Structural engineer missed point load - advice please

[Daniel Johnstone]

Hi guys,

SE originally provided calculations for 2 steel beams. B1 supporting an interior wall, B2 support the rear side of the house. B1 is bolted to the centre of B2 so the steels forms a T shape (pic attached).

The SE forgot to add the point load of B1 on B2. I have now installed the steels and have just noticed this. I approached SE and asked them for clarification. They have resent the calculations, however I have noticed that for B2 the deflection limit has dropped from 374 to 281.

I figure they had to do this otherwise the calcs will fail.

Can anyone let me know if a deflection limit of 281 is ok?

Both beams are carrying plaster, tiling etc.
Any advice would be appreciated.

 

[tony1851]

Steel design codes normally limit the live load deflection to span/360 to avoid cracking of plaster casing to the beam, which would compromise its fire resistance.

If the figures you refer to are the SE's computation of live load deflection, then technically B2 fails (though it won't collapse!). In practice, it may well not be an issue because the live loads assumed (eg for floors it is 1.5kN/m2) are quite high and in practice we rarely load bedroom floors up to anything approaching that figure.

Your SE needs to wake up, btw ......

 

[Daniel Johnstone]

Hi Tony,

Your response confirms my suspicions that something wasn't quite right.
The outside wall B2 is supporting has a pitched roof attached (via, wall plate & roof hangers). Due to the increased deflection calculated @ 18.2mm, do you think the increase deflection may cause an issue with the pitched roof?

I have consulted another SE who has advised to build back 300mm of the wall to reduce the beam span (and tie in to existing column). This then passes due to the reduced span.

Thanks in advance for the advice.

 

[Daniel Johnstone]

Hi Tony, Just to add to the above.

The figures relate to the total load. Figures for beam B2 below.

Deflection
LL deflection = 51.3 x 1e8/(2 x 210,000 x 2,900) = 4.2 mm (L/1211) OK
TL deflection = 221.4 x 1e8/(2 x 210,000 x 2,900) = 18.2 mm (L/281)

 

[tony1851]

If the total deflection is 18mm on a span of 5.1m(?) it might cause cracking above, depending on the material and workmanship.
Are the beams 203 x 133 x 30 by chance? Is the SE assuming the beams will work together and both carry the load 50:50?

 

[Daniel Johnstone]

Hi Tony,
Correct, B2 beams are 203 x 133 x 30.
Correct again, the SE has assumed the beams will work together, please see attached load calculation.
The SE has assumed that the beams will work together. Not sure if this has any bearing but the beam span is 5.1, which include 250mm each side to sit on the column & padstone. This leaves an open span of 4.6 m.

Look forward to hearing your thoughts on this.

 

[jeds]

The only thing I would say is that the SE has the beam carrying roof load across the whole span. If this is a rear knock-through type situation that's not going to be the case. British Standards specifies a load triangle of 45° or 60° depending on the circumstances. Even at 60° the roof load will be less than half the full width.

That said, off the top of my head, you might still struggle to get down to a 203x133x30 beam on the inner leaf. Have you ordered these beams yet?

 

[Daniel Johnstone]

Hi Jeds,
Beams have been ordered and already installed.
I only recently picked up on the miscalculation as it was pointed out to me by another SE who I went to advice for.

Unfortunately changing the beams is not an option. I'm most likely to rebuild a portion of the wall removed to reduce the open span. Not ideal but at the moment it seems like the best option I have.

Initial SE has washed his hands and told me to speak to BC.

 

[tony1851]

@op I doubt an 18mm deflection would cause a problem for the roof, as the timber and tiles will accomodate a degree of movement, but it might cause some cracking in the masonry above. In practice the deflection probably won't be that much as the live load values are quite high - and what's the chance of maximum roof live load occurring at the same time as maximum floor load? And as jeds mentioned, you can take the arching effect of masonry into account, which can reduce the dead load considerably. Id be inclined not to bother with an additional pier.

Small point: had the builder raised the main beam level with the incoming beam, you would have reduced the wall load by 2 courses height.

@jeds, i think you're correct about the beam under the inner skin; the SE seems to be assuming this will only be taking 50% of the point load by sharing it with the outer beam, whereas in practice it will be taking almost all of it. The usual mindless nonsense of connecting beams with lengths of tube and some threaded bar from Screwfix, and expecting them to work as one, is pie in the sky.
Too late now but on that job i would have considered a single 203 x 203 x 46 or 52, with a 250 wide plate tack-welded on top to pick up the cavity wall. The incoming beam could have been notched at the top and sat on the bottom flange (frowned on, i know, but it works!).

 

[Daniel Johnstone]

Thanks for the replies guys.
2nd SE also noted that in practice the inner beam will take most of the load. Would that in practice increase the deflection on the inner beam? as the current deflection calculation is based on the point load being spread across both beams.

If so what would be the best solution (given the current status) to correct?

 

[cdbe]

Can't you just bolt some suitable plates onto the incoming beams to extend them over the outer beam - to transfer either all or part of the load so it's shared equally - seems fairly simple to the untrained eye? Shame to starting building piers and spoil your open plan.

 

[jeds]

How are the beams behaving at the moment. What is the current deflection? Is there a difference in deflection between the two beams?

 

[wessex101]

Initial SE has washed his hands and told me to speak to BC.

I don't think professional liability works like that. You paid him to provide a competent structural design. He screwed up so either he comes up with a suitable solution or he pays for someone else to sort it out.

Is he a chartered engineer or a member of any other professional body? It might be worth looking in to any complaints procedures and apply some pressure on him. I hope his professional indemnity insurance is up to date, this could get expensive.

 

[Daniel Johnstone]

@cdbe, I will definitely query with SE.

@jeds, the inner beam has deflected more than the outer beam. The inner beam deflection is 1.5mm greater than the outer beam. The beams have been installed for 3+ months, that said there are currently no live loads. It's a complete empty house.

@wessex101, I agree, I'm planning to re-approach once the new SE has provided amended calcs. Not sure who is best to refer the complaint to. I'm sure the SE has professional liability, they are quite a large firm with 10+ SE working there so I would hope they have covered themselves. With regards to liability, do you know to what extend they are liable? i.e. covering remedial costs or would it just be to pay back the fee's initially paid?

 

[tony1851]

Be careful if you formally dispute the first SE's calcs. If push comes to shove, they might be able to show that the total deflection for live load (ie udl live load and point live load) is less than span/360, even if only on the inner beam.

While design codes usually set this limit on live load deflection, dead load deflection is usually left to the SE's judgement, and could be open to dispute.

Sure, he made a mistake, but while very annoying and frustrating, it doesn't necessarily invalidate the design.