Specifications for elephant-resistant steel poles

[Peter Apps]

This is probably simple for an engineer, but I’m a zoologist and analytical chemist and so I would rather ask those who know.

I have camera traps out on the African bush, that are monitoring the responses of leopards to artificial scents. The cameras are in steel boxes to protect them from elephants, the boxes are bolted to brackets 150 mm long and the brackets are bolted to steel poles at a height of either 500 or 700 mm above ground level. The poles are 40 x 40 mm mild steel angle, either 2mm or 3mm thickness. The poles are driven deeply into the ground. The mechanical problem is that elephants push the cameras over; they bend the 2 mm thick angle through 90 degrees at ground level, or twist it through 180 degrees. They bend the 3 mm thickness through 45 degrees.

I have videos from the cameras as they are being wrenched around and most of the work is done by the elephants’ trunks, with some kicking and stamping once the pole is bent over. Elephants are alleged to be able to lift 350 kg with their trunks. How hard the they can kick is an open question.

So my question is; what size and gauge angle iron or square tube would I need for a pole that could withstand twisting when 350 kg is applied sideways to a 150 mm lever bolted at right angles to the pole, without exceeding its elastic limit (camera aim is critical so the pole needs to recover after an elephant has tried to bend it) . And, similarly, what pole, standing 1 m out of the ground, could withstand being pushed at the top by an elephant – say 500 kg sustained force.

Thank you.

 

[jedishrfu]

It might help if you provide a drawing or picture of your setup.

My feeling though is that you may need a different kind of deterrent like flashing lights, sound of a predator, some kind of spray deterrent, bushes planted around your pole or something else that will keep them away from it.

What if you disguised your cameras as trees?

More costly but perhaps they won't look at it as something out of the ordinary.

 

[Peter Apps]

Here is a picture of the camera mount. This shows only the back of the camera box, when the cameras are in the field they are enclosed in steel.

It might help if you provide a drawing or picture of your setup.

My feeling though is that you may need a different kind of deterrent like flashing lights, sound of a predator, some kind of spray deterrent, bushes planted around your pole or something else that will keep them away from it.

What if you disguised your cameras as trees?

More costly but perhaps they won't look at it as something out of the ordinary.

Since I am testing the repellent effects of chemicals from scent marks I cannot add any extra repellents to the setup. Elephants knock down trees with the same enthusiasm as they re-adjust my cameras - it seems that they have switched from ecosystem engineering to mechanical engineering !

 

[jedishrfu]

How about a Go Pro attached to the elephant ala elephant-cam? :-)

Could you string a camera up between trees or would that be too low?

Also didn't see your photo in the post.

 

[anorlunda]

Here is a picture of the camera mount.

Use the upload button on the post editor.

How about a Go Pro attached to the elephant ala elephant-cam? :-)

LOL, that would be fun.

But it suggests another possibility. Might a Go Pro be rugged enough to survive without an enclosure? A very small thing might have a better chance of escaping notice by the angry elephant. Smaller things are also easier to camouflage.

 

[mfb]

But it suggests another possibility. Might a Go Pro be rugged enough to survive without an enclosure? A very small thing might have a better chance of escaping notice by the angry elephant. Smaller things are also easier to camouflage.

There is still the steel beam, and it is hard to hide that completely.

If the elephants want to break it for some reason, they'll apply more force to a more durable structure, so we cannot use the current damage as estimate how much they can do.

How difficult is it to test different beam profiles? If you can change them easily, you can do a test series. Use something with twice the thickness, if it works, go down a bit (if cost/weight is an issue), if it doesn't work, go up a bit.

 

[anorlunda]

There is still the steel beam, and it is hard to hide that completely.

Why have any beam or box at all?

The camera could be concealed in a birds nest, or a bee hive, or under a pile of bird poo, anything that occurs naturally in the tree that the elephant is likely to ignore.

Go Pro camera are so much cheaper than professional cameras, that many of them could be concealed in the area to capture different angle. If some are lost, no big deal.

 

[mfb]

I don't know the details of the biology experiment, but it might be favorable to have the camera away from other points of interest.

 

[Peter Apps]

How about a Go Pro attached to the elephant ala elephant-cam? :-)

Could you string a camera up between trees or would that be too low?

Also didn't see your photo in the post.

The camera setup has all sorts of constraints on camera siting and aim that make it impossible to attach the cameras to trees - the trees just don't grow in the right places ! Cameras on elephants has already been done - by National Geographic if memory serves - cameras on animals are generically called crittercams.

I did not insert the picture correctly - I just cut and pasted. I'll try again with the upload button. SUCCESS !

 

[Peter Apps]

Why have any beam or box at all?

The camera could be concealed in a birds nest, or a bee hive, or under a pile of bird poo, anything that occurs naturally in the tree that the elephant is likely to ignore.

Go Pro camera are so much cheaper than professional cameras, that many of them could be concealed in the area to capture different angle. If some are lost, no big deal.

Unlike GoPros the camera traps do not record continuously - they are triggered by sensing moving body heat and shoot 30s of video per trigger. They stay out for weeks at a time in very specific configurations that I have validated to capture and record activities specifically of leopards. They have to be sited to within abut 10 cm over 15 m and aimed to within a degree of arc - hence my use of steel posts that I can place at specific positions, brackets that can be tilted and bolted into place, boxes that can be swivelled and bolted, and protective cases for the cameras - on a previous project we lost five unprotected cameras to elephants in three weeks.

And I am afraid that with the budget constraints that wildlife research operates under the loss of any camera is a very big deal indeed.

 

[Nidum]

Squirrel cage .

 

[Peter Apps]

Squirrel cage .

Apart from it's literal meaning, which I suspect is not the sense you are using it in, I have no clue what a squirrel cage is or how it would help calculate the dimensions of a steel pole.

 

[anorlunda]

The camera setup has all sorts of constraints on camera siting and aim that make it impossible to attach the cameras to trees - the trees just don't grow in the right places ! Cameras on elephants has already been done - by National Geographic if memory serves - cameras on animals are generically called crittercams.

I did not insert the picture correctly

OK, with the additional info you provided it is more difficult. This is a very interesting problem.

I would guess that steel beams even 10x times stronger than the one in your picture could still be knocked down by an angry elephant. Worse, the sharp corners on those beams and angle irons could injure the elephant regardless of whether the mount is damaged. The steel should be encased in concrete to protect the animals; but that's probably not practical either.

I remain skeptical of the brute force/massive structure approach. Surely, even a steel structure might shift one degree during the attack. Therefore, I presume that your criteria is not to continue video recording during and after the attack with one degree accuracy, but rather to rescue the equipment.

I would concentrate on concealment and deception. I would perhaps make the mount very easy to knock down without injury to the animals. Then I would tie the camera to the ground with a security cable to prevent it from disappearing even if the mount was destroyed and the camera knocked down. If the camera didn't look man-made, then the elephant may be less motivated to crush it.

 

[Nidum]

A squirrel cage is a common name for a strong equipment protection cage which is round and made from metal bars .

Something like a ring of metal poles (at say 1 m radius) around the camera pole and tied together with metal hoops .

Solution really depends on required field of view and whether the equipment has to be moved frequently or permanently sited .

 

[jedishrfu]

Found some deterrents that might work here:

http://www.bbc.com/earth/story/20141204-five-ways-to-scare-off-elephants

The interesting one is African bees.

 

[Peter Apps]

Found some deterrents that might work here:

http://www.bbc.com/earth/story/20141204-five-ways-to-scare-off-elephants

The interesting one is African bees.

The universal problem with all the suggestions about elephant repellants is that they will probably also repel leopards (or at least alter their behaviour in ways that have nothing to do with the experiment) and so render the entire exercise pointless.

 

[Peter Apps]

OK, with the additional info you provided it is more difficult. This is a very interesting problem.

I would guess that steel beams even 10x times stronger than the one in your picture could still be knocked down by an angry elephant. Worse, the sharp corners on those beams and angle irons could injure the elephant regardless of whether the mount is damaged. The steel should be encased in concrete to protect the animals; but that's probably not practical either.

I remain skeptical of the brute force/massive structure approach. Surely, even a steel structure might shift one degree during the attack. Therefore, I presume that your criteria is not to continue video recording during and after the attack with one degree accuracy, but rather to rescue the equipment.

I would concentrate on concealment and deception. I would perhaps make the mount very easy to knock down without injury to the animals. Then I would tie the camera to the ground with a security cable to prevent it from disappearing even if the mount was destroyed and the camera knocked down. If the camera didn't look man-made, then the elephant may be less motivated to crush it.

I am not worried if the camera moves during an elephant attack - the aim is not to record elephant behaviour, I get elephant videos as an inevitable spin off of recording leopards. What I would like is for the poles etc to be strong enough that they resist elephant attacks and return to their original position after the elephant has had its fun. An approximate return would be better than the camera being left at ground level looking down at a patch of grass or up at the sky which is what happens with the 40 x 40 angles. I have thought about, and tested, putting the poles on springs - it does not work even in prototype. The present setup returns to position after gentle shoves and nudges by elephants and other big animals like giraffes and hippos, so that approach can work if the mechanics are strong enough.

Making the mount easy to break for elephants means that it is possible for it to be broken by smaller animals, so the camera will spend most of their time lying on the ground instead of collecting data.

The elephants are slow and deliberate in their attacks on the cameras - they do not come charging in from a distance - so they are not likely to injure themselves on the corners of ordinary steel work.

We already camouflage with branches and foliage, it helps a bit but not enough to be practical.

It is possible to design steel work that elephants cannot bend - zoos use it in their elephant exhibits but it is on a scale that is impractical for field work. My question is aimed at finding out whether there is a size of steel tube or angle that will adequately resist the forces that elephants exert on the camera installations and that is small enough for field work, and within our budget.

 

[jedishrfu]

What if you used helical springs like what's used on some childrens playground toys and enclose your camera in a cylnder-like object attached to the springs?

The elephant can hit it or push it and it will pop back up.

 

[anorlunda]

Perhaps you can get some ideas from circus elephant cages.
But I don't see how a cage, or a strong pole in the ground solves the problem of being moved or knocked over by the attack.

I presume that the tubes seen in these cages were cheap galvanized pipe. Here's a table with some strength numbers. The cantilever is closest to the case of a pole stuck in the ground. But don't forget that the bracket to hold the camera and the camera case itself can be the weakest link in the chain.

 

[Peter Apps]

What if you used helical springs like what's used on some childrens playground toys and enclose your camera in a cylnder-like object attached to the springs?

The elephant can hit it or push it and it will pop back up.

I tried that with coil springs from a vehicle. The two main problems are that if an elephant pushes one down with its foot and then steps off it, it will whip upright and probably inflict some damage, and that it needs some damping to stop it jiggling around for several minutes after it has been displaced and released. A heavily over-engineered solution would be an industrial strength door closer that can move on more than one axis.

 

[Peter Apps]

Perhaps you can get some ideas from circus elephant cages.
But I don't see how a cage, or a strong pole in the ground solves the problem of being moved or knocked over by the attack.

I presume that the tubes seen in these cages were cheap galvanized pipe. Here's a table with some strength numbers. The cantilever is closest to the case of a pole stuck in the ground. But don't forget that the bracket to hold the camera and the camera case itself can be the weakest link in the chain.

View attachment 114585

 

[Peter Apps]

The crush cage at the top is proper structural steel, the vertical bars with the mother and baby are probably solid rod and will be anchored top and bottom, I suspect that the the rusty old iron at the bottom is drill casing, which has thick walls.

What I need is for one of you engineers to translate that table into the dimensions of a pipe in cantilever loading that will not go outside its elastic limit when carrying a 500 kg mass 1m from its fixed point. I need to use square tube or angle to resist rotation in the ground (a round tube will just rotate) and to provide a flat surface for the bracket to bear on.

 

[jedishrfu]

Punching bags:

or

 

[Peter Apps]

Punching bags:

or

Brilliant idea, as long as they don't cost any more than 2m of 50x50 square tube

 

[JBA]

This might be condemned by some as "animal cruelty" but what about equipping a few units with small circular battery powered electrified wire ring enclosures and painting all enclosures including the electrified ones a bright distinctive color. The idea being that after a couple elephant encounters with the electrified ones the elephants would begin to assume that all such type and colored enclosures are something to painful to play with.

 

[Peter Apps]

This might be condemned by some as "animal cruelty" but what about equipping a few units with small circular battery powered electrified wire ring enclosures and painting all enclosures including the electrified ones a bright distinctive color. The idea being that after a couple elephant encounters with the electrified ones the elephants would begin to assume that all such type and colored enclosures are something to painful to play with.

Electric fencing that is strong enough to deter elephants works at 80 000 V and makes a loud click every time a pulse fires down the wire - the clicks will deter other animals from coming near the site, so I will get no useful data. Also the fences are power hungry and very difficult to isolate from earth. A bright distinctive colour is exactly what I do not want - the cameras need to be inconspicuous so as not to affect animal behaviour. Teaching elephants to avoid cameras by giving them electric shocks will work only on the elephants that encounter an electrified camera, the rest of them will carry on trashing cameras if the mood takes them. There are over 100 000 elephants in northern Botswana, and probably at least 5 000 in and around our study area - it would take ahuge effort to teach all of them not to mess with camera traps.

 

[JBA]

All points well taken.
What about a structure shape that could not be simply bent and therefore not much fun to play with such as a three legged pyramid made from angle or pipe that is staked down with the camera located inside the structure for protection. The only problem is that the elephants could still grab the supports with their trunks and pull the structure loose. A solution to that could be to make the pyramid with solid metal sheet sides; but, in that case they would probably standout too much in the local environment unless possibly painted in camouflage to blend with their surroundings.

 

[Peter Apps]

All points well taken.
What about a structure shape that could not be simply bent and therefore not much fun to play with such as a three legged pyramid made from angle or pipe that is staked down with the camera located inside the structure for protection. The only problem is that the elephants could still grab the supports with their trunks and pull the structure loose. A solution to that could be to make the pyramid with solid metal sheet sides; but, in that case they would probably standout too much in the local environment unless possibly painted in camouflage to blend with their surroundings.

It's a thought, but has two disadvantages that I can think of. First how to anchor it to the ground - driving in three poles to the depth and angle to allow them to be joined at the top will not be simple. Screw anchors might work, but up goes the cost and complexity. Second - the elephants will probably kick a bend into one of the three poles unless it is as strong as a single pole would need to be to resist being pushed over.

 

[mfb]

Where does the 1 degree requirement for the camera come from? Cameras typically have quite a wide viewing angle, why do you need such a precision?

Simply taking a stronger beam still looks like the easiest option to me.

 

[JBA]

I was thinking more of a rigid welded pyramid structure with pipe collars at the bottom of the corners for retaining stakes to be driven through. You are right about the structure legs needing to be as strong as a single pole but a welded pyramid joined at the top and with base pipe pieces connecting the bottoms of the three legs is an extremely strong structure (but not necessarily a light weight one for handling purposes).

 

[Peter Apps]

Where does the 1 degree requirement for the camera come from? Cameras typically have quite a wide viewing angle, why do you need such a precision?

Simply taking a stronger beam still looks like the easiest option to me.

The 1 degree repeatability in aim comes from tests I did on angle of dip and tilt using a dog the same size as a leopard as a target under controlled test conditions, and from the requirement to have the cameras covering overlapping areas but not dazzling one another with their IR floodlights.

Amen to the stronger beam - but nobody wants to answer that question

 

[Peter Apps]

I was thinking more of a rigid welded pyramid structure with pipe collars at the bottom of the corners for retaining stakes to be driven through. You are right about the structure legs needing to be as strong as a single pole but a welded pyramid joined at the top and with base pipe pieces connecting the bottoms of the three legs is an extremely strong structure (but not necessarily a light weight one for handling purposes).

That would work but would need three times as much steel and three times as much hammering of stakes into the ground.

 

[anorlunda]

nobody wants to answer that question

I think I can get you an answer when I get to my computer tomorrow.

 

[Nidum]

I did a sample case just to see what the magnitude of the problem is .

Simple cantilever / Hollow square steel tube / 1000 mm high / 75 mm across flats / 5 mm wall thickness / Mild steel / 4000 N horizontal load at top

That gives a safety factor of about 1.45 on yield for maximum fibre stress .

 

[Peter Apps]

I did a sample case just to see what the magnitude of the problem is .

Simple cantilever / Hollow square steel tube / 1000 mm high / 75 mm across flats / 5 mm wall thickness / Mild steel / 4000 N horizontal load at top

View attachment 114598

That gives a safety factor of about 1.45 on yield for maximum fibre stress .

Great ! That's what I've been looking for. That is a much heavier pole than I had thought, though not entirely impractical. Hammering it into the ground will be a mission though ! Would you mind looking at what load a 50 x 50 x 3mm square tube could recover from under the same conditions. I already know that I can get 50 x 50 mm into the ground without machinery. Thanks again.