Quite Large Repair

[Brad93]

Sadly i can see no way of identifying this. With the far end hanging on a winch over the sea and sideways movement being taken up by two pillars which house the winch 30m above the sea, there is both the loading force acting downwards, lateral force as the conveyor hits the pillars either side (maybe 30cm of lateral movement), and the torsional force of the conveyor twisting due to the suspended nature at one end and single point of movement at the other end (shaft on two bearings).

 

[TechnicAl]

I honestly cant really follow whats happening but a few curious comments on the second photo
Why are the nuts rusty but the bolts and washers are clean?
Is that the crack between the nuts?. If so, it looks straight.....is it?..........Im wondering if it is a crack or is it lack of fusion along one edge (maybe with a crack filling in the gaps) I cant really see.
Why is the corrosion heavier between the nuts?
Im wondering if corrosion is playing a part.......it weakens

 

[normspanners]

My view of that repair is the bolts are far too close to the fracture, the repair plate needs to taper down either side along its length with bolts equally spaced length wise, welding a crack like that will normally break again because the weld leaves a straight fracture point, unless instead of the bolts they are replaced with plug/slot welds which do allow some flexability

again tapering the plate over a decent length.

 

[normspanners]

Why are the nuts rusty but the bolts and washers are clean?

stainless bolts/ washers but plain steel nuts?
looks to me like a fretting fracture hence the general area rust coz the paint has fractured ? due to the metal fretting and then cracked coz the frame is insufficiently supported at that point.

 

[normspanners]

The other problem is the conveyor lashed to the ship? because the ship will never stay still and will always move forward and astern slightly, as well as traveling up and down which the frame is allowing but the frame will be trying to stop the sideways motion ?

we had one close to us


The ship had been loading limestone from a quarry jetty when it was driven aground by powerful waves and smashed against the concrete wave-breakers that lined the shore.

http://www.telegraph.co.uk/news/wea...und-off-Llanddulas-near-Colwyn-Bay-Wales.html
ship had to be cut up and scrapped

 

[500e]

Would agree with NormS regarding bolts too close to the fracture, + plate is to small.
Would have thought plate should have been at least 3\4 times as long with tapered end to help spread loading.
Is this failure a new thing or a historic problem? if so could be design fault rather than fatigue or plying \ overload damage.
Burning through for the bolts where do they find these people still the were cheap I expect.

 

[normspanners]

Also thinking again the repair plate could have a central rib to form a tee to help with sideways stability ie use universal beam with web cut down using the top /bottom as the repair plate and leaving some of the upright as the rib
a little like this v v v

so the flanges they will form the plates that will be bolted on the side of the conveyor and are made from the top and bottom of the beam sorry bout the drawing I'm rushing the verticals and horizontals are the other way round obviously the span would be along the length of the conveyor and the crack would end up being between the centre four bolts from left to right

 

[octo0072000]

Wonder if it was ever designed to have 8ton dropped into the hopper at once
as it's over 20/30 years old that would cause enough shock loads up the length of the conveyor also it's cracked at a bracing point
could the ripple effect and just good old metal fatigue be playing a part in this

i agree with the bracing idea of Norm
but maybe say bolted over plates say 1.5 metre in length but also could the plate be made as a L shape to continue round the bottom and top of the web or not ?
Just plate is not going to stop any twisting forces on the channel

 

[normspanners]

Yeah totally

but maybe say bolted over plates say 1.5 metre in length

it was just I found that generic photo and used that as an example of a cut universal beam
I think the repair will have to done to allow flex of some sort though whilst still maintaining integrity. eg airplane wing or for not a good eg the cracking (pun) liberty ships ?

 

[octo0072000]

Yeah totally

it was just I found that generic photo and used that as an example of a cut universal beam
I think the repair will have to done to allow flex of some sort though whilst still maintaining integrity. eg airplane wing or for not a good eg the cracking (pun) liberty ships ?

the channel section is think like a wagon chassis it allows twist over it's length
hense bolting over another section of chassis rail over to keep the allowable twist
would like to see a picture of the conveyer with the belt off wonder how many other cracks there are in the crossmembers

 

[tinman123]

I have got M24 bolts, with a pair of plain washers and nylocs for this job. That small plate with 8 bolts with normal nuts is all that is holding that side of the conveyor together at the moment so i have no doubts that my plan will hold it together.

see them plain washers,
drill the hole, put in the bolt, launch the two washers off the gantry and watch em hit the water and then run up the nut and tighten.

the washers arent worth the metal they're made from and the nut and bolt will loosen quicker with them on.
your nyloc nut all be it the best of a bad lot isnt going to do much either but its better than nothing all the same.
to guarantee your nut dosent loosen the only man for the job is a nord lock washer, altho i wouldnt really like to be pricing M24 ones all the same.

 

[Sarto]

Wouldn't a drop or 2 of a suitable Lockite tread locker do any good?
Some variants "should" be very hard locking.

 

[tinman123]

threadloc is a massive help alright, id use loctite branded stuff here but the nut will eventually loosen with it too if the conditions are harsh.
not so much that you would notice it mind but enough to loosen the assembly slightly and in an instance like the op's its not one where you want that to happen.
the only thing ive seen so far that wont loosen is the nord loc's.

altho i have seen a nut and bolt system where using a lock nut on a nut, they were guaranteeing that it wouldnt loosen, they had an expensive looking cunning plan where they had two threads of different pitch on the same bolt while at the same time looking like there was only one thread on it.
it looked like a dear option if its even available.

 

[normspanners]

nord loc's.

yep I use them in competition stuff in critical points, but is it the nuts that loosen in this app , it could be movement of the two sections fretting against each other and wearing thus loosening the tension in the same way truck chassis do hence one reason why they are riveted and replacement bolts can and do slacken.

 

[tomsutton529]

I would be looking at some vibration analysis to see if something is exciting the natural frequency and causing the structure to resonate, this can also be backed up with ODS analysis where a 3D drawing is used to show how the structure is moving. if it been there for 30 years then something has changed somewhere.