to 0.86 * 0.86 by only changing one value.

The 0.73 * 0.73 is: FS calculated adjustment (sigma/theta/whatever) * scalar.


The original data would have worked as a scalar on torque, but not a scalar on corrected torque.

To convert corrected to actual would require a 0.635 scalar if altitude wasn't already accounted for. If that was the case, a scalar on corrected torque would not be sqrt sigma, but actually sqrt sigma times sigma.

0.86 * 0.738 = 0.635

1.0 / 0.86 = 1.163 * 0.635 = 0.738



If it's working... cool... but you would need to find what is compensating, otherwise people won't be able to reproduce your results.

What value are you using instead?
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Perhaps I should have mentioned that I have also set up 1503 and 1504 to have accurate "corrected" values for altitude change and introduced intermediary IAP columns to offset the FS linear interpolation. 1506 and 1507 have little part to play in a helicopter since thrust is minimal, but they are there for completeness.

It could be that 1503 and 1504 are helping the results of 1548, but I was getting incorrect torque values even with them set properly until I changed 1548.

The values I am using for 1548 are:
Record: 1548 Turboprop torque scalars vs Rho
Points: 10
0.000225 0.000000
0.000587 0.458257569
0.000739 0.518652099
0.000891 0.583095189
0.001144 0.65345237
0.001268 0.697853853
0.001496 0.772010363
0.001756 0.85732141
0.002049 0.92141196
0.002377 1.000000

These are simply the square roots of the stock values which were sigma for each altitude, because the corrected values should be sqrt(sigma). I did not bother with the first row because it is 60K alt and I'm using them in a helicopter.

I see no reason why these values for 1548 would not work since they simply fixed incorrectly calculated values in the original 1548. When these values are "uncorrected" to give real scalars, the result is sigma for each altitude which is what I read on my torque gauge as a percent of SSL torque.
The table is working perfectly and that is all I care about.
Joe, the altitude input is the first column which is ambient density for altitudes 60, 40, 35, 30, 25, 20, 15, 10, 5 and SL from top to bottom in 1000's.

Roy
 
Last edited:

mgh

Messages
1,413
Country
unitedkingdom
To confuse the issue further, I attach a page from Issue 5 of the Rolls Royce document The Jet Engine. This goves the corrections to get from observed conditions to standard conditions so need to be inverted to go the other way. Shaft HP is proportional to Shaft Torque and Shaft RPM to the shp correction applies to torque.
 

Attachments

  • ThrustCorrection.jpg
    ThrustCorrection.jpg
    214 KB · Views: 570

jx_

Messages
555
Country
unitedstates
Roy,

Gotcha. What are your values for 1503 & 1504? In practice, the sim should output the correct baseline outputs at 1.0 scalars. Scalars should only need to be used to tweak custom characteristics. If you are only seeking reduction with sigma then you should only need 1.0 all the way across.


Mgh,

I assume in the attachment you were pointing out the portion of jet thrust added? --- if the cfg and airfile are set correctly, FS models that as well. All the other stuff is proportional as you pointed out, so easy to deal with as long as the correction is 'decoded' properly.

Also, I tried to read that post, but it is long and unfocused. Was there a particular part you wanted to point out?



I think where it gets complicated is most FS users don't understand how to use normalized performance data or how a corrected table look-up works, but as long as you understand how to back track RPM, CN, and thrust, at sea level by mach, it should be pretty straightforward.

Using the corrected method, you should never have to set anything more than sea level (or whatever normalization is based on) at all speeds, and any custom characteristics.
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
A few points.
1. My original and subsequent posts concern the way FSXA calculates torque decrease with air density using Table 1548/AIR_80_DENSITY_ON_TP_TORQUE. So far as that table/Token are concerned what I have discussed has proven to accurately simulate HH-3 (Seaking) altitude performance with two and with 1 engine in operation for a FSXA helicopter based on the EH-101. My findings are probably relevent to FSXA turboprop performance since they both use 1548, however turboprops use the propellor tables and helicopters do not. I have not tested turboprop performance using the modified 1548, though I see no reason why it should not work as well as it does with helos in FSXA or FSX, but I can not confirm how it performs in FS9. I should have stated that as an assumption in the original post.

2. There are several ways to calculate "corrected" performance. Most use ambient temperature and pressure ratios compared to SSL temperature and pressure. 1548 uses density ratios, so as far as I am concerned the use of temperature and pressure ratios, though absolutely relevent for most of the turbine tables, is not relevent for discussion about 1548, though it could be interesting.

3. My reference documents are the USN Test pilots Manuals. For fixed wing this is FTM 108.pdf and for heliopters it is FTM 106.pdf. They differ slightly in how corrected SHP or torque is calculated, but FSXA does it more like the 108 method and it works when using the 1548 table I described.

Roy
 

jx_

Messages
555
Country
unitedstates
jx_

The equations that relate observed and corrected thrusts and shps.


Those are the same standard ones we've always used.

The thrust observed one is the inversion of cF = Thrust / Delta
The sfp observed one is the inversion of cSHP = SHP / Delta / sqrt Theta

(For those who may not know, Delta is just ambient versus sea level pressure ratio & Theta is ambient vs sea level temperature ratio.)


You've probably seen them before, they just flipped the denominator and inverted the signs, that's all.


The TEHP is probably good to have.
 

jx_

Messages
555
Country
unitedstates
Roy,

I don't know what you mean. I wasn't suggesting that the table used anything other than sigma, or the like. Just simply stating that FS already does the conversion correctly... If you are getting 0.738 at 10,000 with less than a 1.0 scalar, something elsewhere is compensating --- up scaling.

A 1.0 scalar at 10,000 should produce 0.738 internally.
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Joe,

The table inputs ambient density in slugs/ft^3 and, as originally written, output ambient density ratio. FS then uncorrected that as ambient density ratio squared. So at 10K feet it gave 0.54, not 0.73. By changing the corrected torque scalar output to sqrt(.73), the uncorrected value became 0.73, which is the correct result.

The method used to uncorrect the value was correct, problem was that the value had not been properly corrected in the first place.

Simple as that.

Roy
 

jx_

Messages
555
Country
unitedstates
ok Roy, you're not getting what I am saying so never mind.

Glad it works for you.
 

mgh

Messages
1,413
Country
unitedkingdom
I've only just got round to reading FTM 106.

Equation 4.13 gives ESHPcorr = ESHP / (delta * sqrt(theta))

Equation 4.16gives N1corr - N1 / sqrt(theta)

Shaft torque is given by T = ESHP / N1 and Tcorr = ESPcorr / N1corr

Tcorr = T / delta

This suggests that torque varies with pressure ratio.
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Torque will vary with pressure ratio if the temperature ratio is also taken into consideration. For the same pressure ratio with colder air you get more torque, so it is not just pressure ratio that dictates torque. When you combine pressure and temperature you are looking at air density, I believe.

In the percent torque charts for the HH-3 series helos you input temperature and altitude to get percent torque.

The relationship between ESHP and torque is aircraft dependent

In FS for a helo CN1 varies from N1 by 1/sqrt(theta) and that is how 1503 should be set up for each IAP column. The same applies to 1504 though Mach is low and has a small effect (if the table is set up correctly) 1506 and 1507 have a trivial effect because thrust is usually small.

That leaves 1548 which inputs air density and outputs a scalar for Corrected torque VS density. That scalar when "uncorrected" gives a gauge reading that is squared in FS.

Most "corrected" or "referred" methods use temperature and pressure. They generally uncorrect by pressure ratio. However corrected methods are a convenient way to calculate performance over a wide range of altitude, mach, engine speed etc and they have nothing to do with how a real aircraft actually performs. They are simply, for the test pilot/engineer, a way of generalizing performance from a finite number of test point results.

FS uses corrected methods to establish simulated performance since it is a quick and easy way to compute it. The turbine tables allow one to make adjustments to corrected performance, but they only give the right answer when uncorrected by FS if they have the right numbers in them. One gets the numbers right if they are calculated using equations like you quoted from FTM106.

In the case of 1548, the stock tables input ambient density in slugs/ft ^3 and output ambient density ratio as a scalar on corrected torque. The program then squares these values so it did the right operation on the wrong numbers.The scalars should have been sqrt(density ratio) in the case of this table. Remember the scalars are applied to corrected torque, not real torque, FS uncorrects the corrected value to give real values.

What actually happens in the real world does not always apply when using corrected values like FS does. The equations are not the same in general.

Roy
 
Last edited:

mgh

Messages
1,413
Country
unitedkingdom
Torque should drop off with density as altitude is increased. For example at 10,000 ft torque should be 0.735 of the sea level value.

I must have misunderstood that first line of your initial post, which I thought was the premise for the following discussion.
 

jx_

Messages
555
Country
unitedstates
s = d / t

They are all interdependent and every equation can be re-written to rely on different variables that all lead to the same result, but the technically correct answer is as Roy stated, torque is dependent on density.

Those formulas you posted already account for pressure and temp to reconcile density...


T/T0 = p/p0


PA/PA0 = p/p0 ^M

The sim follows these rules as well.
 

mgh

Messages
1,413
Country
unitedkingdom
The performance of jet engines depends on pressure (P) and temperature (T) separately - not in a 1:1 ratio as in density.

The equation of state gives density (rho) as

rho = P / (T * R)

where R is the Specific Gas Constant which equals 287 J /kg deg K for air.

This means that a 10% increase in pressure and a 10% increase in temperature give no change in density.

Referring back to the formulae I quoted, FTM 106 Equation 4.13 relates changes in pressure (delta) and temperature (theta) to changes in engine shaft horse power (ESHP)

ESHPcorr = ESHP / (delta * sqrt(theta))

A 10% increase in both temperature and pressure results in a 1.1 * s qrt(1.1) = 1.15 or 15% change in ESHP even though there's no change in density. This variation depends on pressure and temperature separately, not on density. As I showed previously shaft torque depends on pressure.

BTW it's usual to explain what the variables are when presenting formulae.
 
Last edited:
Top