What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

RV10 Door came off shortly after take-off

Janekom

Well Known Member
Patron
The scenario is that a new RV10 owner has lost the pilot side door shortly after take-off. His fault and he accepted it, as he simply forgot to latch it properly. As far as I can find out, it was only the Vans spring loaded lock that was holding it.

I know of and has been involved in two other doors that came off locally and in both cases the doors could be repaired easily an re-installed. In both cases the plexi was not even scratched.

BUT

This RV10 was built in the USA and imported to South Africa in 2017 and it has done about 200 hours locally. As an AP (Approved Person) I was asked by the new owner to help with his broken door.

Looking at the door I could immediately see that there was NO ATTEMPT AT ALL to sand/scuff or roughen up (whatever you want to call it) mating surfaces and flanges. The material used to bond also does not look like cotton flox to me.

We now know that it will be a completely new LH door which will be expensive and a lot of hours. Taking into account that we will have to go and remove the wings etc. and trailer to my place to do the repair. I am also worrying about the RH door which will be the same if it has been done by the same builder.

The reason for this message is to bring it under the attention of ALL RV10 builders. Please guys stick to the building instructions which is very clear of what to use and how to build these doors. I believe they are critical parts and no shortcuts should be taken.

Below are some pictures. The first picture is of the piece that has remained on the cabin top. Comments are welcome and I really hope it will be seen by current builders.
 

Attachments

  • IMG_3403.jpg
    IMG_3403.jpg
    1.1 MB · Views: 774
  • IMG_3406.jpg
    IMG_3406.jpg
    749.1 KB · Views: 661
  • IMG_3410.jpg
    IMG_3410.jpg
    1 MB · Views: 523
  • IMG_3417.jpg
    IMG_3417.jpg
    786.1 KB · Views: 528
  • IMG_3418.jpg
    IMG_3418.jpg
    641.3 KB · Views: 377
Sorry to hear of your troubles. This is kind of to be expected with the current erector set level of kits available today. It allows a class of builder to complete a plane without the level of experience and skill found when the kits were more rudimentary and with drawings instead of instructions, though clear here that some don't even read those. Clearly many spend a lot of time here to learn the full breadth of skills needed, but suspect that many do not.

Really a shame to see these kinds of errors/omissions.

Larry
 
Last edited:
To late for a pre-buy inspection now, but I would suspect all phases of construction and want a thorough inspection of the whole plane prior to further flight. That door is representative of shoddy workmanship.
 
Thank you for the comments.
Dave how will a pre buy inspection on a painted door reveal this?

I am hoping for just bone builder to say thank you for the heads up because he has been on the wrong track.
 
Thank you for the comments.
Dave how will a pre buy inspection on a painted door reveal this?

I am hoping for just bone builder to say thank you for the heads up because he has been on the wrong track.

It wouldn’t but there are other things that should be looked at closely. Maybe just an oops on the door but the plans are pretty specific on prep before bonding. There may be other slip-ups, or just shortcuts taken.
 
I no composite expert, but could you tap test the door edges to see if the bond is good? Obviously, you might not be able to identify poor prep work or improper materials, but you would (might?) be able to identify de-bonded areas.
 
+1 for..... tap test the door edges to hear if the bond is good. Obviously, you might not be able to identify poor prep work or improper materials, but you would be able to identify de-bonded areas.
__________________
 
Door hinge reinforcement

I'm just curious if the door reinforcement kit that Air Ward sells would have held the door in place?

For anyone that has installed this kit, what is your experience with it? Also, are the plates aluminum or stainless? How thick? I'm curious if this reinforcement would simply transfer the shear force to the cabin, which may not retain the door either, increasing damage further than just losing the door.

Does a door opening event in flight (and maybe on the ground too) always result in a liberation? If so, why hasn't Van's upgraded the door and mounting? I realize that's a question for Van's, but it's very interesting for me.
 
Does a door opening event in flight (and maybe on the ground too) always result in a liberation? If so, why hasn't Van's upgraded the door and mounting? I realize that's a question for Van's, but it's very interesting for me.

As I recall from posts 10+ years ago----the cabin top is a structural component for rollover protection. The door is sacrificial to protect the cabin top.

If I were ever to build another 10, I would have the Plane Around center door latch as a mandatory upgrade.

https://planearound.com/
 
As I recall from posts 10+ years ago----the cabin top is a structural component for rollover protection. The door is sacrificial to protect the cabin top.

If I were ever to build another 10, I would have the Plane Around center door latch as a mandatory upgrade.

https://planearound.com/

I have read the same posts Mike, but they don't definitively state the structural capacity of the cabin, referring to shearing at the mounts. I've asked of Van's and hope to hear back.

While I do wholly agree with installing the Plane Around door latch, and also placing a placard to highlight it's usage - human's make errors...so I'm still seeking to define whether the door, and/or cabin, can be further reinforced, or just expect to replace a door once it errantly leaves the aircraft.

Noting that the Diamond DA-42/62 has a similar structural arrangement for the door and cabin, and has a door open-in-flight procedure that doesn't suppose liberation, it can be done - just a matter of structure (and weight).
 
I'm just curious if the door reinforcement kit that Air Ward sells would have held the door in place?

For anyone that has installed this kit, what is your experience with it? Also, are the plates aluminum or stainless? How thick? I'm curious if this reinforcement would simply transfer the shear force to the cabin, which may not retain the door either, increasing damage further than just losing the door.

Does a door opening event in flight (and maybe on the ground too) always result in a liberation? If so, why hasn't Van's upgraded the door and mounting? I realize that's a question for Van's, but it's very interesting for me.

Ron

If the door opens in flight, it is gone. No hinge on earth will keep it attached to the a/c. I wonder if the hinge reinforcement is solving a phantom problem.
 
Noting that the Diamond DA-42/62 has a similar structural arrangement for the door and cabin, and has a door open-in-flight procedure that doesn't suppose liberation, it can be done - just a matter of structure (and weight).

Have you ever opened the rear door on a Diamond then compared to an RV-10 door? IIRC, the Diamond's door is massive in comparison. I'd imagine that if the attach points held, the RV-10's door would flutter and fail at some other point. IF (big IF, right) a door unlatches, I suggest it is better that it depart the airframe rather than stay attached and mess up the airflow or slam against whoever is in the front seat.

The real solution is an upgraded cabin/door design with doors that hinge on the A-pillars (like a Cirrus) or along the vertical door-jamb, like a Piper or Cessna.
 
Noting that the Diamond DA-42/62 has a similar structural arrangement for the door and cabin, and has a door open-in-flight procedure that doesn't suppose liberation, it can be done - just a matter of structure (and weight).

I assume the DA42 and 62 have a similar arrangement to the DA40, which I've owned two. There is a secondary latch on the rear door which should prevent the door from fully opening if it does become unlatched.

There have been several door departures on Diamonds as well and as I remember, the hinge mechanism is not all that much beefier than the RV-10. The door latch design is better however.

We have installed the PlaneAround latch on our RV-10 and all of the 10's being built (currently and already flying) at Synergy Air South have/are doing the same.
 
I could be wrong here, but my impression is that all or most all of these incidents are basic pilot error: the pilot took off with one or both latch pins not engaged, and one or both red idiot (door open) lights on. Its a human factors thing: we taxi in hot wx with the door ajar, and get used to seeing the lights on. I wonder if a better warning system - I’m thinking, tie the red lights thru a tach or airspeed switch, then to something like a gear warning horn that is so obnoxious it can’t be ignored - would be an easy fix to 90% of the problem. It wouldn’t fix a passenger actually opening the door in flight, but would fix most other things.
 
I could be wrong here, but my impression is that all or most all of these incidents are basic pilot error: the pilot took off with one or both latch pins not engaged, and one or both red idiot (door open) lights on. Its a human factors thing: we taxi in hot wx with the door ajar, and get used to seeing the lights on. I wonder if a better warning system - I’m thinking, tie the red lights thru a tach or airspeed switch, then to something like a gear warning horn that is so obnoxious it can’t be ignored - would be an easy fix to 90% of the problem. It wouldn’t fix a passenger actually opening the door in flight, but would fix most other things.

I don't like depending on warning lights and checklists when it would be relatively straightforward to re-engineer the doors to be front hinged and eliminate the door loss issue forever. A well designed door should never leave the airplane, regardless of operator error.
 
I wonder if a better warning system - I’m thinking, tie the red lights thru a tach or airspeed switch, then to something like a gear warning horn that is so obnoxious it can’t be ignored - would be an easy fix to 90% of the problem.

Bob, you flew my 10----not sure if you remember how I did the door warning lights. I had a push-to-test button and two green lights to show doors locked.

Check list item on both the initial before start list, and before takeoff list.
 
Engineering

Re-engineering the door for a forward hinge would be a large task, for sure...

Not saying that would be a bad idea but I think the time, effort, and money would be better spent on other things...

It is really very simple to prevent an open door, and as has been stated, human error is likely the largest part of the issue...
 
I assume the DA42 and 62 have a similar arrangement to the DA40, which I've owned two. There is a secondary latch on the rear door which should prevent the door from fully opening if it does become unlatched.

There have been several door departures on Diamonds as well and as I remember, the hinge mechanism is not all that much beefier than the RV-10. The door latch design is better however.

We have installed the PlaneAround latch on our RV-10 and all of the 10's being built (currently and already flying) at Synergy Air South have/are doing the same.

The doors on the da42/62 are similar to the da-40 rear door, but also similar to the Trinidad.

While putting your hope in following a checklist and looking for warning lights are good things to help prevent the event; it’s an administrative cure. Much like power plant safety, engineered solutions are the best ones and then rely on administrative and personal protective measures as a secondary step.

If Vans concurs that the cabin top structure is inadequate to handle the shear force of a door open in flight and that the door is designed to liberate as a protective measure, then I’ll accept that. However, to colloquially state that “no hinge on earth will keep it attached” lacks engineered basis. I assure you, I can design a cabin cover and hinge to do the job. It just may not be cost effective. I need to know more about the design basis of the cabin cover first.
 
I could be wrong here, but my impression is that all or most all of these incidents are basic pilot error: the pilot took off with one or both latch pins not engaged, and one or both red idiot (door open) lights on. Its a human factors thing: we taxi in hot wx with the door ajar, and get used to seeing the lights on. I wonder if a better warning system - I’m thinking, tie the red lights thru a tach or airspeed switch, then to something like a gear warning horn that is so obnoxious it can’t be ignored - would be an easy fix to 90% of the problem. It wouldn’t fix a passenger actually opening the door in flight, but would fix most other things.

I’m thinking along this line Bob. The G3X has the ability to do canopy/door open alerts based on engine RPM; combined with 180 latches probably solves most of the issues, but what about taxiing with door unlatched, maybe even held partially closed, on a gusty day?

Any cure that relies on human intervention is subject to error.
 
The doors on the da42/62 are similar to the da-40 rear door, but also similar to the Trinidad.

While putting your hope in following a checklist and looking for warning lights are good things to help prevent the event; it’s an administrative cure. Much like power plant safety, engineered solutions are the best ones and then rely on administrative and personal protective measures as a secondary step.

If Vans concurs that the cabin top structure is inadequate to handle the shear force of a door open in flight and that the door is designed to liberate as a protective measure, then I’ll accept that. However, to colloquially state that “no hinge on earth will keep it attached” lacks engineered basis. I assure you, I can design a cabin cover and hinge to do the job. It just may not be cost effective. I need to know more about the design basis of the cabin cover first.

Okay, I am not an engineer but it seems intuitively obvious that a door installed per plans will fail at the hinge / fiberglass interface (or the hinge) given the twisting/shear forces on an open door in flight. My comment had *some* hyperbole but not much.

If you move the hinge then that is a different. I had a PA28 for 19 years and had to fix a partially open door in flight on more than one occasion. It was no bid deal. Hinges on the top of the door are a completely different kettle of fish.
 
True

"...Okay, I am not an engineer but it seems intuitively obvious that a door installed per plans will fail at the hinge / fiberglass interface (or the hinge) given the twisting/shear forces on an open door in flight..."

True, however, the designer likely assumed that the door would be closed during flight...If that was the initial assumption, then the current door/hinge assembly design is adequate.

Is it a realistic assumption that the door would be closed during flight? Likely.

Could the door physically come open in flight if both pins were engaged? Not likely.

Now add the human element. This is where it gets harder.

Could someone unintentionally miss engaging one of the pins? Absolutely.

So, now we seek a solution. Re-engineer the cabin top structure for a forward hinge or add a light system and safety latch to alert the human...As no one would intentionally fly with the door open, it would seem that a safety reminder light, and an additional latch would be sufficient to mitigate the risk...without a complete re design of the cabin top and door assemblies.

It will always be a compromise but one thing is for sure, when a human is involved, anything can happen...
 
Could someone unintentionally miss engaging one of the pins? Absolutely.
I still don't get how this can happen. My top and doors are the pink ones and the doors are assembled per the kit instructions save for the flush handles. No special stiffening and it takes significant force to twist the doors enough to latch with only one pin. Missing both pins is easy, though, I suppose.

After seeing the bond come apart in the OP's photos, I realize that it is merely a surface-to-surface bond, nothing mechanical or truly laminated about it. I am thinking of drilling around the perimeter and refilling the holes with flox to make some glass rivets to strengthen the bond on my kit. It wouldn't keep the door on the aircraft but it might keep a door from coming apart due to age or recurring stress of flight.
 
I’m thinking along this line Bob. The G3X has the ability to do canopy/door open alerts based on engine RPM; combined with 180 latches probably solves most of the issues, but what about taxiing with door unlatched, maybe even held partially closed, on a gusty day?

Any cure that relies on human intervention is subject to error.

After a Rhode Island Air National Guard prop blast ripped off my RV-10 pilot door as I was fueling, I created the golden rule:
- Unless the plane is in the hangar, someone is getting in or out, the doors are always shut and locked.

That mean no engine start, no door ajar for taxi, no leaving the door open to fuel, etc.

When not locked these doors are fragile. Even with the door shut but not locked there is risk - as in my case the prop blast bounced the door enough to get under it and away it went.

The door hinges snapped like toothpicks but the hinges (halfs) remained attached to the cabin top and door. No damage at all. The hinge attach screws where installed with #10 tinnermans - as where the other door hardware.

The upside for me was the door did a perfect flip over the canopy, an edge hit the right wing and put a dent in the skin, then landed in the grass. The door had a corner dinged up but other than that is was good to go. I got a couple of hinges from Van’s and flew home.

The Air National Guard eventually paid my claim and although the wing was repairable I built a new wing. The door required some fiberglass work and paint.

So keep the doors shut and locked!

Carl
 
....After seeing the bond come apart in the OP's photos, I realize that it is merely a surface-to-surface bond, nothing mechanical or truly laminated about it. I am thinking of drilling around the perimeter and refilling the holes with flox to make some glass rivets to strengthen the bond on my kit. It wouldn't keep the door on the aircraft but it might keep a door from coming apart due to age or recurring stress of flight.


There's a considerable difference between a bond that has properly-prepared surfaces and one that does not. The doors in the OP were clearly not properly prepared. If prepared according to the plans and Chapter 5, this particular mode of failure is not critical.

"Glass rivets," or more properly, epoxy rivets, are not inherently strong. First, epoxy itself is a lot weaker than the glass/epoxy laminate or even a cotton/epoxy blend, which is what's suggested above. Second, every hole reduces the bond area available for strength. Follow the plans here.

Dave
 
There's a considerable difference between a bond that has properly-prepared surfaces and one that does not. The doors in the OP were clearly not properly prepared. If prepared according to the plans and Chapter 5, this particular mode of failure is not critical.

"Glass rivets," or more properly, epoxy rivets, are not inherently strong. First, epoxy itself is a lot weaker than the glass/epoxy laminate or even a cotton/epoxy blend, which is what's suggested above. Second, every hole reduces the bond area available for strength. Follow the plans here.

Dave

What you say is true but there is also a difference between a laminating bond on 'green' expoxy and one on fully cured epoxy, even if the surface was prepared with peel ply or abrasive surface preparation. Also, there is a difference between epoxy rivets and glass rivets. Flox will stabilize the epoxy quite a bit. Better still would be to put some glass fiber or carbon tow through the hole before the epoxy.

Honestly, I'm not that worried about my doors' bonds because there's not much flex and so little opportunity for the epoxy to craze and delaminate. But the apparent flexibility of other builders' doors does have me thinking a bit.
 
A little O.T., but when I rode as safety pilot in a friend's -10 today, I noticed that his door bowed enough in flight to see daylight beyond the McMaster seal at the upper forward corner of the door. This was true of both doors. He has the Plane-Around latch and did a god job fitting his doors. A little investigation with the back of my hand confirmed that air was leaking out of the door opening in this low pressure/high velocity flow region of the cabin top.

When I go up again I will look for the same phenomenon in my -10. I used a different width of McMaster seal and was careful to angle my pins and locate the holes so that the latches pulled the door downward as well as inward as the pins seated in the door frame. My door skin might be under more tension top to bottom than my buddy's doors. I'll report back. Interested to hear if this is a common finding in the field.
 
doors

I have not notice any air leaks around my doors.

Plane around latch and McMaster seal, 1/4" flange and 3/8" bulb...
 
I have not notice any air leaks around my doors.

Plane around latch and McMaster seal, 1/4" flange and 3/8" bulb...

Same here. Mine is very rigid when locked and I get ZERO draft through the door seal area and I have the small 7/32 bulb. I suspect this could be variable based upon how the door parts are glued up or how the pin holes are located.

Larry
 
The G3X has the ability to do canopy/door open alerts based on engine RPM

My AF-4500 does the same thing. When above idle RPM the audible and visible "CANOPY" warning goes off if any of the 4 door pin sensors are not indicating "Closed". I needed to add two relays to Van's door sensor kit to provide the canopy signal to the AF-4500.

It's pretty hard to ignore an audible "CANOPY" warning. (Stock Van's door system with Van's safety latch)
 
Nice discussion and a lot of good points made, but my reason for posting was to point out what is possibly happening out there and to try and steer builders to a mindset of “let us keep to the construction manual” and not to “this wonder glue will also work”. Ask someone if you are not sure. Maybe the thread title should have been “Follow the instruction manual”.
 
Epoxy or polyester?

Are the 10 canopy shell and doors epoxy or polyester resin? If polyester is it made with wax or without? Either way would require good surface prep to reach the cured resin.
 
Following up my earlier comment

I checked for myself in flight yesterday: my doors are good and tight in flight, no gaps opening up in the seals under air loads. My handles are a good bit stiffer than my friend's doors that have the gaping phenomenon, too. His handles you can close with two fingers; mine take a good bit of "oomph." I oriented my locking pin bevels to face up and out so that as they go into the holes they cam the door bottom down as well as in. I didn't notice how he did his pins. My guess is many of the steps I went through to fit the doors and seals because of advice here, he wasn't even aware of the need for.

I believe the stiffness of my locking handles is a necessary part of how securely they secure the door and I'm glad now that I did them this way. That's part of the difference between building with access to the VAF hive-mind vs. not being part of this community (my friend is not a member here). You guys have kept me from "stepping in it" many times, and I'm grateful.
 
Back
Top