Notching out joists and trusses are a definite no-no due to the stresses of the beam at the outer edges.

But I ask, what about installing lag bolts directly into them, from the bottom?

How does this effect the stress points on the truss?

Any-one know?

I don't believe it would significantly change the stress in the beam unless it's a REALLY big lag bolt or you're hanging a LOT of weight from it. How big of a lag bolt and what sort of weight are you asking about? It's probably only designed for a 10psf live load on the bottom chord.

In a truss, the main "truss" stress is usually tensile throughout the entire lower member (and can be over 4500lbs of tension for an SPF 2x4!), and splitting with a pre-drilled lag bolt is really no worse than having a naturally ocurring knot in the wood, which is taken into account. It would depend on the type of truss, the design stresses, and a lot of other factors to say for certain. It's probably OK as the trusses are expected to be perforated with screws and nails and designed with this in mind, but, if in doubt, ask the engineer who stamped the truss drawing!

FYI, don't ever center-bore or otherwise cut/drill a truss member anywhere without permission from the engineer. For a joist or beam, center-boring through the precise center is the best place structurally to do it, as stress increases with the cube of distance- boring a hole twice as far from the center weakens the beam 10x as much! For TJIs, honestly, it doesn't matter, so long as it's in the web.

[This message has been edited by SteveFehr (edited 01-04-2007).]

When I hang something heavy I run a 2x4 from the top to the bottom chord and hang off that. You do need access to the attic for that but it does get the load up to the top where it was designed to carry it and it stiffens up the whole assembly.

It would depend on the size of the lag bolt. Consider that on angled roof trusses (as in the trusses forming each side of an A) the underside of the truss is under tension, while the outer (top) side is under compression. So (using your notch example) a notch on the underside is much more damaging than one on the top. A lag bolt would similarly decrease load capacity but to a lesser degree than the notch. However drilling into the side of the truss (somewhere near the center) has relatively little effect.

Dnk,
Roof trusses are Engineer designed here with a given load on the top and bottom chords.
It also depends upon the load already on the trusses.
You haven't mentioned what sort of roof these trusses are supporting.
Back when builders used to make trusses on-site, they were made to take a certain degree of load.
Not these days here though, they are fabricated to a formula and that is an un-wavering one, often made off-site from where they will be installed.
My advice would be, if you are going to put stress on any building member, like a truss.
Ask the Engineer that designed the truss and also Skew-nail a support between 2 trusses.
I've walked over a few roofs that have been "spongy" in the past through people putting supports in for their gear.

I too will second spreading the load across two if you can. Other-wise not knowing the particulars I could not say for sure... Generally speaking the lower portion of a truss is tension type load, (unless cantelevered) so long as you are not cutting material out you would be ok - but with a large lag you may be cutting material by installing it..... Generally a fastening method that disturbs the least material is the best. How 'bout a strut between two, over the top of the lower chord - then you'll only need a minimal fastener to hold it in place?

Thanks for the replies...

The question arose from from a guy who is putting buss duct in his garage/shop/lab.

He wants to lag unistrut on the wooden roof truss, and then all-thread down to carry the duct. He has no access to allow mounting of the strut above the truss. His option is to lag to them. I believe he is installing 20' of duct, carried over trusses that are on 2' centers. The load might be a couple hundred pounds over 20'. So 10 trusses need 1/4" lags.

We were curious about the stress to the truss, and I didn't know the answer.

FWIW, it is a metal roof on wood trusses, only supported on the ends.

The HVAC contractors hang air handlers from botton truss chords all the time around here and I haven't seen the structural inspectors blink. Bus duct, running perpendicular to trusses should not be a problem. If it is all hanging on one truss it might be troubling. For some reason putting nails and screws in wood members doesn't seem to bother them either. A hurricane clip will usually have 20 or more nails in it. That looks like you splintered the 2x4.

If it was a new house, you could ask the GC for the truss drawings; he's required to have a stamped copy for final inspection. As it's existing, they probably don't have them sitting around. I have some truss drawings from my new addition at home if anyone's interested, I could upload them to an image host- it shows design forces and stresses and general assumptions the engineer made for generic use by builders. It doesn't really answer the question definitively, but it's interesting if you've never seen one before

In this case, it's a generic non-specific drawing, and 10lbs per square foot live load was assumed for misc fittings, which is considered in addition to the sheetrock and other dead loads. Assuming these trusses were designed with a similar 10psf live load factor, and your bus duct doesn't weigh 20lbs per linear foot, there should be no problem attaching it to the truss. Of course, we don't like to make assumptions in this business, but the risk is pretty low...

[This message has been edited by SteveFehr (edited 01-05-2007).]

Timber trusses do not have a 'neutral-axis', as such, in the chords, ties or struts, the principal stresses are in tension or compression tied and triangulated into to the joint plates - [ which is why it's called a truss! ]. There are additional bending loads across most of the chords, due to the weight of roofing, sheet-rock or man-loads at access, of course. There are therefore NO safe zones to drill holes for bolts or studs in a truss's members. The allowances for material defects, multiple-trusses and all other loads are already used up at the design stage - so there is little additional redundant margin of strength available for modifications. Any hole or notch not only reduces the section area, it introduces a stress raiser [ = bad]. The reason nails or small calibre screws don't ordinately affect the strength of the timber members is that they tend to push the wood-fibres aside rather than shearing them off, thus the continuity of strength of the timber member is retained, [ imagine pushing a knitting-needle through a rope's strands ].
Within certain weight limits then, nailing a timber rail on the upper edges of a set of bottom chords is an acceptable way to provide an anchor for studs to support light loads such as ducting or wiring trays. The chord supports the load, so the nails are just there as locations. A horizontal rail can be drilled for studs/penny washers and nuts, as it's not a critical truss member. In most roofs there will be just such rails, [ longerons ] fitted in order to give the roof lateral stability. 1" x 4" is the usual sort of calibre. I'd consult the/a truss maker anyway to cover my butt; no big deal, the computer programs used to design trusses can rapidly give a yeh or nay to any such proposed changes in minutes.

Alan

[This message has been edited by Alan Belson (edited 01-05-2007).]

As an aside, a few years ago I designed and built two massive trussed beams to fit into an existing trussed roof space in a roof extension for Denise’s brother. They were 32 feet 6” long and 6 feet 6” high, with the upper and lower chords in twinned 2” x 9”s, linked with 5 sets of doubled 2 x 4 struts at 45° set out in 'X' patterns with vertical end and mid ties. No timber was longer than about 10 foot; I jointed the chords too with bolted cheek plates. Joints were with M12mm stainless studding and 60mm diameter spiked timber connectors at every joint, timber to timber, with ¼” thick square plates under the nuts.
Each assembly was 12” minimum thick, 6 timbers meeting at every joint, 3 studs per joint. The timber was 'SS' grade air-dry Douglas fir. I built them on my shop floor, marked each member in best medieaval carpenters method, then dissassembled them for shipping to Maidenhead. On site, we pulled the whole lot up in pieces through the loft hatch- my sister in law is so house-proud we were banned from bashing any holes in the structure!

The beams were assembled along the roof as the extension ‘walls’; we painstakingly wound up an endless stream of long studs with an air wrench normally used to take truck wheels off, taking it in turns to be the sap crammed into the eaves space with the ring spanner. At the end walls the shear loads were transferred to the existing masonry by oak plates set on concrete beams cast into the inner leaves. When in place, after 2 weeks of solid graft, sustained with several hundred mugs of builders’ tea and endless bacon sarnies, we wedged them up to the existing truss upper chords with oak sprags and wedges. Then, with the original trusses ‘unloaded’, we began to saw out the original ties and struts with a sawzall! As each one parted, the roof settled about 1/8” with a groan, quite unnerving! Once they were supporting the roof, we nailed collars at 8 ft head height as ceiling members and fitted floor joists on hangers on the bottom chords, leaving the ceiling below independantly hung to avoid cracking the plasterwork, then cut in the stairwell. I made a bespoke staircase with Pirana pine, a timber I have never used since, as it stinks to high heaven! The upper chords of the original trusses now became the common rafters.

Result: A room, in a trussed roof structure, the full length of the house and about 13 feet wide.
The stressing out took me weeks- not least because the local Building Control Department had to hire in a structural engineer in to check my calculations, as no one on their staff knew how to do the math. Not surprising, the last folks to use 'Wilson’s Method' probably built pole bridges for American Railroads! And check it he did, fair play, we went through every sum, every theory, every line of the calculations in detail, the moment, force and shear diagrams, the specifications of the studs, the connectors Standards, the lot, over 5 days in the Windsor B.C. office. No fee and free tea!

And of course it was vital to get it right - the trussed beams support a huge tiled roof and all the floor loads below.
Several times he expressed the notion that steel I beams would have been easier: True, but then he didn’t know my sister in law, did he?! [ Actually we just couldn’t get RSJs in- the houses each side were too close for a crane.]


Alan

I concur with everything Alan said

Here's an example of a typical computer-spat-out truss drawing:
<IMG SRC=\"http://img219.imageshack.us/img219/721/scissortrussgq8.th.jpg\">