Bolt layout for chain plates

[powerabout]

Guys good engineering says that a long thin chain plate (onto a ply bulkhead) with pairs of bolts in line down its length are poor design yet most boats have this.
Anyone ever put this in an FEA package to see what would be better?

Perhaps a chain plate that is an upsidedown Y shape?

[alan white]

It would be a waste of energy, I think, to concentrate on the chainplate shape. For example, good engineering would be to thicken the plywood to double to a few inches from the metal.

Alan

[BHOFM]

I don't have the knowledge to give all the figures, but the
plates are like 100 times as strong as the cables, and if they
had 12 bolts in each one, they would still tear out what ever
they we bolted to anyway!
Or just pull the frame up through the deck! Or push the mast
through the keel!

If you have doubts, add a second set of stays! Let it break
something we can fix, not destroy the boat!

[Knut Sand]

Check this:
http://www.sponbergyachtdesign.com/A...ngineering.htm

scroll down to fig 7.

He states that the main stress are caused by shear forces in the bolts.

He's right. He also mentions the other issues to consider, stress in the smallest cross-section of the plate, and the shear forces for pulling out the bolt at the top. (He also has drawn the bolts in a zigzag pattern to increase the stressed area, with horizontal bolting, in pairs, you will reduce the cross-section of the area in the chain plate that take the stress, which will make a weaker design).

(hope he doesen't mind me putting a link to his side).

Also the reason why they run like that, pretty much straight down, is probably to (normally) take advantage of hull reinforcments inside (frames). If you have a hull skin, with enough thickness to take the stress, you might go for an upsidedown "Y" shaped chain plate.

So, in my opinion a straight simple plate like normal along the line of attack for the force from the stays, will be the easiest solution to engeneer and also to manufacture.

Critical parts: these should be considered as pretty critical... also the wires.... I've heard this "bonk-pink-swusj" sound once (followed by more sounds) and do not plan to hear it again, and will do any needed action to avoid that..... Included selling (sold) the sailboat....

[powerabout]

Knut
True enough but I'm sure all the B-ENG's out there would say after the first 2 bolts that are usually perpendicular to the stay and in the same plain athwart ship, the rest do almost nothing and then hence neither the rest of the plate.
If you witness a failure it seems quite easy to pull a slot in the bulkhead when the ply has lost its original strength especailly as the ply failure corresponds with the hole through the deck for the stay attachment hence my surmation that a Y type plate would in most cases place bolts in an area of bulkhead that would most probably be in better condition.
I just thought BoatDesign is a good place to start a discussion on this.

I wonder if out readers are aware the last AC boats had neither keel nor deck stepped masts and they hung in space on a beam that was held by the shrouds so there was only the healing force applied to the hull and the rig load was not applied to any part of the hull.

So there is a whole lot more to chain plates than the average stone age yacht designers would have us believe

[Knut Sand]

Powerabout,
I do not claim to be an expert....

If we start at the top, going down, the wire will be let's say 8 mm thick, high quality, certified. The rest of the components will normally not be certified....

I found some data at Wirator (121-049-0800) for 8 mm 316 wire, cross section area is 30 mm2, rest of the table I didn't understand, but it's not unusual to say that 316 wire has a yield of approx 500 MPa (N/mm2). That would say that the wire can take approx 15000 N stress (max, yield, no factor of safety). the rest of the design should be better or equal.
So; the bolt in the top of the chain plate shall handle that, Ok normally there are not supplied with a certificate, let's assume 200 MPa max allowable stress, shear stress should then again be reduced, Normally (I use) a factor of 0,7 for shear stress. 200 x 0,7 = 140 Mpa.
15000/140/2= 54 mm2 needed area; 8,3 mm2, that mean that a bolt there with dia 10 mmm will do ok.... (if you'r unlucky with the shape and the gap widens, you may get bending, and shear over one cross section may ruin your day, but hey.... blind men do normally not sail regattas).
The chain plate, cross secion for shear (pulling straight up/ out) should also have at least 2 x 54 mm2 area to work with. The connection piece (Sjakkel) dont recall the name, but short version, will need a geometrical shape of that of twice the hole, I.e. 20 mm plate, that is 3 mm plate minimum. 3 mm plate is often difficult to get hold of, so 4-6 mm is better, and will keep the shape better... 316 is much like juicyfruit, easy to bend out of shape....
Now, to the holes fixing this **** to the ply/ hull: As for the bolt on top, a total cross section of 15000/ 140 = 107 mm2 (for the bolts) will do fine, so; the cross section in the threaded area of a bolt is smaller than the bolt diameter, M6 will have a cross section of approx 5 mm M8 will have 6,8. That give us a minimum no of bolts:
M6: 107/ 19,6 = 5,5 i.e. 6 bolts.
M8; 107/ 36 = 2,9 i.e. 3 bolts. 3 is bad, rotation around the center bolt will not keep against anything....
Then you should consider the side pressure in the ply, if you use 3 M8 bolts, 15000 N/ 3 = 5000 N on each, add some uncertainty due to uneven hole pattern... 500 kg sideways minimum, hammering, pounding, waves...
If one of these bolts get too much sideways force, it can start bending, add that stress to the shear forces, it'll snap, the two remaining will then immedeately be over their limit.

The factor of safety here sucks.... I'd multiply by 2, at least. I guess....

[Knut Sand]

Forgot the other factor here...
What part do we want to give, in case an stupid ass fixes this mast. When tightening the wires, he'll probably be placed face down, on his knees, face above the chain plate. If the wire snaps, chance are, it'll probably whistle past his ear in a hurry, if the chain plate give way......

Next thing a lawyer calls you.....

btw.... I'm a conservative guy, when it comes to things like this.....

[powerabout]

how did we end up with an M8 bolt shear being less than an 8mm wire in stretch?
You didnt mention which type of wire?

It looks like we need to have a ratio of wire size to chain plate bolt size to bulkhead thickness, but this gets complicated depending on the lamination on the plywood bulkhead.
Are there any standards for this?

[BHOFM]

Let us look upwards for guidance! No,, not him, the top of
the mast, where the cable is held by a plate, with one, two
screws?? Does anything need to be stronger than this point??

The only failure I have ever seen was at the top of the mast,
due to loose rigged and going over chop and the mast was
whipping violently back and forth over each wave! The sails
were not up, it was under power at the time!

[Knut Sand]

Originally Posted by powerabout

how did we end up with an M8 bolt shear being less than an 8mm wire in stretch?
You didnt mention which type of wire?

It looks like we need to have a ratio of wire size to chain plate bolt size to bulkhead thickness, but this gets complicated depending on the lamination on the plywood bulkhead.
Are there any standards for this?

Think I mentioned Wirator, and the part no, just as a wild guess example.

I assumed that the wire had a (test) certificate showing at least a yield of 500 MPa, not unusual (www.matweb.com, type 316 threads). I assumed that we had no (test/ material) certificate for the chainplate/ bolts, neither so unusual, but the world is improving..., hence the chosen max yield of 200 MPa (200 MPa is very low, I would have no problem using real certified values, if they're proven, but then... Ok I' am conservative kind a fella..).

And then, stressed available cross section area of the wire is approx 30 mm2 (It will not be 50 mm2; there's some air in there, between the threads).
Available stressed area for an M8 bolt should be considered for the minimum cross section, (if you manage to place the shear area in a full diameter area of the bolt, take that as a bonus, someone, will probably later replace the loosened bolt with one that have full threads.....).

Internal tension for shear forces will he higher than if the material is exposed for clean pulling forces, picture that the force will attack in the max unlucky directions, that's 45° to each side, in the molecules(?)/grains in the material, allowable stress for shear forces must then be reduced by a factor of 0,7...

Area for a Ø8 bolt no threads; 50,27 mm2
M8, Reduced area due to the threads; 36 mm2, these bolts are placed 90° in the angle of the wire, shear stress. (i.e. 200 MPa x 0,7)

Life has so far learned me this; wires for things to lift heavy things with, are delivered by concerned people, they deliver with certificates, material and testing...and they charge you for it.... On the assembly line/ shop floor, there's always a monkey (foreman or "craftsman") pulling out parts and material of uncertain origin and quality... If the monkey don't show up, on the shop floor, he'll pop up later during some maintenance session of the product.... In some designs, I'll consider this monkey factor. That's my right.

"this gets complicated depending on the lamination on the plywood bulkhead.
Are there any standards for this?"

Ply, if lamination/ fibers are unlucky orientated, you will have less strength than if the fibers are orientated for optimum stress.

Standards 1; I believe that ABS do have some standards for rigging details. Many of their standards are available for download, you'll have to register and cope with a plenty "Welcome to ABS messages" on the email afterwards, but it’s at least free.

Standards 2; I believe that's why it can pay to contact a naval architect with a full drawing/ spec, and get input, Sponberg seem to be pretty serious.

Btw; Nobody has ever asked me to design something that's meant to be airborne/ fly.... Strange....

[Knut Sand]

Originally Posted by BHOFM

Let us look upwards for guidance! No,, not him, the top of
the mast, where the cable is held by a plate, with one, two
screws?? Does anything need to be stronger than this point??

The only failure I have ever seen was at the top of the mast,
due to loose rigged and going over chop and the mast was
whipping violently back and forth over each wave! The sails
were not up, it was under power at the time!

"Does anything need to be stronger than this point??"
No, but can we assume that the angle of attack will be as optimal as this for all designs when we're closer to the deck? If we make a design with close to optimal angle of attack for all situations, I seen no problem with this. But if bending forces occur, its can be a problem.

"The only failure I have ever seen was at the top of the mast,"
Well then we're pretty covered... On a 24' "daysailer" (GRP) Sailed Solo, I had the experience that the holding point for the aft stay, in aft deck area said "thank you for this journey", I didn't actually see it happen, as I was looking forward.... Got actually a bit stressed there for a moment......
We can safely conclude that most rigging parts are pretty critical.

[powerabout]

Anyone know where I can get shear strength of plywood with bolts?

[Tad]

Wood is neither homogeneous or perfect, mechanical properties vary with species and grain direction. Ply is great for engineering because it's dimensionally stable, but poor because only half the grain will be aligned with your load. Thus a bulkhead laminated of diagonal layers is better than the 0-90 degree ply.

Average mechanical properties of various woods are available from a few sources, and there is variation between them, depending on who and how testing was done. A reasonable table of properties is presented in the Gougeon Brothers book. The Wood Handbook, from the US Department of Agriculture is a somewhat dated source as well.

The first thing you need to know is wood species, often marine bulkheads are one of the mahoganies. We use 4000-5000 psi (compression parallel to the grain) as a maximum for mahog. Cut that in half as a working load. Then remember only half the ply thickness is doing the work, ignore the other half. All the load must be born by the first part of the bolt next to your plate. We use one bolt diameter of depth for calculations.

Bolts should not be in a line, spreading them in a diamond pattern with none directly aligned is better. See LF Herreshoff mast fittings for an example of good practice. No bolt should be closer than three diameters to another.

You can increase the bolt bearing surface with epoxy annulus's (oversize hole filled with epoxy and threaded for bolt) or with metal "donuts" that increase the surface area of your fastener.

[powerabout]

Tad

Yes it all makes sense, how come I rarely see it?

Powerabout

[Tad]

I should also mention that I like to see a safety factor of 3-5 for the chain plate (attachment to the boat) over the breaking strength of the wire/rod rigging. I want the rig to fall down and not have to worry about the chainplates at all....ever!

Of course this is for cruising boats, In a racing boat or multihull you might go down to a safety factor of 1.2-1.5.