Short Strips / Decision Point

[OpenCirrus619]

Obviously one does the appropriate performance calculations before attempting a take off.

A long time ago I was told that, as an additional check, I should pick (and mark if necessary) a point X% along the runway. If I hadn't reached Y% of my expected flying speed by then I should abandon the takeoff. This would (should) always give me a safe distance to stop in - especially in taildraggers.

My problem is that I always seem to fly aircraft with a performance meaning that I know there is a problem if I am not airborne halfway down the runway. Combined with the fact I was given this gem "a long time ago" I cannot remember X and Y. :ugh:

Can anyone give me any suggestions?

Thanks

OC619

[Lurking123]

Something like X 30%, Y 50%.

[hugh flung_dung]

85% of lift-off speed by 40% of TODR.

HFD

[sycamore]

OC619, the figures are 71% of L/O Speed by 50% TORA. It is an acceleration `checkpoint` and derived from the equations of motion; if you can`t measure/estimate 1/2 Tora,you can time it by `backworking `the POH distances.Works fine. Authority-D.Stinton ,Ex-CAA TP.

S=u+v^2 x t
S=ut + 1/2at*
V*=U* +2aS
*=sqd.^=divided, S=dist,a=accel,u=initial velocity,v= final velocity,t=time; 60
kts =100 ft/sec.
The other 50% TORA is SDA...

[Keygrip]

I haven't put ANY thought into the mathematics - but does this also cover you for stopping distance required?

Can you abort at that point and guarantee not to overrun (mathematically)?

['India-Mike]

Sycamore - something wrong with that first equation. It's dimensionally non-homogenous so can't be right.

These kinematic equations presume uniform acceleration ie uniform or constant force. Not quite so sure how that would generalize for all aeroplanes.

I thought I knew Stinton's books quite well - can you point me to which one and which page no's for the equations? Ta

[sycamore]

I-M,K-G, et al, sorry ,corrected the second one;you can check WIKI,of course; developed,says D-S, by the RLM at Rechlin,(Farnborough `s equivalent in Germany) in WW2, when testing captured a/c,as a simplistic way of measuring T/o perf; weigh a/c,calculate wing area,therefore loading,use a conservative Cl,and calculate approximate Vs;use Vs x1.1 as Vlo,measure out strip,open up and go..!!Once airborne you can then verify/adjust speeds accordingly...
Obviously a little more to it,but that`s about it.
The `other 50%` will not necessarily ensure you stop in time,but you can hope that the brakes have better decel than the accel. You can of course do the calcs.yourself,by timing to a fixed speed,at varying weights/different surfaces/different power and check against a POH/or make your own...I would advise a suitable long runway to start...It`s how the performance charts were developed; I did see somewhere recently,on Pp.or elsewhere,a `howgozit` accelerometer,which may have been used on earlier big jets,to show if you had the correct accel..if not..Stop...
D-S book `Flying Qualities and Flight Testing the Aeroplane` isbn 0-632-05056-X..Sect.4,p247

[OpenCirrus619]

Thanks for the replies. They started me thinking and, having thought back many years, I managed to put together a couple of formula...

Assumptions:
Constant power: I know the power delivered (CS prop aside) will increase with speed.
No friction / air resistance: In real life this will reduce acceleration as speed increases.
No other factors: e.g. As weight comes onto wings reducing wheel drag vs increased aerodynamic drag


s = distance
t = time
P = power
m = mass

Energy at an instant = 1/2.m.v^2
Energy used from rest = P.t = 1/2.m.v^2
Thus: v = sqrt(2.P.t/m)

From rest: distance = Integral, between 0 and n, of sqrt(2.P.t/m) with respect to time which is (4/3.P.t^1.5)/m (time to the power of 1.5).

A graph of these 2 (normalised to 100% on each axis - 100% speed = lift off, 100% distance = end of take off run available) looks like:
http://i16.photobucket.com/albums/b18/andykmoore/SpeedDist.jpg

Combining into a Speed over distance graph gives:
http://i16.photobucket.com/albums/b18/andykmoore/SD2.jpg

Anywhere you fall below the curve you are below your target speed to achieve lift off.

I've added 3 dots:
50% speed after 30% distance
71% speed after 50% distance
85% speed after 40% distance

I believe the wrong assumption for 71/50 suggestion is that it assumes constant acceleration - whereas, in reality, we have constant power.

Anyway - that's my latest theory.

OC619

[Lurking123]

That's me bu&&ered if I ever try to fag-packet my perf. :(

[Jumbo Driver]

Fascinating calculations, OpenCirrus619, and, to my mathematically unpracticed eyes, they seem good as far as they go; however, they address only part of the question.

The crucial aspect is at what percentage of the TORA should you make your decision (based on your second graph), so that you can stop within the remaining distance? Also, does that point not vary with the power/weight ratio of the particular aircraft? Surely, any calculation must also depend on braking effect and take into account the type of surface and conditions?

With so many variables, I wonder whether you really can arrive at a rule-of-thumb answer that is safe to use across the board?


JD
:)

[hugh flung_dung]

I was "taught to teach" 85% of lift-off speed by 40% of TODR when I did my MEPL FI upgrade many years ago. The person who taught me this was a senior ex-RAF test pilot (known also to JD); I can't recall which of the performance documents it comes from but I recall stumbling-over the source again a few weeks ago. I can only think that it's in JAR-OPS because I (sadly) was looking through that recently. I would be eternally (well, until next week) obliged if someone could put me out of my misery and tell me where it is.

Of course JD is absolutely correct that there are many factors which affect the real world accelerate-stop distance, but if the manufacturer doesn't provide the data one needs something on which to hang one's hat.

HFD

[Oktas8]

This is all very interesting reading.

At the risk of being dull & boring, if you want to know whether you'll reach obstacle clearance height by the end of the runway, why not compare TORR with TODR? Most manufacturers provide both those sets of data. Or, if accel-stop is your desired option, why not compare TORR with LRR, again data provided by most manufacturers. This will get you close to the proportions required at rotate speed, which will in turn enable an estimate by (for example) 10kts less than rotate speed.

If my manufacturer did not provide that data, I would find an aircraft of similar design that did have the data and carry the data across. Absolute distances would not be the same, but proportions tend to be similar across types of similar design.

85% by 40% sounds very accurate. Sounds very RAF though (not insulting - RAF pilots being very intelligent people) - many pilots would struggle with the mental arithmetic for that. Two thirds speed by one third distance is more accessible to Joe Street.

[OpenCirrus619]

I forgot the most important bit....

Having done the maths I like hfd's 85% at 40% - so that's what I'll use.

With respect to 2/3 at 1/3 that appears to be much closer to the line (if not on it).

As far as Mass & Power go it doesn't make any difference to the shape when the graph is normalised to 100% of takeoff speed at 100% distance allowed.

Finally, the comment about obstacle clearance, is well taken - my observations above only apply to getting to a sensible speed at the end of the take-off run.

OC619

[B2N2]

With so many variables, I wonder whether you really can arrive at a rule-of-thumb answer that is safe to use across the board?

At the risk of sounding un-educated, there is a rule -of-thumb:
You should be at rotation speed by the time you reach the half way point of the runway.( which is kind-of-the same as 85% and 40% but easier identifiable)
Better yet:
T/O distance ground roll + landing distance ground roll + 20% = runway length required. Using the values published in the POH.

[Jumbo Driver]

You should be at rotation speed by the time you reach the half way point of the runway.( which is kind-of-the same as 85% and 40% but easier identifiable)

But this makes no provision for the type of surface, condition of the runway, braking effect, etc. ...


Better yet:
T/O distance ground roll + landing distance ground roll + 20% = runway length required. Using the values published in the POH.

I don't see this how this is relevant to the question that has been asked. Your answer seems to be to quite another question, i.e. "how might you calculate an accelerate-stop distance if you have performance figures in a POH but no specific graph for ASDR."

The thread question is quite different and relates to a visual "go/no-go" check during take-off of a certain speed against percentage runway distance. For aircraft using short strips, it is often the case that there is no POH performance data available anyway.


JD
:)

[bjornhall]

The thread question is quite different and relates to a visual "go/no-go" check during take-off of a certain speed against percentage runway distance.

But the go/no-go decision can be made based on such calculations, can't it?

Takeoff ground run: D1
Takeoff air distance (total takeoff distance minus ground run): D2
Stopping distance (approximately equal to landing ground run): D3
Runway length: L

This gives:
Earliest lift off distance: D1
Latest lift off distance/latest abort distance: min(L-D2,L-D3) = D4

The decision point can therefore be chosen anywhere between D1 and D4.

On a short runway, half way between D1 and D4, rounded to some convenientely observed percentage of runway length, seems to work well.

On a long runway, it might be reasonable to put the decision point not too far beyond D1 to also catch any engine power problems...

Landing decision point can be calculated similarly. Touch and go margins as well, but that tends to get a little scary... :ooh:

Takes two minutes to calculate from the POH. If there is one...

[Jumbo Driver]

bjornhall, that's all absolutely fine if you have a POH or Flight Manual to work from but many (if not most) operations from short strips do not.

As I understand OC619's original question, he is looking for a rule-of-thumb which is separate from any performance calculations (if indeed there are any!), in order to assess likely take-off performance by comparing %-age take-off speed against %-age runway used, for use on-the-day and in the general case.

Obviously, if there are performance figures available in a POH or FM, these should be used before any attempt to take-off. Clearly they could then be translated by proportioning and measuring achieved against calculated distances. However, I don't believe that is what OC619 was asking. He wants a general-case check of speed-v-distance as he hurtles along the runway in an attempt to (hopefully) leap into the air, and in a form that can be used on-the-day, as-it-happens, quite independently of any specific performance calculation.

That's how I read it, anyway ...


JD
:)

[bjornhall]

JD, see what you mean, and that makes good sense!

From those who do have a POH/AFM , I have heard many who still prefer or suggest rules of thumb instead of POH figures or, even worse, do not have a decision point at all... On the basis that it is not needed, or too cumbersome to do, or that the POH does not contain the data anyway... Therefore I wanted to illustrate that not only is it possible, but quite straightforward!

[B2N2]

But this makes no provision for the type of surface, condition of the runway, braking effect, etc.

And that is why I called it a rule-of-thumb and not exact science.
There is no formula that allows for all those variables.

You can practice on a longer runway till you have consistent results.
Problem is these results only apply to that specific runway if it is anything else but concrete.
Grass runways have too many variables to use anything else but a rule of thumb:


type of grass
length of grass
presence of moisture such as dew
moisture content of the soil
upslope/downslope
variations of all of the above over the length of the runway

and last but not least, pilot performance is not consistent either.

At some point , after practice, I was able to land and T/O this particular C172
in 1000' concrete. So a stop 'n go and be airborne prior to the 1000' point on a 5000' runway with 10 kts of headwind. That doesn't mean that I tried it at on a 1000' runway or ever have the inclination to do so.
It is perfectly safe to do so on a 5000' runway but not on a 1000' runway.

And let's not forget the human element here.
The majority of people can perform very well if there is a (perceived) large safety margin present. If that safety margin is taken away performance deteriorates. Definitely the case in my stop n go example.
Some more examples:
Most people can walk on the edge of the sidewalk. Most people can't or won't if a balance beam is now suspended 5 feet in the air. Same width.
Why is performance different? Perceived safety margin.
No difference in aviation since the human element is still present.

Back to the original question:

If I hadn't reached Y% of my expected flying speed by then I should abandon the takeoff. This would (should) always give me a safe distance to stop in - especially in taildraggers

So the question is ASDR.
If a POH is present, use the numbers with a fudge factor or don't, your choice.
If no POH is present (is that even legal?) you'll have to use your own numbers. Own numbers as in what you can do in that particular airplane, not what somebody else can do in that airplane. Also once you have experimented enough that you have consistent numbers don't assume that this now applies to any aircraft of the same make and model.
I doubt there is much difference between eg 2004 Cirrus and a 2005 Cirrus but older (30+) airplanes of the same make and model can have significant differences between them.
Significant in the sense that on a (too) short runway you will make it with one and trying the other one ends in tears.
Wonder which formula this individual used:
http://www.pprune.org/private-flying/358815-da42-twinstar-crashes-lands-end-merged.html

[Big Pistons Forever]

I found the theoretical discussion in this post very interesting but at some point it has to be brought back to a practical application in flight training.
The POH numbers are a guide and only apply to the exact conditions present during the test flight (and a new airplane and consistantly correct control inputs throughout the test). So assuming we are starting with a 30 plus yr old, hard used rental, and a pilot who may be safe but just a little bit rusty and you have allready invalidated the assumptions in the POH. Plus of course you have the imponderables like the fact the first half of the field may be nice and firm but the second half isn't as well drained and therefore is very soggy.

I tell my PPL students to calculate from T/O distance from the POH and then add a 50% additional safety factor. I also teach then to be wary of the takeoff trap of thinking that that even though the end of the runway is approaching I am "almost" at takeoff speed and so if I continue it will work out. I like the 2/3 of flying speed at 1/3 distance as a working rule of thumb as it is easy to calculate. I also make sure they pick a definite visible marker at the 1/3 point alongside or on runway as estimating the 1/3 rd point during the actual takeoff is not really practicable. Finaly I emphasize the importance that the aircraft must be accelerating throughout the takeoff. If the airspeed stops increasing than the takeoff must be aborted.