Match maker: Power availability should match expected demand

The recent trend in Foundation Fieldbus power supplies and conditioners has been to develop units that have higher voltage and higher current capacities than needed. This is ironic, since the exact opposite trend is happening to the field devices at the other end of the cable. These devices require less power, lower current, and they are able to operate closer to the fieldbus-specified minimum of nine volts.

Let’s take a look at what might be considered a reasonable current load for a fieldbus system. As a rule of thumb, use a conservative value of 20 mA for each device when designing a fieldbus system. Most installations do not use more than 12 devices on a single network; not because of physical constraints, but rather to manage the risk of losing so many signals in the event of a single point of failure.

The maximum current load on a network is in the range of 240 mA, plus additional room for connecting handheld diagnostic tools (10 mA) and 50 mA (60-13) in the event of a short circuit. The maximum current required in a fully loaded segment of 12 devices at 20 mA/device + (60-13 mA) is 300 mA (12 x 20 + 60). A reasonable maximum current demand for most systems is therefore 300 mA.

The other half of the power calculation is voltage for which a minimum of nine volts is required at any point in the network. The worst-case scenario is one in which the maximum length of 1,900 metres of cable is installed. Fieldbus Type A cable has a nominal resistance of 100 Ohms/km, so to keep the calculation simple we can assume an overall system resistance of 200 Ohms.

Using Ohm’s law that V=IR, you can calculate the worst-case voltage drop – ∆V= 0.3 A x 200 = 6 V.

Therefore, if the power conditioner can supply a minimum of 15 volts at its terminals, the minimum nine volts will be available anywhere in the system. To provide a margin of error, a reasonable value should be 16 to 17 volts.

The main reasons for specifying a power conditioner with more than the above voltage output are to either ensure more voltage at the end device, or to have a larger margin for the voltage drop due to cable resistance. Cable resistance will change with temperature, cable resistivity (function of manufacturing and cable diameter) and length. Therefore, any power conditioner greater than 300 mA and 17 volts is in most cases over-specified or not needed to overcome power consumption elsewhere in the system. This analysis is based on the installation of traditional fieldbus power conditioners and does not always apply when specifying fieldbus barrier systems.

Because fieldbus barrier systems include transformers and a number of components to convert the high-energy trunk to the low energy of the intrinsically safe (IS) spurs, the barriers themselves consume some power and also have an associated voltage drop.

It is these fieldbus barriers that drive the need for larger power conditioners. Each fieldbus barrier consumes a maximum of approximately 250 mA per unit (depending on manufacturer), supporting a maximum of four IS spurs and devices. Again, to refer to our 12-device network, the maximum-sized power conditioner should not exceed 750 mA. This is approximately 2.5 times what is required when installing a more traditional fieldbus system.

What happens to the extra power that is installed in the cabinets with the power conditioners? It becomes heat. It does not make sense to use electricity to heat a cabinet that is likely already having temperature control problems, or to spend more money on cooling to dissipate the extra heat.

Referring back to the single point of failure, with today’s redundant systems, the single point is either the sensor itself or the home run cable, just like with traditional analogue systems.

If you purchase the largest power conditioner as the “easiest out,” you – as an engineer or designer – are not doing your job. The easy way out is to order the biggest so you can be sure you will always have room to spare. Remember that the power conditioner is part of the system and must match the demands it is expected to experience once installed.

In the next issue, we’ll talk about maintenance practices–corrective, preventative and predictive.