828a

Notso Fantastic;

I can not understand why you or anybody else would consider checking the time to 100kts as being an added work load. BOAC"s system on the Britania was for the F/E to extract the time in seconds from the Flight Manual and write it on the T/O data card which was then placed on the pedestal for all to see. The timing was done by the Captain only so that when the wheels started rolling he started the aircraft clock's sweep hand [ which was adjacent to the ASI ] and at 100kts he simply noted that the sweep hand was near the time . It was a silent operation as there were no calls from other crew members. If you happen to know of anything in aviation that requires less brain power than this then please tell me as I'd like to know about it too. 828a

6100

NotSo

You are right, it is added workload, but I know of two instances in my airline where incorrect data was fed into a computer which spat out more incorrect data, none of which was picked up by the crew. This had the result of a very heavy aircraft taking off with a significant derate in place. By the grace of god and the fat built into aircraft certification, both planes flew when they really should not have.

Had these guys had an acceleration check the problem would have been picked up early and would have been nothing more than a low speed reject with safety report to follow, as opposed to a very near major disaster.

So yes a distraction, but one that would probably save lives somewhere one day.

DOVES

Dear Sirs.
Perhaps my 19.000 Hrs, 38 years, and 12 different liner of flying gave me the right to insist on my opinion.
The CHECK "Time to 90 kts" (not to 100 as I said in a preceeding post) could have been of some help in so many occasions.
I found, among my many manuals the SE 210 Performance Chart.
I don't know if it's possible to include in this kind of mail an image file (TIF, JPEG, BMP); if yes will somebody please let me know how?
By the way: entering in it with Temp., P.A., A/C Weight, Wind Comp., Rwy Slope, Eng. bleeds, Flaps position, Rwy contamination,
it was possible to calculate the acceleration time (sec.) to 90 kts.
P.S.: At that time it was used to Take Off with the engines at full throttle, but should not be a big deal to introduce in the chart another parameter: the engines derating due to a certain assumed temperature.

square leg

Today we took off with 73.0 tonnes. The time to 100 kts was 23 seconds. It was a FLEX T/O.

I thought it was not that difficult to glance at the stopwatch at the "one-hundred" call. My eyes are usually scanning all over the cockpit screens and the runway in various patterns, depending on the actual conditions obviously.

Justathought

RatherBeFlying

Those without charts giving time to V1 / 100 kt./ 90 kt. can begin charting the weight, derate etc. against time taken on the stopwatch -- and start checking the airspeed when that time comes up.

Be careful about using the time measured at MTOW with derate when you're light as you will likely be much farther along the runway when that time comes up.

mcdhu

Interesting thread!

A low tech facet to this is the fuel flow during t/o. On the minibus, the ff on a max flex t/o at sl should be around 2x3150kg and at Toga shoul be around 2x3600kg. With that, n1, and the speed trend arrow, which should be off the top of the pfd, you have a pretty good idea about how its all going.

Cheers all,
mcdhu

None

As Jabbara and Flight Safety discussed, I like the IAS at one of the TDZ markers that are on most runways. In the US, they are the 1,000 ft "fixed distance" markers. For ICAO, it's the "Aiming Point" markers, 400m (1,312 ft).

I wonder how difficult it would be to determine minimum acceleration check speeds for these points based on the selected takeoff performance (thrust and flap settings), and assumed weather and runway conditions?

In theory, all that would need to be achieved is the minimum acceleration. If you reach that speed prior to the check distance markers, you need not pay any further attention to that aspect of the takeoff.

I like that I would be looking at two things I already look at on the takeoff roll; IAS and outside. I also like knowing the progress early in the takeoff roll. This would likely avoid an abort above 100 knots

FullWings

I've just read this thread with great interest. The whole TOPM idea is something I've been thinking about ever since I started powered aviation.

I agree with many posters about the 'scatter' in normal aircraft performance and the difficulty in separating something 'wrong' from acceptable variation.

How about this:

The FMC knows the runway, the weight and configuration of the a/c, etc. This can be passed on the the TOPM device, along with airspeed/groundspeed/temperature/pressure, whatever.

During the takeoff roll the TOPM continuously monitors the position and acceleration and gives a warning only if you are not going to get airborne and/or clear the initial obstacles. That level of performance is quite far removed from a 'standard' all-engine departure, so shouldn't lead to many false alerts.

Once the technology has been proven and accepted, we can move on to TOPM II which will warn of much smaller shortfalls/anomalies.

Most glass PFD's have a 'trend vector' on them - as someone else said, it's not exactly rocket science to have an 'expected vector' beside it. The bigger the difference, the greater the problem...

farqueue

Can I point everyone to
http://www.casa.gov.au/avreg/fsa/dow...p/fsa12-15.pdf

Summary: Wasps nested in EPR probe of 737. The nice relaxing start to the take-off got worse in a big way! Includes some discussion of monitoring time/speed to distance.

john_tullamarine

These sort of devices have been on the agenda for many decades. If I recall correctly, Grover discussed such a device of his development in his performance book.

FYI, the author of the article mentioned in the previous post is a very active PPRuNe contributor ...

OverRun

Updated link to article mentioned by farqueue. The Halifax MK thread and preliminary report has spurred interest in this topic.

http://www.casa.gov.au/fsa/2000/sep/FSA12-15.pdf

Confabulous

Hang on - isn't deceleration being checked by computer during braking? Surely all you'd need to do is modify that system to output positive as well as negative accel values to the crew - green/amber/red lights, as well as a trend vector - I believe the Rafale uses an Nx value to monitor takeoff performance instead of slam checks?

AirRabbit

White Knight wrote:
I completely agree that some kind of “acceleration check timing” should be a rather simple thing to accomplish in the cockpit. However, I think there would be some parameters that should be taken into consideration that will ultimately affect that timing; and I believe that the airplane/engine manufacturers would be able to supply that sort of data within minutes if it were to be required by the regulatory authority or if it were profitable for those manufacturers to do so. Further, I believe such timing information would be applicable with the older "steam gauge" airplanes as well as the modern FMS equipped and/or electric airplanes.

And, most respectfully, once again, I would point out that the “Palm 90” accident in Washington, DC, while there was, indeed, an acceleration issue (the crew very likely set takeoff EPR with the PT2 probes blocked, yielding an inaccurate reading) it was not the acceleration (or lack thereof) that caused that particular accident.
________
AirRabbit

sky9

The problem with stopwatch and acceleration is that it doesn’t tell you if you are on the wrong runway or started from the wrong position on the right runway. GPS is now sufficiently accurate for a computer to know what runway the aircraft is on and the distance to the end on the tarmac.
I could see no problem with having a predicted ASI and an actual ASI tape, with the actual laid over the predicted so a warning would only show if the acceleration was inadequate.
With a bit of refinement it could also be used for landing.

AirRabbit

Surely, with the technology that exists in today’s environment, there is bound to be a way by which the information being discussed here could be made available to the flight crew. Notwithstanding the accuracy of the many varied GPS receiver units out there today (and the very accurate ones are considerably more costly), there would remain a requirement to upload all of the then-present, pertinent information – things like runway slope, head/cross/tail winds, runway contaminants, etc. – all of which affect acceleration. Additionally, the system would necessarily have to take into consideration gross weight, configuration, and engine performance values – the resulting performance for which might be best supported by the airplane and engine manufacturers. My opinion is that this would present an additional burden (read, cost) to the airline. Finally, display of the results of the calculations (even along the lines of a “predicted” vs. “actual” acceleration tape or graph, as suggested by sky9) probably would not be terribly difficult in many of the modern cockpit displays, but would, I think, require some substantial retrofit for all of the earlier machines.

As I said earlier, the motivation to incorporate (let alone retrofit) such systems could be accomplished by direct regulatory requirements – but would likely incur at least some (and more likely) substantial resistance from those who would have to bear the cost. The argument would likely be, “The statistics just don’t show that this costly system would prevent any of the deaths recently occurring in aviation accidents. Why punish (insert name of objector) by requiring this additional, costly system, into an already overly busy cockpit?” The system, at least in the US, would have the regulatory authority defending their position to a congress (representing the more influential financiers of their on-going political careers) demanding to know why they should back such a costly and potentially impotent system. The bottom line would be the accuracy and viability of such an acceleration measurement system and who would bear the burden of liability in the event of a malfunction of the system or an incorrect operation or interpretation of the system display.

Solve those problems to the satisfaction of everyone involved (pilot, airline, airplane manufacturer, engine manufacturer, regulator, government, and public) and you’ll probably have a useful system. Otherwise, who knows?

My personal belief is that the best opportunity to get such a system into the cockpit, would be for the airplane and engine manufacturers to develop a relatively simple, low cost option that could be ordered when the airplane was delivered – preferably one that could be easily and cheaply retrofitted to earlier aircraft. This would take the burden off of the airplane and engine manufacturers (as they would be providing such a system). It would take the burden off of the regulatory authority and its respective government (as there would be such a system available). The liability burden would be placed somewhat on the pilot (to operate and interpret the results correctly and take appropriate action – but the pilot already has a myriad of identical responsibilities in the cockpit now – one more would hardly be prohibitive). I believe that the airline would bear the lion’s share of the responsibility to purchase and/or install such a system. If they did not, the liability incurred by the airline after an incident or accident that could be linked back to their failure to purchase/install such a system is likely to be substantial. This potential may just be enough of an “incentive” for them to “bite the bullet” and appropriate the necessary time and funds to equip the fleet -- before something happens (we would all hope).

_______
AirRabbit

sky9

I don't know the cost of a 747 accident, however there have been recent cases of aircraft departing on the wrong runway and aircraft landing late with a tailwind and contaminated runway.
Personally I suspect that such a system would pay for itself within a couple of years.

alf5071h

A significant issue with takeoff acceleration checking systems is the nature of their failure. Simple systems – accel / distance / timing may have a failure rate that would increase the number of rejected take offs, which in having their own degree of risk, may increase the overall risk of operation.

More sophisticated systems are also subject to error, notably human mistakes. A computerised acceleration check is of little use if the system depends on an accurate aircraft weight, but the crew miss set the value, as may have happened in Halifax. Of course with additional complexity weight input could be automated, but this and the associated cost has to balance the overall risk in operation.

Most operations depend on appropriate human behaviour, this cannot always be guaranteed, but with crew crosschecking, and simple gross error checking systems (time to 90kts) the risks should be constrained to manageable levels. Then there is ‘experience’ and ‘seat of the pants’ feelings which should not be discounted, but are often absent in today’s operations. Where risks cannot be contained then technology has to be used e.g. GPWS / EGPWS for CFIT.

Overall the problem is still with the human, s/he, we, all make mistakes, but equally humans can be a powerful error detection system – providing that they are fit, not tired, or fatigued, etc, etc.

Loose rivets

As mentioned early on in the thread, all those years ago, a retired air-force colonel here in Texas had proposed and patented a system that announced ‘vocally' the distance traveled down the runway. I think it was called "Tell Me." The pilot would have heard one one one one Two two two two, etc. etc.

I immediately realised that if one used the marker receiver, the kit was in place on almost all aircraft...it just needed fan typ aerials at say, one thousand foot spacing--and switched to increase in the correct direction. It was some days later that I realised that this signal could also be encoded with data for a performance computer. Needless to say, it would soon have become obsolete with the ILS DME installations just over the metaphorical horizon.

I always felt that it required a ground-based reference. So my competitors Doppler system (which was frequently dropping out) and indeed subsequent attempts with INS systems, would not have had that very solid reference to the runway that ground transmissions would offer.

I'm surprised that performance computers have not become a reality years ago. I never felt that it would cause too much mental clutter during the acceleration run, indeed watching one's speed increasing ahead of a scheduling indicator, would I fancy, be rather reassuring.

LGB

For any aircraft, assuming a constant acceleration, calculating the time to accelerate to a given speed within a certain distance is quite simple. Knowing that the acceleration is really not linear (especially with engine failed after V1), this is just a rule of thumb of course, but it beats purely relying on your sixth sense.

Note: The formula below have to be corrected for the decreased acceleration as speed, and thus drag, increases


Given: Runway length, V2

Example: 3500 meters, 180 knots

(Knots to m/s is a factor 1.944, 180 knots = 93 m/s)

Time to accelerate is

2 x 3500 / 93 = 76 seconds

Average acceleration is
93^2 / (2*3500) = 1.23 m/s^2

Knowing the time to V2 is too late in the take-off roll for the calculation to be useful for a possible reject.

In our company we use a call of "80 knots". If I want to be at V2 of 180 knots by 3500 meters, 80 knots should latest be reached about 34 seconds after brake release (for a linear acceleration). (80 knots is 41 m/s, divided by 1.23 m/s^2)


A formula to calculate time to 80 knots (or whatever speed you want) for any aircraft, assuming constant acceleration is:

(41 x 2 x runway dist) / V2^2

If you are at 180 knots in V2, the following rule applies for all runway distances:

Runway dist/100 = seconds to 80 knots, i.e. 2800 meters = 28 seconds.

A simplified formula (3% off, but a lot easier), using knots and meters directly:

The formula has V2 divided twice, as some simple calculators does not have the square function. You can even do this calculation with your mobile phone calculator! (If you do have a sqr button, divide by V2 only once, but use that button)


(Vtc = Velocity TimeCheck)

Time = Vtc x TORA x 4 / V2 / V2

Example:

Vtc = 80 knots
V2 = 140 knots
TORA = 2400 meter

Time = 80 x 2400 x 4 / 140 / 140= 39 seconds


Prefer feet instead of meters for runway length?

Make it times 1.2 instead of times 4 above

If you use the same speed every time you calculate, multiply that speed by 4 (meters) or 1.2 (feet) once and for all, to make an even more simple formula, like for time check speed of 100 knots:

400 x TORA / V2 / V2 (meters)
120 x TORA / V2 / V2 (feet)

One factor yet to be corrected is the issue with the acceleration being non-linear. If you really reached the time check speed exactly at the calculated time, you probably would not go airborne in time.

Would some of the wise guys here say if this is possible - can a rule of thumb be made that fits most aircraft? Like two thirds of the time calculated above?

I am only after a rule of thumb formula, not a precision calculation, is it possible?

FearlessFreep

Just a quick perusal of these posts I see a few things that are not mentioned with any frequency and that is runway slope and friction coefficents. Since there are many other contributing factors to takeoff performance other than the raw factors of meteorlogical conditions and the basic physics of acceleration it sounds that this "time to speed" calculation should be something that is addressed in runway analysis specific to each airport/runway. I would not think that it would be too difficult to implement those calculations into the total results.

Where I work we do not have the benefit of fancy FMS systems. Everything is done the old fashioned way - by hand!

I think that this is an excellent point and one that is not stressed enough. While rolling down the runway I endeavor to check the "acceleration rate" of the airspeed needle and the corresponding position on the runway. It's all just a feel kind of thing. It would be handy to have a timing to back things up with.

It is interesting to see that this acceleration issue was addressed in the past. More times than not we can find solutions to problems taken care of in the past.

In my own operations I will calculate the V speeds for takeoffs in the "Normal" (Reduced Thrust) category and "Max" Thrust. I include in my takeoff breifing these speed differences, that if we need to we will apply "Max" thrust if the takeoff being performed is a "Normal" one and if during the takeoff roll there is insufficent acceleration past V1.

I tend to be conservative in the utilization of "Normal" thrust in conditions that would be detrimental to our takeoff performance but do not negate the use of "Normal" thrust such as low barometric pressure, runway slope, relatively high gross weights.