Blown-in cellulose insulation for a flat roof

Hello everyone,

we are planning soon the renovation of a typical bungalow from the late 60s. Regarding energy efficiency, we are open, but for budget reasons we want to focus on quick wins if possible. The windows have mostly already been upgraded by the previous owner to very decent triple glazing, the façade is clad with bricks and in good condition, unfortunately without an air gap – we definitely want to keep our hands off that and maintain the nice condition.

On the other hand, the roof seems to have potential. Currently, we do not yet know the condition – technically the roof was refurbished about 15 years ago (new sealing on top), but we fear that the insulation was not the focus then. So we assume a cold roof with moderate insulation above the air gap, which was probably installed only because of the pitch. Oh, and it is a wooden beam construction, the membrane on top is covered with gravel.

Now a craftsman company, which will be there for other work, has proposed cellulose blown-in insulation with simultaneous sealing of the attic. The whole thing sounds surprisingly cheap in terms of cost (four-figure).

What are the pitfalls here? How effective is something like that? Will it have structural significance in terms of weight? I fear that hardly anyone will guarantee the exact condition of the beams anymore; other than a visual inspection, this can hardly be determined. Or is that irrelevant because cellulose, quite naively said, is insignificant in terms of weight?

Sorry if this topic already exists, but I have looked back to 2017 and found nothing. And technology does change...

I’m pushing the topic up again, unfortunately there have been no answers so far, but it’s worth a try.

We have made some progress in the inventory analysis - old drawings were found and a visual inspection has more or less confirmed everything:

From inside to outside, the structure is as follows: plaster, 5cm Heraklith (wood fiber insulation board), 2.5cm glass wool, then the ventilation gap from outside - the space between the beams. Above that it becomes diffuse. From below you can see a corrugated cardboard, which is not found in the drawing, maybe it’s the old formwork. Above that presumably the “new” roof with an insulating mortar, which due to the ventilation probably has no energetic function.

If I model the whole thing in Ubakus (as mentioned - everything above the ventilation has no function), I get a not bad U-value of about 0.4x. That’s not top, but clearly better than “uninsulated”.

If I blow in a wood blown insulation there now (unfortunately cellulose is not available at Ubakus), the U-value drops to about 0.16, but the dew point then lies below the upper layer in the blown insulation. With 133 days drying time, 25% wood moisture and over 2kg water per sqm, Ubakus strongly advises against it. Makes sense, basically we now have an interior insulation, for which the rest of the structure is not designed.

A vapor barrier under the Heraklith would provide remedy here, even the thinnest one would push everything into a green zone - but that is not installed and tearing off all the plaster would probably drive the costs toward a conservative roof renovation.

What do you think? This problem should affect everyone who treats their roof this way, otherwise this blown insulation wouldn’t exist. Or can cellulose handle this strong moisture input?

We have a new bungalow with a half warm roof (open living area) and a half cold roof for storage. If you do the warm roof, insulate the roof from the inside, if you do the cold roof, insulate the ceiling between the rooms. Both are different, which one do you want?

If I now blow in a wood blown-in insulation there (unfortunately cellulose is not available at Ubakus), the U-value decreases to about 0.16, but the dew point then lies below the outer shell in the blown-in insulation. With 133 days drying time, 25% wood moisture, and over 2kg of water per sqm, Ubakus strongly advises against it. Logically, in essence we now have an internal insulation, for which the rest of the construction is not designed.

A definite statement can be given to you here (and also by no other neutral person). The conversion from ventilated insulation to "internal insulation" is always critical if the diffusion to the outside is not clarified. For blown-in insulation in masonry, this is usually quite easy to assess and leaks towards the masonry are also easy to detect.

Flat roofs are doubly risky. On the one hand, because the outer layer (bitumen or EPDM) is usually diffusion-tight and therefore moisture introduced can hardly escape. On the other hand, due to the inhomogeneous structure of the ceiling with penetrations and connections, room air more frequently enters the insulation layer. This leads to more moisture ingress.

Personally, I would keep my hands off in this case and, if it is to be done anyway, in any case check the ceiling sealing and, if necessary, renew/repair it. In particular, connection points at the walls and cable penetrations should be checked to ensure that everything there is really airtight. The latter does no harm even if no insulation is blown in. So far, the moisture can then still dissipate, but the heat loss is still there.

Alternatively, installing 30mm PIR + plasterboard on the underside would be a solution. Not ecologically perfect, but physically harmless.

We have a new bungalow with half a warm roof (open living space) and half a cold roof for storage. If you do a warm roof, then insulate the roof from the inside, if you do a cold roof, then insulate the ceiling between the rooms. Both are different, which one do you want?

Thanks for the reply, but sorry, I don’t quite understand the question. The roof is as it is, it’s not a matter of choice. I think this roof construction is generally referred to as a cold roof (roof surface cold, since insulation and ventilation are underneath). The question was now, where the risks lie if I fill the ventilation with blown-in insulation.

Nobody here (nor any other neutral person) will be able to give you a definite statement on this. The conversion from ventilated insulation to "internal insulation" is always critical if diffusion to the outside is not clarified. With blown-in insulation in masonry, this is usually quite easy to assess and leaks to the masonry are easy to detect.

Flat roofs are doubly risky. On the one hand, because the outer membrane (bitumen or EPDM) is usually diffusion-tight and therefore moisture introduced can hardly escape. On the other hand, due to the inhomogeneous structure of the ceiling with penetrations and connections, indoor air more frequently enters the insulation layer. This introduces more moisture.

Personally, I would stay away from this in this case and, if it has to be done, in any case check and possibly renew/repair the ceiling sealing. Especially connection points to the walls and cable penetrations should be checked to ensure everything is really airtight there. The latter does no harm even if no insulation is blown in. Up to now the moisture can still dissipate, but the heat loss is still there.

Alternatively, installing 30mm PIR + drywall on the underside would be a solution. Not ecologically perfect, but physically safe.

Okay, thank you very much! So basically more of a classic internal insulation from below? Why then are these blown-in insulations for roofs offered on a large scale at all? Friends of ours also have a bungalow and were offered that as well. But you can’t find any experience reports about it online.

Why then are these blown-in insulation materials for roofs even offered on a large scale?

Whoever has a hammer finds a nail everywhere. You can do it, and in many cases, it is a cost-effective and simple solution. Everything can go well in your case too. But no one can really look precisely inside the roof, especially when it is blown in. If you install PIR on the underside or even just connect a vapor retarder airtight, you can have everything blown in there as well. Physically safer is the current construction.