Contents

Testing times (added: 9/2/2012)
To PLC, or not to PLC, that is the question.(added: 16/1/2012)
Saving Energy with Temperature Control (added: 3/1/2012)

Testing Times!
added: 9th February 2012

At Top Hex, we like to say that everything we supply is second hand. By this, we mean that every controller that goes out of the door has been functionally tested. I wanted to explain why we do this, the most common reasons that a controller will fail and what we then do to put it right before it gets to the client.

During the manufacturing process, there are various stages of inspection and test. However, the final test that is carried out at Top Hex is a functional test. During the functional test, the controller is powered up and run in a mode which is as similar as possible to the operating conditions for the product. There is a certain poetic license in this - for example we have a 9.0kW heater controller (see news story) and it is not safe or feasible to put 40A @230V through the output circuit, but we do simulate a load.

There are several reasons why a controller may have a fault. Although all circuit boards are tested before they are shipped to us, there is a tiny possibility a fault has not been picked up. Electronic components are pretty reliable, but there is still a small chance that a component is faulty, maybe from the manufacturing stage or possibly if handled badly in the supply chain. There is also the possibility that the wrong component has ended up in the wrong place or that the solder joint is poor - which can either be a human or a machine error. With a skilled workforce, this doesn't happen very often either.

So why bother? In testing a batch of 100 controllers, it would not be unusual for all 100 to pass first time. Although we are able to automate much of the testing process, it can be tedious and it is certainly time-consuming.

From our point of view, we accept that although we may be supplying a £100 controller, it may be fitted to a £100,000 machine and be shipped overseas. The implications of a controller being faulty can often be horrendous and we would rather not take this risk.

If a controller does fail the test, the solution is simply to swap the faulty part or correct the soldering and to completely re-test, in case the repair has created a further fault. The possibility of this happening is remote, in fact we can't actually remember it having happened. However, when testing, you can't be too careful!

To PLC, or not to PLC, that is the question.
added: 16th January 2012

A question that I am frequently asked is whether a Programmable Logic Controller or 'PLC' is the best solution to controlling a machine or system or if it is worth considering a custom designed electronic controller. At Top Hex, we use both solutions and in my opinion, the decision as to which is the most suitable is usually clear-cut. In this blog, I have set out some golden rules which I use and hope will point you in the right direction.

The most important consideration is whether you will be supplying the control system again in the future. If the system is a "one-off" then a PLC is undoubtedly the best solution. I would suggest that if you can imagine ordering 3-5 controllers at a time, then it's time to consider a customised electronic controller. The problem with an off-the-shelf PLC is that you will paying for features that are of no benefit to you but may be useful to somebody else. For example, most PLCs have a real-time-clock which adds several pounds of cost - great if you want to know what time it us - otherwise a waste of money!

Often the best approach is to rationalise your design wherever possible and look for similarities between systems. If you can treat each new system as a "variation on a theme" then you can potentially be running standard software and saving yourself a great deal of effort. Many of the electronic designs we produce for our clients allow enormous variations between systems. For example. we have recently designed a temperature control system which allows for between 1 and 255 channels - this should cover most eventualities!

The other advantage of a customised controller is that it is possible to make some significant cost savings by including external components on the circuit board. Power supplies, safety relays, temperature controllers and even power switching contactors can all be included in the design and will cost a lot less if they are fitted to the circuit board. Best of all, you can save time and money on wiring as there will be no cables required between these components.

If you are looking to interface with a PLC manufacturer specific communications protocol then most likely, the PLC will be your best option. If you want to include a standard interface such as an Ethernet port or a USB port then you are likely to find that the customised electronic is going to be more flexible and indeed less expensive.

Ultimately, there is very little difference in technology between a PLC and a custom electronic controller. I like to think of a customised controller as being a PLC with exactly the right combination of inputs and outputs for your system. From experience, it doesn't really matter whether the decision is made on a technical or a cost basis as the answer is usually the same! Please let me know if I can help provide you with the tools to make this.

Saving Energy with Temperature Control
added: 3rd January 2012

We have recently developed a controller which controls an electric heater in an enclosed space – e.g. an office or industrial unit. The controller takes a temperature reading from a room temperature sensor and also close to the heater. The temperature control software uses a version of our self-tuning temperature control algorithm which adapts the control parameters to achieve the optimal possible temperature control.

The system was tested in a typical environment and our client carried out comparative tests using a leading competitors BMS controller. We knew our adaptive temperature controller was good, but we didn’t realise quite how good!

The Top Hex controller consumed 20% less electricity than the leading competitor in maintaining the required room temperature. After patting myself on the back for a few moments, the engineer inside me kicked in and I am able to substantiate the dramatic efficiency savings we were able to achieve.

The competitor’s controller uses a PID based algorithm which works reasonably well at warming up the room and minimising the overshoot, to an extent at least. However, the real differences were clear when the environment changed – for example when a door was opened or when the sun shone through the window. The Top Hex temperature controller adapted its response to cope with the changes parameters. It is almost like it is carrying out a critical assessment of its response and learning from its mistakes.

 However, the competitors controller doggedly stuck to its guns using a set of parameters that suited a situation which no longer existed.   As the Top Hex temperature controller was continually optimising its response, it used less energy in the process, minimising the over and undershoots that dog conventional controllers. It is also worth pointing out that unlike the competitor’s product, the Top Hex designed controller does not need to have separate winter and summer settings – it simply adapts to suit the season.

We know from experience that the Top Hex self-tuning algorithm works equally well in multi-zone heater control applications, fast and slow responding systems (particularly in cooling applications) and even in non-temperature control for say humidity or pressure. If you have an application you would like to discuss, please email me at simon@tophex.co.uk