HOME

PRODUCTS

INFORMATION

CONTACT 

Risk Reduction for Greenhouse Growers 
Summary of a talk by Jeff Broad,   Autogrow Systems Ltd at the 2003 AHGA conference

As a supplier of greenhouse control equipment Autogrow staff  have visited a lot of growers over the years.  Many have surprised us at the high level of risk that they seem to be willing to carry when with a little planning and ingenuity they could considerably reduce their exposure to risk.

Perhaps they have good insurance companies?  I hope so!  

Managing Risk
Managing risk really comes down to carefully assessing what problems or failures may arise, what the consequences would be, how the risk may be reduced, and planning how you would cope if the worst really did happen.  

There are many risks faced by the grower including:-

• Electricity supply failure (blackout and power surges)

• Equipment failure

• Water supply

• Weather

• Disease

• Pests

• Market

We will take a look at only the first two of these as they are in the area of my expertise.  However, all risk factors need to be evaluated and planned for.

Electricity supply 
We need to consider both the possibility of a complete outage and also the consequences of voltage surges.

Complete Shutdown of Electricity Supply
Obviously all pumps, vent movers, controller etc will stop and the consequences are pretty obvious.

What can you do? 

A few ideas…

The obvious solution is standby generation but this is very expensive. You need to evaluate the consequences of shutdowns for various amounts of time and then assess the cost of standby generation against this.

The alternatives are to have a large fully automatic generator that is capable of working the complete system with multiple motors and systems running simultaneously or a smaller hand-started generator capable of running only one sub-system at a time.  

For a 1 Hectare greenhouse a standby generator capable of running everything might be a 70KVA generator at a cost of around A$30,000.  If you believe the risk of a prolonged shutdown is small then you may well come to the conclusion that this is a bit extravegant.

A cheaper alternative would be to have a smaller manually started generator set that is able to run one sub-system at a time ie either one set of pumps/valves OR one set of vents OR one fogging system and then to manually cycle the generator between the systems.  Perhaps a manually started 2 KVA generator costing about A$2000 would suffice for this purpose.  Of course, you will need to be there continuously and might have to stay throughout the night.  For a 1 in 5 year event, this might be acceptable.

NFT systems are the most vulnerable to short power failures as the roots dry very quickly in warm weather.  For these systems you could have a large reservoir tank raised so that you can allow a small amount of water to dribble through the gullies.  This can even be automated by using a normally open valve (closed when power is applied and open when power is removed).  Of-course if you have access to town water then obviously this could be used in place of the raised tank

If at the end of the day you have absolutely no power then you should know how to move the vents manually. Most motorized vents can be driven by turning the motor spindle.  A cordless electric drill with the appropriate socket spanner and a ladder are useful for this but make sure that the power is completely disconnected as it could come on at any time and cause a nasty accident.  Make sure you keep your drills batteries fully charged.

Power Surges or Transients

Power surges are often caused by line switching or by nearby lightning strikes.  Some of these induce high voltages onto the power lines with a very high frequency component.  This means that it doesn’t behave like ordinary electricity but almost like a radio wave and will jump across gaps and insulation and find its way right into the electronics.  Here it can cause considerable damage which sometimes only starts to show after a few months when the equipment becomes generally unreliable.

The experts agree that the main defense to this kind of damage should be in the following order of importance:-

• Divert the lightning away from the power lines or data cables (required in high risk areas)

• Fit a large surge arrester at the main power board and smaller ones at sub-distribution boards.  This is absolutely essential.  I hope you will all go back to your farms after this and check that these are fitted and are still operational.  To check that they are still functional, look at the window on the front of the diverter.  If it is green it is still working but if it is red it should be replaced.

• Buy equipment with inbuilt surge suppression and power filtering.

• In lightning prone areas you may also need external filters or even a UPS.

Equipment failures

Equipment can fail, and usually does, at the most inconvenient times such as on public holidays when you just can’t get any assistance.   Even the best equipment can fail.  Last year I visited a new installation and asked if I could take a peep inside the newly installed large expensive controller.  Off came the cover and to everyone’s surprise mice droppings were everywhere.  Even on the delicate circuit boards.  I am sure that mouse excrement is very corrosive and I suspect that that system will fail in the near future.  The system controlled four large greenhouses and when it fails the grower is likely to have a disaster of significant proportions.  I certainly hope that his insurance cover is good.

I believe absolutely in a modular system in which each controller has only a small area of impact so that if it ever fails the possibility for damage is limited.  If the cabling to each module is pluggable then it is also possible to significantly reduce repair costs and also speed up repairs by getting a swap unit couriered to site.  Swapping it over and sending the faulty unit back to a central repair station where faults can be found efficiently and repair quality can be assured.  Making repairs on-site is not easy and often the most appropriate equipment is not available.

A short story:- A greenhouse in Taiwan had a large expensive control system installed to provide both fertigation and environmental control.  The system failed after only three months of use and a technician was promptly sent from Europe to investigate.  The technician established that the fault was due to a power surge and consequently the repair would not be covered by warranty.  A quotation was produced which included spare parts to be couriered down from Europe and, of course, the bill for technicians time, air fares, hotel accommodation etc.  The owners decided that they could not afford it and the system was scrapped.  Since then they have operated both the venting and irrigation manually and are only now upgrading to automatic control equipment.

If the system had been modular and pluggable, they could have returned the faulty unit to the factory for efficient and cost effective repair.  This would have saved the major cost of the failure, namely the travel, hotel and salary for the technician to travel to site.

Another story:-  Also about ten years ago an Auckland grower installed a new glasshouse and computerized control system.  This system actually used the PC for doing the control and we all know how unreliable PCs can be.  Over the weekend there was an electrical storm and the dying act of the PC controller was to open all of the vents.  The greenhouse was devastated. 

Of course nothing can protect you from a direct or even nearly direct lighting strike but I was left wondering if enough had been done to protect and make the system fail-safe.

The layout and organization of the greenhouse plant can also have an influence on the impact of a control system failure.  For instance, if the dosing system fails, perhaps because of a faulty pH electrode, dirty EC probe, faulty solenoid or relay stuck in, what is the worst that can happen?  Well if you have a 1000Ltr tank full of strong acid or nutrient stock connected then maybe by the morning your plants will not have any roots left on them.  On the other hand if small “day tanks” are used that only hold enough for one day then even if the whole lot is dumped in one go the crop will not be harmed.  The day tanks can easily be arranged to fill automatically each morning from a larger tank

The use of “day tanks” can minimize the impact of a faulty sensor, controller pump or solenoid valve overdosing the mixing tank.  

Similarly, if the heating system is oversized then a faulty temperature sensor could switch the heating fully on and cook the plants.  By sizing heating, cooling, injection dosing components correctly so that they are only just big enough then if the worst should happen and they switched fully-on, the resulting damage is minimized.

Doubling up on solenoid valves or pumps (ie putting two pumps in parallel) has been suggested as a means of adding some “redundancy” to the system so that if one fails the other will probably continue.  This approach has some merit but needs to be done with care as the extra complexity can in itself have a negative effect on reliability.  Personally, I do not recommend this approach.

Remote support
In my experience, the majority of reported faults are not faults at all but are often the result of incorrect controller setting.  To reduce the need for expensive callouts to get the settings corrected it is vital that you select equipment which can be accessed remotely from a distant computer either by telephone link or over the internet.  Good equipment will allow all readings and settings to be viewed and changed remotely.  

Monitoring
There is absolutely no substitute for totally independent monitors with an alarm system for both the nutrient and environment.  Having computer monitoring of the controllers is very useful BUT you should never believe what the controller is telling you.  The reason is simple; if a sensor on the controller is faulty – say reading low – then the controller will raise its output to try to correct the situation and to get the reading from the faulty sensor to be correct.  So if the temperature sensor is reading ten degrees Celsius too low the heater will come on and raise the temperature by ten degrees.  Even though the plants are cooking it will report that everything is fine.  Only a totally independent monitoring system with its own sensors will safeguard you from this situation.

Remote monitor/data logger suitable for monitoring nutrient supply and/or run-off.

Similar units are available to monitor/log environment variables such as air temperature, humidity, solar irradiance, CO₂ etc.
All are capable of sounding alarms at the PC in the event of a variable going too far out of limits.

To sum up….

      1)      Do you have electricity supply backup?
2)      Do you have adequate surge protection?
3)      Do you have the ability to swap out a controller yourself or with local support?
4)      Does your local agent carry a swap unit?
5)      Do the suppliers offer telephone or internet remote access support?
6)      Can you (and your staff) work everything manually?
7)      If a sensor or controller was to fail would you have adequate warning and would the risk be limited (alarms, day tanks, small areas under control)?
8)      Do you have independent monitors connected to an alarm system?
9)      Do you have suitable insurance?  Just in-case!