Announcement

Collapse
No announcement yet.

Optimising miles per litre

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • tomshep
    replied
    Ok Joan, I get it now. I've a little project I am working on which might be of interest...

    Leave a comment:


  • Joan Walsh
    replied
    Originally posted by tomshep View Post
    ... snip...

    ? (goes off to scratch head.)
    Tom, I suspect in your closing remark you have misunderstood. The speeds in the graph are IAS, not GS

    Originally posted by mikehallam View Post
    "Velly interesting ! "

    I think we four stroke flyers had better bow out of this zone as there's virtually no correlation between out sleek speedsters and your draggy bundles of plastic & cotton. ... Snip ...
    Two factors to consider there. First, I think the curves would be similar but stretched out over a higher range of airspeeds; and secondly, yes, from what I could tell from data I found, the 912 has a flatter fuel burn/rpm curve. From the two data points you gave earlier, I think the crossover point for you would be 25kt headwind; so above 25kt on the nose you'd save both fuel and time by flying faster.

    What I don't have is error factors for the figures (nor do I intend to bother working them out) so I can't be sure that the numbers are any good. The trend however seems clear - at the speeds I intend to fly, in the winds I intend to fly in, and in the aircraft I'll be bimbling in, I'll go further on limited fuel by flying slowly, just as Paul said near the start of this thread. At least now I think I understand a bit more why, and it is indeed counter-intuitive.

    Thanks everybody for an interesting discussion.
    What next?

    Leave a comment:


  • mikehallam
    replied
    "Velly interesting ! "

    I think we four stroke flyers had better bow out of this zone as there's virtually no correlation between out sleek speedsters and your draggy bundles of plastic & cotton.

    As tomshepherd points out - and thus - I made a cross ref. using his thoughts to my findings with a 912 Rans S6-116.

    e.g At 80 kt into no head wind, petrol consumption in litres per n.mile = Z. Into 20 kt mpg it increases 100/80 = 25% to 1 1/4 Z - [or if only 10 kt + 12 1/2%.]
    Simply put, does a 20 kt increase in air speed cost more or less than 25% fuel. [Similarly 10kt and 12 1/2%]

    FWIW the Rans' mpg is pretty level with my WD ground adjustable two blade prop running at what I call sensible engine rpm's, i.e. 4,600 to 5,000, For me normal is 4,800. Too low and it's a regime that Rotax don't recommend, though they claim 5,500 continuous is O.K.
    For that higher rpm yet keeping speeds within the Rans useable form factor, could possibly mean fining off the prop. even more - which I feel may increase wear and be even noisier.

    mike hallam.

    Leave a comment:


  • tomshep
    replied
    So at 35 Knot headwind, you use the same fuel at 45 as at 55 (so cane it,) but in still air, the extra ten Knots increases consumption by more than fifty per cent. That validates the fly slow theory. The results from the fuel computer in the MiniMax which has now been calibrated seem to suggest 7 litres an hour at 52 Knots 7.4 MPL and 12 at 63 or thereabouts.5.25 MPL. Broadly comparable. (447 and not quite as draggy as the Thruster.)
    24 MPG v 33. The car does three times the lower figure at 60 MPH. Fat lot I care!

    Still air in the Thruster 4.3 MPL at 55
    45 into ten Knots gives 5.6 MPL at what looks like it ought to be the same power setting.
    ? (goes off to scratch head.)
    Last edited by tomshep; 17th April 2018, 16:21.

    Leave a comment:


  • Joan Walsh
    replied
    Well, my dearly-beloved took the TST up for a jaunt yesterday and came back with three approximate data points for airspeed vs rpm. I would have liked two more, one each above and below this range but given the thermic weather he restricted the speed range to 10knots above stall and below Va. These three do the job to show the principle though.

    I fed the numbers into my fuel flow vs rpm graph which I posted here [link] and the result is in the graph attached... and what do we see? Nothing other than what Paul W originally said, we're generally better to fly slower. Only at real really strong headwinds when we'd probably be tying the windsock down to prevent it blowing away would it be worth increasing airspeed to cover distance with limited fuel.

    The attached graph shows that according our data litres per ground-mile increases with airspeed up to about 30 knots headwind. At about 35knot headwind our TST would cover the same miles per litre at any of our chosen speeds. Above that(!) the optimum speed increases as the miles per litre reduces towards nothingness.

    Warning: these data are from one flight, in one aircraft, in thermic conditions so need to be taken with a large pinch of salt. I played with the achieved airspeed used in the graphs and even a one knot difference moved the optimum point for litres per nm, but I reckon the general principle matches Paul's prognosis: optimum airspeed for fuel burn per mile is slower than you'd think. I'd add that the greater the headwind, the lower the fuel benefit of flying slowly so increasing airspeed into a particularly strong headwind wind will at least reduce the journey time if that is an important consideration, otherwise fly slower and enjoy the time in the air.

    TST headwind fuel calc.jpg

    Leave a comment:


  • Lauriehurman
    replied
    Hi Joan, I've just found this thread and have enjoyed reading it. It demonstrates very well how rubbish FB is and how good a forum can be. Thank you for starting it.

    You mentioned Competition flying. As Paul and Paul have said Flying slow is the preferred tactic for a couple of reasons that don't relate to the performance of the plane/engine.

    We don't – usually – fly long straight legs. When you fly a curved track any calculation would be continually changing.

    The slower you go the easier navigation (map reading) is.

    We don't know what the headwind is because we're not carrying a GPS.

    Laurie (2)

    Leave a comment:


  • tomshep
    replied
    Bound to be. He did the calculations and produced the graphs that demonstrated the surprisingly large amount of fuel saving that could be made by streamlining the struts on a MiniMax. I was gratified to find that my last hour in the air was made on 7.4 litres. I really am going to have to thrash it more because when the struts are streamlined, that will come down even more!

    Leave a comment:


  • Martin Watson
    replied
    Coincidentally there is a letter from Jeremy Harris in April's MF mag about specific fuel consumption and the power needed for level flight. This tends to support the idea that an empirical and pragmatic approach is simplest.

    (Incidentally I wonder if this the same Jeremy who used to post on here many moons ago?)

    Leave a comment:


  • Paul Dewhurst
    replied
    The way to calculate precisely is very similar to what Steve says - but rather than a speed/ sink polar what you need is a speed / MPG polar. Then move the axis up and down as Steve says to move the tangential point.

    The easiest way to produce this curve is to fit an accurate fuel computer - and get the fuel burn for a range of speeds.

    I used to do this when I was a keen competition pilot.

    But as Paul Welsh ( my long standing comp partner) says what it generally reveals for most microlights is that you donít want to speed up too much - unless the wind is very strong indeed.

    Indeed most of the time we cruise well above speed for best range, and if range is truly an issue you may find the calculations suggest you want to fly slower than Ďnormalí ( Joanís 45 knot floating excepted!)

    Leave a comment:


  • Martin Watson
    replied
    Thanks for those references Joan. So his rule of thumb is to speed up by a quarter of the headwind (based on a close approximation to the academically rigorous answer). That gives you a much simpler problem - all you need is to measure real fuel burn in your aircraft at various airspeeds and do the time/distances calcs for different headwinds.
    (I'm sorry to keep harping on about it, but I still think that the Rotax data is of little help to you because you need to know the power required curve for your aircraft for it to work - even just for guesstimating).

    Leave a comment:


  • Joan Walsh
    replied
    Dead right you're missing something Martin... what I think you're missing is that I'm not treating this with academic rigour, I'm experimenting and testing the result against intuition in an attempt to get a rule of thumb; thus:
    • From Rotax data, I go from rpm v power to via fuel v power to fuel v rpm.
    • I then do a simple fuel mass to volume conversion and get a fuel volume rate v rpm curve that I don't believe because it doesn't match experience.
    • Then comes the cube law prestidigitation to deliver a fuel volume rate v rpm curve that more-or-less matches both experience and figures I found on various forums via Google (other search engines being available)...
    • in other words I guessed, but I'm willing to run with it because neither my life nor my career depend on it - and it looks plausible for my purposes.

    Having derived a curve which I'm prepared to play with for fuel flow vs rpm, I now intend to record - for my specific aircraft - a relationship of level flight airspeed v rpm which I can trace across to the plausible fuel flow v rpm curve to give me fuel flow v airspeed (which I'll do for one-up flight and 2-up). I can then translate this into curves for mpg vs airspeed over a range of headwinds, and voila! an answer to the question posed in my OP. My guess is that with the 2-stroke there is no fuel benefit in speeding up into a headwind, perhaps there'll even be a penalty. Either way it would be nice to have an inkling of its magnitude when deciding whether to plug on or give up and divert downwind.

    I think what I'm trying to work out is simply: is it simply sufficient to fly at the speed implied by the standard methods using the gliding polar curve or does the engine fuel consumption v power curve distort that to give a significantly different best into wind cruise airspeed? At the moment I'm tempted by the Hallam/Uzochukwu formula "for fuel economy, cruise at the book best-glide-angle airspeed plus 50- 75% of estimated headwind component", bearing in mind the Welsh principle that we generally cruise faster than the fuel economy speed anyway so increasing speed further isn't necessary or even worthwhile.

    Or if anybody want a truly rigorous approach, I found: http://www.nar-associates.com/techni...al-flying.html especially Range for a Piston-propeller Powered Aircraft and Wind Effects on Maximum Range
    Last edited by Joan Walsh; 2nd April 2018, 17:35. Reason: Apologies to Paul Welsh for misspelling his name

    Leave a comment:


  • mikehallam
    replied
    Originally posted by Martin Watson View Post
    I think I'm missing something still though, Joan. How do you know what power the engine is delivering at the combination of rpm and airspeed that you are flying at? Without that I don't see how you can calculate fuel consumption from the Rotax data.
    Martin,

    I no longer have the 447 or the Rotax manual, but in my 4-stroke book (and its pdf is available above in one of Joan's attachments) they give several curves, WOT power but also the propeller power and below it one can see fuel used.

    As my 912's Warp Drive ground adjustable pitch prop has been changed for seeing what it does at various times from 15 tip degrees through to 13 I can't say it has ever had the same loading as Rotax'. However from my real life consumption one can work back to their power curve and I've slightly adjusted (in pencil) alongside their curve to get my best estimates. In practise the finer end gives better take off (not surprising) but cruise is just as good and consumption very slightly better, so pulling a coarse prop isn't always a good thing. One thing, my figures above are mostly from a dip stick reading and tail level affects the accuracy combined with if one uses start engine to OFF (when it has miserly consumption) to my own take off to touch down plus half the taxy and warm up time - usually 3.5 minutes).

    However as all you guys say the two-stoke is a different beast combined with its tuned exhaust system designed for a narrow optimum rpm range which BTW enables surprisingly high specific continuous power output (e.g. the 447 is something like 100 bhp per litre ~ a 1950's 'cammy Norton'.
    The expanding and contracting 'silencer' cones are at lengths from the exhaust port to assist scavenging as well as blocking new charge loss via the open port all happening with pulses at the speed of sound for that temperature and gas composition. At something approaching WOT at the correct rpm I suppose it ought to be at its most efficient, perhaps some curves can be found to relate rpm, partial power & WOT power versus lb/hp or similar ?

    mike hallam.

    P.S. this forum wouldn't let me add a J-Peg of my own prop curve but it doesn't tell much more.

    Leave a comment:


  • Martin Watson
    replied
    Originally posted by Joan Walsh View Post
    I'll not bother doing long flights and measuring fuel, I'll just find speeds vs rpm and feed that into the graphs I derived from Rotax data.
    I think I'm missing something still though, Joan. How do you know what power the engine is delivering at the combination of rpm and airspeed that you are flying at? Without that I don't see how you can calculate fuel consumption from the Rotax data.

    Leave a comment:


  • Joan Walsh
    replied
    Thanks guys, I agree too that the fuel difference over practical flying speeds isn't worth the fret with a 4-stroke engine. It may well be a different matter with 2-strokes and I'll try it behind the 503 when I'm next up - I'll not bother doing long flights and measuring fuel, I'll just find speeds vs rpm and feed that into the graphs I derived from Rotax data.

    I only wanted a rule of thumb and never expected to get this deep into an analysis, but hey as Mike said, it's been interesting. Another thing to consider would be comparison of the calculated speed to fly into a headwind vs best glide angle speed into same wind using the glide polar curve.

    Tom's right too - if you want to get absolute best speed and fuel economy out of an aircraft, redesign it FAR103 and similar designs have an airspeed limit in their requirements so drag is desirable, but that doesn't apply in the UK and SSDR owners are free to tinker if they wish.

    Meanwhile, back at the hangar;
    Headwind Rule of Thumb part 1 - with a 4-stroke engine cruise speed doesn't significantly change mpg.
    Headwind Rule of Thumb part 2 - with a 2-stroke engine... TBD

    Leave a comment:


  • tomshep
    replied
    Fuel burn is never a problem because there is always more money but range is very much an issue because a tank has a finite capacity.
    Mogas is plentiful and readily available but not on the remarkably few airstrips that a 20 litre fuel tank can get you to with a 447.
    My aircraft is currently getting a major aerodynamic cleanup because reducing drag is the best way to get a 2 stroke to burn less fuel (at 510g/KWh) Reduce the power required to maintain speed by four horsepower and that is two litres an hour. Streamlined struts, for example on a MiniMax usually get this sort of improvement when cruising at 60 MPH. A good fuel computer is the best way to set the aircraft up for the best range, which is, I think, the crux of this discussion.

    Leave a comment:

Working...
X