arches

Caught out

I have been caught out by the NHS. Got a phone call on Thursday to say come in on Sunday to have half your pancreas removed. Needless to say, things have got behind and will get more so.
Hope to be back here in about two weeks.

Bill

The Dean's Eye again

What with one thing and another I haven't managed to visit Lincoln for some years and I was aware that there was some resistance to these ideas. Hamish pointed out that the Lincoln Cathedral web site has a lot on the Dean's Eye. It's really rather sad if it is being built in an incorrect form. I wonder how they will make it stand up?

More on the Dean's Eye

I promised more on the Dean’s Eye and then got distracted. Let’s try again before 9am comes and I have to start real work.

I said that the tracery was too thin to withstand wind load. That isn’t entirely true. There are several ways it could be made to work, but which ones are more likely.

For some time, there have been two iron bars spanning diagonally across the window and providing supports at difficult points. That is certainly not a mediaeval solution.

I got to thinking about what I would want to do if I was a master mason wanting to create something spectacular. I think the answer is I would have tested it. The window only stands a chance if it can develop sufficient compression to span as an arch. I think it is possible that the spanning is broken down into sections.

That brings us to the Plastic Theorems which essentially say that the structure will try every way it can to stand up before it gives way. Many engineers find it hard to accept the application of the plastic theorems to masonry, but if it is good enough for Jacques Heyman it is good enough for me.

If we begin by thinking about the centre piece of a quatrefoil surrounded by four large circles. There is a construction joint between this section and the outer ring of circles, so the mason obviously constructed it, and though of it, as a separate section. If he trial assembled it on the ground (and he surely would have done that) he could not have carried out any tests because there would be nothing for it to push against.

Could he have used other available technology to make a test possible? I think he could have got the wheelwright to hoop it with iron, shrunk on to pre-compress it. With the hoop in place, the whole section could be lifted off the gound, supported on a ring wall and loaded with bags of sand to give confidence it would work..

What is more, that bound ring of stone would only weight about 1 ton and would be well within the capacity of available lifting technology. Could it be that the window is still prestressed?

Perhaps an architectural historian somewhere has some ideas on this?

The Dean's Eye

Many years ago, Peter Ross, then at Arup R&D challenged me to produce a thrust line analysis of a rose window that showed it would work. I never quite got to that point. It is a difficult three dimensional problem. I will get there one day though, because they do work so it MUST be possible to find the load paths. I do have some ideas and they will be expounded below.

A couple of days ago, while working on the cathedral bridge, I found myself talking to Bruce Duncan (Lazenby Chaplain at the University of Exeter) about other things I do and that got me back to thinking about roses and about the Dean's Eye at Lincoln in particular.

The Dean's Eye is huge. I have a photocopy of a photogrametric survey which shows the outer diameter as about 8.5m and the main diameter of the tracery as 7.3m. Converting those to feet gives 28 and 24. So, there is a 2ft chamfer on the walls, reducing the thickness to that of the tracery which is only 8in (200mm). We have an open network of stone only 200mm thick and 7.3m diameter. It is high in the north wall of a cathedral which in turn is on top of a hill. There is nothing between it and the north pole to soften a northerly wind. It has stood up for the better part of 700 years, albeit with some repairs along the way.

What is very strange about the layout of the window, though, is the use of iron bars within the stone and bronze bars outside it in some places. The drawing I have done of the window as it is now shows the bronze bars crossing at the centre. The iron bars are in the straight members which bear on the hollow faces of the stone surrounding the centre quatrefoil.

I was uncomfortable with the bronze bars from the moment I saw them, but when I saw the photogrammetric survey, I realised that the window could not, originally, have been built like that. If we take out the spurious, relatively modern, parts (we know the stone isn't original because it comes from a different place) the picture looks like this.

It is, surely, obvious that the original window was built like the final drawing. Here, there is at least a relatively direct path through the stonework across the cente of the window.

I think I will stop there and leave further thoughts on Roses for a separate post.

What do we really know about arches?

I am just joining a bid with Gifford for a project with CIRIA to write a best practice guide on arch bridge assessment and maintenance. reading the brief has left me with a hightened impression of our lack of knowledge.

Arch bridges represent about 60% of the bridge stock in the UK. That proportion is falling as new bridges are built and neglected arches are replaced with (often cheap nasty, and short lived) concrete. I've heard it said that bridges represent 60% of the asset value of the UK railways so Arches may be as much as 25%. I have been looking at arches seriously since I started research on Wade's Bridge in Aberfeldy in 1981. What I mostly see is neglect. I will dig out some photos soon and post them here as a sample. On the research front, we have reached the stage where we have a number of tools capable of predicting the behaviour of simple arches, but nothing for everday use that will take into account the whole system of an arch bridge.

The real problem is serviceability. There is rarely any true problem with the strength of arches (though the simplistic analyses we are limited to do sometimes suggest a problem [actually, that isn't quite true, of which perhaps more later]). They just deteriorate until they can no longer do the job. Most of that deterioration could be halted with a little care. In many cases it could even be reversed. Because the costs cannot be quantified (ie what will it cost of I don't do this) the inkies and bean counters say the work isn't needed because we can't show a significant cost benefit. (oh, and by the way, no you can't have the staff you would need to work it out either).

We owe it to our children to sort out these layers of problems. We must understand the bridges better. We MUST sort out the economics properly and we MUST MUST MUST conserve what we cannot or will not afford to replace.

If we don't really know the static load capacity of an arch bridge, what chance do we have of putting limits to serviceability and predicting when they will be reached. The only possibility is professional judgement and no one in serious control of money is prepared to accept that any more

A fun job but a serious one

I am trying to design a "knock down" arch bridge for use in a church service. The design constraints are pretty severe.
 It has to be built by amateurs
 It has to cost almost nothing
 The parts must be light enough to position by hand
 It must be strong enough to carry two people.
 It must be big enough to be seen from about 50m in a crowded Cathedral.

So far, it is looking like £200+, which is certainly not nothing. There are some bits that will need to fit pretty closely so it will have to be jigged if it is to work.

Arches don't like to be light. They need weight to deal with concentrated loads. The best solution I have come up with so far is prestressing with a lorry strap. I have put a pdf picture of the design so far on the picture gallery. Each voussoir will be a box of 6mm ply with 50x50 timber for the main bearing pieces. The handrail pieces double as locators for the joints. I will add a50x25 rail between them on top of the deck to provide a lip and a tread to prevent slipping. A tie over the top of the hand rail posts will provide a rail and also stress everything together to make it stable.

I will need to put wheels on the bases, but they will also need jack ups to make them level on an uneven floor.

Any thoughts on the design so far ,or on sources of material? There is a reply address for private coms or you can go public by responding on the site. My client (the Christian chaplains at Exeter University) have approached some constractors on the site and we may be able to make things from cast-offs.

How I got into this.

I have been working on arches now for 22 years. That is a lot of accumulated thiking, to say nothing of the pictures and data. I keep trying to find ways to make some of that more puclicly available. Trouble is, it has to be relatively painless or I won't do it. My son Hamish started a blog some time ago and it looked interesting but didn't get me going. Last week I saw an article in the Guardian on Typepad and decided it was time to make a concerted effort.

At the same time, I am trying to finish off a few jobs before going into hospital and to clean up the OBVIS web site and create a proper one for Bill Harvey Associates. No doubt something will give, but at least there is a start here.

One thing I wanted a place for was attempts at writing for a guide to arch bridge faults I have been producing for Network Rail. Some of the stuff I have put together didn't really fit and I expect to post it here or on the BHA web soon.