Continental O-200

[bomber]

So, a test bed is a nice idea, but since you can't test without propeller in JSB, not feasible in FG. A combination of engine and propeller is always hard connected here for a given plane.

well that's simply not true I'm afraid... you use a prony brake within FG and as there are issues with just adjusting the pressure and temp you have to 'float' the plane up to various alts to complete the static test.

Simon

[bomber]

There are two things to keep in mind:

1.) The engine simulation, while in many aspects more accurate than in other FDMs, it is not perfect in JSB either. Your fuel consumption for exmaple can be calculated either simply via a set BSFC or by engine specifics (especially volume, horsepower, rpm). Now, since you give both, your fuel consumption will be higher than expected because well, the simulator does some things double. Same with sparkfaildrop. you add sparkfaildrop and at the same time BSFC and you added a fuel loss to the bsfc while the manufacturer data usually already include that factor. So your numbers will look in the end differently because if you want to set those parameters all, you have to use something like BSFC-faildrop-(additional_loss_at_Engine_temp)) and this is already simplified.

I find this quite illuminating and you've completly lost me... could you elaborate.. are you saying I don't need BSFC ?

also sparkfail drop is when a single spark plug fails... the examples given usually give a figure of 10% power reduction

Simon

[jwocky]

Okay, the BSFC you find in the manufactuer's documentation already includes this 10% loss
now you set sparkfaildrop to get 10% loss ... which is actually ANOTHER 10% loss

so either you have to reduce the BSFC accordingly or let sparkfaildrop out because it's already in BSFC included.

The other thing is, if you provide BSFC, you override the fuel consumption calculation in JSB which, without BSFC would be based on the engine data like bore, volume, etc. However, you need this data if you plan on playing with a cooling system later.
So the trick is here, to provide BSFC and let sparkfaildrop out (because the manufacturer has it already in his data included), but to provide cylinders, bore, volume ... for the cooling system.

The other thing, I found is, those tables, aerodynamics creates are generally on the low side. Okay, I personally have a tendency to tune them up on the high end then, because of the discussion we had about reality vs. fun, but somewhere in between is a sweet spot.

About prony brake, you talk about a standardized test propeller? That would be doable, but for the purpose of testing different engines for different propeller planes, it would be hard to standardize it. Not even talking about the problem of different pressures yet unless we build the test bed as pressure chamber to simulate different altitudes. I have no idea how, yet, but that is maybe because I'm operating on a low caffeine level yet.

[bomber]

the prony brakes doable here ya are... its not mine it's been around for years.

Code: Select all

<?xml version="1.0"?>
   
   
<!-- Generated by Ron Jensen

With a constant speed propeller you can achieve the same RPM with a variety of horsepower.
Use a prony brake as test load.
Set the advance-cmd-norm to 1/10000 of the RPM you want to achieve
for 2100 rpm set 0.21).
You can then sweep the throttle to see what effect throttling has on the model.
Or you can hold the throttle constant and sweep the RPM via the advance-cmd-norm and
see what effect that has on power.   
Finally, you can look at the prony fdm's 'propeller-pitch' to see what coefficient is
being used for a given power setting.   
-->
   
   
<propeller name="Prony">
   
  <ixx> 2.4 </ixx>
   
  <diameter unit="IN"> 72.0 </diameter>
   
  <numblades>  2 </numblades>
   
  <minpitch>   0 </minpitch>
   
  <maxpitch>  90 </maxpitch>
   
  <minrpm>     0 </minrpm>
   
  <maxrpm> 10000 </maxrpm>

<ct_factor>0.0</ct_factor>
   
<cp_factor>1.0</cp_factor>

  <table name="C_THRUST" type="internal">
   
    <tableData>
   
         0    10   20   30   40   50   60   70   80   90
   
    0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0
   
    1.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0
   
    </tableData>
   
  </table>
   
  <table name="C_POWER" type = "internal">
   
    <tableData>
   
          0        10      20      30      40      50      60      70      80      90
   
     0.0  0.00002  1.0000  2.0000  3.0000  4.0000  5.0000  6.0000  7.0000  8.0000  9.0000
   
     1.0  0.00002  1.0000  2.0000  3.0000  4.0000  5.0000  6.0000  7.0000  8.0000  9.0000
   
    </tableData>
   
  </table>
   
   
</propeller>

[bomber]

Okay, the BSFC you find in the manufactuer's documentation already includes this 10% loss
now you set sparkfaildrop to get 10% loss ... which is actually ANOTHER 10% loss

so either you have to reduce the BSFC accordingly or let sparkfaildrop out because it's already in BSFC included.

The other thing is, if you provide BSFC, you override the fuel consumption calculation in JSB which, without BSFC would be based on the engine data like bore, volume, etc. However, you need this data if you plan on playing with a cooling system later.
So the trick is here, to provide BSFC and let sparkfaildrop out (because the manufacturer has it already in his data included), but to provide cylinders, bore, volume ... for the cooling system.

but without the sparkfaildrop how can you simulate the power when only a single spark plug is working ?

[jwocky]

You can use both, but then you have to reduce BSFC by the 10% you lose with sparkfaildrop.

[bomber]

well thats not a problem as the BSFC needed to make the engine perform to spec is way off compared to manufacturers data......

0.554 manufacturers data compared to 0.265 to perform as expected.

Simon

[jwocky]

Yeah, a part of the difference is that sparkfaildrop is basically already in the manufacturer's data

When I have a problem with fuel consumption, I have a list of possible suspects, not all are actually engine related, but that is what I look at:

- bsfc/Sparkfaildrop (yeah, I fell for that too occasionally)

- propeller pitch, propeller gear ration
Propeller pitch means usually, you need a lot of rpm to hold the speed, gear ratio shows often in too high propeller rpm and actually less throttle used by the AP to compensate for thie more in rpm (you still use a lot of fuel because the engine has to work hard to spin the propeller at those rpms)

- manifold pressure (especially if you have turbos)

- plane balance (pressure point/CoG)
That needs a little bit of explanation because it works around two corners: A plane that in cruise altitude sits always nose up has a higher drag. So, if you are on autopilot, the AP compensates for it with more throttle, more throttle, diminished by mixture setting, means more fuel consumption. Now, in a plane like the Pup, where the heavy motor sits in front, a more or less heavy pilot sits still relative far forward and everything that comes after that is build light, you have more of a dart behaviour that is compensated by the vertical stabilizer. The lesser you have to compensate with the elevators, the less drag. The trick is, that those small planes have often a much bigger relation between vertical stabilizer area and wing area or a thicker stabi to produce a little overproportional lift. The other way is to set the CoG accordingly, but that sucks a little because the CoG changes over the time of a flight when fuel is burnt.

- drag per se
A lot of planes in FG get a lot of drag. Usually, it brings a better fuel consumption to have less drag on the regular hull and wings and more on flaps and speed brakes (if there are some of those)

- lift per se
The usual aeromatics lift calculation is a bit on the weak side in many cases. It calculates with kind of a standard wing profile, but in modern planes, there are sometimes more optimized profiles used which prude more lift out of smaller area (because the wing profile is thicker at the thickest point, because manipulated wing tips prevent disruptions of the stream profile). Too little lift leads to too high pitch, which leads to more fuel consumption at cruise.

- and in my case the usual typo and I have not seen it list
Drag coefficient for retracted wheels sits on 9.0 instead of 0.0. same for flaps, spoilers, speed brakes
Lift ... same thing only vice versa.
Flaps 0 became accidentally flaps 1

Bottom line is, if everything is right, the BSFC is usually (with use of sparkfaildrop) somewhere around BSFC from the manufacturer *0.88 give or take, in my experience. With such a big difference, I suspect fligh attitude or something with the propeller.

[bomber]

The engine looked good so far, so I was searching the reasons for the too high fuel consumption in propeller and or airframe respectively the balance.

Why do you think the fuel consumption is too high ?

Because I had to reduce the BSFC to get it to the correct value... ?

Do you suspect that will have an effect on something else in the engine code ?

remember that with a prony brake on there is no propellor and it's strapped to a test bed so no height either (sea level)

[bomber]

https://www.flightglobal.com/FlightPDFA ... 200659.PDF

reading this i've spotted...

best cruise Tas/fuel consumption,

These two are both full rich mixture

95kt/5.4 Imp gal/hr (2,500 r.p.m. 72.5 percent power at 2,000ft),
82kt/4.2 Imp gal/hr (2,300 r.p.m. 55 per cent power at 4,000ft);

This figure is for leaned mixture (above 4500ft)

93kt/4.9 Imp gal/hr (2,500 r.p.m. 65 per cent power at 6,000ft),

so we know how some hard facts about this engine... the only problem being that changing the RPM changes the Volumetric efficiency of an engine.