Somebody lose a door?

IDENTIFICATION
Regis#: 78MU Make/Model: EXP Description: RV-10
Date: 09/26/2009 Time: 1827

Event Type: Incident Highest Injury: None Mid Air: N Missing: N
Damage: Minor

LOCATION
City: JEFFERSON CITY State: MO Country: US

DESCRIPTION
AIRCRAFT ON TAKEOFF, PASSENGER DOOR SEPARATED FROM AIRCRAFT, LANDED WITHOUT
INCIDENT, JEFFERSON CITY, MO

INJURY DATA Total Fatal: 0
# Crew: 1 Fat: 0 Ser: 0 Min: 0 Unk:
# Pass: 1 Fat: 0 Ser: 0 Min: 0 Unk:
# Grnd: Fat: 0 Ser: 0 Min: 0 Unk:

WEATHER: SCT029 BKN045 260/08G15 10SM A2995

OTHER DATA
Activity: Unknown Phase: Take-off Operation: OTHER


FAA FSDO: KANSAS CITY, MO (CE05) Entry date: 09/28/2009

Yeah... There is a description from the pilot on matronics.

I'm surprised that something so minor necessitated an FAA incident report.

Years ago I was departing Dallas Love in an old Cessna 402, ferrying to a paint shop in Mena, AR. Shortly after takeoff 13L, the right side emerg exit hatch blew off. Oops. I circled back and landed... the tower told airport ops guys to keep an eye out for a missing door. They found it on a taxiway at the far end; no damage to it, just a bent corner. Had a mechanic straighten it out, re-install and secure with lots of duct tape and off I went to Mena. All this happened at Class B towered field... no body said nuthin' and no Feds came calling. Maybe I just got lucky?

Can someone copy the pilot's description over here? I'm just curious if it was a construction issue, operational issue, or both.

double sided tape failure?

Sorry, I could not resist. Re thread a while back on trim tab temporary mounting methods.

Moderators may show no mercy, give no quarter.

Operational issue. The pilot stated he failed to ensure the aft pin was engaged:

http://forums.matronics.com/viewtopic.php?t=63161

Quite a lot of discussion on RV-10 door latch design of late. Seems to be a weak link in that, if you aren't careful you can not have it fully latched with potentially dire consequences on departure.

In any event - if interested you really need to search the RV-10 matronics archives but unless you have actually built or are building an RV-10 you probably won't get much out of it as the door on RV-10 isn't much akin to any other RV configuration.

When I get to that point I'll have some head scratching to do. I know what mods I'd like to make but as in everything the devil is in the details. Since modern RV plans are really just an assembly manual I"ve got to wait to get actual parts before I can reverse engineer them so I can design some mods that will fit.

The key to the whole RV-10 door problem is having closing sensors on all four contact points. That is two on each side. We have a light for each side and if the door is not secured properly the lights will stay on. Proper pre-take off chack list should ensure doors are closed properly. Do I always remember? No but I do check the lights to see if they are off.

We installed the Hendricks door system and fitted the Cleveland catches, 4 magnets and sensors, a circuit for left and right with alarms into the Dynon D180. Seems to work well.

I do notice our Right Hand Door does bow outwards slightly in fligt at the centre of the bottom edge. Left does not.

Need to get some formation friends to take some close up pics, but there must be a large DP around that point.

The left side has high pressure from the prop wash, where the right side side has low pressure from the prop wash.

Door problem or not?

I have not flown the -10 yet, but Van has flown theirs for hundreds of hours
and no door problem so far.
The most recent departure of someone's door was due to improperly closing it and should not raise the red flags on the door closing mechanism.
While I did add the aluminum door guides and the bullet shaped magnetic tips,
the rods still protrude more than an inch past the aluminum frame.
The actuating mechanism is stock as per Van's instruction.
Not disputing your observations, simply stating that a properly closed door
with pins protruding past the aluminum door frame ought to stay closed.

Todd

I had wondered if that was the cause..... and might well be!

N427EF said:

I have not flown the -10 yet, but Van has flown theirs for hundreds of hours
and no door problem so far.
The most recent departure of someone's door was due to improperly closing it and should not raise the red flags on the door closing mechanism.
While I did add the aluminum door guides and the bullet shaped magnetic tips,
the rods still protrude more than an inch past the aluminum frame.
The actuating mechanism is stock as per Van's instruction.
Not disputing your observations, simply stating that a properly closed door
with pins protruding past the aluminum door frame ought to stay closed.

Click to expand...

Why don't you call Van's and ask about their RV-10 and Sun-n-Fun a few years back? No it's not the door when it is properly closed. It is the proper closing that is the problem.

TSwezey said:

Why don't you call Van's and ask about their RV-10 and Sun-n-Fun a few years back? No it's not the door when it is properly closed. It is the proper closing that is the problem.

Click to expand...

I think this is at least the 3rd door "issue" I've heard of on RV-10's. It wouldn't be a surprise if others went unreported. My point is that the RV-10's doors are not as failsafe as I'd like. Also, you have the complication of the passenger probably being the person to close/latch the right door.

Losing or damaging a door is bad enough, but I worry about a failed door taking off or severely damaging a horizontal stab one day. This is a problem in search of a robust (i.e. idiot proof) solution.

with an experienced pilot in the passenger seat, there is no excuse for both of them.
no idiot proof solution can replace due diligence.

RVnoob said:

with an experienced pilot in the passenger seat, there is no excuse for both of them.
no idiot proof solution can replace due diligence.

Click to expand...

True, but people make mistakes every day and it is the engineer/designer's responsibility to eliminate (or at least minimize) the risks inherent in any of those mistakes.

The failure to latch a door properly *shouldn't* be a potential threat to the safe conclusion of a flight. Look at most GA types - leave the door unlatched and it simply opens a few inches in flight and you land the airplane at the nearest airport. Leave a door unlatched (or improperly latched) on a -10 and something very bad could happen.

TSwezey said:

Why don't you call Van's and ask about their RV-10 and Sun-n-Fun a few years back? No it's not the door when it is properly closed. It is the proper closing that is the problem.

Click to expand...

Todd,
That was a totally different situation. Remember... N410RV was the initial prototype airplane...it had a prototype (different handle /latch design) that slipped and didn't produce full stroke of the pins. All kits delivered to customers have been supplied with a different handle/latch design.

TSwezey said:

The left side has high pressure from the prop wash, where the right side side has low pressure from the prop wash.

Click to expand...

This has not been my experience.

I believe there is low pressure on the lower exterior of both doors.

Mine work fine but not stock Van's either...Hendricks locks and handles, Rivethead pins and recepticles, properly rigged door switch warning lights..I CAN'T close my doors with the aft pin out of alignment and if I did my door closed light would stay on as well as tweaking my door trying to make it lock.

Lot's of choices guys and gals but if you don't have a method of verifying that the pins are engaged and USE it....like CIGAR or IMSAFE or GUMPS... the best locking system in the world won't save you.

Like I posted previously, we have the hendricks system and the aluminium door catches, we installed 4 reed switches and they do work.

It is possible to close the door with the front pins only engaged, however if you set the warning system up and in our case the Dynon "EMS ALERT" will bug you to death until you fix it.

I would like to see less flex in our RH door is all.

I am not so concerned about the pins being all the way in as I am about a passenger getting a shirt/coat sleeve or a book or headset cable caught on the door lever and opening the door in flight!

Cheers!

DB

One thing I think contributes to the problem is the hold open strut.

Due to the fact the strut is located at the back, and the door handle at the front of the door, there is already a twisting force applied to the door as it is closed, and that force wants to keep the rear from closing all the way.

My doors are set up with a pair of magnetic reed switches, and a push to test button on the panel. You get a green light when door is properly closed.

Accidentally opening?

DB,

I agree that the closing mechanism is not fool proof but the opening mechanism is indeed fool proof at least with the factory supplied door locks.
There is no way you could even purposely open the door latch with a shirt or hand bag or headset cable.
You have to use at least two fingers to depress the lock lever and then turn it.
I do not know about after market door locks but if those mechanisms don't provide a positive lock on the handle you have created a problem that did not exist with the factory provided hardware.
I will admit that any aftermarket handle looks better than the square tube handle provided by Van's.

read back a bit......... we use the hendricks door system!

There are pro's and cons either way!

I think we have the lessor of the two evils!...................I hope!

Here's my $0.02 coming from the perspective of a nuclear engineering director/senior reactor operator.

There is nothing inherently wrong with the door design. When it is installed per the plans and properly operated the door will not open in flight. However the multiple incidents of this happening lead you to the conclusion that there is something wrong or something not working as intended. The probability of recurrence is high and the consequences are also high. This translates to high risk which is unacceptable in my world of nuclear power plants.

I think the human factors consideration in the door design is poor. The normal motions of closing and latching a door should have a virtual zero probability of failure. It is something we all learned to do at age 5 and you cannot expect the human brain to approach this task in the same manner as brain surgery 100% of the time. It is also too easy for a passenger to momentarily open the door for some reason and relatch it, possibly unnoticed by the pilot.

The warning lights are a bandaid to a poor design. While it helps to prevent a failure to properly latch it does not work 100% of the time. The recent incident is a case in point.

The real fix is a door that NATURALLY closes properly - no special thought or actions required. An alternative is to design an interlock that prohibits takeoff unless the door sensors are a go.

In my industry we would not tolerate this many failures. Something would be done about it. In this case I'm not yet sure of the answer. In my 10 I am modifying the plans slightly such that the pins extend an additional 5/16 inch. I am doing this by cutting the geared racks slightly longer. My next challenge is to get the door to seat properly such that the pins lock in with no special effort. In the interim I am looking for other good ideas. Most of the ones I have seen are more bandaids.

I think sometimes we get distracted, or get into a situation where the door closed check item can be overlooked or taken for granted. I know firsthand, as I had this happen on my first RV-10. No excuses, but it was a multi-stop day, and while I had always checked the door multiple times earlier with the same passenger that day, on this time I called for a door check and got a yes. Well, on climbout I noticed that the cabin noise was louder than usual, and I first checked my ANR battery status. That was green, and then the chill went down my spine faster than you can read this. I immediately noticed that the passenger door was not latched at the rear pin. I reached over and put a death grip on the door handle, reduced powerand speed, and told the tower I needed to come back for an open door. I was over a very densely populated area and did not want to lose that door and perhaps hit someone with catastrophic results. I never let go of the door until all three wheels were on the ground, completely flying the airplane, adjusting flaps, power, and making the radio calls with my left hand.(nice flying characteristics that 10 has!)
Yes, this was prior to the door warning switch service bulletin. I have installed them in the RV-10 now under construction, and I have added one change. I have wired the power to the door warning circuit through a micro-switch on the throttle quadrant. That way, if I go to power and the doors aren't latched, 2 big red lights will come on. I think I have a better chance of noticing them that way than having them on all bright and red while we are taxiing out with the doors open in HotLanta during the summer (no A/C).
It's a thought, anyway.
I know there's lots of discussion on the design of the doors, etc. It is my opinion that the doors are not unlike retractable gear airplanes. Piper even once tried to make the retracts fool proof, and even the FAA made them disconnect that circuit for a while. The old saying about retracts may apply here... without using checklists and visually checking when possible could lead to unintended consequences no matter what the design is.

Vic

I know not everyone has the Dynon system, however those who do can configure a warning light for both left and right, plus you get a loud EMS ALERT in your headset and a warning bar on the screens, if you clear it and the fault is still present it reappears again a minute or something like that later.

I do not think the chances of taking off with a proper light/alarm system is any greater than you forgetting to put fuel in and not checking the guages......so its mitigated quite well with the right system in place.

Trouble is we do not know enough about the incidents that happened. One readily admits the aft pin was not in........ well what do you expect? Its going to fail. Some builders door alignment may be worse than others, making the problem more likely.

Properly set switches and wired in a fail safe manner with a good alarming/indicator is the answer. You may make door improvements but these in no way take the place of a good detection and warning system.

DB

Roll Pins

Here is one other thing to watch for... the door handle locks into place when a spring shoves a roll pin into a slot. On one of my doors, I found that the roll pin did not always fully engage into the slot. In this condition, the door could be unlatched simply by a bump of the handle.

The roll pin would stick primarily if the handle was latched slowly/gently. A little WD-40 helped and it was less likely to stick when the handle was closed firmly.

I agree that proper latching and checks will virtually eliminate the possibility of a door opening. Therefore, I am willing to call the stock door adequate. However, it is also unfortunate that improper latching can result in a dangerous situation when so many other designs do not. I'd much prefer a fail-safe door over the stock one.