Dispatch V inflight

[Mansfield]

Safetypee, many thanks, particularly for the link. That presentation is very similar to the one I have entitled “Landing on Slippery Runways” by Paul Giesman and Jim Ratley of Boeing, but it always helps to see similar graphics in more than one presentation. Both presentations attempt to depict scaled comparisons, in which the medium braking action distance, unfactored, comes in just under the wet cert distance. This is not the case with the MD80 data that I have. However, I looked at our company data for both the 737 and 757, and it is indeed exactly that way. This might be a Boeing North/South thing, but it also occurred to me that the MD80 cannot crack the reversers until the nose is on the ground, whereas aircraft with sleeve type reversers can. Additionally, the MD80 is limited to about 1.3 EPR in reverse, lest one blank the rudder a' la LaGuardia last winter.

My operator provides us with a table containing six columns. (No AFM in the States…ever). The first two are titled “Dry” and “Wet” and contain no correction factors for inoperative reverse. According to the textual description in the performance manual, these two columns represent the cert data, and the text specifically refers to the respective factors, i.e., 1.67 and 1.92.

The next four columns are titled, “Good to medium”, “Medium”, “Medium to Poor”, and “Poor”. This is in alignment with the RCAM. The textual description for these columns claims that these data are based on the manufacturer’s estimated landing distance, plus a 15% factor, using normal reverse, Vref+5 knots and a 0.5% downslope. We also have the corrections for approach speed, winds, temperature and no reverse. The description further states that the air distance is predicated on 7 seconds at Vref. These descriptors lead me to believe that this approximates the TALPA template.

The raw data provided is very useful. The more-or-less linear progression of distance numbers across the page, from the dry column, to good, to good-to-medium, to medium, etc., suggests a linear increase that appears pretty logical. If you understand why the first two columns contain no factor for inoperative reverse, and the rest of the columns do, you’re one step ahead. However, it is only in the fine print that the rather substantial shift in safety margin is explained. Even then, the check airman who gave me my LOFT last week was under the impression that all of these numbers included the 1.67 safety factor.

The range of touchdown for narrow-body aircraft is typically stated as between 800 and 1500 feet beyond the threshold. The touchdown zone, of course, is defined (and illuminated) as the first one third of the length, up to 3000 feet. Policy is very clear that a go-around must be initiated if one has not touched down before floating across the touchdown zone. Safetypee’s two examples of how one uses the safety margin highlight how we seem to straddle these criteria. In other words, the 800-1500 foot target is designed to reserve the margin for other happenings, while the go-around policy tolerates some use of the margin in normal operations. On most 12000 foot runways, either one works seamlessly and invisibly.

Until we get to a runway limited case with less than optimum braking action. The certification margins are huge. The consensus seems to be, and is certainly recorded in much of the literature, that such margins are necessary because line pilots cannot duplicate the profile flown by the test pilot. I disagree to some extent with that notion, although not entirely. Most of the test pilots I have known are pretty good pilots, but I’m willing to bet that many of us who fly three or four legs a day every day can do some fairly precise flying as well. On my commutes to work through Washington National, for example, I see almost everyone, from the old salts to the fresh-faced kids, hit the thousand foot mark almost every time, even coming off the turn following the river visual.

Rather, the large margins in cert exist to protect against a wide range of possibilities, which is essentially an average risk approach. Looking at how the Boeing medium braking action raw data still appears to be less than the wet dispatch data, it occurs to me that this large margin, average risk approach has served a bit too well over the years, masking some nasty realities regarding actual braking coefficients and so forth.

So now we have much better data regarding the actual braking conditions. At the same time, it seems to me that we have shifted from a large margin, average risk approach to a narrow margin, specific risk approach. Through the use of approach speed additives, the RCAM, temperature adjustments, slope corrections, etc., we are tweaking each specific risk that we can identify, and therefore justifying a substantial reduction in the applied safety margin. There is likely a good argument for this change, both economically and from a purely safety point of view. However, it is a substantial change in philosophy that does not appear to be articulated at all.

As Safetypee points out, many normal landings eat into the old margin. No one notices because the margin is huge and most runways are long. But we can’t have a policy that allows you to continue trying to land as long as you are over the touchdown zone when half of that touchdown zone comprises your entire safety margin.