Optimal girder setback (pre-engineered hip roof)

[StrEng007]

Question for all you truss nuts.

Is there a way to determine the optimal girder truss set back for a hip roof? I'm specifically referring to the hip girder that catches all the jack trusses at the end of the roof.

I've consistently called for 5'-0" setback from outside face of 8" wall (CMU, because... Florida). Roof pitch is typically 2:12 up to 4:12 and the girder setback is honored 95% of the time with no questions asked. I don't know where I got that 5'-0", I picked it up and have been using it for as long as I can remember.

The question is: IS THERE A BETTER WAY?

 

[XR250]

Seems like 6 ft. would be better as it is a multiple of 2 ft - which is the typical truss spacing (at least around here).
Keeps things on layout.

 

[jdcollins]

I've always used 7'-0". I've seen 7'-0" often, sometimes 5'-0", other times 9'-0". From what I've noticed, the steeper the roof pitch, the smaller the set back. This seems to be due to the fact that you can get a deeper girder truss "quicker" on a 5/12 pitch as opposed to a 3/12 where you might need a bigger set back.

I was told early on to use an odd number so that the first jack truss is 1'-0" back from the edge of the wall. If you use an even number the first jack is just connected to the hip jack, rather than both the hip jack and the wall.

 

[XR250]

I was told early on to use an odd number so that the first jack truss is 1'-0" back from the edge of the wall. If you use an even number the first jack is just connected to the hip jack, rather than both the hip jack and the wall

Interesting - good to know

 

[StrEng007]

If you use an even number the first jack is just connected to the hip jack, rather than both the hip jack and the wall

If you have an overhang, won't you always have at least one jack that connections to hip jack only and not wall?

 

[jdcollins]

I guess you technically can depending on how long your overhang is.

 

[jayrod12]

Where I practice it's either 6 feet or 8 feet. Depends on the supplier.

 

[lexpatrie]

When I did these at 6', they wanted 8'. When I did 8', they wanted 6'.

I stuck a note on the plans that if they change it it's hourly to redesign the support and any affected headers, which they usually threw the girder truss onto the window header.

If you want guidance, place it so it's on a wall, not a window header.

 

[RontheRedneck]

Florida is in its own world. They don't seem to do it the way the rest of the country does. So I can't tell ya much about that.

I would never use a setback that short. It would mean the hip girder was very short, which might cause deflection problems.


I'm not claiming that the way we do it is the best way or the right way. But I'll explain the thinking behind it.

We use a setback of 8' or less most of the time. That way our stock hangers will work. No special orders.

Residential trusses - Let's use a 24' 0" hip system to make the math easy. Half the span, or 12' - Would be from the end of the building to where the hip ridge lines meet. Step down 2 spaces at 2' OC and you'd have a setback of 8' 0".

From there, we step it down another 3/4" for a setback of 7' 11 1/4". (To the face of the girder) We do that so the hip jacks are spaced 2' OC across the end of the building. The plywood goes down easier for the framer.



On commercial buildings that doesn't always work. If you have a 60' clear span girder that's only 4' deep you can have a hell of a lot of deflection.

Some engineers and truss designers pay attention to that. Some don't. I always try to remember it, but I'm human.

 

[dik]

8' is common in the industry.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Eng16080]

One other thing to consider, which I think Ron might be referring to in his comment above, is the differential deflection between the girder truss and the end wall (wall parallel to the girder truss). For example, if the offset is only 4 ft and the maximum girder truss deflection is 1" at midspan, considering that there will be zero deflection at the wall, the differential deflection between the end wall and truss will only be L/48. This could lead to some problems in terms of the ceiling finish.

I don't think most engineers usually consider this in their designs, but it's perhaps a good idea to at least keep in mind. This wasn't really on my radar until I read a recent post about this issue here.

 

[Brad805]

Ask 20 truss designers and I suspect you will get 10 or more preferences for any number of reasons. Pick something reasonable and be happy. No matter what we do the truss guys like to try and do their own thing. Maybe it is better in some parts, but keep in mind who does 90% of the design. That group varies immensely.

 

[StrEng007]

Thanks for all this information.

What's interesting to note is that I've asked this question to several other engineers and the answers seem to line up with locations (maybe it's not that interesting after all).

Most Florida engineers have 5'-0" and 7'-0" as their go to numbers. Seems like northerners (which is everyone else by comparison) are saying 8'-0".