Hybrid roof diaphragm

[WWTEng]

Please see the attached two sketches. Half the building framing is plywood over light gage trusses and the rest of the half is metal deck on bar joists. I have enough lengths of cmu walls on both ends to get masonry shear walls to work for lateral loads. But the problem is the connection between the plywood and metal deck. I really don't want to use a steel moment frame in the middle and would like to keep the steel frame along the middle as gravity only.

So the question is, can the two types of roof diaphragms be connected in a way that they act as continuous piece from shear wall to shear wall?

 

[jittles]

I would just say do what you would normally do even if they were both wood - provide appropriate attachment of both diaphragms to a continuous collector element below.

Based on your sketch, I would say it is easier said than done. You might need a couple of different elements connected together to make it work, but it seems doable.

 

[msquared48]

I agree. The connection of the steel deck to the steel beam is a cakewalk.

Fore the connection of the wood to the steel beam,, add a 3X nail plate under the wood truss to the top of the steel beam, connected with Nelson sill plate anchors, and add blocking between the top of the steel g=beam and the plywood diaphragm over the trusses. Make appropriate connections of the blocking to both.

Mike McCann
MMC Engineering

 

[BAretired]

A diaphragm needs two chord members continuous from end to end designed to resist the calculated tension or compression. The transition of chord members at the junction of the two roof types may require special attention.

BA

 

[SAIL3]

You may already have taken this into consideration.....from a quick look at your setup, I would expect the metal roof diaphragm to be much more rigid than the plywood portion, say 5 to 10 times more rigid....so how would this affect the distribution of the shear to each shear wall?....assume the cmu walls run N-S...I would expect that you may need a shear wall/bracing in the E-W dir for a portion of the metal deck to account for this redistribution of shear loading in addition to investigating the connection you originally mentioned...could be dead wrong on this, but, this is my initial quick look at it....

 

[WWTEng]

SAIL. Both wood and metal deck are flexible diaphragms but I guess what you are saying is to account for the relative rigidity of the two roofs. I'll have to think about that some more.

 

[BAretired]

If the only elements resisting lateral forces are the masonry shearwalls, then a difference in rigidity between the two diaphragms should have no effect on statics.

BA

 

[RFreund]

If the only shear walls are the bearing walls on each end (nothing at the junction of the 2 diaphragms), then I agree with BA that the difficult part is the detailing of a continuous load path across the 2 diaphragms for the compression / tension chords.

EIT

http://www.HowToEngineer.com

 

[SAIL3]

BA...I would have to agree with your statement, but when I look at it as a horizontal girder, there are some non-typical situations here.....
1. to act as a diaphragm one needs a top and bottom chord(top & bott fla)....ok as far as the metal deck is concerned, but when we get to the plywood portion there is no web connection to the chord(fla)...in fact the web is 45 degrees to centerline of girder...so I may not be able to treat this as a continous
diaphragm(girder)....at first glance, it is like having a girder with a continous web for ,say, half the span and then a prctically open web area for the other half....granted, this web zig-zags out and back in this area and I have never tried to analyze something like this before.......so to get an upper bound of the problem, I assumed the metal deck cantilevered of one wall and carrying it's own horizontal shear load and half the horizontal shear load of the plywood area, ofcourse, one needs bracing/shear wall for a portion of the metal deck in the other orthoganal dir....

 

[BAretired]

SAIL3,
The pitched roof portion of the structure must be made into a semi-rigid box capable of carrying the required wind shear from the end of the steel deck to the west shearwall.

If the plywood deck is a competent diaphragm at 45^o to the horizontal plane, it provides an indirect path which can carry shear from the west wall to the steel beam. If it is not adequate or is too flexible, horizontal diagonal bracing can be placed between the bottom chords of the pitched light gauge trusses to complement diaphragm action. Alternatively, a ceiling diaphragm could be introduced.

It is still necessary to provide a pair of continuous chord members between shearwalls. The bottom chord of the edge truss and the top chord of the edge open web steel joist could form a chord, but they are at different elevations, so a detail must be devised to connect them in such a way that they are capable of transmitting the calculated force taking into account the eccentricity.

One way to do this would be by means of a column at each edge of the roof designed to resist a moment of F*e at the top where F is the force and e is the eccentricity.

Another way could be by introducing a parapet at the flat roof portion with the chord member raised up to eliminate the eccentricity.

BA

 

[WWTEng]

Yes, the only elements resisting shear are cmu only on each side.

 

[SAIL3]

BA...you addressed most of the questions that I had...your idea of introducing a plywood ceiling diaphragm would certainly eliminate alot of the uncertainty as to the actual behavior of the system and , as you mentioned, the importance of a continous top and bottom chord....it's an interesting problem, but in this particular case the shear loads are probably so low that it would not warrent spending awhole lot of engingineering time chasing the loads thru the structure....