As part of our "green" build, we have a steel portal (house shaped and holding up the ridge beam) bolted down on to the uninsulated concrete ring beam. I've just asked our architect to come back with solutions to this perfect cold bridge, but wonder if anyone has any knowledge of same. I wonder if there are plastic heavy duty blocks that the steel could sit on?

I think what you are looking for is Foamglas which has a really high compressive strength so you should be able to put a block or slab of it between the base of the steel and the concrete ring beam. Although im nowhere near knowing if this would actually work and even then you would still have thermal bridging through the massive bolts needed.

We're using Farrat TBG300 for a beam flying through the wall.

http://www.farrat.com/products/tbg_300.html

Foam glass has a compressive strength of 2.5n/mm2 and a conductivity of 0.055 w/m.k

TBG300 has a compressive strength of 300n/mm2 and a conductivity of 0.18 w/m.k

So 120 times the strength and 4 times the conductivity. If you can get the contact area down (by using the cut out methods shown on the web site) you should be able to get a lower loss than the foam glass, esp if you combine with insulating washers and stainless bolts.

You could also increase the thickness by using 2 sheets stacked.

I was quoted about £40 for a piece 175x100mmx25mm custom cut to my drawings.

hope this helps.

Ahh that’s new to me, quite interesting to read about that product indeed it’s one to remember.

Thanks beelbeebub

I recon everyone ignores problems like this, I designed all my thermal bridges out at planning stage but planners dont care about this stuff though by the principles of CSH they should.

blind eyes everywhere -- how come building regs dont pick up on this?

nice to find someone who cares!

Best solution would be/have been to bring the ring beam into the thermal envelope by putting the insulation on the outside/underneath of it. Then the slab and steel would be at ambient temperatures providing thermal mass at low levels of the loss.

The Viking solution is essentially this, as is: http://www.aipassivhaus.com/blg-june09.html

Is it too late for this?

Beelbeebub: "Foam glass has a compressive strength of 2.5n/mm2 and a conductivity of 0.055 w/m.k

TBG300 has a compressive strength of 300n/mm2 and a conductivity of 0.18 w/m.k

So 120 times the strength and 4 times the conductivity. If you can get the contact area down (by using the cut out methods shown on the web site) you should be able to get a lower loss than the foam glass, esp if you combine with insulating washers and stainless bolts."

Sorry beelbeebub but a thermal conductivity of 0.18W/mK is over 3 times worse than Foamglas which is not great at 0.055W/mK.
In coldstores of old beneath the steel columns supporting mezzanine floors they used hardwood as the thermal break that would have a thermal conductivity similar to TBG3000.
Extruded polystyrene (XPS) with a compressive strength of 350kPa was used over the rest of the floor.
The hardwood has generally now been replaced by high compressive XPS like Dow's Styrofoam HD300 which is a 700kPa foam with a thermal conductivity of 0.025W/mk (0.029W/mk to the 90/90 rule). I think Jackon/Gefinex also has a 700kPa XPS foam

Dickster, I'd ask an engineer to check out the loadings and see if this foam would work in your situation, maybe with spreader plates?

The point loads should be taken no further than the wall plate if possible and spread out through the walls at 60 degrees to give more spread out loads at foundation level, unless you are building a sun room?
You can still dig 1m x 1m x 500mm deep holes/pads, line them with dense EPS or XPS underneath and at the sides like a biscuit tin. and pour concrete into these to give you insulated point load pads. each pad can take 3.5 tonne if you use XPS.
Its always better to keep steel either warm or cold, if the steel is passing through the insulation line in a well insulated house it will be the coldest surface in the house so will be prone to condensation and may constantly weep in cold weather, possibly giving ice inside the house in extreme cases. I think you should otherwise try to wrap the beam in XPS sealed at the joints to keep water vapour out.

Sorry for resurrecting an old thread but struggling with my structural engineer and wanted to float a potentially daft idea/solution.... drum roll... A diy multifoil insulation bootie for the steel column and base plate.
TBG 300 pricey and only insulates the bottom of the plate
VH ideas excellent but Bregs and Structural engineer unhappy for several reasons (novelty and trees mostly)

Steel post goes through 100mm steel reinforced slab, to DPC, through to 300mm EPS, through 250mm compacted crushed stone to base plate onto pad foundation.

Bitumen painted steel to guard against condensation.

Could it work?

Thanks

Zombie thread, I know, but...

Posted By: dicksterHowever small steel section means that steel bolts form 25% ish of joint anyway, so thermal bridge reduction is limited.

But if it's ordinary steel section with stainless bolts the bolts will have noticeably less than half the thermal conductivity of the section so thermally about 10% of the joint.

http://www.engineeringtoolbox.com/thermal-conductivity-d_429.html

Posted By: Novice1Hit a brick wall on all 3 counts with structural engineer..... Dismissive says Insulation will move

Only by 0.5% during loading!

Posted By: Novice1Thermal bridging insignificant.

The steel would probably read 15oC-16oC when its -10oC outside, which is colder than non thermally broken window frames.

Posted By: Novice1 Need help finding a second opinion from a structural engineer experienced with passive type details.

Help is available from experienced structural engineers.

>cladding the steelwork with insulation in its entirety up through the house...

That's what we just did.
You need to watch this closely as it happens though. It sounds easy but it's amazing how many tricky spots you come aross in-practice that were not noticed on the plans. Like a lot of things discussed here you need to be practical about it and fix them as you go. We still ended up with a few areas that were less insulated than I would have liked.

> TBG300...

Thanks. Interesting stuff now noted and filed for future reference.
It looks a lot like what we used (together with a tube that isolates the bolts) where the external wall anchors bolt to the steel frame.

Hi dickster. A few thoughts, for what they're worth

Idea 1 Use insulating pad (plastic, Foamglas or hardwood) and bolt through as normal using stainless bolts for lower heat transfer

Idea 2 Spray plastic foam insulation on the steelwork for at least the first couple of metres. Even if the steel is cold by conduction from the concrete, there will be no cold steel surface available for condensation

Idea 3 Use plastic top-hats or insulating sleeves and washers for the bolts, rather than direct contact with the steel.

Use all three together

Posted By: SprocketWhat am I missing (that fits on my beermat and would make a big difference)?

The rate of change maybe. Steel conducts heat quickly, insulation does not. So, and I may be wrong here, if you get a short term temperature difference, it conducts a disproportional amount out heat away. Though as you say, it is still a small fraction of the other losses.

Well that's about 600kwh per year lost to the pads.

So at 14p/unit that's about £84 a year.
Still sounds pretty insignificant (for this building) to me.