OWSJ Reinforcement vs Midsupports

[TORCHMAN]

Working on a project where the loading is being changed on OWSJ members. I plan on proposing 3 options to my client:

1) Reinforce webs - as it stands, I calculated its compressive resistance to be 10kN
2) Add 1 or 2 supports at bottom chords - I didn't think this was an option but saw KootK recommended it on a thread years ago (single support still loads multiple webs higher than 10kN in compression unless more web members are added)[sup][/sup]
3) Add 2 supports at top chord

Below is an image but it is uploading blurry so I also attached it.

Option 1 is a lot of work, also, I will need to check footings.

I am thinking that option 2 is the optimal since the client could get a beam under bottom chord more easily than under the top chord and no modification to webs is required. As I understand it, using option 3, 2 new web members would be required at each OWSJ. Below is a KootK sketch from some time ago showing a good detail for option 3.

Note that there are 2 floors that need to be reinforced.

Base on your experience, what are the Pros & Cons?

 

[HTURKAK]

You haven’t told us if the top and bottom chords need reinforcing .

I will not agree with you. The web elements which require reinforcing are around supports and i have estimated 16 elements .I would still prefer the reinforcing of the diagonals.

...


He is like a man building a house, who dug deep and laid the foundation on the rock. And when the flood arose, the stream beat vehemently against that house, and could not shake it, for it was founded on the rock..

Luke 6:48

 

[jerseyshore]

Been a while since I did anything other than #1. Reinforcing webs is usually pretty easy. Welding an angle to each side of the existing web usually does the trick.

 

[TORCHMAN]

You haven’t told us if the top and bottom chords need reinforcing .
Quote (Option 1 is a lot of work, also, I will need to check footings.)

I will not agree with you. The web elements which require reinforcing are around supports and i have estimated 16 elements .I would still prefer the reinforcing of the diag

No need for reinforcing the chords. Since the loading on the entire floor plate is changing on 2 floors there are over 30 of these OWSJ that will need reinforcing. So a lot more than 16 web members to be reinforced. Which is why I am wondering about a beam running perpendicular to the OWSJ.

 

[jerseyshore]

Seems pretty rare to see a web-only reinforcement unless it was a localized HVAC/ point load near the ends. I would double check your chords in the model. If you need to reinforce more than the first 4 end webs then the chords usually become overstressed as well.

What are your options besides reinforcing or re-supporting the joists? Can you add beams between each joist to split the load in half? Why is the load changing?

 

[TORCHMAN]

Thanks for the insight jerseyshore.

The loads are changing as the use of the building is going from office to self storage.

Adding beams in between each joist is another option but I imagine more costly than a mid support.

I agree that the web-only reinforcement is a bit strange. These are Anthes OWSJs from the 1950s and typically, as I understand it, Anthes joist webs were made with 1-1/4" diameter rings - at least that is what I have gathered from the 1973 manual that I got from jayrod12 on another thread a couple of weeks ago. These joists, however, have 14mm bars for webs and after performing an analysis, I found that at least the last 3 compression webs on each side have to be reinforced.

 

[jerseyshore]

Not sure what slab & decking you have, but when loads are increasing it is common to see the deck fail also, so adding intermediate beams is a 2 for 1 type reinforcement.

Are your joists wall bearing or beam bearing? Do the beams need reinforcement as well? If you add perpendicular supports would you need to add columns as well?

9 out of 10 times the cost of labor (i.e. welding) is the highest cost on these jobs in my area and what we always try to minimize. Unless we have physical constraints, most owners will prefer to do whatever requires the least amount of "messing around" with the bar joists.

 

[bones206]

With such a significant load increase, I would lean towards adding joists rather than reinforcing or breaking up the existing joist spans. Two-piece retrofit OWSJ's are a good option depending on the available support conditions.

 

[JLNJ]

Office to storage. Doesn't that double the load? I'm surprised that a complete overhaul isn't required.

 

[TORCHMAN]

office 1.9kPa
Storage 4.8kPa but the NBCC has a clause - 4.1.5.8 Variation with Tributary Area which allows for a factor to decrease load on storage buildings based on the area. Live load resulted in 3.6kPa. So the live load increased but did not quite double.

Dead load remained at 1.8kpa

So the total load went from 3.7kPa to 5.4kPa: a 50% increase in total load

 

[jayrod12]

I'm still shocked that with a 50% increase in load that the chords don't require reinforcing.

I've previously done the new mid-span (ish) steel beam, created new bearing seats and reinforced/supplemented the web members as required. It actually worked pretty slick. This was the preferred option as there was so much existing electrical and mechanical to reinforce the joists would be a huge undertaking, versus the new beam and web work was relatively localized.

It just survived it's first winter, so far so good. It did require new piles and columns but that worked pretty well with the redevelopment of the space that was going on as well.

 

[TORCHMAN]

I'm still shocked that with a 50% increase in load that the chords don't require reinforcing.

I wonder if the original design in the 40s/50s had higher live loads than today's minimum requirement by code.

When you did your midspan new beam, I imagine you sat the the top chord on the new beam? If so, do you know the procedure for the work i.e. did they have to shore both sides of the joist to fit the beam in through the web?

 

[jayrod12]

Our contractor was able to fish the new beam up into the space without cutting the bottom chord, so they left the entire thing intact during the reinforcing/repair work.

I had originally envisioned two lines of shoring to facilitate cutting the bottom chord, sliding the beam up and doing all the welding of webs and new beam seats, but they were able to do the modification work before lifting the beam in, then fished the beam into the joist space. Installed columns and welded the beam to the new seats and installed column braces. Then cut the bottom chords out.

 

[bones206]

I'm not too familiar with the NBCC, but it doesn't seem logical to me that they would permit a 25% live load reduction for a floor joist subject to storage loading. Maybe for a column trib area, but for a joist or beam it seems really unconservative.

 

[jayrod12]

Agree with Bones regarding live load reduction. For 4.8 kPa areas it's a minimum of 80 square meters before it kicks in.

Also, storage loads are often non-reducible

 

[TORCHMAN]

Below is the clause and a reference from the clause. As I read it, this OWSJ is a "structural member supporting a tributary area of a floor."

Bones206, I wonder if you would be correct in interpreting only columns as being a "structural member supporting a tributary area of a floor."

The floor plate is 320 sq ft per floor.

 

[jayrod12]

The area used in that calculation is per joist, not the entire floor plate. Therefore for an assumed spacing of these joists of 6 feet (1.8m), they'd need to be 146 feet (44m) long. And that's just to get to the 80 square metres. To get a LLR = 0.75, you'd need to get to 320 square metres of area per joist.

 

[TORCHMAN]

Great point jayrod12, not how I had read it.

When it comes to a load bearing wall, column or footing, I wonder if then you would use the area of the floorplate which is more likely to add up to a number higher than 80m2.

 

[jayrod12]

You'd use the tributary area associated to that specific item under design. If you're designing your wall on a per metre basis, then getting to 80 is tough if it only supports one or two floors. Beams and columns you get to 80 far faster. Columns you're there right away with a 9m x 9m grid.

 

[human909]

I think the wording is quite clear and jayrod12 is correct. Your joist in question is a structural element that supports a tributary area which is NOT the entire floor.

When it comes to a load bearing wall, column or footing, I wonder if then you would use the area of the floorplate which is more likely to add up to a number higher than 80m2.

Absolutely. A column that might have a tributary area of 50m2 per floor. But if it supports 20 floors then it has a live load tributary area of 1000m2.

I personally don't do many structures of 30m high. I could apply a similar reduction allowed by my code by I generally don't for a new design. But I'd happily use it if I had to squeeze more out of an existing structure. That said I might question the suitability of applying it to storage areas, no matter what the codes says.