Friday, March 13, 2015

The F-35 and the Infamous Transonic Acceleration Change

Part 3 : Updated and Bumped


(Part 1 here: Part 2 here, and 'Bonus' Block 60 Comparison Here)

I'm still working on the F-35A vs Block 60 F-16 comparison, but wanted to add a reference to a comment I made at Breaking Defense in response to one 'Peter Goon': a former FTE (best known as a talking head for Air Power Australia ). I'm repeating it here because for some reason, the graphics accompanying my response did not post, and one is an update of a chart shown already in this series.  Mr Goon dropped a series of snarky comments, any of which can be discarded, but none so readily as this one:
RE: Q: How does an aircraft that takes over 60 seconds to accelerate from 0.8M to 1.2M at 30k ft ISA have "at least" the maneouvrability of a Block 50 F-16C which does this in less than 30 seconds, with 4 missiles on board?
My response reads thusly:
Well Skippy, your first big problem is that the only F-16C Block 50 that can go from M.8 to M1.2 in "30 seconds" can only carry two missiles and....its about out of fuel.
Your second big problem is that you are comparing your infeasible F-16 loadout with an F-35 KPP configuration that is carrying 60% (of a much larger capacity than the F-16) fuel load AND 2 tons of bombs.
To answer your question more fully, and as a object lesson to others that it is YOU who are really pedaling the "indifference to reality" with hyperbolic nonsense, the F-35A at KPP weights and loadouts SPANKS the Block 50 F-16 in transonic acceleration for all but at the very lightest weight/drag combinations identified in the F-16's operating manual.
I dropped two graphics with data extracted from the Hellenic AF Block 50/52 Change 8 Operators manual, but they didn't 'take' for some reason. First, the F-16 Block 50 basic weight/drag data:
This is the heaviest a Block 50 F-16 can get without adding drag by hanging stuff off the outside 
...and then an updated table (data and comment added for clarity only: no existing data changed) of the Block 50's transonic acceleration times:
Only at the lowest weight/drag combinations can a Block 50 transonic performance best the F-35A KPP configuration. Remember the F-35A is carrying 60% internal fuel AND two 2K Lb JDAMs. 
Conclusion?

YES, the F-35A has:
  "at least" the maneouvrability of a Block 50 F-16C. 
Deal with it.


Original Post, as posted on 1 March below this point...... 

KPP Requirement or no, ‘What If’ the F-35 will someday really need to get rid of all or part of extra seconds in the new KPP enroute to Mach 1.2?

(Part 1 here: Part 2 here, and 'Bonus' Block 60 Comparison Here)
The F-35 Transonic Acceleration KPPs were ‘changed’. This change was almost immediately decried far and wide as a failure by the ill-informed and self-serving critics. I’m certain some have forgotten or missed the discussion in Part 1 where I referenced the DoD documentation describing the process by which KPPs are established, and changes are made, and that they are only changed if the changes are validated as being operationally acceptable. I’m certain others will miss the references in Part 2 where it was highlighted that they changes were always going to happen, if only because you cannot violate the laws of physics. (Those people may be sputtering at their computer screen soon if they aren't already). That the F-35 KPP changes were implemented at all proves the changes were acceptable by those responsible for the integrity of the requirements and supporting system engineering process.

This post and series is not for those people.

It has been for people who want to understand more about the changes and their ramifications. Agreement or approval by me or anyone else as to the acceptability of these KPP changes is irrelevant. The agencies to which they are relevant have already decided.(Right about here those ‘other people’ would mentally register a need to utter a Circumstantial Ad Hominem against those responsible that when you strip away all the unsupported assertions, logically translates into little more than “those people are the experts, what do they know?...oh and “they must be lying!” or “what are they hiding?”)

BUT!... probably, someday, someone flying one of the F-35 variants is going to REALLY need to accelerate M.8 to M1.2 as quick as the original KPP specified or even more quickly (just not often enough to pursue it as a KPP). Truth be told, there’s probably going to be a time or twelve where somebody is going to wish ‘instantaneous’ translation to Mach 10, but you have to draw the line someplace within the laws of physics.

IF there is some operational need to shave all or part of 8 seconds (or the longer times for the other variants) going from .8M to 1.2M, can it be accomplished? If so, HOW would it be done, and is the remedy needed so onerous that it will adversely impact the F-35’s viability or operational utility? With the F-35A needing only to overcome 8 seconds, the program office could have easily just specified a lower weight to accomplish the feat, and a checklist item added to the effect of “if you want to do ‘this’, then ensure fuel and weapons on board do not exceed X lbs”. But as fuel on board at a mid-mission point tends to be a valuable commodity, and weapons carriage is a fighter aircraft’s reason for being, I presume everyone can see why the users, JROC and program office didn’t take that route.

“One Weird Trick”

If, by chance, “8 seconds” is a big deal, even for an aircraft with a ‘greater initial acceleration than an F-16’ and it still means those 8 seconds crossed some breakpoint minimum needed operationally, then the good news is it is rather easy to accommodate. In fact, fighter pilots have been using ‘this one weird trick’ (Man, I hate those ads) for decades to squeeze extra acceleration out of the transonic region.

It’s called ‘Unloading’

Hat tip to Pat ‘Gums’ McAdoo, USAF Lt Col (Ret), at F-16.net for first reminding me of this ‘way back when’:
“RE: transonic accel....... I'll bet that the profile was st-and-level and then gofor it. We who have done it know that all ya gotta do is reduce AoA by lowering the nose a tiny amount and shazaam!”
You can also bet any future opponent that the F-35 runs into WILL also be doing the same thing. Sometimes ‘unloading’ will be the best thing for a pilot to employ, sometimes something else will be the best thing to do. Aeronautics has advanced to exploit the medium to its maximum. There is no magic airplane that does everything better than any other airplane within each generation or two: there's always a 'catch'.

Unloading the wing during an acceleration run through the transonic region pays off in increased acceleration (or reduces the decline of acceleration if that is all that is desired) immediately. It is also the key part of employing the Rutowski Profile (Ref #10 in Part 2) in reducing climb times for the same reasons. When a pilot completely 'unloads' the wing, all of the wing wave drag coefficient contribuion due to lift goes to zero, and the aircraft ‘sinks’ (altitude decreases). This is a relatively slower rate of descent compared to an actual ‘dive’, by the way. But a dive is also an option for even greater acceleration, and as Shaw observes (Fighter Combat, p. 407) it is even more effective the heavier and ‘cleaner’ an aircraft is-- assuming the dive does not take the aircraft past its structural limits. When that much wave drag suddenly disappears, the acceleration rate increases dramatically (Shaw recommends a “sharp pull-down” to a dive attitude then an ‘unload’ as particularly effective). Higher acceleration rates are like compound interest: the more you have earlier, the more the distance you cover and the more speed you have at the end: If the wing is unloaded early in a run when acceleration rate is already high but late enough that the aircraft is near Critical Mach, the time the plane would need to be unloaded might be exceedingly small for an F-35A to achieve 1.2M those eight seconds faster. It might help people to be able to visualize how little difference there is between M.8 and M1.2 at 30K feet if we use a graphic (we're keeping the numbering of figures as beginning in Part 1):
Figure 20: Every tenth of a Mach is about 40KIAS at 30K ft Altitude
An eight second difference for the F-35A in all probability translated into achieving something like Mach 1,1-Mach 1.18 in the original 55 seconds. That's small potatoes as far as overall speed goes. 

Here’s the example of a straight wing drag profile shown earlier:


Figure 21: Typical Drag Curve for a 'Loaded' Straight Wing 
Now, here’s a representation of what the ‘straight wing’ drag curve looks like if unloading the wing reduces the wing wave contribution by a conservative 40%:

Figure 22: Typical Drag Curve for an 'Unloaded' Straight Wing 
The total peak drag coefficient in this instance is reduced by 25%. Observe in the next figure that the peak drag coefficient ‘unloaded’ at around M1.1 is about the same as just before Mach 1 ‘loaded’, and that even in the typical Critical Mach range for fighter aircraft between M.8 to M.9 (Shaw, “Fighter Combat”, Pg. 399) the drag coefficient is reduced. 


Figure 23: Straight Wing Drag, Loaded vs Unloaded
IMHO, the F-35 Critical Mach Speed, like most advanced fighters, should be well into the upper half of that range IF the performance measures in the Bowman paper are to be believed (We’ll get to that in a moment). 

The actual amount of drag reduction will vary depending upon the total aircraft design (and remember it will also vary by airspeed and altitude). Shaw’s “Fighter Combat” provides a convenient graph highlighting the increases in acceleration that can be expected at different altitudes over a wide speed range for legacy aircraft. It is modified here to illustrate the altitude and Mach range of interest:

Figure 24: Typical Benefits of Unloading on Acceleration (Original source as noted, modified by Elements of Power)
We see in Shaw’s chart that for legacy fighter designs flying from .8 Mach to Mach 1.2, unloading increases the acceleration rates between ~17% at M1.0 to as much as ~20% at Mach 1.2 and ~25% at .8 Mach at the 30K ft KPP altitude (Ref #9). This appears consistent with what we would expect from the differences in the drag curves shown in our examples. It should also be readily apparent that both the F-35B and F-35C could employ the ‘unload’ technique in the same manner as the F-35A. Obviously, the F-35B would only need a small ‘bump’ compared to the F-35C. The smaller the time/speed difference needing to be closed, the more flexibility on timing and duration is available for the pilot to choose when and how long to ‘unload’.
Figure 25: Unloading and/or Diving to Achieve Original KPP times.
The F-35C’s obviously much greater wave drag due to lift than the other two variants (expanded upon in Part 2) also means it would just as obviously benefit MOST OF ALL from the unloading technique. If a dive would be necessary for the F-35C (as the ‘worst case’) to meet the old transonic acceleration standard, it could be able to carry that acceleration beyond 1.2 to perhaps even M1.6  (F-35 Max Speed in level flight. We do not know what the 'never exceed' speed is, so we’ll stick with what we know the F-35C can do. If ending the acceleration run at the same altitude (30K feet in the KPP) is important, the F-35 pilot can accelerate beyond M1.2 and so after climbing at the end of the run, trading smash for altitude, the plane is still flying M1.2 or better.

This might be a good time to interject the observation that if there was ever evidence the original KPPs were established before the aircraft design was even undertaken, it is the original common KPP of 65 seconds for the F-35B and F-35C: two airplanes with vastly different weights and wing areas. This tells me the trade space for the F-35C design might not have been fully understood when the Navy laid down the requirements or the KPP really was a 'let's see what we can get' figure.
We should also note here that the KPPs were/are against a predicted weight standard and that all indicators are that the predicted weights are being beaten, with all variants coming in (for now) below the modeled weights used for the KPPs.  Those lower weights mean lower wave drag to begin with, and in turn better transonic acceleration.   
Shaw asserts that unloading for best transonic acceleration is done close to ‘Critical Mach’. Bowman (Part 2 Ref #9) identifies the F-35A KPP threshold for max speed at 30K ft altitude without afterburner at >.96 Mach. This speed cannot be too far above or below F-35A Critical Mach because “at speeds faster than the Critical Mach number the drag coefficient increases suddenly, causing dramatically increased drag”. Afterburners (and/or unloading) would soon be required to go any faster once Critical Mach is reached.
Bowman BTW, is now a Navy Captain, and as far as I know continues to NOT be an F-35 'fan'. I'm still hoping for some karma backlash that will tie him to the F-35 program as a developer or operator, just for the entertainment value alone.   

Conclusions?

Other than changing the KPPS evidently was not a 'big deal', there's not much to conclude: just a lot of open questions remain.
That the program was warning far ahead of the KPP change that the spec was unrealistic (Ref #3 and #4 in Part 2), makes the initial acceleration KPP looks more like it was a ‘show us what you can do and we’ll revisit the issue later’ placeholder. An entirely likely possibility is that as the program matured, different advantages were seen in areas of aircraft capabilities that shifted design emphasis to 'someplace else'. A KPP may be set based upon an assumption of X capability would require KPP X to be one thing, and then as the design matures, and operational concepts evolve or even threat perceptions shift, it becomes apparent that X capability isn't as important as Y capability, which if is addressed would require more or less out of the KPP X.  

Examples? Program news over the years seemed to talk a lot about range and carrier approach handling as being 'big deals'. Did range become more important than acceleration? If so, fuel could have been added which increased weight and that reduced acceleration. For the C model, did the Navy decide a bigger, higher drag wing meant fewer carrier landing mishaps? It is also entirely possible one or more of the variant KPPs started simply as a wish-list item.

We've heard before about he F-35 KPPs being based upon aerodynamically 'clean'  legacy aircraft specs. Was that a 'miss' where the initial KPP-setters failed to recognize the physics involved, or they did recognize it but since there was no design to evaluate at the time, they let it go: leaving it as an admirable goal but also knowing the KPP could always be changed later? That last explanation makes most sense to me. I've dealt with the acquisition system bureaucracies for decades, and that kind of development  seems totally in character for how the system works. But whatever the reason, we can only 'guess'. We don’t really know 'why', and I wouldn't be surprised if no one is still around within the program who remembers the 'why'. It would make a terrific question for some politico to enquire about, especially if there was some dark secret behind the change as F-35 detractors seem to often insinuate.  

The Bottom Line, Again

For the F-35A model, the 8 second difference between new and old KPP appears to be trivial. The F-35B would probably have to unload a few seconds earlier or for a few seconds longer to meet the old KPP and the F-35C may have to unload a lot earlier and for a few more seconds to do the same.

But it is hard to say for certain, because the B and C early acceleration profiles may be just as good as pilots flying the F-35A assert about the CTOL version. The F-35C might have the toughest time making up the acceleration difference because it has a much larger non-lift contributor to wing wave drag to go with the larger lift contribution. But as it is also slightly longer overall and especially longer in the wing and tail surface areas, the F-35C shape in its entirety may (I don't think so but I can't rule it out) be lower drag than the others above Mach 1.1+. Or the F-35C may have other lower drag advantages due to something like its ‘cleaner’ wing attachment transition on the bottom surface. Who knows?

The newer 'changed' KPPs are as far as we know, still based upon modeled F-35 performance using “end-of-life, worst-case” scenario relative to the F135 engine’s power capacity” (Ref #7 Part 2) and “two per cent thrust” penalty (Ref #8 in Part 2). These ‘wedges’ against the F-35's performance could also be the difference between meeting the old KPP and needing the new KPP in the case of the F-35A, and part of the difference for the F-35B and C. 

The changes might also (almost certainly for the F-35C) have been due to encountering an ever-so-slightly higher than planned/predicted peak transonic drag coefficient, or some combination of the above. Contrary to what some might think, computational fluid dynamics and wind-tunnel testing do not prevent small surprises when the aircraft finally flies, they just lower the chances and severity. Sometimes those surprises go undiscovered for years (Bitburg Roll anyone?), and frankly I'm STILL surprised at how little difficulty the F-35 has had in some performance areas compared to legacy aircraft.  Beginning at about .8 Mach, even the tiniest drag [design] differences affect an aircraft’s [drag] performance out of proportion to the differences. The ability to cruise in military power at Mach 1.2 for some distance (Part 2, Ref#2) indicates that for the A and the B model at least, that ‘peak’ in the Drag Coefficient around Mach 1.1 is a fairly narrow one. 

The importance of the KPP change all comes back to what Tom Burbage was quoted as saying in Ref #4:
“...the biggest question is: are the acceleration characteristics of the airplane operationally suitable?”
The people buying and flying the airplane apparently say ‘yes’. And they've got the data and required knowledge and judgement to make the call. The voices outside the program do not.


Acceleration Sidebar: Is Acceleration More Important to Offensive or Defensive capabilities?

If I had to make a guess as to whether or not F-35 transonic acceleration was more critical in the A2A arena for closing on an enemy or escaping an enemy, I would say it is more important as an offensive attribute for closing on an enemy. Why? For two main reasons. First, I seriously doubt there is a fighter in existence or on the drawing board that can be expected to consistently attempt to overtake an F-35 without hesitation: or any aircraft that will competently shoot missiles over the shoulder at their pursuers. Second. The F-35 is being fielded as an optimal pack hunter: any aircraft pursuing one F-35 will always be a little more cautious if the pilot has reason to expect there to be another F-35 or two that he has no clue as to where it or they are.

Next Up: 

Stimulated by Commenter 'Tim A.' in Part 2, I'm going to do a quick parametric examination  of transonic acceleration differences between the 'ultimate' F-16 (Block 60) and the F-35A at A2G and A2A weights and loadouts. Nothing too extensive-- just something to ponder.

  

4 comments:

Unknown said...

Looking forward to part 2. Thanks SMSgt Mac!

S O said...

Not sure late F-16C agility is something to write home about, though.

This up-engined 1970's design has been surpassed by Typhoon and may be unfit to dodge many of today's AAMs.
Modern anti-g-suits allows for up to 11 g turns (according to published info), something the F-16 was never meant to do (same).

Maybe there's a critical agility threshold in modern combat aircraft:
Either you exceed it and benefit from agility or you're not agile enough. This applies especially to missile dodging (supported by warning sensors which can detect low signature rockets in time and reliably).

Statistical Mechanic said...

I definitely understand a lot of the negativity towards the F-35 is just negative hype, and I have really enjoyed reading your breakdown. Nevertheless, on the subject of unloading to gain acceleration, doesn't this miss the point? The big deal of the transonic acceleration KPP being whatever it is is not that there is some combat situation where that specific time is important. The big deal is how this compares to aircraft it might meet. So, if the baseline for the F-35 KPP is the F-16C block 50, I think comparing the straight and level performance of the F-16C to the unloaded performance of an F-35A is cheating. Can't the F-16C just unload too?

SMSgt Mac said...

Hello SM, thanks for bringing this up.
As I didn't take the time to really drive certain points home that I've made elsewhere before and since when I wrote the piece, I appreciate the opportunity to do so here.
First, KPPs are not requirements, they are literally "Parameters" for evaluating the weapon system against the Requirements within a trade space. Those parameters can be adjusted as needed to ANY value so long as the overarching requirement involving multiple KPPs is not violated or requires change. It is not unusual to change KPPs for a reason or several reasons: from discovering they're not representative of what the requirement's intent is to the requirement itself being unachievable (not unheard of in new weapons system development where the reason the new weapon system program even gets going is because existing tech will not suffice for the future). The latter is ONLY a problem if a reduction/change would subvert the reason for the program existence. If there was a set specific time that the F-35 had to meet without external stores that could not be violated (for some unknown reason) to ensure its classified Lethality and Survivability requirements would be met, the transonic acceleration time KPP would not be raised above that time and a redesign would be undertaken. If a redesign could not solve the problem, a program could be cancelled if the problem made the value of the program unworthy of pursuit. You can’t “cheat” on a KPP if the KPP values still support the requirements.
Second, and as it happens, it was publicly acknowledged years before the KPP change that the KPP was established based upon CLEAN F-16 and F-18 performance and that adjustments would have to be made to recognized that the F-35 acceleration ability with internal stores was not degraded as it was for the F-16 and F-18 (http://www.f-16.net/forum/viewtopic.php?t=16745). I would re-emphasize here that NOBODY goes through the transonic regime easily, which means whoever accelerates better early getting to the transonic region has the edge, and pilot reports to date point to the F-35 as superior in that regard to an armed F-16 and F-18. I suspect further that the loaded F-35A may get to M1.1 much faster than a clean F-16, but that the last little ‘nth’ of V costs a few or more seconds due to higher drag rise than predicted, or perhaps incurred due to design features that support other KPPs.
Third, the ‘unloading’ advantage would not benefit an armed F-16 or F-18 to the extent that it benefits an F-35 because of their lighter weight (less drag from lift) and greater frontal area (non-lift) drag increases compared to the frontal area drag in their ‘slick’ state. I should have also emphasized more about ‘unloading’ not necessarily involving a ‘dive’ as well. Simply unloading the stick allows acceleration to increase speed and lift, which allows more unloading that gives more speed and lift, up to the point where minimum AoA is reached to generate the lift needed to fly level at the weight and speeds flown. Theoretically, you could unload carefully without losing any altitude at any time. Diving on the other hand involves a tradeoff: converting more potential energy to kinetic, that might have to be recovered later, but possibly on the other side of the transonic drag region peak (at a lower cost in V). Both are techniques already well known and in use. They also certainly allow greater latitude for designers and programs to meet hard requirements.