Chimney Design

[MortarThePoint]

I'm a bit nervous to share this post, but hopefully people will think the design sensible. I designed most of it and I'm not a Structural Engineer so here's hoping I've not made a massive mistake.

 

From the Architect: The design calls for a chimney breast that narrows from 2409mm to 1228mm width. The fireplace opening is 1805mm wide and 1200mm above finished floor, 1125mm above top of hearth.

 

 

The Flue Arrangement

 

To avoid complex corbelling, discussions with the brickie about how all that is done as well as to allow brickwork to come up, I have decided to make the offset in twin wall flue (i.e. Schiedel ICID). It's a bit nuts, but the flue arrangement is single wall vitreous to twin wall flue to Isokern pumice. Considering the labour etc, the twin wall doesn't actually add much cost and looked to solve a lot of problems. It means the Isokern flue is entirely vertical and can start much higher up. I checked with Schiedel as well as our intended HETAS installer that this is technically possible.

 

 

Granular fill: I could create a smaller box out of blockwork to be filled with the ganular material (Leca or a perlite substitute) but that would either need much longer ties or increase weight of blockwork. The plan is to construct as in the Architect's drawing and use lots of granular fill.

 

The Lintel Arrangement

 

Well if you thought the flue was complex, there's quite a nest of lintels.

 

House Inner Leaf Lintel: Above that opening there will be a concrete lintel which will be timber dressed. This concrete lintel will be the same width as the house wall inner leaf that it bears on which is 140mm. I chosen to treat it as only bearing on the blockwork rather than the brickwork that actually defines the edge of the reveal. The gap in blockwork is 100mm wider on each side than the fireplace opening, so 2000mm. I calculated the loading on this lintel under two different assumptions:

 

1) It does triangulate: assuming the blockwork forms an arching action and so I can consider everything in a 60 degree triangle above the lintel. Unfortunately the floor does  intersect this triangle so I needed to consider the load from that. The two loads to add are the weight of the 60 degree blockwork triangle and the load from a section of floor bearing the same width as the width of the triangle at that height. Below is the calculation I did. The eagle eyed among you may notice I haven't included the lintel self weight, but that shouldn't tip the balance much. Floor loads from the Structural Engineer's original calculations.

 

2) It doesn't triangulate: should there be no arching action. I drew a rectangle above the lintel up to and including the floor, but no higher. The maths is much easier:

• The wall creates a UDL of 1.275m * 0.14m * 1450kg/m3 * 10N/kg = 2.6 kN/m
• The floor UDL is (4.710kN/m2 + (0.9 * 3.700kN/m2)) * (4.2m / 2) = 16.9 kN/m    [4.2m is the floor span]
• TOTAL UDL: 2.6kN/m + 16.9kN/m = 19.5 kN/m

 

Apparently the rule of thumb is that if the window is over 600mm from the opening then it can be neglected and assumed to triangulate. The distance is 630mm, so I wanted to consider both cases. There are various lintels that could be used to meet the requirement. Some lintels are on crazy lead times these days, but I have chosen a Supreme R21240 which, when used the 21A orientation 140w x 215h), has a load capacity of 47.40 kN/m. Deflection at 1/3 capacity (15.8 kN/m) is 3.41mm, so deflection at 19.5kN/m should be 4.2mm. The triangulated case could perhaps have been met by an R15240 in R15 orientation (140x x 100h) with load capacity of 9.92 kN/m, but I wanted to go beyond that and fancied the R60 fire rating of the R21 lintel. 

 

Stack Lintels: this is much more complicated as there are various loads to consider. There need to be lintels at right angles to support the inner leaf of the chimney stack. The main lintel (PURPLE) has to support lintels all these lintels as well as a section of the house's outer leaf brickwork. [NOTE: the GREEN lintel is the one considered above]

 

Listing all the loads on the PURPLE lintel from left to right:

• 0 to 1805mm : 100mm blockwork, 385mm high (BLUE) : UDL 0.56 kN/m  = (0.1m*0.385m*10N/kg*1450kg/m3)
• 0 to 881mm : house outerleaf brickwork (ORANGE) : UDL 5.78 kN/m  = (0.1m*3.5m*10N/kg*1650kg/m3)
• at 927mm : left hand lintel that supports outer leaf brickwork return : point load 3.12 kN  = (0.1m*0.631m*6m*10N/kg*1650kg/m3) / 2
• at 1126mm : lintel bearing that supports the side of the stack inner blockwork (GREY): point load 2.74 kN  = (0.1m*0.624m*6m*10N/kg*1450kg/m3) / 2
• at 1273mm : lintel bearing half the weight of the flue liner and surrounding granular fill : point load 2.28 kN  = (9.1kN / 2) / 2
• 1176mm to 1805mm : stack inner blockwork (GREY) : UDL 8.7 kN/m  = (0.1m*6m*10N/kg*1450kg/m3)
• at 1713mm : lintel bearing other half of the weight of the flue liner and surrounding granular fill* : point load 2.28 kN  = (9.1kN / 2) / 2

 

* weight of flue liner and granular fill:

• weight of liner : 1.1kN  = ((11kg/unit * 6m * 10N /kg) / (0.6m/unit))
• weight of granular fill : 8 kN  = (0.429m*0.624m*6m*500kg/m3*10N/kg)  [assumed 500kg/m, Leca seems to be about 400kg/m3 I think]
• TOTAL : 9.1 kN  = (1.1kN+8kN)

 

Total load on lintel: 22.0 kN  = ((0.56*1.805) + (5.78*0.881) + 3.12 + 2.74 + 2.28 + (8.7*0.629) + 2.28)

Quick worst case calculation by putting full load at lintel mid-span. The maximum moment from a point load mid-span is double the moment of the same load spread as a UDL I believe. Therefore equivalent UDL is 24.4kN/m  = (22.0kN * 2 / 1.805m)

That's a bit scary!

 

A more considered calculation:

 

 

I want an equivalent UDL loading to compare with lintel specifications. A UDL across a beam has a maximum bending moment of M_max = q * L² / 8 . rearranging gives q = M_max * 8 / L² . Putting in the maximum moment from BeamGuru gives q = 5.32kNm * 8 / (1.805m*1.805m) = 13.1 kN/m. That's much better ?

 

A 2100mm Naylor R9 lintel has a stated Allowable Load of 25.78 kN/m with 1800mm free span. This is very nearly double what is calculated here. I expect there would be some load spreading onto surrounding walls as well reducing the loading on the lintel, but I don't understand all the mechanics of that, so have tried to keep the calculation worst case.

 

There would be some load spread by the blockwork between the BLUE lintels and the PURPLE one, but I wanted to consider the case similar to not having that blockwork. I might repeat the calculation assuming 45 degree load spreading under the BLUE lintels. These lintels have short free spans (424mm) and are chosen to match masonry of help support the flue base.

 

The Structural Engineer hasn't shown much interest in all this when I have tried to get him to take a look.

 

Tools: QCAD, OnShape, BeamGuru.com

[ToughButterCup]

I feel for you. 

Quote

 

The Structural Engineer hasn't shown much interest in all this when I have tried to get him to take a look.

 

I have come to understand professionals' reluctance to take over someone else's work... It's a bit like being handed blocks of JavaScript written by someone else and being asked to sort the 'issue' out. 

 

Q. What's the issue? 

A. It doesn't work. 

 

It takes longer to unpick the error than to rewrite it ab initio. 

 

Might something analogous be happening here? 

 

PS. Some interesting reading in your OP for me this evening. Thanks. Can't understand it yet, but.... 

 

Edited September 26, 2020 by ToughButterCup

[MortarThePoint]

1 minute ago, ToughButterCup said:

I feel for you. 

 

I have come to understand professionals' reluctance to take over someone else's work... It's a bit like being handed blocks of JavaScript written by someone else and being asked to sort the 'issue' out. 

 

Q. What's the issue? 

A. It doesn't work. 

 

It takes longer to unpick the error than to rewrite it ab initio. 

 

Might something analogous be happening here? 

 

 

I know what you mean there, but I tried to get the Structural Engineer to do it all at first and then did it my self out of despiration. I should have said he didn't show much interest in the whole area.

[ToughButterCup]

Phone call time? 

Or Yellow Pages? 

[MortarThePoint]

I described the setup and gave some of the images to lintel suppliers. I shared more about my thoughts on the GREEN lintel, but didn't want to 'lead the witness' on the PURPLE lintel so just shared images and load numbers.

 

For the GREEN lintel, one supplier said a 140w x 140h lintel would be OK, whilst another recommended a 140w x 215h lintel saying it would triangulate, but did his calculations based on it not to be on the safe side.

 

For the PURPLE lintel, the same first supplier recommended a 100w x 215h lintel and the other said to consult a Structural Engineer.

Edited September 26, 2020 by MortarThePoint

[MortarThePoint]

One thing I am not so keen on is the PURPLE lintel bears on a 100mm wide wall effectively. It must be stabilised a lot by the corner with the 140mm blockwork, but it's always nice to think of something bearing on a fat wall. The bearing itself can be improved by widening the wall slightly as shown below.

 

 

Edited September 26, 2020 by MortarThePoint

[Mr Punter]

Who did the design?  At first floor level, why are there are four leaves of masonry and two layers of wall insulation?  Also, what is the point of using the pumice and twin wall on the same chimney?

[MortarThePoint]

3 minutes ago, Mr Punter said:

Who did the design?  At first floor level, why are there are four leaves of masonry and two layers of wall insulation?  Also, what is the point of using the pumice and twin wall on the same chimney?

 

Its always struck me as a bit nuts that the chimney stack inner blockwork is insulated from the house. That's how the Architect designed it. It would take a lot of load off the PURPLE lintel if that blockwork tied in to the house inner leaf.

[joe90]

Gosh, with mine it was inside the house so no need for insulation. We used pumice from base plate, via two bends like yours, hauntched with weak Mortor and up the right hand side just like yours. Our bottom lintel was oak (although it only holds up brickwork to first floor. In first floor isokern surrounded by blockwork supported on lintels like yours and back filled with vermiculite. So very similar but slightly different due to being internal only. (We did no calculations and BC did not question it.

[MortarThePoint]

12 minutes ago, joe90 said:

Gosh, with mine it was inside the house so no need for insulation. We used pumice from base plate, via two bends like yours, hauntched with weak Mortor and up the right hand side just like yours. Our bottom lintel was oak (although it only holds up brickwork to first floor. In first floor isokern surrounded by blockwork supported on lintels like yours and back filled with vermiculite. So very similar but slightly different due to being internal only. (We did no calculations and BC did not question it.

 

I'm increasing thinking the insulation between the house and the chimney is pointless and results in a lot of complications.

[Mr Punter]

Could you do it like this and maybe even have it step in a couple of hundred mm?

 

[MortarThePoint]

4 minutes ago, Mr Punter said:

Could you do it like this and maybe even have it step in a couple of hundred mm?

 

 

It does seem more sensible doesn't it. I don't know why the Architect has insulation there.

 

It would have to transition back to the Architect's plan at wall plate so that the wall of the chimney on the house side is back aligned with the outer leaf of the house wall. Otherwise the chimney would end up much squarer above roof level than originally planned. That would involve some lintels at wall plate level though which doesn't seem an issue.

 

I have shown the outside edges of the chimney brickwork above roof level in RED:

 

[MortarThePoint]

The tiredness is getting to me, a six month old and self building is a killer combo.

 

Measuring an elevation drawing (at 1:50 scale) it has the chimney wall on the house side above the inner leaf of the house wall, not the outer leaf. Ignore the drawing in my last post.

 

My comment was correct as it's referring to the blockwork of the chimney needing to align with the house wall outer leaf.

 

Up to wallplate through, there shouldn't be any reason the chimney can't join the house's inner leaf as @Mr Punter suggests.

 

I still need the PURPLE lintel to be strong to support the house wall outer leaf where it bears on it.

 

Joining the chimney stack blockwork to the house inner leaf below first floor level would increase the load on the GREEN lintel a bit.

 

Elevation View:

Plan View: