Purlin problem

[Roger]

I was in my spare bedroom late last evening looking for an item of long lost equipment (a 40 year old climbing rope if you must know) when I happened to look up and noticed that the horrible brown paint on one of the upper purlins (the ceiling is in the plane of the roof) was flaking off near one end and cracks previously hidden by plaster infill were starting to gape. Closer inspection 9 feet up suggested that the purlin was indeed close to collapse. While the wood of the lower half of the purlin where the horizontal gaping cracks are is as hard as a very hard substance the upper surface is disfigured at one point by a vertical crack in wood so soft and can insert my penknife blade a good inch into the beam (away from the crack) without exerting myself. Presumably the upper surface is wormy sapwood while the lower surface is heartwood.

The beam itself is a rough dressed tree trunk and I know from earlier work that the butt ends of at least some of these purlins still have traces of the original bark. The roof is at approximately 30 degrees and there is about 5 feet horizontal between the failing purlin and the lower purlin that side and another 5 feet to the ridge pole above. The roof is thick Yorkshire stone slates and I estimate there is at least 13 cubic feet of stone in that small section.

Any remedial work is going to involve removing part of the roof so that will have to wait till late spring at the earliest but I do need to plan ahead. The 2 considerations at present are:

Where can I source a seasoned oak mildly dressed tree trunk roughly 8 inches square and some 11 feet long? It might even be an advantage if it is not particularly straight. The beam it is going to replace certainly isn't.

To what extent do building control need to be involved? It used to be the case that simple (simple??) repairs like for like didn't require any BC involvement but the nanny state is getting everywhere and if I take this opportunity to improve the insulation of the roof ISTM that BC will insist on getting their fee.

Meanwhile I have stuck an acro prop under the suspected break and, as the purlin is not exactly under one of the major floor beams below, another acro prop and spreader across the ceiling under the minor floor beams. The props are not particularly in the way as the suspected failure point is only about 18 inches from the wall.

[Andrew Gabriel]

Or it's gone rotten through moisture.

Personally, I'd get a structural surveyor to come and have a look, as a matter of some urgency. I rather imagine that if an area of the roof did collapse, it would quite possibly kill someone indoors or nearby outdoors.

A few years ago, a house up the road from me was suddenly cordened off with police "do not cross" tape. There was no sign of anyone there, not what you might expect if it was a murder crime scene. On the way past a second time, it looked in again, still no change. Then I happened to glance up, and realised why -- the roof had done the splits. The ridge was sagged right down in the middle, and the lower edges sticking out way past where they should be. Didn't look like any of the tiles had come off, so probably no one injured. House was only 5-10 years old -- I never heard what the cause was, but it took best part of a year before repairs finished and scoffolding finally removed.

[Roger]

The message from [e-mail] (Andrew Gabriel) contains these words:

Wormy sapwood and solid heartwood are typical of old oak. There has certainly been moisture ingress in the past. The roof was only felted for the first time in 1978 and I suspect that that wasn't 100% successful as the roofing laths weren't counterbattened first and the slates are nailed rather than pegged.

I tend to act as my own (rather rusty) structural engineer.

Given the positioning it is unlikely that any roof fall would would be external although there is a slim possibility that slates on the periphery might slide. The internal dividing walls are stone and the lower purlin looks sound so there is no reasonable possibility of the damaged area spreading.

The building as a whole is a hovel that was jerry build some time before

1841* (where some time might possibly be a century or two) so I wouldn't entirely discount the possibility of an external wall falling down at some time in the future, but I have long since rebuilt from the ground up the 2 sections that were most at risk.

*Date of first OS map of the area.

[Roger]

I am a bit disappointed at the lack of response to my original posting. No one has as addressed either of the questions I posed.

Anyway I have been thinking a bit more about the practicalities of doing the work myself and even whether it is too much for a 64 year old weakling.

The first problem is getting access to the failing purlin. Much as I would like to avoid the expense I think I will have to have at least part of the house elevation scaffolded. At a minimum I will have about a ton of stone slates to lift and getting them down a ladder one at a time (and even worse back up) is a daunting prospect even if I don't reach the 3 foot wide slate somewhere near the centre of the roof. With a scaffold at least some of the slates could be left topside. When the roof was done back in 1978 none of the slates actually left the roof but the end result wasn't particularly satisfactory and I had to relay several sections in the following decade. The ridge stones are about 30 inches long and carved from solid stone. They will have to stay balanced on the ridge, with a supporting framework if necessary. I replaced one years ago and had to cut the replacement in half before I could carry it up the ladder.

The second problem is whether or not I can get the old purlin out and the new purlin in without removing the rafters. If not it would mean that all the slates in that section would have to come off and all the lath and plaster ceiling come down.

The third problem (and the one that really bothers me) is the weight of the purlins. Getting the old one out will be easier than getting the new one in as I could, if necessary, cut it up in situ. If my calculations are correct a replacement 8" x 8" oak beam 11 feet long is going to weigh about 250 lbs which is more than double what I could lift directly but probably not too heavy to suspend via a hoist from the rafters. Back in 1990 I had some considerable difficulty reinstalling a 5 foot x 1 foot by 6" stone lintel when I rebuilt a section of wall. It was as much as I could do to tip it on end and it needed a trolley jack to raise it to position from the window sill. That probably weighed about 400 lb but it was a much more manageable size and I was 18 years younger then.

All other problems seem to pale into insignificance by comparison at the moment but no doubt more tricky ones will surface in due course.

[Roger Mills]

If the damage to the purlin is at one end, couldn't you support it on steelwork, cantilevered off the end wall and extending under the sound part, without needing to replace it?

[mail]

Dear Roger It would be helpful to have more information in terms of dimensions and extent of decay. What is the span of the purlin? You imply it is 11 feet but is this its lenth in total of its span What supports the purlin at each end? Presumably one end is a wall and the other a truss? What is the total depth of decay in the eight inches? Are you able to determine it goes all thru the whole CSA?

From a read of you post I get the impression that you have a typical oak or similar hardwood timber which has been rougly squared off but will plenty of wane left and some bark In such timbers it is very common for the sapwood to be eaten away with Death Watch or CFB Ap (wood worm) and what is left is normally fine to carry the load.

What I think is the most important question you need to answer is "What is the deflection?" Can you establish this and to verify you concern about iminent collapse ( and I am sceptical) you should consider a crude form of measurement of any condinuing deflection. Ideally a digial movement meter from a stable base or failing that get a vernier scale from a fixed datum and mesure it! Wood is a truly marvellous support material and will not catastropically collapse unless decayed and you appear to have a strong tension zone (and ipso facto some compression zone) intact so I would only fear collapse if there is evidence of serious decay or serious deflection.

What to do? First you need to make a realistic assessment of its deflection If this is not moving and initial assessments assure you that there is sufficent meat on the timber (say greater than 75% CSA then you are likely to be able to do a repair in situ which is what I guess is most likely. EVEN if there are considerable shakes a repair is possible as the tension zone is intact.

Send me a photo and I can advise better but I imagine a repair along the following lines

1) prop from below with angled fillent to take the 30 degrees down to a suitable surface for load bearing and in as many places as needed 2) investigate by drilling with a Siebert drill the CSA and make a profile of the timber along its length c) calculate if remaining timber will do the job if so cut out the rotten (not insect affected only decayed) timber and support rafters above with folding sliding wedges d) if not run a RSC along with plates above and below to take the beam load (this may have to go in the wall as well) (a timber might be ok instead of a steel

None of this involves removal of the roof covering

If it transpires that the timber does need replacing consider a vertically laminated CCA treated composite beam as an honest modern repair If you insist on a big original one piece timber consider the old trick of cutting it in half vertically and swapping one half upside down and bolting together with or without flitch plate - this spreads the effects of any defects Chris

[Roger]

The message from "Roger Mills" contains these words:

Thanks for the input.

The centre of the damage is about 18 inches out but the gaping cracks extend at least the same again.

Unfortunately there is no loft, the ceiling is in the plane of the roof and any support would stick out like a sore thumb. If it were in a loft I would be sorely tempted to put in something crude but practical.

It might be easier to put an RSJ immediately below but that wouldn't be visually much more attractive than a wall support.

To put in a loft in that room would change the character of the house and might be considered a complete bodge given the circumstances. And I and not sure it would be any less work.

[Roger]

The message

from [e-mail] contains these words:

Thanks Chris

No, that is the length of the timber required. The span is marginally over 9 feet, probably not much more than an inch.

One end is in what would be an outside wall if it wasn't for an attached barn but the wall does project a little way above the barn roof. One end of the purlin is buried in the wall above barn roof level. The other end is buried in an internal wall. Both walls are rubble filled stone walls. The outside is some 2 feet thick while the internal wall is about 1 foot

8 inches.

Not clear how far the decay extends into the inaccessible face of the beam but the upper accessible face is mostly spongy to a depth of about an inch. (But see further below).

I think it is oak and some of the wood is certainly very hard. Any bark would be on the buried ends. There is some chamfering but the beam looks more like a peeled branch than a regular beam in places.

Plenty of evidence of ancient woodworm but not AFAIK of Death Watch. The house came complete with a 30 year treatment guarantee that a previous BS had insisted on but that has run out.

Deflection at present is not much which is why I didn't run screaming from the room when I first noticed it. I will have another look but I was basing my prediction on imminent collapse on the way in which the paint has flaked off the stressed areas round the horizontal cracks and the way the horizontal cracks have widened enough for the plaster infill to fall out. If the infill hadn't moved I would be far less worried. Almost every major beam in the house, both purlin and floor have a good many horizontal cracks.

I have no idea how long the situation has taken to develop. It could have been overnight but equally it could have been several years and just gone unnoticed. I have a piece of timber between the top of the acro prop and the beam which can still be twisted by hand and if that tightens up I will know it has moved. If the rafters are sound they will be providing a degree of support.

It was the vertical crack in the sapwood that suggested to me that the beam was close to breaking but that could have been a red herring. I might try chiseling off the soft stuff to see what it is like below. I reckon at least 25% of the CSA is not contributing anything to the current strength of the purlin.

[e-mail]?

Is this a broadband connection?

Sorry but I have no idea what a Siebert drill is.

As I mentioned in my reply to Chris this is an exposed beam and I don't want any repair/replacement to look out of place.

I have just had another close look at the damage. Ignoring the orientation of the beam (closer to the plane of the ceiling than the horizontal the bottom face is hard and relatively flat. Only about 7" wide at this point. The section tapers slightly towards the ceiling and closer to the hidden face on which the rafters sit there is some rounding. On one side of the beam there is prominent crack about 18" long and about 1" at its widest. This crack goes at least half way through the beam. The acro prop is positioned about midway along this crack. Above the crack and from about the same position the other side the wood quickly becomes soft to a depth of about 1.5 inches, not the 1" I previously suggested. The vertical crack (actually a square wave form) doesn't appear to be any deeper than the surrounding soft wood.

There appear to be other areas of infill plaster that haven't moved and which I haven't tried to dig out. Replacing some of the fallen plaster pieces suggests that the large crack has gaped rather more than an eighth of an inch since the cracks were filled.

[somebody]

In message , Roger writes

[snip]

There are lots of timber repair systems - usually a mixture of resin and steel. Certainly end decay is fairly straight forward to repair, decay along the length a little harder.

I saw a video at some point where a groove was cut into the top of a beam, steel rods inserted and then filled with resin. I'm afraid I can't find it though to point you to.

Crack repair can also be dealt with by resin injection.

Google around for more specific info.

Hth Someone

[Roger]

The message from somebody contains these words:

The crack that is giving most cause for concern is some 18" long and has a 3.5" x 1" triangular section at its widest point. That would take some filling and, given the the ancient nature of the visible internal surfaces I think there many well be an adhesion problem. However I am all in favour of any solution that would avoid the necessity of taking part of the roof off, at least for the moment. Some time in the future the roof needs relaying anyway.

I will try. I haven't made a very good fist of finding solutions so far. :-)

[mail]

Roger resin repairs are to be avoided where possilble as they are not reversible but can be of use as a gap-filling adhesive. If, as appears to be likely, you need an unobtrusive repair of the compression zone and a connection to the tension zone over 18" of cracking - you need an engineer to design a repair. I now envisage (ignore the ~RSC idea in my first post) a steel plate - recessed or otherwise to suit - along the bottom of the purlin with bolts running up to the top where you will have replaced the comrpression timbers with new timber and gap filling resin. All nice and unobtrusive albeit fiddly to do.

I will answer questions on my first post separately

chris

[mail]

=46rom chris Try to quantify it compared to other purlins and with itself over time

which is why I didn't run screaming

=46rom chris Good plan but see if you can get a measurment system going

If the rafters are sound they will

yes mail@

Depends where I am! but both my regular location are bb

=46rom chris See below

=46rom chris now understood

=46rom chris In oak this is not uncommon - it is a shake cure is to stitch top and bottom together with steel bolts

The acro prop is positioned about midway along this

not really much of a problem compared with the whole of a 8 x 8 timber but as it is mid span maybe worth doing something much depends on the conjuctioin with the shake and its location

do not really understand this I wait for the photo

That is a lot but not if the cracks were filled 300 years ago! Chris

A Siebert drill is a specialist 1 mm (ie yes 1000th of a m wide that goes 300 mm into timber and produces a graph of density. The holes are so small that very little structural damage is done by the drill and you can get an accurate profile of the CSA by drilling every 150 mm longitudinally and say 50 mm across the CSA.