Lightning Protection in Greenhouses
The
frequency and severity of lightning discharges appears to be increasing
world-wide. Countries that often took no precaution against lightning are now
finding that they need to address this issue, at least at a basic level.
Countries that have used some lightning protection in the past are now finding
that they need more sophisticated and powerful methods of protection. In
addition, to the installation of adequate lightning conductors which protect
against direct hits, it is becoming increasingly important to protect electronic
devices from indirect or secondary effects of lightning. Of course, nothing will
protect electronics from a direct hit.
Secondary lightning surges enter systems by three main routes;
1) via the incoming mains supply,
2) by induction directly into data and sensor cables and
3) by hitting the ground nearby and raising the ground potential causing a ground potential gradient in the surrounding area. Any two grounding points within this area with therefore experience a brief potential difference during the lightning discharge. What all this means is that there are two main areas within a system where surge protection is needed.
Mains
surge protection
Firstly, protection is needed on the incoming mains power supply – especially where this is by overhead power lines. The rule here is that the most powerful protection must be applied as close as possible to the point of entry of the power lines onto the site. This is normally at the main distribution board. Secondly, lighter power surge arrestors should be fitted on sub-distribution boards in each greenhouse. Finally all modern electronic products will have some surge protection within it. Where this is not sufficient, there are many proprietary plug-in surge arresters or even better, UPS devices, on the market.
Data cable protection
The second place where surges can enter is by induction (inductive or capacitive) into any long lengths of cable – particularly sensor cables and data cables. Any current (or lightning bolt) flowing parallel to a data cable will induce voltages into it. For example, lightning discharge between clouds often travels horizontally; this will tend to induce voltages into long cables running horizontally, for example, cables running to enviro sensors positioned along the greenhouse. Lightning discharging to ground is more often vertically orientated and so it will tend to induce voltages into data cables running vertically – for example the cable running down a pole from the wind sensors.
Not
only does the lightning induce voltages directly into these cables but it also
raises the ground potential where it strikes the ground. The ground potential at
one greenhouse can thus be raised many hundreds or even thousands of volts
relative to the ground potential at the next greenhouse. A cable running between
greenhouses may therefore have a large voltage difference at its two ends.
Potentially we might have 100s of volts or more applied to the terminal of the
little electronic transceiver chips. Bang!!!!
What we do to mitigate this risk is use optical isolation so the two
transceivers can float at a voltage above ground. This also largely solves the
problem of induced voltage but only if the voltage induced in both the A and B
wires is EXACTLY the same. That is why we use TWISTED PAIR data cables to try to
expose each wire to exactly the same induction forces. However, when 100s or
1000s of volts are induced, there is bound to be small differences in voltages
between the two wires eg a 1000 volts on one and 1010 on the other. If this
difference exceeds more than about 10V then again, BANG! Using screened cable
will further help but this is still not the complete answer and some voltages
will still be induced during violent lightning events. To try to clamp these
smaller voltages down suppression diodes
(transorbs) and gas discharge tubes may be used. If an induced surge is
received, these devices should clamp the voltage to an acceptable level but if a
very large hit comes in they may fail – normally as a short circuit (ie fail
safe). However, although they fail in a way which will protect the data bus they
also prevent the bus from operating. For this reason our new designs use
replaceable modules , which include the transceiver chip together, suppression
diodes all fully optically isolated. The new modules will even self detect if
they have a fault and will automatically disconnect themselves from the bus
minimizing disruption.
For lightning prone areas, we also have available a data cable surge suppressor module that should be installed outside the equipment it is protecting – ideally 1m away. These have some fairly heavy duty gas discharge suppressors as well as smaller high speed Transorbs. Unfortunately, these heavy duty suppressors will degrade the signal quality to some extent and so their use should be limited to only particularly vulnerable positions – eg at the entry point of a cable into a greenhouse or at a piece of equipment that is frequently affected by lightning.
The following diagram shows a possible method of fitting these PROTECTOR suppressor boxes.
Note that the stub of cable connecting between the suppressor module and the local controller(s) should not exceed 1m to 2m at the most. Also note that the cable screen is grounded separately at a single point in the system, normally at the master device.
DELAYED LIGHTNING DAMAGE
Unfortunately, even after applying all of these safeguards, damage from lightning is still possible and even worse, this damage is not always immediately apparent. Sometimes it makes pin-prick holes through the semiconductor material inside the chips which then gradually degrades and may fail anytime in the following 9 months. In high security situations any equipment that is even slightly damaged by lightning should be completely replaced.