Difference in Factors of Safety

[Seano]

I fear that it would put the cat among the pigeons good & proper if a punter asked me the SWL of a rig ..

 

One of the nice things about being a rigger is that this never happens - all too busy telling the FoH noise person they're mixing the show all wrong. ;)

[dbuckley]

One thing was very clear though: a 10:1 factor of safety is way way higher than was used to spec the beam, the building or the stage it's hanging from. (Quite likely to be as low as 1.2:1)

 

The University of Canterbury has an annual bridge building competition, where civil engineering students build a bridge across the river that goes through the campus, and it has to hold two people, but they have to get wet when they add a third. Holding a third person is a fail.

 

Read all about it with a picture.

 

Back on topic, I suspect that suspending people engineering and building engineering is a bit like electricial engineering, in that there are often a large number of circuits at the final distribution level, but they sum upwards using rules of thumb and engineering experience called diversity, such that 10 x 32A circuits don't actually have a 320A supply. As one goes up the chain and the numbers get bigger gets bigger the diversity factor decreases.

[Superpants]

Over the years I’ve been involved in writing a number of British and European standards as part of committees, albeit to do with caravans, not stage equipment. All the comments about how these standards are developed are pretty much right, with one major addition- the standard usually starts life with a single author writing an initial draft, and that can have a profound effect on the direction of the standard. It is also quite surprising what can end up in standards and law that you never expect to;

 

I was working for a trade association when a new set of European caravan standards were published, primarily based off the British ones. In the years of development, the industry had been given many opportunities to comment, and for practical testing of the defined tests in the new standards to be performed, but little input was made. Once the standards were published, and we started to apply the tests, we soon started to find issues, but it was now too late to change the standards so we set up a UK working group to address the issues. I started the ball rolling by writing interpretations that the addressed the issues, which were then discussed and altered by the working group, ultimately being adopted by the industry as a defacto standard.

 

In turn, other European countries were finding the same problems, so the European trade association adopted our interpretations across all its members, albeit translated into different languages.

 

The French have a system of trailer inspections, like our MOT, but based on the standards, so in time the French government adopted the interpretation into use, eventually enacting it in law. I therefore find myself in a strange position of having text I wrote making its way into law in a different country, primarily due to the way the standards are developed.

 

The post script to this is that the standards eventually got updated, but more than 5 years later….

 

 

[sjb88]

One thing was very clear though: a 10:1 factor of safety is way way higher than was used to spec the beam, the building or the stage it's hanging from. (Quite likely to be as low as 1.2:1)

 

Hang on a second!

 

Before comparing factors of safety, especially between different industries with different sets of standards (e.g. construction and rigging), it is important to make sure you are considering "like for like" in terms of the sorts of loads and strengths being compared.

 

In structural engineering, loads and material strengths are normally (nowadays) given (in standards, data sheets etc) as "characteristic" values. The characteristic value of a load is the value which there is no more than a 5% chance will be exceeded. The characteristic strength of a material is the value which 95% of samples of the material will be stronger than.

 

The engineer will then multiply the characteristic loads by a load factor ("partial safety factor") to gain a "design" load which there is an extremely small chance of exceeding. The factor will depend on the type of load and what other loads it is being combined with (the latter to reflect the fact that on a building you might be exceptionally unlucky to see both an unusually high floor live load and gale force winds, say, at the same time - not because there is any reason why these cannot coincide, but because of the balance of probabilities). A typical load factor for a live load on its own would be 1.5. They will also reduce the material strength by a "material factor" (another partial safety factor) which reflects the variability of the material. For example, a material factor for structural steel is usually 1.0 (in the UK) as steel is produced in a factory, and factories will produce steel with strength higher than that quoted, 100% of the time. For concrete a value of 1.5 is more appropriate as it is so dependent on site conditions.

 

The check an engineer does is therefore Characteristic Load * Load Factor < Characteristic Strength / Material Factor.

 

A total "Safety Factor" based on this can be calculated as Load Factor * Material Factor. For an example of a live load on a concrete structure as above this would work out as 2.25. However it is important to note that this is based on loads and material strengths worked out based on the probabilities described above. If you have defined your loads and strengths in another way it is meaningless to compare them. For example if you were (foolishly!) calculating based on the median material strength you would need a much larger material (and hence total) safety factor.

 

I would be interested to know how the methods for calculating loads and strengths in the rigging industry (before safety factors are applied) compare.

 

Background: I am a structural engineer.

[trussmonkey]

Seano I was in the same PLASA session (sat in front of you I think )..

 

I took out of the that 10:1 probably came, as you said, out of the machinery directive which in turn prob came from the mining industry. Starting at 4:1 someone thought that isnt enough safety so lets up it to 5:1. Then performer flying come about and LOLER specifies that all kit used for performer flying should be double the FoS hence 10:1.

 

It was a bloody good lecture and eye opening as a whole room of engineers and industry experts all could not really agree as to exactly where our FoS come from.

 

Half of the time when we are specifiying kit/systems we select kit which is compatible with other bits of kit i.e. a 3.25t shackle with a 1t hoist as a 2t/1t shackle is a dont fit at all. So we are overrating our systems with out even thinking about it.

[Seano]

Before comparing factors of safety, especially between different industries with different sets of standards (e.g. construction and rigging), it is important to make sure you are considering "like for like" in terms of the sorts of loads and strengths being compared.

 

No doubt. Possibly not a comparison to be taken *too* seriously. However, the speaker and the panel for that talk were a bunch of structural engineers very familiar with all kinds of construction projects, and also with the entertainment industry as consultants for venues, temporary structures, all that. The gist of what they were saying, and the point of the comparison was that the FoS we're working with (especially when it gets right up to 10:1) often seems almost ludicrously risk-averse from their point of view.

 

I would be interested to know how the methods for calculating loads and strengths in the rigging industry (before safety factors are applied) compare.

 

Regarding loads:

I spend a fair bit of my time doing the 'house rigging' thing, see a lot of touring shows' rigging plots and sometimes have to think quite carefully about whether they're going to be likely to break the arena roof.

 

I think it's safe to say the loads are much more predictable than, say, the live load on the floor of a residential building. No need for statistics to figure out how many of the residents will have a heavy bookshelf or a huge fish tank. By the time you come to rig a lighting truss, for example, you generally know exactly what you're going to be hanging and how much it weighs.

 

Methods vary, sometimes the initial calculations have been worked out to the nth degree and or generated from a CAD package, other times they're more or less an educated guess. Often, by the time a touring production has left rehearsals the show has been weighed and the loads indicated on version 2 (or 3 or 4..) of the plot are actually empirical measurements.

 

Much of the time individual loads of multiple points lifting a single structure will be statically indeterminate though, so however carefully the loads have been predicted a bump of this hoist or that hoist can easily transfer a few 100 kilos from this point to that point. It's increasingly common these days to see load cells used in that situation and the loads monitored. Reassuringly, the results of the load monitoring generally bear out what's predicted. Turns out the educated guesswork was, by and large, actually pretty good all along.

 

Regarding strengths:

Apart from a few bits of very specialist kit used for performer flying and wotnot, the equipment we're using is the same lifting gear used across many other industries. Shackles, slings etc. have strengths determined by the manufacturer, and are marked accordingly. Something else that was interesting to hear at that talk was that a few of the folk there have had old gear destructively tested, and that the steel wire ropes we're using tend to break pretty much at their rated breaking load even when they look as if they've had a long hard life being horribly abused. Also fairly reassuring. Particularly today - I'm off to work in a mo to hang a show using some of the most kinked and curly-wurly steel currently on the road. ;)

 

Seano I was in the same PLASA session (sat in front of you I think )..

<snip>

It was a bloody good lecture and eye opening as a whole room of engineers and industry experts all could not really agree as to exactly where our FoS come from.

 

Aha. Quite likely - I was there with the Bolton contingent. And yes, it really was, very interesting stuff. :)

[kerry davies]

Interesting thread. IMO we have inflated FoS for a few reasons.

One is the build-up-of-tolerance factor in temporary construction. A tent might be 3:1 and a rig 2:1 but that might add up to a 6:1 final FoS.

Another is that we tend to "idiot-proof" as much as possible because of the numbers of people who have access to the kit these days so 6:1 becomes 12:1.

 

We are an industry which is very much a "work in progress". FoS may become more realistic and lower but only when we have some further self-regulation, as the permanent buildings have Building Standards. We erect far more structures in far more varied circumstances than the construction sector do and with no idea what the skills levels of those doing the work might be. In perfect conditions we could use far lower FoS but we have yet to do one single job in the history of TDS where everything and everyone was perfect.

 

The main reason I think we are very small-c conservative in these matters is our experience. Just about anything that could possibly go wrong has gone wrong in our game and continues to go wrong but the consequences are far more serious and publicised than in the permanent construction game.

 

Building sites do not normally open their doors to Joe Public and then ply them with "substances" and work them into a frenzy. I think construction FoS would be as high as ours if that were the case.

[librarian28]

Then performer flying come about and LOLER specifies that all kit used for performer flying should be double the FoS hence 10:1.

Could you point to the para in LOLER that says this? Just to save me having to wade thru' to check it.

 

I think it's safe to say the loads are much more predictable than, say, the live load on the floor of a residential building. No need for statistics to figure out how many of the residents will have a heavy bookshelf or a huge fish tank. By the time you come to rig a lighting truss, for example, you generally know exactly what you're going to be hanging and how much it weighs.

I understand (I think) that you might need to calculate forces on a bridle, vectors, turning moments, wind sheer, snow on the roof and such stuff. And best to understand that even a lighting truss has curious counter-intuitive fulcrums and forces (according to Chris Higgs' books) which might affect how much one end of a truss 'weighs' but.....

 

If you are suspending / lifting a load of, say, 75kg the only extra force (I don't know if it's predictable or unpredictable) is the accelerating / de-accelerating force of gravity during a snatch (either planned or unplanned). Now, when I was a lad, someone picked a random number and called it "ten" and then announced that if you used a shackle, say, which breaks at more than 750kg to attach the load, then it is 'safe'. Is that an acceptable rule of thumb?

[Seano]

I understand (I think) that you might need to calculate forces on a bridle, vectors, turning moments, wind sheer, snow on the roof and such stuff. And best to understand that even a lighting truss has curious counter-intuitive fulcrums and forces (according to Chris Higgs' books) which might affect how much one end of a truss 'weighs' but.....

 

Forces on a bridle - actually, it's rarely necessary to calculate those forces. Most of the time it's sufficient just to know that it's less than 100% of the load and the bridle isn't too 'flat'. (An included angle of about 60-90 degrees is generally about right.)

 

Wind sheer and snow on the roof are above my pay grade mate. ;) Very much in the domain of the structural engineer.

 

If you are suspending / lifting a load of, say, 75kg the only extra force (I don't know if it's predictable or unpredictable) is the accelerating / de-accelerating force of gravity during a snatch (either planned or unplanned).

 

An unplanned "snatch" sounds to me like the kind of thing that would probably be called a "near miss" (or a "dangerous occurrence" as far as RIDDOR goes - so reportable, probably). A fixture dropping off it's clamp and being caught on a safety bond perhaps? Or a twisted shackle in the top rigging righting itself with a 'bang'. Something to be avoided, definitely.

 

Planned (or perhaps better to say, unavoidable) - the obvious source of that is starting and stopping a fixed-speed hoist. The traditional rule of thumb is to allow 25% of the load for that with a 4m/minute hoist, which is actually very conservative. It's unlikely to be that much in reality - even with video and/or sound types trying to get fiddly pins in or out and doing the bump bump bumpity-bumpity-bump thing.

 

Flying set and/or performers involves "dynamic loads" of course - generated by the acceleration/deceleration of things (including, perhaps, the jiggling up and down of human bodies). There is much misunderstanding about 'dynamic' loads generally I think.

 

Now, when I was a lad, someone picked a random number and called it "ten" and then announced that if you used a shackle, say, which breaks at more than 750kg to attach the load, then it is 'safe'. Is that an acceptable rule of thumb?

 

Yes. That's what is meant by a "factor of safety" of 10. The maximum "safe working load" is defined as the minimum breaking load divided by 10. Actually the random number that was picked would have been a 5, which was then arbitrarily doubled for lifting people, or lifting above people. So the process wasn't "think of a number", it was more "think of a number.. now double it". ;)

 

In practice with a shackle it's neither here nor there. A 75kg shackle will almost certainly be too small to use. A 75kg bit of set would be more likely to end up on a 500kg shackle. 1T rigging is generally done with 3.25T shackles, because 1T shackles are physically too small to use to connect 1T slings and steels.

[librarian28]

As Seano & I both seem to be sat at home with time on our hands.............

Forces on a bridle:

The NAA bumpf says: "For example, to suspend a 2 tonne load on a three-legged bridle the SWL on the master link ring would need to be 6 tonnes" So, you may not need to calculate it, but you need to use one that is strong enough to satisfy the NAA. And, I suppose, multiplying a number by three is (strictly speaking) doing a calculation.

 

I'm pleased that you recognize that calculating wind shear on a ground support structure is beyond your pay grade. Mine too. I'm trying to get my head round my simplest example before we get complicated.

 

If we could avoid unplanned stuff, we wouldn't need safety factors. You may install something expecting to do a bit of bumpity bump and I may install expecting a couple of shackles to right themselves. We will always be planning for the unexpected? So, whether it's a shackle righting itself, a bumpity bump, or a performer jiggling, they are all putting the same 'extra' force on the rig. F=Ma. So, is (in my example) a shackle which breaks at 750 kg OK for all eventualities?

 

I think that in the olden days, it was just always 10. Full stop. For everything. Probably chosen because in previous centuries, we only had our fingers or slide rules to do calculations. And who has got time to learn how to use a slide rule?

 

What size shackle to use may not be here or there to you, but I've got a budget to work to. So, I want to know the smallest I can get away with.

 

(Actually, it's not really about cost. I don't take much notice of mis-quoted regs or opinions about whether it is (4:1 x 2) or 25% of the load. Or whether I'm 'lifting' or just 'suspending'. Or whether it's a 'lifting machine' or 'lifting accessory'. I just multiply by 10 and bugger the extra cost - it saves my brain for more amusing tasks).

 

How come the BR clock says I posted this at 3.10 PM? Did someone put the clock forward twice instead of back once?

[Seano]

The NAA bumpf says: "For example, to suspend a 2 tonne load on a three-legged bridle the SWL on the master link ring would need to be 6 tonnes" So, you may not need to calculate it, but you need to use one that is strong enough to satisfy the NAA. And, I suppose, multiplying a number by three is (strictly speaking) doing a calculation.

 

I'm vaguely aware of that bumf. On the subject of 3-legged bridles and O-rings it's wrong imo. Using a master link to bring three legs together directly is not good practice anyway - better for all the loads on a ring to be in the same plane I'd say.

 

Quite easily achieved with a 3-legged bridle by combining two of the legs in a shackle first thusly:

 

http://www.deepsoup.f2s.com/BR/3legs.jpg

 

When using master links in accordance with the manufacturer's guidance, I see no reason not to also accept the manufacturer's stated SWL. (Which is likely to be somewhat academic anyway since, as with shackles, a usefully large physical size will generally imply a somewhat larger SWL than is actually required.)

 

I think that in the olden days, it was just always 10. Full stop. For everything. Probably chosen because in previous centuries, we only had our fingers or slide rules to do calculations. And who has got time to learn how to use a slide rule?

 

I rather think it was 4. (Which was also mentioned in the talk that Trussmonkey and I attended as the FoS historically favoured by Victorian engineers working with iron.)

Or 5. (Which is still the FoS most often applied to general purpose lifting kit.)

 

In previous centuries I think things more often tended to come in multiples of 12 rather than 10 to ease the arithmetic. More divisible. Hence 12 inches, 12 hours, 60 minutes, 360 degrees and so on.

 

I'm old enough (just) to have learned to use a slide rule at school. Worthwhile as they were allowed in O level exams and calculators weren't. I've not seen it for some time, but pretty sure I still have my old slide rule kicking about somewhere.

[librarian28]

Or 5. (Which is still the FoS most often applied to general purpose lifting kit.)

Perhaps they had already doubled up to 10 as the usual rule before I got my first protractor. And I grew up in a brief (but golden) age when you could do most things using your fingers. Or, maybe, this is the better time when you just need your thumbs.

 

Maybe it was 4 until they realised that it barely allowed for the extra force of a serious snatch. Especially when we went metric and gravity went up to 10m/s.

 

(Please don't post photos like that without a warning - I went dizzy and nearly fell over).