FSXA Superchargers and engine power

[NDORFN]

I guess if you tell the sim that the BHP is X and then change the type from turbocharged to supercharged, the sim might not take account of the power cost. Be interesting to do it the other way, start as supercharged with say 20% power cost then change to turbocharged and see what happens
Roy

I think the base BHP is determined entirely and only by cylinder_displacement, compression_ratio, and number_of_cylinders. Pretty sure the max_rated_hp is a sound scalar and nothing to do with physics (I've discovered that it warps sounds because it calculates negative HP as well as positive resulting in a compressed sound scale from 0-full HP, which is why people were doubling the max_rated_hp parameter to try and stop their sound pitch from peaking below highest RPM).

When turbocharged= 1 it ignores the supercharger parameters and when supercharged= 1 it ignores the turbocharger parameters, so there's no way of applying a power cost to the turbocharger (which makes sense). What I have worked out though is that if you set it up for turbocharged and set the max_design_mp as high as you like, it still gives the same BHP output at the same intervals of manifold pressure as the supercharger does. The only difference is there's a lag function with the turbocharger, the basis of which I've yet to discover. The other thing is, if you set critical_altitude to zero, aside from the lag, it responds to altitude exactly the same way as the supercharger. So I think from that and from what I've learned from you, the critical altitude for a supercharged set up would need to be tuned in the .air file somehow which would necessitate a supercharger_power_cost.

The thing that's really messing up my understanding of the whole thing is how supercharged engine are controlled in relation to critical altitude. Is it true that throttle has to be increase gradually between sea level and critical altitude to sustain power output? Or do they sustain it without increasing throttle up to critical altitude and start falling off after that the same way a turbocharged engine does?

 

[Dutcheeseblend]

I think the base BHP is determined entirely and only by cylinder_displacement, compression_ratio, and number_of_cylinders. Pretty sure the max_rated_hp is a sound scalar and nothing to do with physics (I've discovered that it warps sounds because it calculates negative HP as well as positive resulting in a compressed sound scale from 0-full HP, which is why people were doubling the max_rated_hp parameter to try and stop their sound pitch from peaking below highest RPM).

Yes, the max_rated_hp does not influence the power calculation by FSX. At least, I have not discovered any influence yet. max_rated_rpm however does. I have set it to 2600, and while tweaking with the max_rpm_friction_scalar, I came up with this graph (NB:test on ground level, zero speed):

I made it because I discovered a certain RPM-MP relation on changing scalar. Apparently, the sim cuts off all RPM 'growth' when it passes the max_rated_rpm.
I don't know if this graph makes any sense. I was creating it while trying to get 2600 rpm without power change, instead of the 2400 rpm.

When turbocharged= 1 it ignores the supercharger parameters and when supercharged= 1 it ignores the turbocharger parameters, so there's no way of applying a power cost to the turbocharger (which makes sense). What I have worked out though is that if you set it up for turbocharged and set the max_design_mp as high as you like, it still gives the same BHP output at the same intervals of manifold pressure as the supercharger does. The only difference is there's a lag function with the turbocharger, the basis of which I've yet to discover. The other thing is, if you set critical_altitude to zero, aside from the lag, it responds to altitude exactly the same way as the supercharger. So I think from that and from what I've learned from you, the critical altitude for a supercharged set up would need to be tuned in the .air file somehow which would necessitate a supercharger_power_cost.

I believe (but I should retest it to confirm), that there is indeed a power difference between a turbocharged engine and supercharged engine with power cost. However, this powercost is useful for tweaking the max rpm.

The thing that's really messing up my understanding of the whole thing is how supercharged engine are controlled in relation to critical altitude. Is it true that throttle has to be increase gradually between sea level and critical altitude to sustain power output? Or do they sustain it without increasing throttle up to critical altitude and start falling off after that the same way a turbocharged engine does?

AFAIK, one should increase the throttle gradually. There isn't a single notion of 'critical altitude' or something alike in the T.5 documentation. There is however a notion of a limiter (previously denoted by me as an 'automatic MP control'), aka a 'wastegate'. This wastegate (an article in Flight on the Pegasus also comments on this) keeps the max pressure at a set magnitude. Now, the challenge is to get this working in FSX. Or, might there be another undocumented CFG variable on this? I think it is pretty much the same as a critical altitude.

 

[Dutcheeseblend]

Yes, the max_rated_hp does not influence the power calculation by FSX. At least, I have not discovered any influence yet. max_rated_rpm however does. I have set it to 2600, and while tweaking with the max_rpm_friction_scalar, I came up with this graph (NB:test on ground level, zero speed):
View attachment 29520I believe (but I should retest it to confirm), that there is indeed a power difference between a turbocharged engine and supercharged engine with power cost. However, this powercost is useful for tweaking the max rpm.

Ok, just tested it. With the following settings:

Test 1:
supercharged=1
supercharger_boost_low_end=1.0
supercharger_boost_high_end=2.376
supercharger_power_cost=0.1

At full throttle, we have 40.15 inHg and 771.1 SHP.

Test 2:
turbocharged=1
max_design_mp=40.15
min_design_mp=0
critical_altitude=10000.0000

At full throttle, we have 40.15 inHg and 885.9 SHP.

Test 3:
supercharged=1
supercharger_boost_low_end=1.0
supercharger_boost_high_end=2.376
supercharger_power_cost=0.0

At full throttle, we have 40.04 inHg and 898.7 SHP. To achieve the same MP, I run another test (#4).

Test 4:
supercharged=1
supercharger_boost_low_end=1.0
supercharger_boost_high_end=2.38
supercharger_power_cost=0.0

At full throttle, we have 40.15 inHg and 902.9 SHP.

Conclusion:
With zero power cost, the supercharged engine produces more power than a turbocharged one, given they produce the same MP.
Interesting is also the difference in MP between #1 and #3.

 

[Roy Holmes]

Have a look at "Aircraft_Sim_Tech_Zyskowski.pdf"
Here is an extract that explains where BHP comes from:

"Piston Engine Simulation
One of the most remarkable aspects of the SimEngine is
the high-fidelity piston simulation module employed.
This module is designed to simulate the piston engine
Otto cycle (see Figure 8).
The Otto cycles shown are:
Intake (F-E)
Compression (A-B)
Combustion (B-C)
Power (C-D)
Exhaust (D-E)
Through the use of efficiency tables, flight model
defined variables and propulsion theory, the engine
brake power can be determined.

Each component of the Otto cycle is
simulated on a per-cylinder basis, then multiplied by
the number of cylinders to obtain the total shaft torque
available.
Table look-ups are then performed for the engine
friction and mechanical efficiency for a given RPM,
which is then used to modify the shaft torque output.
Finally, the engine brake power can be found by
multiplying the engine shaft torque by the engine
omega, or engine speed, in the form of radians per second."

The engine tables are:

;Piston engine mechanical efficiency table (max 5 entries)
;IN: RPM
;OUT: Power Efficiency scalar
TOKEN_BEGIN AIR_61S_ENG_MECHANICAL_EFFICIENCY
dd 5 ; Number of Entries
REAL8 400.000000, 0.9000 ;
REAL8 750.000000, 0.67200 ;
REAL8 1725.000000, 0.672000 ;
REAL8 1840.000000, 0.640000 ;
REAL8 2300.000000, 0.640000 ;
TOKEN_END

;Piston engine friction table (max 4 entries)
;IN: RPM
;OUT: Torque due to Friction applied in resistance to RPM (ft-lbs)
TOKEN_BEGIN AIR_61S_ENGINE_FRICTION
dd 4 ; Number of Entries
REAL8 -300.000000, -41.00 ;
REAL8 300.000000, 41.00 ;
REAL8 900.000000, 40.00 ;
REAL8 2300.000000, 95.0 ;
TOKEN_END

If your engine revs higher than 2300, you probably want to reflect that in these tables.( AKA 508 and 509)

It is also worth studying the P-51 Racer, since it is the only stock aircraft that is supercharged. Here are some key features of the [piston_engine] section of its aircraft.cfg.
turbocharged= 0, but critical_altitude= 24100 . This despite the fact that the SDK explanation of critical altitude only mentions turbocharging. Also the explanation of critical altitude
"Altitude to which the turbocharger will provide max design manifold pressure (feet)" max_design_mp= 150 You could substitute "supercharger" for turbocharger.
I guess that critical alt not referring to superchargers is yet another SDK omission. BTW the max RPM in 508 and 509 for thee Racer is 2700, It's max rated RPM is 3450!

supercharged=1 ; on/off
supercharger_boost_low_end=1.00 //multiplier on manifold pressure at low RPM
supercharger_boost_high_end=7.38 //multiplier on manifold pressure at high RPM
supercharger_power_cost=0.22

BTW, the waste gate is a sort of throttle valve in the exhaust that controls the turbocharge pressure, it opens more if pressure is too high and lets more exhaust divert round the turbocharger. That is basically how over boosting is prevented.
look at (A:RECIP ENG WASTEGATE POSITION:index, percent)

What to do with the throttle in a climb depends on how it was handled by the specific aircraft. In general you should monitor/adjust the boost according to a climb schedule, but that is an unrealistic requirement for a fighter in combat, so the Spitfire had an automatic boost limiting control.
The turbocharger lag is there because you do not get more boost until you have more exhaust whereas the supercharger does not have that restriction. Cars have the same issue.

My sole experience with supercharged engines was in the Piston Provost. It had a 5 minute 3000 RPM +8 boost setting (which gave it 550 BHP on the ground and up to 5000 ft) but there was a restrictor which stopped the throttle at 4 1/2 boost setting unless you broke a tell-tale wire. The difference between the +8 and 4 1/2 settings was big. For climb we used 2900 RPM and 3 1/2 boost.
Roy

 

[Dutcheeseblend]

If your engine revs higher than 2300, you probably want to reflect that in these tables.( AKA 508 and 509)

Thank you, will have a look in these.

It is also worth studying the P-51 Racer, since it is the only stock aircraft that is supercharged. Here are some key features of the [piston_engine] section of its aircraft.cfg.
turbocharged= 0, but critical_altitude= 24100 . This despite the fact that the SDK explanation of critical altitude only mentions turbocharging. Also the explanation of critical altitude
"Altitude to which the turbocharger will provide max design manifold pressure (feet)" max_design_mp= 150 You could substitute "supercharger" for turbocharger.
I guess that critical alt not referring to superchargers is yet another SDK omission. BTW the max RPM in 508 and 509 for thee Racer is 2700, It's max rated RPM is 3450!

Hm, interesting, so you state that this variable will also apply for superchargers? We'll need to find out.

BTW, the waste gate is a sort of throttle valve in the exhaust that controls the turbocharge pressure, it opens more if pressure is too high and lets more exhaust divert round the turbocharger. That is basically how over boosting is prevented.
look at (A:RECIP ENG WASTEGATE POSITION:index, percent)

I understand. This value changes from 100% at/near critical altitude to about 70% at sealevel with a turbocharger. With a supercharger it remains constant, at the last value. So the supercharger does not have a wastegate in FSX.

I came to the conclusion that I better use the FSX turbocharger instead of the supercharger because of its automatic control/limiter. That works correct. Now the other things left are the rpm and hp stuff.

 

[NDORFN]

I came to the conclusion that I better use the FSX turbocharger instead of the supercharger because of its automatic control/limiter. That works correct. Now the other things left are the rpm and hp stuff.

Test 2:
turbocharged=1
max_design_mp=40.15
min_design_mp=0
critical_altitude=10000.0000

At full throttle, we have 40.15 inHg and 885.9 SHP.

Test 4:
supercharged=1
supercharger_boost_low_end=1.0
supercharger_boost_high_end=2.38
supercharger_power_cost=0.0

At full throttle, we have 40.15 inHg and 902.9 SHP.

What happens if you set the critical_altitude in Test 2 to 0.0?

The documentation mentions an 'automatic manifold pressure control' which is native to the Bristol Pegasus. This system limits the maximum MP to 101.9 cmHg or 40.12 inHg. Probably, this is also of importance.

Based on videos, other posts and Roy's account of the Provost, I've concluded that my aircraft uses stops to limit maximum MP at takeoff, which can be pushed past at altitude without going over max stated allowable manifold pressures, so the FSX supercharger works best.

I can't quite figure yours out. According to the Engine Rating chart you should get ≈825BHP, 2600RPM(min pitch) & 5lbs boost. Yet at 10,000ft, same 2600RPM(presumably min pitch) & same 5lb boost, yet MORE BHP!, ie. ≈910 (for 5 minutes max). I don't understand how you'd make MORE power at 10,000ft than at sea level with the same control settings. It should be less. So is it possible that the 'automatic manifold pressure control' system is related to that lever in the middle of throttle controls, and limits manifold pressure to 40.12 inches at takeoff automatically but when adjusted or switched allows natural manifold pressure to take place in flight? That would mean that it is set to START at takeoff and MAX SNELH at 10,000ft explaining the power increase. Just an idea.

 

[NDORFN]

Also... is that a bridge that joins the two throttle levers which would press against the middle lever when pushed all the way forward?

 

[Dutcheeseblend]

@NDORFN Ok, let's first write down all documentation as given by Fokker, regarding engine (control). I cite the handbook:

Engines
The engines are equipped with Claudel Hobson carburettors with automatic manifold pressure and mixture control, and Eclipse electrical starters.

Engine controls
Between the throttle levers, there is a pin lever. Before the throttles can be put in the "max. speed" or "take-off" mode, this pin lever has to be moved backwards.

Checking manifold pressure
Having selected 100% airscrew and mixture at normal, the manifold pressure should reach 94 cmHg when the throttle levers are set to "max. speed" (pin lever should be lifted). When setting the throttle levers to "take-off", the manifold pressure should reach 101.8 cmHg (pin lever should be lifted again).

Automatic manifold pressure regulator
This equipment limits the maximum obtainable manifold pressure to 101.8 cmHg, at which mode the throttle is in take-off setting. Mixture lever is set to normal and will remain here.

Throttle lever in "max. speed" results in 94 cmHg and in "cruising" 83.2 cmHg.

The latter manifold pressure is also the maximum allowable manifold pressure when using 'poor' mixture.

The manifold pressure is therefore being regularly increased by advancing the throttle lever from "idle" to fully open, and the correct manifold pressure is being obtained automatically at every random setting of the throttle lever. In "take-off" mode, an extra-rich (fuel jet?) will be automatically enabled to obtain an extra rich mixture. The construction of this automatic MP regulation is such, that in the event of failure of this regulation, still sufficient engine power remains to continue flight.

Automatic mixture regulator
The mixture lever can be set in only two modes. The first mode will result in a "normal" and the second in a "poor" mixture. In both cases, the regulation for variating altitude is fully automatic.

There is a link between the throttle lever and the mixture lever. The mixture lever can be set in the "poor" mode only when the throttle is set in cruise mode, in other words: cruising with a 'poor' mixture. Moving the throttle lever beyond this mode, will move the mixture lever automatically back to the "normal" mode.

When the throttle lever is being closed, the mixture lever will also be automatically moved to the "normal" position. The pilot can choose to select "normal" mixture in all throttle modes.

Controlling throttle and mixture levers
During take-off and climb, the throttles should be fully opened in mode "take-off", with the airscrews on take-off rpm (830 HP with 2600 RPM and 101.8 cmHg).
The first movement will automatically ensure a rich mixture.

Once an altitude of 300 m, or a time period of 3 minutes has passed, the throttle lever should be moved back to "max. speed" mode and the RPM should be regulated at 2250 RPM for routine flights or 2475 RPM for war flights (during climb 83 HP at 2475 RPM and 94 cmHg). The mixture lever will remain in "normal" mode.

Cruising with normal mixture
Once the desired cruising altitude is reached, the airscrew should controlled to 2250 RPM, while the throttle lever should be set to "cruising" (MP 83.2 cmHg). The mixture lever should remain at "normal". The cooling gills should be closed.

Here's a picture of the throttle quadrant again:

Throttles: yellow. Modes: DICHT (idle, closed), KRUISEN (cruising), MAX: SNELH: (max. speed) and START (take-off).
Mixture (denoted as H:GAS): blue. Modes: NORMAAL (normal) and ARM ('poor').

And here's a schematic drawing from the handbook (though I doubt the use of it now):

I hope I'm not disturbing to you all..! It's pretty difficult for me, but I'm very thankful for all help so far.

By the way, @Roy Holmes, regarding my other thread about automatic mixture control, I should add a provision that enables the enrichening during take-off.

Kind regards,

Daan

 

[Dutcheeseblend]

Ok, before I continue: things get complicated, see http://www.nederlandseluchtvaart.nl/forums/showthread.php?35088-Fokker-T-5-FSX&p=68349#post68349

I'm only starting to learn about these Peggies.

 

[NDORFN]

I'm still following this. Just letting my brain catch up. It's a STEEP learning curve.

Are you using AirEd at all?

 

[Roy Holmes]

Since I'm more familiar with what jets do in the sim rather than what pistons do, I consulted my guru Jerry Beckwith on that subject.
He said that the FSXA supercharger model is a single stage single speed type and that for anything more complicated it would be better to use the turbocharged model.
He also said that the critical_altitude parameter has no effect on manifold pressure when a supercharger is configured and that manifold pressure varies with altitude. I was wrong on that score.
So Daan you are on the right track for sure.
Roy

 

[Roy Holmes]

NDORFN said

According to the Engine Rating chart you should get ≈825BHP, 2600RPM(min pitch) & 5lbs boost. Yet at 10,000ft, same 2600RPM(presumably min pitch) & same 5lb boost, yet MORE BHP!, ie. ≈910 (for 5 minutes max). I don't understand how you'd make MORE power at 10,000ft than at sea level with the same control settings

Power is a function of air density. At 10,000 ft ambient temperature is lower than sea level, so for the same boost pressure the air entering the engine will be more dense.
Roy

 

[Dutcheeseblend]

Are you using AirEd at all?

No, I prefer to use AAM.

Since I'm more familiar with what jets do in the sim rather than what pistons do, I consulted my guru Jerry Beckwith on that subject.
He said that the FSXA supercharger model is a single stage single speed type and that for anything more complicated it would be better to use the turbocharged model.
He also said that the critical_altitude parameter has no effect on manifold pressure when a supercharger is configured and that manifold pressure varies with altitude. I was wrong on that score.
So Daan you are on the right track for sure.
Roy

The T.5/Pegasus throttle system is rather complicated indeed. The link I posted yesterday points out that the throttle valve of the engine is not directly controlled by the throttle lever. Instead, the system is 'geared' in some way, where the supercharger pressure co-operates with the throttle lever to control the throttle valve. That is, that's how I understood. My understanding on this has to be confirmed yet by my Dutch supercharger/Pegasus/Mercury experts

So, after Jerry's conclusion, I'm getting even more sure that a turbocharger is the way to go, despite it having a slightly laggy behaviour. But, is that a problem? No, I'd say.

Well, as said, the max_rated_rpm parameter does make sense in the engine behaviour, as it actually limits (or better: cuts off) the maximum achievable RPM. That's a good thing, so I could get to the phase of checking all RPMs with MPs.
But, when the engine is run static (no forward velocity), the maximum RPM will be approximately 2400, while it should achieve 2600. Well, guess that's not a problem either, as long as the available power is OK. When flying at ground level the RPM will get 2600 so I'll live with that.

So:
Ground level, static: 2400 RPM, 40.12 inHg, should produce 810/840 HP. I adjusted the mechanical efficiency (TBL508) at that RPM until I got some 825 HP. On a side note, I made an exception on the automixture gauge, providing 100% mix with full throttle and sim on ground.

Ground level, flying: 2600 RPM, 40.12 inHg, should produce 810/840 HP as well. I adjusted table 508 accordingly.

Now:
10000 ft, flying (eheh): 2600 RPM, 40.12 inHg, should produce 890/925 HP. This is however not the case, because the sim only produces approx 830 HP, which is an average shortage of 80 horses. Can this be solved? Not a big problem though, because the potential virtual pilots of the T.5 won't even know the real life numbers to fly on. But, if there could be a solution for this...



I keep going! There are a few things that need correction still.

 

[WarpD]

Just a side note... while reading all of this... lag in a turbocharger is actually correct behavior. It is the primary difference between a turbocharger and a supercharger. A supercharger has no lag because it is driven directly off the engine's RPMs. A turbocharger has lag because it is driven off the exhaust flow which will indeed lag behind throttle movement.

 

[Dutcheeseblend]

Just a side note... while reading all of this... lag in a turbocharger is actually correct behavior. It is the primary difference between a turbocharger and a supercharger. A supercharger has no lag because it is driven directly off the engine's RPMs. A turbocharger has lag because it is driven off the exhaust flow which will indeed lag behind throttle movement.

I know, but it isn't correct when the sim should replicate supercharger behaviour. Never mind, it works fine now.

 

[NDORFN]

I think you need to look at the gear_reduction_ratio. You mentioned it was 0.666. Is the prototype geared? A small increase in this should get your RPM from 2400-2600 for take off with only a tiny increase in BHP if you're using the turbocharged method. For me using the supercharged method, it means a huge BHP gain (obviously, going from 2000RPM to 2300RPM is a significantly more drive to the supercharger) but I used supercharger_power_cost and eventually came up with a set up at a power cost of 35%. It's starting to make sense.

If the prototype IS geared at 0.666, check the propeller diameter. It could be too big.

 

[NDORFN]

Just a side note... while reading all of this... lag in a turbocharger is actually correct behavior. It is the primary difference between a turbocharger and a supercharger. A supercharger has no lag because it is driven directly off the engine's RPMs. A turbocharger has lag because it is driven off the exhaust flow which will indeed lag behind throttle movement.

I'm beginning to understand how superchargers and turbochargers work in more detail, but should the lag on a turbo be so bad that if I push full throttle from idle, manifold pressure jumps to 75% of it's range but lags for around 5 seconds before jumping to 100%. 75% of it's range seems to it's max without turbo, so does this mean the turbo is doing nothing for 5 seconds then kicking in? 5 seconds seems like a long time for exhaust flow to build especially when the engine itself is humming at it's full unboosted power.

 

[WarpD]

The delay is all about design, and I have no idea what turbocharger they modeled.

 

[WarpD]

I know, but it isn't correct when the sim should replicate supercharger behaviour. Never mind, it works fine now.

But if you are using turbocharger, it isn't replicating supercharger... so that statement is confusing.

 

[NDORFN]

It's ALL confusing.