Re: Cobra top hinge plate bolt failure

[Steve Koerner]

On May 18, 11:45*am, JJ Sinclair > wrote:
> On May 18, 10:23*am, Andy > wrote:
>
>
>
>
>
> > On May 18, 9:44*am, Steve Koerner > wrote:
>
> > > I have 3 Wing Riggers that I need to get shipped out today, but as
> > > soon as I get a chance I intend to investigate changing the bolts to
> > > the next larger size, 3/8" grade 8. *Of course that will mean drilling
> > > the holes larger and it will also mean that the head will not be
> > > neatly counterbored into the handle. *If that plan works out, I will
> > > post back here.
>
> > > GW
>
> > I'd be interested to hear how that works out. *There is a least one
> > spacer block inside the frame extrusion and it looks like it would
> > have to be extracted, drilled, and replaced. *There may be two each
> > side. *Given the limit access to the extrusion opening it could be a
> > bit awkward to work the inner blocks.
>
> > If I was going to larger bolts I think I'd relocate the handles and
> > keep them on 5/16 socket cap bolts. *They wouldn't need the spacer
> > blocks as long as a moderate torque was used.
>
> > Andy (GY)
>
> The plot thickens, I moved both wings aft in order to get a good look
> at the forward corner of the fiberglass top. The top has an aluminum
> rail where it meets the lowar trailer and this "L" member overlaps the
> cross member (where the hinge plate is bolted). There are 4 large
> aluminum rivets on each side that bind these members together. Well,
> all 4 rivets on the right side were sheared off and 2 on the left side
> were sheared off! The forward 'shove' from the struts would try and
> shear off these rather weak fastners as they are loaded in "shear".
> The thing I don't understand is why the large steel bolts are shearing
> off, because they are not loaded in shear?
> I replaced the corner rivets with 1/8" steel rivets.
> JJ- Hide quoted text -
>
> - Show quoted text -

JJ: I don't think that the big bolts are shearing off. They are
breaking due to a gradual fatigue process in tension. What I'm
thinking is that it is the fore and aft forces due to road bumps and/
or washboard acting against the inertia of the top that is putting the
cycled load on the bolts that eventually is their unduing. The
fatigue thing was definitely confirmed by Howard Banks based on the
observations on Andy's broken bolt. When you said yesterday that you
checked your bolts, did you really check them for incipient fatigue
damage?

Very interesting that there are sheared rivets going on as well. I'll
have to look for that too.

GW

[JJ Sinclair[_2_]]

.. * When you said yesterday that you
> checked your bolts, did you really check them for incipient fatigue
> damage?

I didn't remove and inspect the bolts because I was afraid the struts
would move the top, but thinking about it now, I don't think the top
would move if it was fully open and struts extended all the way. I'll
give it a try................If you hear a loud scream from no cal, it
means the top moved on me! Dave wants pictures, but some of us don't
know how to post pictures along with a whole bunch of other things
having to do with computers!
JJ

[Andy[_1_]]

On May 18, 2:22*pm, JJ Sinclair > wrote:
> . * When you said yesterday that you
>
> > checked your bolts, did you really check them for incipient fatigue
> > damage?
>
> I didn't remove and inspect the bolts because I was afraid the struts
> would move the top, but thinking about it now, I don't think the top
> would move if it was fully open and struts extended all the way. *I'll
> give it a try................If you hear a loud scream from no cal, it
> means the top moved on me! Dave wants pictures, but some of us don't
> know how to post pictures along with a whole bunch of other things
> having to do with computers!
> JJ

The top will move with the struts fully extended and the bolts
removed. Even with one hinge outboard bolt removed that corner of the
top rises and it look 5 of us to align the parts to get the
replacement bolt in. (people were available and clamps were not)

Please clamp the parts together so they can't move and then only
remove one bolt at a time.

Andy

[JJ Sinclair[_2_]]

> The top will move with the struts fully extended and the bolts
> removed. *Even with one hinge outboard bolt removed that corner of the
> top rises and it look 5 of us to align the parts to get the
> replacement bolt in. (people were available and clamps were not)
>
> Please clamp the parts together so they can't move and then only
> remove one bolt at a time.
>
> Andy

Yep. the inboard bolts came right out and looked fine, but the
outboard bolts didn't want to come and I wasn't inclined to force the
issue. A gap between the hinge plate and cross-member did appear as I
loosened the nut. I believe the outboard bolts fail after the bolt is
loose and repeatedly banged around (up & down + for & aft). Once the
outboard bolt gives way, the inboard isn't long for this world. I
think we're OK if the bolts are tight and kept tight.
Cheers,
JJ

[Andy[_1_]]

On May 18, 12:30*pm, Steve Koerner > wrote:
> On May 18, 11:45*am, JJ Sinclair > wrote:
>
>
>
>
>
> > On May 18, 10:23*am, Andy > wrote:
>
> > > On May 18, 9:44*am, Steve Koerner > wrote:
>
> > > > I have 3 Wing Riggers that I need to get shipped out today, but as
> > > > soon as I get a chance I intend to investigate changing the bolts to
> > > > the next larger size, 3/8" grade 8. *Of course that will mean drilling
> > > > the holes larger and it will also mean that the head will not be
> > > > neatly counterbored into the handle. *If that plan works out, I will
> > > > post back here.
>
> > > > GW
>
> > > I'd be interested to hear how that works out. *There is a least one
> > > spacer block inside the frame extrusion and it looks like it would
> > > have to be extracted, drilled, and replaced. *There may be two each
> > > side. *Given the limit access to the extrusion opening it could be a
> > > bit awkward to work the inner blocks.
>
> > > If I was going to larger bolts I think I'd relocate the handles and
> > > keep them on 5/16 socket cap bolts. *They wouldn't need the spacer
> > > blocks as long as a moderate torque was used.
>
> > > Andy (GY)
>
> > The plot thickens, I moved both wings aft in order to get a good look
> > at the forward corner of the fiberglass top. The top has an aluminum
> > rail where it meets the lowar trailer and this "L" member overlaps the
> > cross member (where the hinge plate is bolted). There are 4 large
> > aluminum rivets on each side that bind these members together. Well,
> > all 4 rivets on the right side were sheared off and 2 on the left side
> > were sheared off! The forward 'shove' from the struts would try and
> > shear off these rather weak fastners as they are loaded in "shear".
> > The thing I don't understand is why the large steel bolts are shearing
> > off, because they are not loaded in shear?
> > I replaced the corner rivets with 1/8" steel rivets.
> > JJ- Hide quoted text -
>
> > - Show quoted text -
>
> JJ: I don't think that the big bolts are shearing off. *They are
> breaking due to a gradual fatigue process in tension. * What I'm
> thinking is that it is the fore and aft forces due to road bumps and/
> or washboard acting against the inertia of the top that is putting the
> cycled load on the bolts that eventually is their unduing. * The
> fatigue thing was definitely confirmed by Howard Banks based on the
> observations on Andy's broken bolt. * When you said yesterday that you
> checked your bolts, did you really check them for incipient fatigue
> damage?
>
> Very interesting that there are sheared rivets going on as well. *I'll
> have to look for that too.
>
> GW- Hide quoted text -
>
> - Show quoted text -

I think there is significant shear loads on the bolts. The top of the
trailer mates to the lower half with a rubber seal. The top has
considerable mass and inertia. Any time the trailer floor moves up
and down, as on a washboard road, the vertical acceleration is
transferred to the top by the gap seals which will yield and then by
the forward hinges. There will therefore be a cycling shear load on
the hinge plate bolts any time the trailer moves over a rough
surface. Given that the plain shank of the bolt is not long enough to
engage in the rear wall of the front frame extrusion, or the spacer
plate, or the hinge plate, it's really not surprising the bolts fail.

Another factor is the the bolts compress the thickness of the 2 layer
grass top. Over time the top yields reducing the bolt clamping force
and allowing relative vertical motion between the parts that were
intended to be clamped together, but now are not.

I found deep compression of the glass outer shell under the handle
ends where the bolts failed. This could have be avoided by using a
solid spacer inserted into the shell.

Andy (GY)

[Steve Koerner]

Andy:

I certainly agree that once the bolts become loose they will be seeing
shear load. I also agree that compressing the fiberglass shell over
time is a certain way for them to get loose. That sounds like a
pretty good theory. Of course, once they get loose, they'll be under
increased tension cycling too. As JJ has said, the key thing may be
to not let them get loose.

I can recall that my left rear latch had become a bit troublesome
prior to the failure. I now realize that meant that the bolt was
probably loosening in the left front.

My latest observation is that the hinge plate bolts on my trailer are
marked 8 8 which is not a high strength bolt. So I'm now thinking
that my corrective action will be to upgrade to high strength and
periodically monitor those bolts to keep them tight.

GW

[JJ Sinclair[_2_]]

> I can recall that my left rear latch had become a bit troublesome
> prior to the failure. *I now realize that meant that the bolt was
> probably loosening in the left front.

Yep, nothing new here, my ash-25 trailer was doing the same thing, it
just didn't get to the bolts before I sold it to John in Australia
(John, check your bolts). I do remember the latches had moved forward
as my present latches have, Andy is right, the forward pressuer from
the compressed struts will shear the aluminum rivets then apply
pressure under the hinge bolts which will degrade the fiberglass or
gall the aluminum top which will lead to LOOSE BOLTS over time. Loose
bolts = sheared bolts! My trailer closes and latches better, now that
I have replaced the sheared off aluminum corner rivets with 1/8" steel
rivets. BTW, I can see a slight forward bend in the right side of my
fiberglass top just outboard of the hinge plate.
Cheers,
JJ

[Eric Greenwell[_4_]]

On 5/19/2011 7:04 AM, JJ Sinclair wrote:
My trailer closes and latches better, now that
> I have replaced the sheared off aluminum corner rivets with 1/8" steel
> rivets. BTW, I can see a slight forward bend in the right side of my
> fiberglass top just outboard of the hinge plate.

My metal top Cobra trailer has had a forward bend on both sides for at
least 6 years, probably longer. I also just noticed the pivot holes for
the top are elongated fore and aft. I may add a bushing to them, which
would move the top back 1/8" or more, and make it easier to close.

--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)

[MKoerner]

I suspect this problem has nothing to do with loads imposed by the gas
springs or bumpy roads or inertia. I think the tension in these bolts
that leads to failure results from “bridging” the trailer across dips
in the road.
Our trailers can be viewed as two beams, the floor section and the
lid, setting one on top of the other. These beams are pinned together
at each end; at the front by the hinge pins and at the rear by the
latch pins. During normal use the lower beam, the floor section, is
loaded along its length by the sailplane parts and supported in the
center by the axel. The trailer sags at each end. The resulting
bending loads put the top edge of the lower section in tension. These
bending loads also pull down on the front and rear of the lid.
Meanwhile, contact with the lower section along the claim shell joint
pushes up on the center of the lid. As a result the lid also sees
bending loads in the same direction as the floor section. This puts
the lower edge of the lid is in compression, not tension. As a result
the front bottom edge of the lid is pulled toward the rear, away from
the mating surface of the lower section. Loads in this direction would
be transferred to the hinge brackets and not to the bolts which clamp
the brackets to the lid. Furthermore, the deflections would probably
be small relative to the play in the pins.
On the other hand, when you tow the trailer across a dip, such as
starting up a steep driveway, the trailer may form a bridge across the
dip, with the back end dragging on the pavement and the front end
supported by the tow vehicle’s hitch. This causes the trailer to sag
in the middle, opposite to the normal loading. This puts the upper
surface of the lower section in compression. The lid is still pinned
in place at the front and the back, but it doesn’t carry much load so
it does not bend nearly as much. This would be especially true with
the fiberglass tops which are probably much stiffer than the aluminum
version. The lid effectively takes a short-cut across the gap as the
lower section sags away in middle. As a result the front bottom edge
of the lid is pushed forward, past the mating surface of the lower
section. Loads in this direction would be transferred through the
bolts, putting them in tension.
Some years back a number of us, led by JJ, myself included, found
substantial cracks on our trailer tongues, opening from the bottom.
Same story: Bridging.
The odd thing about the current problem is the sudden onset of
multiple events… unless you all drove across the same ditch???
Mike Koerner

[Andy[_1_]]

On May 20, 1:44*am, MKoerner > wrote:
> I suspect this problem has nothing to do with loads imposed by the gas
> springs or bumpy roads or inertia. I think the tension in these bolts
> that leads to failure results from “bridging” the trailer across dips
> in the road.
> Our trailers can be viewed as two beams, the floor section and the
> lid, setting one on top of the other. These beams are pinned together
> at each end; at the front by the hinge pins and at the rear by the
> latch pins. During normal use the lower beam, the floor section, is
> loaded along its length by the sailplane parts and supported in the
> center by the axel. The trailer sags at each end. The resulting
> bending loads put the top edge of the lower section in tension. These
> bending loads also pull down on the front and rear of the lid.
> Meanwhile, contact with the lower section along the claim shell joint
> pushes up on the center of the lid. As a result the lid also sees
> bending loads in the same direction as the floor section. This puts
> the lower edge of the lid is in compression, not tension. As a result
> the front bottom edge of the lid is pulled toward the rear, away from
> the mating surface of the lower section. Loads in this direction would
> be transferred to the hinge brackets and not to the bolts which clamp
> the brackets to the lid. Furthermore, the deflections would probably
> be small relative to the play in the pins.
> On the other hand, when you tow the trailer across a dip, such as
> starting up a steep driveway, the trailer may form a bridge across the
> dip, with the back end dragging on the pavement and the front end
> supported by the tow vehicle’s hitch. This causes the trailer to sag
> in the middle, opposite to the normal loading. This puts the upper
> surface of the lower section in compression. The lid is still pinned
> in place at the front and the back, but it doesn’t carry much load so
> it does not bend nearly as much. This would be especially true with
> the fiberglass tops which are probably much stiffer than the aluminum
> version. The lid effectively takes a short-cut across the gap as the
> lower section sags away in middle. As a result the front bottom edge
> of the lid is pushed forward, past the mating surface of the lower
> section. Loads in this direction would be transferred through the
> bolts, putting them in tension.
> Some years back a number of us, led by JJ, myself included, found
> substantial cracks on our trailer tongues, opening from the bottom.
> Same story: Bridging.
> The odd thing about the current problem is the sudden onset of
> multiple events… unless you all drove across the same ditch???
> Mike Koerner

To the best of my knowledge my trailer has never been "bridged". The
rear of my trailer is high and it is towed by a vehicle with a short
overhang.

Steve may have sent you my photos that support my suggested failure
mode. If not, send me an email and I'll copy you.

Andy (GY)

[Steve Koerner]

But Mike, the bolts have failed by fatigue. What you describe is a
rare occurence that would, if the force were sufficent, cause the
bolts to fail at once. I know my trailer had not done any "bridging"
in a very long time as I've purchased taller tires that prevent that.

I have documented my repair process in a PDF file with numerous
pictures. The document includes a design for a backing plate that
goes under the handle. The document also gives suggestions for
clamping the plate while removing the bolts. This document provides a
recommended replacement bolt and a procedure for repairing the likely
damage to the base of the handles as well. Eventually I'll post this
at my Wing Rigger site (which I can't modify at the moment because
I've replaced a computer with the web tools); for now it is posted
here:

http://www.box.net/shared/9dvnt7qgs8

Steve Koerner (GW)

[db_sonic[_2_]]

On May 20, 7:59*am, Steve Koerner > wrote:
> But Mike, the bolts have failed by fatigue. *What you describe is a
> rare occurence that would, if the force were sufficent, cause the
> bolts to fail at once. *I know my trailer had not done any "bridging"
> in a very long time as I've purchased taller tires that prevent that.
>
> I have documented my repair process in a PDF file with numerous
> pictures. *The document includes a design for a backing plate that
> goes under the handle. *The document also gives suggestions for
> clamping the plate while removing the bolts. *This document provides a
> recommended replacement bolt and a procedure for repairing the likely
> damage to the base of the handles as well. *Eventually I'll post this
> at my Wing Rigger site (which I can't modify at the moment because
> I've replaced a computer with the web tools); for now it is posted
> here:
>
> http://www.box.net/shared/9dvnt7qgs8
>
> Steve Koerner *(GW)

nice write-up.

[Tuno]

Unfortunately that box.net doesn't work from behind my firewall, but I
look forward to reading it this weekend.

One thing I've learned with certainty from this thread is that Mike
and Steve fell out of the same tree!

tuN7o

[Eric Greenwell[_4_]]

On 5/20/2011 7:59 AM, Steve Koerner wrote:
> But Mike, the bolts have failed by fatigue. What you describe is a
> rare occurence that would, if the force were sufficent, cause the
> bolts to fail at once. I know my trailer had not done any "bridging"
> in a very long time as I've purchased taller tires that prevent that.

My trailer has a 30' body,a 4' tongue, and carries 850 pounds of glider
(total trailer weight: 2450 lbs). It's towed almost exclusively by a 23'
motorhome. "Bridging" is the trailer's pastime on every trip, as it goes
in and out of gas stations and parking lots. It's bridged hundreds of
times during the 160,000 miles it's been towed, yet the bolts have not
failed. I won't know if they are damaged until I remove them; however,
they are still very tight. Or, maybe the metal top makes the trailer
more tolerant of bridging.

I doubt the tongue cracking a lot of us, including me, experienced
several years ago was due to bridging; instead, the situation and
symptoms were classic examples of fatigue failures induced by welding
stress concentrations.

In any case, the basics of the "bending" theory should be easy to check
by putting jack stands under the rear of the trailer, a floor jack on
the tongue, then jacking the tongue up while measuring the deflection of
the trailer bottom and the movement of the trailer top. My guess is the
deflections will be very small.

--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)

[David Salmon[_2_]]

At 19:51 20 May 2011, Tuno wrote:
>Unfortunately that box.net doesn't work from behind my firewall, but I
>look forward to reading it this weekend.
>
>One thing I've learned with certainty from this thread is that Mike
>and Steve fell out of the same tree!
>
>tuN7o
>

Surely when the top is up the gas struts are at the end of their travel,
and therefore are in effect a solid, weight support, somewhere near the
middle of the top. This would imply very little loading on the hinge in
that position, just whatever the unbalanced load is.
If , as it seems, the problem is fatigue in the bolts, as many last 15/20
years, they cannot be stressed much above the fatigue limit, so a small
amount of strengthening should do it. So my thought was to drill and add a
third bolt from the inside, with the nut and substantial washer inside the
handle. Once in place, the existing bolts could be replaced one at a time.
Can't remember using the handles anyway.
However Steve's is an elegant solution.
Dave

[Andy[_1_]]

On May 20, 7:59*am, Steve Koerner > wrote:
> But Mike, the bolts have failed by fatigue. *What you describe is a
> rare occurence that would, if the force were sufficent, cause the
> bolts to fail at once. *I know my trailer had not done any "bridging"
> in a very long time as I've purchased taller tires that prevent that.
>
> I have documented my repair process in a PDF file with numerous
> pictures. *The document includes a design for a backing plate that
> goes under the handle. *The document also gives suggestions for
> clamping the plate while removing the bolts. *This document provides a
> recommended replacement bolt and a procedure for repairing the likely
> damage to the base of the handles as well. *Eventually I'll post this
> at my Wing Rigger site (which I can't modify at the moment because
> I've replaced a computer with the web tools); for now it is posted
> here:
>
> http://www.box.net/shared/9dvnt7qgs8
>
> Steve Koerner *(GW)

Nice write up but a disappointing solution. I was sure you would have
bored out holes in the glass top under the handles and inserted
machined spacer bosses. Those would carry the bolt clamping force
directly to the front face of the trailer frame extrusion.

I'm disappointed, not because you did it this way on yours, but
because I was hoping you'd make a second set for me ;)

Andy

[Andy[_1_]]

On May 21, 12:10*am, David Salmon > wrote:
> At 19:51 20 May 2011, Tuno wrote:
>
> >Unfortunately that box.net doesn't work from behind my firewall, but I
> >look forward to reading it this weekend.
>
> >One thing I've learned with certainty from this thread is that Mike
> >and Steve fell out of the same tree!
>
> >tuN7o
>
> Surely when the top is up the gas struts are at the end of their travel,
> and therefore are in effect a solid, weight support, somewhere near the
> middle of the top. This would imply very little loading on the hinge in
> that position, just whatever the unbalanced load is.
> If , as it seems, the problem is fatigue in the bolts, as many last 15/20
> years, they cannot be stressed much above the fatigue limit, so a small
> amount of strengthening should do it. So my thought was to drill and add a
> third bolt from the inside, with the nut and substantial washer inside the
> handle. Once in place, the existing bolts could be replaced one at a time..
> Can't remember using the handles anyway.
> However Steve's is an elegant solution.
> Dave

If you decide on the third bolt method don't forget that each existing
bolt has a spacer block inserted inside the front frame extrusion.
Without that spacer the main function of the third bolt may be to make
the other two loose.

You are right about gas strut forces with the top open. The force is
zero and the struts just form one arm of a triangle that would like
very much to collapse. The way it would like to collapse is to move
the front of the top forward and up.

It's much easier to change the bolts before they fail than to get the
parts into alignment after the bolts fail.

Wear marks on my hinge spacer plate indicate the hinge plate had been
moving in an arc about the inner bolt until the outer bolt failed.
This suggests that only the outer bolt is likely to be subjected to
the cyclic shear loading that is assumed to have caused the fatigue
failure.

It took a while to find the nuts and bolt ends. They didn't just drop
on the trailer floor, they were projected over the top of the front
compartment dividing wall into the wing storage area.

Andy

[Steve Koerner]

Come to think of it, my bolt ends were under the wing area too. Mine
must also have projected. It seems to me that would happen if the
bolts failed in tension but would not happen if the bolts had become
loose and were sheared.

The other thing that occurred to me yesterday afternoon is that Andy's
bolts failed at or near the juncture of the threaded and unthreaded
portion of the bolt. That would be a location about .31 inches inside
the front extrusion frame and not where the main shearing action would
be expected to occur due to the plate sliding .

So, although I had accepted Andy's theory just yesterday morning, now
I'm having doubts. I'm back to thinking that the fatigue failure must
me primarily in tension.

GW

[Andy[_1_]]

On May 21, 10:43*am, Steve Koerner > wrote:
> Come to think of it, my bolt ends were under the wing area too. * Mine
> must also have projected. *It seems to me that would happen if the
> bolts failed in tension but would not happen if the bolts had become
> loose and were sheared.
>
> The other thing that occurred to me yesterday afternoon is that Andy's
> bolts failed at or near the juncture of the threaded and unthreaded
> portion of the bolt. *That would be a location about .31 inches inside
> the front extrusion frame and not where the main shearing action would
> be expected to occur due to the plate sliding .
>
> So, although I had accepted Andy's theory just yesterday morning, now
> I'm having doubts. *I'm back to thinking that the fatigue failure must
> me primarily in tension.
>
> GW

The bolts are always in tension with the top closed even if the
clamping force has reduced to zero. I don't think the fact that the
bolts were under tension when they failed is inconsistent with the
theory they had fatigued due to cyclic shear loading.

Maybe to fully understand the failure one would have to know the
diameters of all the holes the bolt passes through. It's possible the
bolt experienced bending rather than true shear loading. The hinge
plate outer hole is 8.8 mm. To the best of my recollection the wear
marks on the hinge plate spacer were 7mm in height at the outer end.
I didn't record that measurement but photo IMGP6653_edited-1 (email
5/20) shows both the hole and the wear marks and 7mm seems reasonable.

Since our bolts failed in different places it's possible the failure
more was different. Unfortunate that you discarded yours. It would
have been interesting to see if the fracture surface looked like mine.

GY

[Tuno]

Nice writeup Steve. You should fabricate and sell repair kits, with
the value-added bolts and backing plates.