Plywood Gusset Plates on Lumber Rafters

[woodman1967]

Hello all

I realize there are similar question asked previously but I think this one may be a bit different. I have been asked by a manufactured housing company to design a plywood gusset plate for a rafter (not a truss). The rafter is 12/12 and spans 7' (its a small module). When designing a gusset plate for a truss you don't usually take moments into account as the members of a truss are analyzed for tension/compression forces only.

Can this assumption be used for a rafter as well, my feeling is no. How would moments be accounted for in the gusset plate design?

Also, different from trusses where there is no outward thrust, will there be an outward thrust from this rafter? In this case the rafter is hung from an LVL ridge beam.

Any help would be appreciated.

 

[CJJS]

What is the gusset plate connecting? If it is hung from a ridge beam and bearing on an exterior wall, I am not sure what you are trying to connect.

 

[Einsteim]

Exactly what are you connecting? Is there a collar tie or ceiling joist?

 

[msquared48]

I am confused too - Are you trying to splice two rafters together with gussets to maintain the moment capacity of the rafter member?

Regarding the second question, technically yes, but this can be averted by notching the rafters at the head and tail for flat bearing at the ridge beam hanger, and at the wall top plate. You can also avert the force by installing a metal strap across the top of the ridge beam holding the opposing rafters together.

Mike McCann
McCann Engineering

 

[woodman1967]

Sorry forgot to add that the rafter itself will be cut at 24" for shipping. Onsite the rafter will spliced together again with the plywood gusset. It will similar setup to splices of a standard truss which are connected with metal plates. So the rafter will be spliced at 24" then connected with the plywood gussets.

Thanks again

 

[Einsteim]

What do you mean "cut at 24""? 24" from the bearing end or from the rigeline?

Still confused in Michigan....

 

[woodman1967]

Sorry, I guess I am not being clear (my wife would agree

Anyway, the cut at 24" is measured along the length of the rafter stating from the bearing end and not the ridgeline.

Thanks again

 

[Lion06]

That is an awful lot of splices for a beam. I would shy away from this. One of the problems with this kind of connection in wood si that even if you can get it to work for strength it will likely not have the rotational fixity you are looking for. The creep effects of wood will cause the moment rotation curve of the connetion to slowly move down toward a semi-rigid connection, maybe further. This could result in larger than expected deflections. Why does it need to be spliced at 24"? How long is it?

 

[woodman1967]

OK, this rafter is spanning 7 feet and is 12/12. There is only one splice at 24" from the outside bearing wall (not the ridge). The reason for the splice is the roof is partially built at the factory then completed on site.

Thanks again everyone, I appreciate the interest so far.

 

[HouseBoy]

It seems that the splice connection will definately need to be designed for the bending moment. It might be easier to lap two pieces of rafter. 7 ft. span is not too great so maybe the bending is low. How deep is the rafter?
Is there a ceiling joist or is it a vaulted ceiling space?
Can you post a picture or sketch?

Also, not sure how/why this "modular" or panelization works. How is the roof "panel" configured?
Again, pics or sketches would be ideal.

 

[woodman1967]

Ok, I will try to clear this up some more. I have attached a PDF showing what we use for most situations (you will see that the splice is made basically at the bearing point). The second cross-section shows what they want me to design, a plywood gusset to splice the rafter at 24" on one and 36" on the opposing rafter. I haven't seen much info on designing a nailed plywood gusset plate for moments just for lateral loads. What is the consensus on this?

 

[HouseBoy]

I don't think the system is stable without bending capacity being provided in the plywood gusset.
I also wonder about the stability (bracing) of the 4 play trusses. It doesn't look like the rafter 'extensions' would do much to brace the compression chord of these trusses. Perhaps it will be a simple matter to improve this but...don't overlook it.

If the 4 ply trusses provide sufficient vertical capacity (and they are braced adequately) I would think the bending in the plywood will be maneagable but there will be bending and I would guees that the plywood plates will need to run up on the rafters much more than you show so you can develop bending wiht the connection to the rafter..

 

[woodman1967]

Thanks Houseguy, but I still may not be clear. The first drawing isn't really part of the question, it is attached to show how the plywood gusset had been designed for another project. My question pertained to the second page of the attachment where the splice is located away from the bearing point. I am wondering about designing it to handle the moment at the points specified. Also, is there a good place to find info on desiging a nailed gusset plate (w/ adhesive if necesary) to handle moment forces.

Thanks as always

 

[CJJS]

Have you ever considered using roof trusses with hinge plates?

http://www.mii.com/site/frameset.aspx?siteid=32&section=3722

 

[woodman1967]

Hello CJSchwartz

This company uses hinged plates for all their trusses. But, this is a rafter and the hinge plates won't work. They actually use hinge plates in this case only to hold the folded roof together...once it is on site the roof is rotated into place and then the splice would have to be reinforced with a plywood gusset.

 

[HouseBoy]

Here are some thoughts to get the discussion going. I'm not trying to be 100% complete necessarily and some of simplifying assumptions may not be universal but they may be useful to start discussion.
I would look at the bending in the rafter at the point of the splice.
The plywood gusset will need to develop that bending moment on the fastener group on EACH SIDE of the splice.
For bending in plywood I have considered material thickness of one helf the plywood thickness and calculated section properties for that. You might need two plays or something similar. Often when there is a decent amount of bending, I find it better to use a scab piece with greater length (over which to design the moment in the faster group.

 

[OldPaperMaker]

Woodman 1967,
You have at least 2 major problems to deal with:
1. The rafters will have a shear, moment and compressive load at the splice. The forces on the nail or bolt couple in this connection will probably be very high.I would never try to make a connection like this with a wood member.
2. It doesn't look like there is ridge beam on this building. Therefore,the structure itself is unstable when the "temporary ceiling framing" member is removed. You will have a hinge at the top of both walls and at the ridge. There is nothing to keep the walls from spreading out when snow or wind loads are applied to the rafters.

 

[Stillerz]

I think it was stated that there is a Double LVL ridge beam which, when designed properly, should eliminate the need for collar ties.
In general, the whole idea of this splice is terrible and will result in a horribly constructed house with virutally no craftsmanship involved whatsoever.
You can "engineer" just about anything you want. I could run an FEA on a shoe box to determine it's strength, but we all KNOW a two year old girl can step on it and crush it.
This is not practical. Any carpenter in the field constructing this house would scoff at this detail. It is fundamentally unsound and any "design" of this splice connection would be based on horible construction techniques.
Garbage in -> Garbage out.

 

[HouseBoy]

To Stillerz:
I do not agree that it is a GI/GO scenario (necessarily). Also, even with a suitable structural ridge, there will be bending and shear at the splice loation.

I think that ultimately the simplest solution might be to add a "sister" to the rafters in order to achieve sufficient lap so that a connection with moment capacity can be made as well as to provide a member with sufficient and knowable bending capacity at the splice location. That wasn't the original question though, so.....

 

[GT1878]

I'd have to agree, a scab rafter extending the entire length is your best solution. I don't think you have enough room to develop the bending into the gusset them back out of the gusset into the rafter at the wall bearing side.

Truss repair doesn't directly address bending in the plywood gusset in alot of cases. Min guesset sizes and min nailing spacing in truss chords based on testing is how the bending is embedded into the design