Structural welds on residential property

[colnerov]

And those welds are plenty strong enough as I've proved to the naysayers, should pack my gear and come do these beams as definitely better than the builder's guy!

That's what happened last time and earned much praise. (No names, no pack drill). Not for me to say, up to someone to volunteer.

 

[Munkul]

Structural engineer did look at the bolting option but after doing the calcs, he determined the loadings were too high for bolting

I've seen a structural engineer spec the opposite for the same reasons, sometimes I think they just can't be bothered to use their brains to design a practical joint.

 

[tom2207]

And those welds are plenty strong enough as I've proved to the naysayers, should pack my gear and come do these beams as definitely better than the builder's guy!

lol ,,,, your really not ,,, a row of tacks and some savage inclusions are not whats needed ...
as Mr Eastwood once said ... A man has to know his limitations .

And before the row starts ,, [ Im not interested and cant be bothered with it ] I do have the relevant CSWIP quals to say that , and no I wont be posting my reg number either ,,,,, next.

 

[8ob]

Structural engineer did look at the bolting option but after doing the calcs, he determined the loadings were too high for bolting

As with the welder I would say he should also be given the dcm .

Bob

 

[Brad93]

Been done with Spezials

They haven't chipped the black slag off.

Chip the slag then wire wheel it and inspect again.

I've seen a lot worse. And that doesn't look that bad.

 

[Brad93]

On site welding isn't covered by structural steel standards.

The vertical welds look like they've been done uphill, which is a good sign. One of those beads looks like the slag is still attached over the top.
I'm not saying the welding is good, but it looks like it won't fall apart at least.
However, it's your house, if you want it "right" then someone will have to redo it.

I agree. It's been done uphill which is a good sign.

And Specials are deeply penetrating.

 

[Brad93]

I think as mentioned above, whoever supplied & detailed the structural steel wants a sharp rap with a stick.

The beams should have come from a company with CE mark BS EN 1090.

They should have been bolted together onside.

There is no need to weld those onsite.

 

[tom2207]

Been done with Spezials

They haven't chipped the black slag off.

Chip the slag then wire wheel it and inspect again.

I've seen a lot worse. And that doesn't look that bad.

looks like some of it has been missed all together , I did say in about the second or third post that it wasnt chipped in places , and that alone would bother me , I dont think Ive ever seen an actual welder who doesn't look at a weld they have just done , they day that your pride drops that low is the day you take up woodwork.

 

[Brad93]

looks like some of it has been missed all together , I did say in about the second or third post that it wasnt chipped in places , and that alone would bother me , I dont think Ive ever seen an actual welder who doesn't actually look at a weld they have just done , they day that your pride drops that low is the day you take up woodwork.

Yep. Not chipping the slag is a weird one.

What worries me is the number of arc strikes next to that vertical.

Whilst it has been welded uphil, if he has restarted the weld 100 times it's cold and likely lacks penetration.

 

[tom2207]

Yep. Not chipping the slag is a weird one.

What worries me is the number of arc strikes next to that vertical.

Whilst it has been welded uphil, if he has restarted the weld 100 times it's cold and likely lacks penetration.

that hits the nail on the head ,,, anyone who needs that many goes over such a large area to get an arc going is likely only using what rods were in the back of his car ,,, unless of course he welded it in the dark and was actually looking for somewhere to weld .

 

[Bki26]

lol ,,,, your really not ,,, a row of tacks and some savage inclusions are not whats needed ...
as Mr Eastwood once said ... A man has to know his limitations .

And before the row starts ,, [ Im not interested and cant be bothered with it ] I do have the relevant CSWIP quals to say that , and no I wont be posting my reg number either ,,,,, next.

Hardly a line of tacks, my welder is a pain to get started for some unknown reason so I use the full rod each time and they came out perfectly once amps/mask issues were sorted.

Good for you the splitter ain't part of a car/boat/plane etc so don't need your cswip approval that you love to bring up, if you don't want an argument then stop being an ar*e with the sly remarks it's pretty simple just move on about it

 

[tom2207]

Hardly a line of tacks, my welder is a pain to get started for some unknown reason so I use the full rod each time and they came out perfectly once amps/mask issues were sorted.

Good for you the splitter ain't part of a car/boat/plane etc so don't need your cswip approval that you love to bring up, if you don't want an argument then stop being an **** with the sly remarks it's pretty simple

see above dont overestimate your own abilities , someone will get hurt ,,,, [ take a close look at some of the pics you posted , then apologise if your man enough ]
Id post the pictures for you , but this isnt the thread for it ,,,, this is about someone else's issues .

 

[Bobtail4x4]

As the house owner you are ultimately responsible, you need a competant builder, and he needs a competant welder,

contact Donny building control,ask to speak to the area manager not the guy who comes out on site,

 

[Pete.]

Structural engineer did look at the bolting option but after doing the calcs, he determined the loadings were too high for bolting

What?

Is he related to the welder?

 

[Jelly_Sheffield]

On site welding isn't covered by structural steel standards.

BS EN 1090-1 (Eurocode) is definitely still applicable, and the CDM regs create an implicit duty of care for the Principle Contractor to ensure the work is in conformity with that standard.

Whether the construction products regulations (which requires UKCA/CE marking) are applicable for on site welding hinges on how you interpret the word "delivered".

As far as I am aware the industry as a whole seems to interpret that as meaning "delivery to site" meaning that a FFW job is out of scope of UKCA marking, but the steels should still be marked, and the execution of the welding still has to comply to Eurocode.

I always look at this sort of thing and wonder which dinlo designed it, those beams should have had the end plates welded on and the beams bolted into position, you can’t get between two narrowly spaced beams and get a proper weld.

Bob

The design doesn't inherently create any insurmountable access issues, assuming it's done in a sensible order.

The top left annotation states:
- the plate should be 10mm larger than the height of the left UB
- the plate is welded to the left UB above and below the beam with a 6mm Fillet.

The bottom right annotation states:
- the right UB should be welded to the plate on both sides of the flanges and the web.
- the welds should be 6mm fillets.

Looking at the plan view, there's space to have options.

Easiest thing to do would be complete the welding on the ground, immediately before lifting the steel in:

- the long span beam sits through the door opening, and the short span stays in the garage
- you have access everywhere, and with a bit of manipulation everything can be done in a Flat position.
- the assembly is then lifted in so that the short span is inserted from the garage side of the wall, piece of cake.


Alternatively if you must weld in situ:

- first weld the plate to the left hand UB (short span) on the floor where you can do it all in flat position.
- the lift the short span beam in and bed it in place.
- then lift the longer span in and fit up the joint.
- weld the easily accessible surfaces first to get the joint secured.
- the only weld with potentially poor access is the one between the plate and the top of the top flange of the web,
- but that can be addressed by ensuring sufficient material above the workpiece has been removed, in principle you could even pull a couple of floorboards above it to get at it from the next floor up.

I think as mentioned above, whoever supplied & detailed the structural steel wants a sharp rap with a stick.

TBH they didn't do an amazing job of detailing the joint design in the first place.

It's clear enough, but for the sake of 60 seconds more time taken to add 3 extra leaders, so each weld is indicated using the correct symbols, it would have been crystal clear, without relying on the narrative description.

I have to assume the engineer in question is working for a practice which specialises in churning out simple beam calcs at a rate of knots for cheap price per job.

No shame in that but they're not going to be the kind of practice that offers the client any value added services like design for installation or providing a full method statement for the works.

I would agree with your opinion they both could and should have given proper guidance on how the works were to be delivered... But I think we all know that when cost is on the line the world doesn't work like that

The beams should have come from a company with CE mark BS EN 1090.

They should have been bolted together onside.

This is the bit that doesn't make much sense to me.

Even taking OP's comments into consideration that:

Structural engineer did look at the bolting option but after doing the calcs, he determined the loadings were too high for bolting

There's no magical upper limit to the loading you can make with a bolted connection, you just design it differently, Right?
- specify larger or higher strength class bolts to allow for greater preload, or
- add more bolts to increase the total joining force applied, if increasing the preload would cause the components you're bolting together to reach their yield stress.
- use a connector piece with a bigger surface area to adequately distribute the additional preload or accommodate the bolts.
- specify connector pieces made from thicker steel if the downward sheer stress from loading would cause the connector to yield itself.

There's a practical limit to how big the connector pieces could physically be, but I can't see how you'd get anywhere close to that without the UB's themselves being massively over-stressed.

Kinda supports making the assumption the engineer was basing the decision not to use a bolted joint on standard parts programmed into a beam calculator.

None of the standard designs for a bolted joint works, so go straight to a welded design, a direct result of them working to a price-point.

There is no need to weld those onsite.

Looking at the plan view there's also more than enough space for it to have been be fabricated off site, carried inside, manipulated into position and installed as a single piece.

Assuming it's drawn to scale with 700mm door openings, the short span is around 1100mm, and the long span is 4400mm, so doing it as a single piece adds about 25kg to a beam that already weighs 101kg...

That's already heavy enough to be awkward, and the extra 25kg isn't enough to make it meaningfully more awkward to do it as a team lift, or makes basically no difference if using mechanical means.

Welding in-situ was by far the worst installation option to choose.

 

[pawsnizzy]

Ask him if he used a helmet....

I think he IS a helmet!

 

[Gareth J]

Yep. Not chipping the slag is a weird one.

What worries me is the number of arc strikes next to that vertical.

Whilst it has been welded uphil, if he has restarted the weld 100 times it's cold and likely lacks penetration.

Looks to me like they may have done what I sometimes do on galvanized or other heavily contaminated stuff that I'm too lazy to clean off properly - a series of stacked tacks. Higher power than you can get away with on a continuous weld, the previous bridge/tack helps hold up the next weld.

It's a cheat and almost inevitably ends up with slag inclusions and a bodge of a weld. Would explain the multiple restarts on that vertical.

 

[Brad93]

BS EN 1090-1 (Eurocode) is definitely still applicable, and the CDM regs create an implicit duty of care for the Principle Contractor to ensure the work is in conformity with that standard.

Whether the construction products regulations (which requires UKCA/CE marking) are applicable for on site welding hinges on how you interpret the word "delivered".

As far as I am aware the industry as a whole seems to interpret that as meaning "delivery to site" meaning that a FFW job is out of scope of UKCA marking, but the steels should still be marked, and the execution of the welding still has to comply to Eurocode.


The design doesn't inherently create any insurmountable access issues, assuming it's done in a sensible order.

View attachment 522876
The top left annotation states:
- the plate should be 10mm larger than the height of the left UB
- the plate is welded to the left UB above and below the beam with a 6mm Fillet.

The bottom right annotation states:
- the right UB should be welded to the plate on both sides of the flanges and the web.
- the welds should be 6mm fillets.

Looking at the plan view, there's space to have options.

View attachment 522880

Easiest thing to do would be complete the welding on the ground, immediately before lifting the steel in:

- the long span beam sits through the door opening, and the short span stays in the garage
- you have access everywhere, and with a bit of manipulation everything can be done in a Flat position.
- the assembly is then lifted in so that the short span is inserted from the garage side of the wall, piece of cake.


Alternatively if you must weld in situ:

- first weld the plate to the left hand UB (short span) on the floor where you can do it all in flat position.
- the lift the short span beam in and bed it in place.
- then lift the longer span in and fit up the joint.
- weld the easily accessible surfaces first to get the joint secured.
- the only weld with potentially poor access is the one between the plate and the top of the top flange of the web,
- but that can be addressed by ensuring sufficient material above the workpiece has been removed, in principle you could even pull a couple of floorboards above it to get at it from the next floor up.


TBH they didn't do an amazing job of detailing the joint design in the first place.

It's clear enough, but for the sake of 60 seconds more time taken to add 3 extra leaders, so each weld is indicated using the correct symbols, it would have been crystal clear, without relying on the narrative description.

I have to assume the engineer in question is working for a practice which specialises in churning out simple beam calcs at a rate of knots for cheap price per job.

No shame in that but they're not going to be the kind of practice that offers the client any value added services like design for installation or providing a full method statement for the works.

I would agree with your opinion they both could and should have given proper guidance on how the works were to be delivered... But I think we all know that when cost is on the line the world doesn't work like that


This is the bit that doesn't make much sense to me.

Even taking OP's comments into consideration that:


There's no magical upper limit to the loading you can make with a bolted connection, you just design it differently, Right?
- specify larger or higher strength class bolts to allow for greater preload, or
- add more bolts to increase the total joining force applied, if increasing the preload would cause the components you're bolting together to reach their yield stress.
- use a connector piece with a bigger surface area to adequately distribute the additional preload or accommodate the bolts.
- specify connector pieces made from thicker steel if the downward sheer stress from loading would cause the connector to yield itself.

There's a practical limit to how big the connector pieces could physically be, but I can't see how you'd get anywhere close to that without the UB's themselves being massively over-stressed.

Kinda supports making the assumption the engineer was basing the decision not to use a bolted joint on standard parts programmed into a beam calculator.

None of the standard designs for a bolted joint works, so go straight to a welded design, a direct result of them working to a price-point.


Looking at the plan view there's also more than enough space for it to have been be fabricated off site, carried inside, manipulated into position and installed as a single piece.

Assuming it's drawn to scale with 700mm door openings, the short span is around 1100mm, and the long span is 4400mm, so doing it as a single piece adds about 25kg to a beam that already weighs 101kg...

That's already heavy enough to be awkward, and the extra 25kg isn't enough to make it meaningfully more awkward to do it as a team lift, or makes basically no difference if using mechanical means.

Welding in-situ was by far the worst installation option to choose.

Agree with much of what you've said.

I'd still have bolted the short piece on to the long piece. Maybe have gusseted the connection plate.

The problem is very few of these structural engineers are good at steel detailing. They can read in a book ten different ways of doing things but they've probably never had to do the welding or the assembly onsite.

Having the connection plate overlap the beam by 10mm top and bottom is certainly a bit of a new one

I'd weld a gusset to the centre back of the connection plate on the bench, then have the connection plate sized just right to sit on the very edge of the top and bottom flanges of the beam.
When welded to can do a 6mm fillet in one pass or 8mm in two passes.
You can then reach in (especially with a 450mm electrode) to weld the gusset vertically.

 

[addjunkie]

BS EN 1090-1 (Eurocode) is definitely still applicable, and the CDM regs create an implicit duty of care for the Principle Contractor to ensure the work is in conformity with that standard.

Whether the construction products regulations (which requires UKCA/CE marking) are applicable for on site welding hinges on how you interpret the word "delivered".

As far as I am aware the industry as a whole seems to interpret that as meaning "delivery to site" meaning that a FFW job is out of scope of UKCA marking, but the steels should still be marked, and the execution of the welding still has to comply to Eurocode.


The design doesn't inherently create any insurmountable access issues, assuming it's done in a sensible order.

View attachment 522876
The top left annotation states:
- the plate should be 10mm larger than the height of the left UB
- the plate is welded to the left UB above and below the beam with a 6mm Fillet.

The bottom right annotation states:
- the right UB should be welded to the plate on both sides of the flanges and the web.
- the welds should be 6mm fillets.

Looking at the plan view, there's space to have options.

View attachment 522880

Easiest thing to do would be complete the welding on the ground, immediately before lifting the steel in:

- the long span beam sits through the door opening, and the short span stays in the garage
- you have access everywhere, and with a bit of manipulation everything can be done in a Flat position.
- the assembly is then lifted in so that the short span is inserted from the garage side of the wall, piece of cake.


Alternatively if you must weld in situ:

- first weld the plate to the left hand UB (short span) on the floor where you can do it all in flat position.
- the lift the short span beam in and bed it in place.
- then lift the longer span in and fit up the joint.
- weld the easily accessible surfaces first to get the joint secured.
- the only weld with potentially poor access is the one between the plate and the top of the top flange of the web,
- but that can be addressed by ensuring sufficient material above the workpiece has been removed, in principle you could even pull a couple of floorboards above it to get at it from the next floor up.


TBH they didn't do an amazing job of detailing the joint design in the first place.

It's clear enough, but for the sake of 60 seconds more time taken to add 3 extra leaders, so each weld is indicated using the correct symbols, it would have been crystal clear, without relying on the narrative description.

I have to assume the engineer in question is working for a practice which specialises in churning out simple beam calcs at a rate of knots for cheap price per job.

No shame in that but they're not going to be the kind of practice that offers the client any value added services like design for installation or providing a full method statement for the works.

I would agree with your opinion they both could and should have given proper guidance on how the works were to be delivered... But I think we all know that when cost is on the line the world doesn't work like that


This is the bit that doesn't make much sense to me.

Even taking OP's comments into consideration that:


There's no magical upper limit to the loading you can make with a bolted connection, you just design it differently, Right?
- specify larger or higher strength class bolts to allow for greater preload, or
- add more bolts to increase the total joining force applied, if increasing the preload would cause the components you're bolting together to reach their yield stress.
- use a connector piece with a bigger surface area to adequately distribute the additional preload or accommodate the bolts.
- specify connector pieces made from thicker steel if the downward sheer stress from loading would cause the connector to yield itself.

There's a practical limit to how big the connector pieces could physically be, but I can't see how you'd get anywhere close to that without the UB's themselves being massively over-stressed.

Kinda supports making the assumption the engineer was basing the decision not to use a bolted joint on standard parts programmed into a beam calculator.

None of the standard designs for a bolted joint works, so go straight to a welded design, a direct result of them working to a price-point.


Looking at the plan view there's also more than enough space for it to have been be fabricated off site, carried inside, manipulated into position and installed as a single piece.

Assuming it's drawn to scale with 700mm door openings, the short span is around 1100mm, and the long span is 4400mm, so doing it as a single piece adds about 25kg to a beam that already weighs 101kg...

That's already heavy enough to be awkward, and the extra 25kg isn't enough to make it meaningfully more awkward to do it as a team lift, or makes basically no difference if using mechanical means.

Welding in-situ was by far the worst installation option to choose.

Great explanation, when I did both my builds with steel, those drawings are far far better than I got from both the engineers used. Was all fairly vague to be honest, not even weld sizes, just a hint to use a competent onsite welder.

First time back to back u channels with a vertical up to support a roof joists at 90 degrees. Building inspector was interested and insisted the two back to back u channels were bolted. Once they were in place, I put a few welds to stop any movement while drilling, he then said, just weld the lot. As council folk go he was old school and was actually very helpful through out, my Mrs feeding him cakes helped.

Second time, new guys attitude was (mid covid) your a bit far out, just send me some pictures, it was only weld on some drilled tabs to hold joists in place, not really necessary once they were noggins all went in.

 

[carbonarc]

Hi ash1356. Just a bit of information to provide an insight as to strength of the welds you have.
A fillet weld with a leg length of 1/4" will provide a shear resistance based on the throat distance of around 1 imperial ton for each 1" of weld run.