The model does not need to be elaborate. It needs to contain the lines that vary between the options being compared, and to apply the same assumptions to each.

Purchase price is one line of the model

The purchase price is the easiest line to obtain and usually the smallest in proportion to the total. What it does not capture is anything about how the equipment behaves once it is in service.

The lines that vary between configurations are the ones worth modelling. A heavier set costs more to transport and takes longer to assemble. A set with a wider range may need larger accessories. A configuration with more channels requires more calibration and more training.

Applying the same assumptions to each option is what makes the comparison valid. A model that applies the buyer’s actual site conditions to one option and the supplier’s optimistic assumptions to another produces a comparison of assumptions rather than of equipment.

Calibration and verification over the life

Calibration is a recurring cost with three components: the interval, the scope and the availability. The interval determines how many calibrations occur over the modelled life, the scope determines the price, and the availability determines whether the equipment is out of service and for how long.

Availability is the component most often omitted. Equipment that has to be shipped to a distant laboratory for calibration may be unavailable for weeks, and on a maintenance programme with fixed windows that unavailability has a cost. Equipment that can be calibrated locally, or that comes with a service exchange arrangement, carries less.

The verification activity between calibrations also belongs in the model, because it consumes reference equipment and staff time. Where an in-house reference is required, its own calibration cost is part of the picture. The measurement system requirements that determine how a high-voltage measuring system is qualified are set out in IEC 60060-2, and the traceability framework the calibration has to satisfy is described in the NIST handbooks and by the accreditation bodies such as ILAC.

Consumables, accessories and wear parts

Consumables vary between configurations in a way that is easy to miss. Test leads, clamps, filter elements, oil samples, gas cylinders and calibration gases are all recurring purchases, and their cost depends on the equipment design as well as on the price list.

The measure that matters is availability rather than unit price. A consumable that costs less but has a long lead time can stop a planned outage, and the cost of the delay exceeds the saving. For equipment used in critical maintenance, the specification should include the availability and lead time of the consumables it depends on.

Wear parts that are specific to one manufacturer are a long-term dependency. Where a part is available only from the original supplier, the buyer’s cost over the life depends on that supplier’s pricing, and the model should state that assumption rather than leave it implicit.

Software updates and licence terms

Test equipment increasingly depends on software for control, analysis and reporting. The terms under which the software is licensed and updated are part of the total cost, and they vary between suppliers.

The questions worth answering are whether updates are included, whether a subscription is required, whether the software continues to work if a subscription lapses, and whether the data files remain readable outside the software. The last question matters most over a long retention period, because a record that can only be opened by software the buyer no longer licenses has effectively been lost.

Where the equipment is part of a test data system, the integration cost belongs in the model as well. An instrument whose data has to be transcribed into the buyer’s system reintroduces manual work on every test, and that work recurs for the life of the equipment.

Service, downtime and mobilisation cost

Mobilisation is often the largest single line over the equipment life and the one least likely to appear in a purchase comparison. It includes transport, lifting, assembly and connection time, and it is incurred on every job rather than once.

The cost depends on the weight, the number of pieces, the assembly complexity and the supply requirement. Two sets with the same electrical capability can have very different mobilisation costs, and on a programme with many short outages that difference dominates the model.

Downtime for repair belongs in the model as an expected cost. The question is not whether the equipment will ever fail but what the consequence is when it does: whether a replacement is available quickly, whether a service exchange arrangement exists, and whether the outage can proceed with the equipment unavailable. Service arrangements of this kind are described under testing services and OEM/ODM solutions.

Training and competence replacement

Training is a recurring cost rather than a one-off. Staff change, and each new operator needs training on the equipment and on the methods it supports. Where the operation is unusual, the training burden is higher and the dependency on the supplier’s support is greater.

The model should include the initial training and an allowance for replacement over the equipment life. Where the equipment supports a method that requires specific competence, the cost of maintaining that competence across the team belongs in the picture.

A configuration that is simpler to operate can be cheaper over its life even when it costs more at purchase, because it reduces the training and support burden. That comparison only becomes visible when both are in the model.

Residual value and obsolescence

Residual value is a credit at the end of the modelled period, and it should be applied only where a realistic market for the equipment exists. For specialist high-voltage equipment the second-hand market is thin, and an optimistic residual assumption is a way of making the model show the answer the author wanted.

Obsolescence is the other end of the same question. Equipment whose software or spares stop being supported becomes unusable regardless of its physical condition, and the risk of that happening earlier for one option than another belongs in the comparison.

The useful treatment is to state the assumed life and the assumption behind it. Where the assumption is that the equipment remains supported for fifteen years, that assumption can be tested by asking about the support policy for the current model generation.

Building a TCO comparison across two bids

The comparison should use a common period, common assumptions about the workload and common treatment of the uncertain lines. A five-year and a fifteen-year model produce different rankings, and the period should reflect the buyer’s planning horizon rather than whichever period favours a preference.

The output of the model should be a cost per year, so that options with different assumed lives can be compared on a common basis. Where the conclusion depends on a single assumption, that assumption should be stated so the decision can be reviewed on it.

The model’s real value is not the number it produces but the conversation it forces: which lines vary, what the buyer’s actual mobilisation conditions are, and what the equipment will be required to do. Those questions are worth answering whether or not the final figure is used. Independent acceptance practice that informs a maintenance programme is published by NETA, asset management practice is coordinated through CIGRE study committees and published by EPRI, and the equipment range the model applies to is listed under all electrical testing equipment, with partner supply arrangements described under become a partner and documentation under product documents for download.