T-38 down

Yeah I'm still here. It's been a long day. It was already a long day dealing with some personal matters with my parents and that whole disaster we call the CONUS apathy towards Puerto Rico. Then this thing happens on the last go of the day. Tough tough couple of days for the immediately affected families. Our hearts absolutely go out to them. Our main focus right now is to provide emotional support and rally as a flying community.

Not at liberty to comment on the specifics of what transpired and what we know so far, but for my part, I am constantly humbled and reminded of the nature of this business we're in. Control is somewhat illusory in life it seems. But you can't coward in a corner. So we sack up, kiss our loved ones, and hack it to the best of our abilities. As aviators we analyze, thank our lucky stars, and hopefully apply lessons going forward.

Trust me when I say I am as interested as everybody else here on the safety aspects of this incident; after all I place my life on this equipment every day. Equipment you all have paid for and entrusted us to employ effectively. I just hope it ends up being something other than what I privately dread, because no amount of political correctness is gonna keep me from saying my peace if this thing points that way. We all have lines in the sand, and I know mine. But this is not the time, so I digress.

Being alive today to hug my family, raise the nose and break ground, and chase after life with a sense of purpose; I'm thankful for that opportunity. A brother in arms will not get to do that today. A Nickel on the grass....
touch-the-sun.jpg
 
Sorry to hear about your loss out there.
 
Seen a few of those when I was at ATC bases. Always sobering and so sad.

RIP
 
Wow...my PPL instructor just soloed in the T-38 a few weeks ago. He said he might of been flying from Texas to Mesa, AZ this past Friday...wonder if this is a related incident.
 
Sad to hear about this :(. A few years ago we had a 38 crash that killed both here on base. It was sobering to say the least and really sad how it happened. They grounded the fleet for a few days but came down to an aileron torque tube bolt that broke on take off. Was not cool at all. I haven't been to work yet today so don't know if they grounded the fleet again.
 
That's terrible :(

I'll never forget hearing about the Brit. Test Pilot that was in my brother's class that was killed on his first test flight out of TPS...
 
Woke up this morning to a bunch of nickel in the grass posts by friends on FB, and read about this here on PoA. Ugh.

RIP

Keep yourself safe out there, @hindsight2020
 
Sad to hear about this :(. A few years ago we had a 38 crash that killed both here on base. It was sobering to say the least and really sad how it happened. They grounded the fleet for a few days but came down to an aileron torque tube bolt that broke on take off. Was not cool at all. I haven't been to work yet today so don't know if they grounded the fleet again.

Yeah I remember that one, though I might be confusing it with another incident where the ailerons had been reverse rigged......ie left stick gave right roll.....they were too low and fast to figure it out before impact occurred.
 
Yeah I remember that one, though I might be confusing it with another incident where the ailerons had been reverse rigged......ie left stick gave right roll.....they were too low and fast to figure it out before impact occurred.

"The accident investigation board determined the cause of this mishap to be mechanical failure of the right aileron, which failed in the full down position before takeoff"

http://www.af.mil/News/Article-Disp...als-release-report-on-columbus-t-38-accident/

I remember there being talk of a bolt that broke which caused the failure but could be wrong. It was sad because they ejected while inverted.. ugh! It happened so fast not sure they could have done anything at that point.
 
Been assigned to ATC pilot training bases 3 times, and the worse I experienced was a T38 in the late 60s where both student and IP were decapitated. I was standing at the rear of the hospital when the "meat wagon" rolled in with the two pilots. Three body bags came out, confusing me. Hospital sergeant told me it was the two bodies each in a bag, and their heads in the third bag. Very gruesome and I was screwed up the rest of the day thinking about those poor souls and their families. Been to accident scenes a few times and it ain't fun.
 
Tragic. My condolences.

I drive by Laughlin often on Hwy 90 and frequently pull over in an area the public uses for viewing the ballet in the sky just before the main gate. When the young men are training the skies are buzzing. My place in WTX gets buzzed all the time by B1-B's at low level. Last visit a few weeks ago, two B-52's came over! The whole area north of Big Bend and south of Alpine is an MOA, specifically IR178 which my place is directly under.

I silently thank all the young men and women serving every time they fly over!
 
And here it is guys, as promised. AIB finally out. And for the AIB reading challenged, here's the pedestrian cliff notes (so so job at capturing the essence). This is what we're dealing with every day. Don't ask my wife her opinion of the USAF, it might get me booted. I'm barely 3 months out of DLF but it's fresh as yesterday. It's public now, so spear away.

I had a huge rant going to post but decided to redact it. I'm just gonna go krav a wall in my garage for a while until I calm down. At any rate, bottom line, you gotta check your own six in this life. Words I live by. Aaaaaaand I digress.
 
Yeah I'm still here. It's been a long day. It was already a long day dealing with some personal matters with my parents and that whole disaster we call the CONUS apathy towards Puerto Rico. Then this thing happens on the last go of the day. Tough tough couple of days for the immediately affected families. Our hearts absolutely go out to them. Our main focus right now is to provide emotional support and rally as a flying community.

Not at liberty to comment on the specifics of what transpired and what we know so far, but for my part, I am constantly humbled and reminded of the nature of this business we're in. Control is somewhat illusory in life it seems. But you can't coward in a corner. So we sack up, kiss our loved ones, and hack it to the best of our abilities. As aviators we analyze, thank our lucky stars, and hopefully apply lessons going forward.

Trust me when I say I am as interested as everybody else here on the safety aspects of this incident; after all I place my life on this equipment every day. Equipment you all have paid for and entrusted us to employ effectively. I just hope it ends up being something other than what I privately dread, because no amount of political correctness is gonna keep me from saying my peace if this thing points that way. We all have lines in the sand, and I know mine. But this is not the time, so I digress.

Being alive today to hug my family, raise the nose and break ground, and chase after life with a sense of purpose; I'm thankful for that opportunity. A brother in arms will not get to do that today. A Nickel on the grass....
View attachment 57919

bsbd :(
 
Regardless of the mistakes that were made, they took into account potential civilian casualties before attempting egress.
 
The report doesn’t explain why the torsional loading on the left shaft was increased...what was the cause of that (since loss of that increased the loading on the right, which was already heavily loaded due to a generator short on the right and caused the right shaft to be finally overloaded which gearbox is said to have already been compromised due to wear).

The arming issues may have caused the death, but they did not cause the crash. Were it not for the failure of shaft on the left side, the (over)loading of which was either not explained or I missed, the accident chain would not have begun. Am I totally missing something? I did read it pretty quickly and may have.
 
Last edited:
Regardless of the mistakes that were made, they took into account potential civilian casualties before attempting egress.

A heartfelt sentiment no doubt, but what may escape the gallery is that they didn't effect any control once dual hydraulic failure ensued. You see, the T-38 is hydraulic controls with no RAT. Even windmilling hydraulics are insufficient to effect control sufficient for attitude change of consequence. In this case, even that is moot since the connection to the drive was severed on both engines. In other words, the aircraft was effectively semi-ballistic with two good engines and no control surfaces.

They had no way to affect impact point control, beyond power inputs affecting pitching moments. Thence, the decision to delay ejection may seem altruistic to you, but was borne more out of suspension of disbelief (the deceased attempted to continue to fly even after the qualified IP had given up on it). To be clear, the IP demanded the ejection in spite of the deceased continued attempt at bartering with the aircraft for control, as they were about to go LV below the horizon. The deceased was simply not mentally prepared of getting out of that airplane that day, and he paid the ultimate price for it. Yes, the IP made procedural mistakes relating to the crew coordination items in the checklist. That was very much contributing to the deceased finding himself panicked or incapacitated, either way fatally unable to do the actions of arming the seat like he should have accomplished before takeoff. I'm not casting aspersions, just relating how they found themselves in the position they did.


The report doesn’t explain why the torsional loading on the left shaft was increased...what was the cause of that (since loss of that increased the loading on the right, which was already heavily loaded due to a generator short on the right and caused the right shaft to be finally overloaded which gearbox is said to have already been compromised due to wear).

The arming issues may have caused the death, but they did not cause the crash. Were it not for the failure of shaft on the left side, the (over)loading of which was either not explained or I missed, the accident chain would not have begun. Am I totally missing something? I did read it pretty quickly and may have.

No, you're not missing anything. You did it in 6.9 seconds flat. Cigar for you sir.

So let me fill some gaps in what you're looking at, as SGOTI who may or may not have flown the accident airplane a time... or a dozen.

First, people need to understand how the system works. These airframe mounted gearboxes are simple transmissions designed to take the high RPM input of the turbine and mate it with one generator and one hydraulic pump per gearbox, and obviously one gearbox per engine.

Torsional loads transients reach a peak as noted in the report. Remember, even though these torsion load mainly deal with the generator, the hydraulic pump relies on the same gearbox to turn. The aircraft *WAS (*important editorial on my part) designed to be able to handle the entire electrical load on one generator, and handle the entire hydraulic demands of the flight control surfaces on one hydraulic pump.

Here's the answer to your question, and that will never be in that report. This isn't 1955.... It's 2017. What changed? A lot.

In that time, we've made hundreds of changes to the electrical demands of the aircraft, on ever increasingly worn cabling and ever increasing losses that increase the overall demands on the system, which ultimately boils down to a higher baseline torque load on both generators and therefore both gearboxes' drive shaft. That's my opinion, but what do I know.:rolleyes: What I do know is we haven't done **ck all to the voltage regulator, crossover relays, and most importantly...our craptastic airframe mounted gearboxes.

The gearboxes have also not gotten any younger, and as listed by the report, the gearbox on the right engine had 7 failures in the preceding year and a half. 7 failures. You know how that makes ME feel when I have to strap into the thing? You guys would cry bloody murder if your alternator belt snapped 7 times in 18 months. Imagine now if further, your steering was dependent on that belt not snapping and not merely your alternator.

So, torsional loads. Understand that under the 2017 paradigm these shafts are running, as a baseline, at a much higher torque than they were intended for. Meaning, they're running closer to the shearing value as part of the normal course of business, something they would never accept in 1955. OK, so why does that matter if it's still under the threshold you might ask. Well, that means that when the gearbox shifts (this happens in the 65-75% engine RPM band, moving the throttle in either direction by the way) the box shifts like a tranny, and the load spikes momentarily. So then, as these things are already running at a higher baseline, hello momentary overtorque city. In the case of this airplane, you're now relying on the sole remaining, already-documented multiple offender, POS gearbox that they kept fielding. And amongst us friends we know that thing had zero, nada, zilch, kaput shot in hell of getting them home without failing almost immediately the second they shifted that right throttle through the shift range.

So to put it in blunt terms: If neither component has the BONA FIDE ability to handle the entire demand in the course of normal field conditions as defined today, it is not accurate to assert you actually have redundant components in earnest. This undercuts the notion of redundancy to the point where it demands an evaluation of the outright redesign of the component and a new performance threshold established.

...And this is the problem.
 
Last edited:
Strong words, well -spoken.
 
snipped

...And this is the problem.

Those gearboxes are far from simple, they are very complex, maybe not in the mechanical, Rube Goldberg sense, but in an engineering sense, and in what they are asked to do, they are very complex and under some serious stress, where everything must be perfect in order for them to be reliable. Even today engine manufacturers struggle designing these gearboxes. From what is written in your report above I'm not sure I would get into one of those aircraft.
 
In other words, the aircraft was effectively semi-ballistic with two good engines and no control surfaces.

They had no way to affect impact point control, beyond power inputs affecting pitching moments.
Wow... we have to fly near Sheppard AFB a lot and a couple of other places where there are T-38s and I grew up near Randolph AFB. Love the shape and grace of that plane but never knew that about it. Hope something is done on a serious level, really soon about that.

Also note to self that if I hear a T-38 with ANY sort of emergency situation to be extra vigilant and get out of the way as necessary.
RED_3143.JPG
 
So to put it in blunt terms: If neither component has the BONA FIDE ability to handle the entire demand in the course of normal field conditions as defined today, it is not accurate to assert you actually have redundant components in earnest. This undercuts the notion of redundancy to the point where it demands an evaluation of the outright redesign of the component and a new performance threshold established.

...And this is the problem.
Wow, just wow..:(:( Fact is fact, but I find it hard to believe they could not come up with a new fix for the problem you described. As @PaulS said "I'm not sure I would get into one of those aircraft".

Years ago I worked on the J-85 engine that went in the T-2 Buckeye, I think that's what the T-38 uses. The Buckeye did not have an afterburner were the T-38 does.

Thanks for your comments about the troubled systems the T-38 has.
 
And here it is guys, as promised. AIB finally out. And for the AIB reading challenged, here's the pedestrian cliff notes (so so job at capturing the essence). This is what we're dealing with every day. Don't ask my wife her opinion of the USAF, it might get me booted. I'm barely 3 months out of DLF but it's fresh as yesterday. It's public now, so spear away.

I had a huge rant going to post but decided to redact it. I'm just gonna go krav a wall in my garage for a while until I calm down. At any rate, bottom line, you gotta check your own six in this life. Words I live by. Aaaaaaand I digress.

Damn hindsight, damn. So senseless and unnecessary loss of lives. Thanks for the update.
 
So I guess it's less expensive to lose the occasional aircraft, and sometimes crew, vs. re-designing the gear boxes? Cold way to look at things...
 
It is always -- always -- a question of money.

The T-38 is a legacy system that already has gone through three major SLEP/upgrade programs in the last 15 years (C model, PMP, ejection seats), and the T-X replacement program has been bubbling and steeping this whole time.
 
AThey had no way to affect impact point control, beyond power inputs affecting pitching moments. Thence, the decision to delay ejection may seem altruistic to you, but was borne more out of suspension of disbelief (the deceased attempted to continue to fly even after the qualified IP had given up on it). To be clear, the IP demanded the ejection in spite of the deceased continued attempt at bartering with the aircraft for control

If the fellow had tried to eject when first requested, do you think he would have had enough time to debug the situation, arm the seat, and get out? Also, with the reputation of these gearboxes, I'd think it would be ingrained in all pilots who fly these birds that a dual gearbox failure is an immediate get out of the airplane event.
 
As and Army MTP for 20 plus years the gearbox failure seven times in a year and a half is so wrong...we had similar issues with components on various Rotor Wing aircraft as added fixtures and new mods were produce using the same aircraft infrastructure....the story is repeated over and over...also the reason and need for recurrent training unfortunately made its case that day as well. We often laugh at the Army's Warrant system...and I was both Commissioned and WO...but our Army Warrants fly their whole career without gaps to fill other command duties...I wish we could learn from our mistakes and we are better at it now than 35 years ago when I first flew a Army Aircraft but that's not much solase when this happens...my comments are simple and out of frustration for those who undertake this duty...
 
Also, with the reputation of these gearboxes, I'd think it would be ingrained in all pilots who fly these birds that a dual gearbox failure is an immediate get out of the airplane event.

The checklist has always mandated ejection as the final step of a dual gearbox failure.

Realize that the situation that @hindsight2020 is referring to regarding the difference in load on the gearboxes, and that causing increased fatigue and increased failures, is not something that has been known, understood, or canon for long. The T-38Cs have been flying operationally since about 2001, and that's ostensibly what has created the greater load/demand on the gearboxes, and there wasn't some immediate increase in gearbox failures (or dual gearbox failures) when the C models became operational. Honestly, I never heard that set of information discussed once in my association with the T-38C between 2003 and 2015, and over 1,100 flying hours in it over those years (and never, FWIW, experienced a gearbox failure in those hours).

If what he's saying is true, then it is a problem which has developed over the last decade and is still developing.

Regardless, this information should have no impact on what a crew needs to do with a dual gearbox failure, which as has been mentioned, results in a complete inability to move the hydraulically-actuated flight controls immediately.
 
Last edited:
The checklist has always mandated ejection as the final step of a dual gearbox failure.

Realize that the situation that @hindsight2020 is referring to regarding the difference in load on the gearboxes, and that causing increased fatigue and increased failures, is not something that has been known, understood, or canon for long. The T-38Cs have been flying operationally since about 2001, and that's ostensibly what has created the greater load/demand on the gearboxes, and there wasn't some immediate increase in gearbox failures (or dual gearbox failures) when the C models became operational. Honestly, I never heard that set of information discussed once in my association with the T-38C between 2003 and 2015, and over 1,100 flying hours in it over those years (and never, FWIW, experienced a gearbox failure in those hours).

If what he's saying is true, then it is a problem which has developed over the last decade and is still developing.

Regardless, this information should have no impact on what a crew needs to do with a dual gearbox failure, which as has been mentioned, results in a complete inability to move the hydraulically-actuated flight controls immediately.

All due respect, do you consider 5 and 7 failures each within 18 month for both gearboxes par for the course in purporting the position that "we didn't know we had a problem..."? You already gave the answer in post 25. You think those were the only two components in the DLF fleet that had that history? I could keep going but I can't, as I'm sure you understand.
 
If the fellow had tried to eject when first requested, do you think he would have had enough time to debug the situation, arm the seat, and get out? Also, with the reputation of these gearboxes, I'd think it would be ingrained in all pilots who fly these birds that a dual gearbox failure is an immediate get out of the airplane event.

Yes, more Warnings were put in the Tech Order, they were imo repetitive. They're always written in blood, and this one is no exception.
 
All due respect, do you consider 5 and 7 failures each within 18 month for both gearboxes par for the course in purporting the position that "we didn't know we had a problem..."? You already gave the answer in post 25. You think those were the only two components in the DLF fleet that had that history? I could keep going but I can't, as I'm sure you understand.
Hindsight2020, an appropriate name for this unfortunate incident. Were the gearboxes repaired on sight or sent back to the manufacturer? What failed? Sorry if it's in the long version of the links, I didn't read those, if it is please let me know and I'll find it.
 
All due respect, do you consider 5 and 7 failures each within 18 month for both gearboxes par for the course in purporting the position that "we didn't know we had a problem..."? You already gave the answer in post 25. You think those were the only two components in the DLF fleet that had that history? I could keep going but I can't, as I'm sure you understand.

No, I'm not discounting what you are saying at all -- in fact, I'm listening with quite a lot of interest.

I was trying to provide some context for people who read your post, and think that this is some long-known, long-established issue with the T-38C and can't understand why the USAF hasn't done anything to fix it.
 
Right is “flight controls” and left is “utility.” Can the left supply pressure for the flight controls if the right fails first?
 
Whenever we do cause analysis, the end goal is to generate corrective actions for prevention.

So strictly from a cause analysis standpoint (based on limited internet info), the shaft was designed to shear at a certain point, considering certain design safety factor/buffer. Over time, system changes may have impacted the suitability of the formerly designed failure point (the shaft), a designed in relief point for whatever legitimate engineering reason at the time of original design or even some later mods (not any or all from what I’m hearing).

From a corrective action and future prevention standpoint (THE MAIN REASON FOR DOING A CAUSE ANALYSIS) - need to figure out if the shaft failure point needs to be increased or a redesign/investigation into the evolution of all the parasites on that shaft need to be evaluated, or both (probably both).
 
Right is “flight controls” and left is “utility.” Can the left supply pressure for the flight controls if the right fails first?

Yes.

29qcpea.png
 
Last edited:
Back
Top