Most hybrid power enquiries arrive as a sentence: the site needs power, there is a load, and there is a rough idea of what the installation should achieve — a solar array of some size, storage to go with it, and the existing generator kept as backup. That sentence is a starting point. It is not something a system can be designed from, and treating it as one is the most common reason a project loses months between the first message and the first drawing.
A design is a calculation, and a calculation is only as good as its inputs. When an input is missing, the engineering does not stop — it substitutes an assumption. An assumption costs nothing on paper and a great deal on site, because by the time it turns out to be wrong, equipment has been ordered and foundations have been poured.
What follows is the information that has to be established before a solar, battery and diesel hybrid system can be designed. It is grouped the way it is normally collected, and for each group it sets out what the answers change in the design. The purpose is not to fill in a form. It is to know, before the first design conversation, which parts of the picture are established facts and which are still being assumed.

Why a Missing Input Becomes an Assumption
Design work on a hybrid system proceeds through a chain: load, then generation, then storage, then conversion and distribution, then control. Each link is calculated from the one before it, and a gap early in the chain propagates through everything downstream.
A peak demand that is estimated rather than measured leads to an undersized inverter, an oversized generator, or both. A daily energy figure without a load profile leads to a battery sized for average consumption that empties in the first three hours of the evening peak. An existing generator described only by its capacity — without its age, condition or minimum loading — leads to a control strategy that either shuts it down too often or never lets it run in the range it was built for.
The cost of a missing input is therefore not paid at the design stage. It is paid at commissioning, in change orders, in equipment that arrives and does not fit the duty it was bought for, or in a system that works but performs worse than the calculation promised. Collecting the information is a deliverable in its own right, not a preamble to the real work. The engineering conversation that follows — the one described in how an enquiry is reviewed and developed into a power system design — can only move at the speed of the information it has to work with.
Site, Location and Use
The first group is the easiest to collect and the one most often left vague.
Location. Latitude sets the solar resource and the seasonal swing in it; it also sets the temperature range the equipment has to survive, which is a design input on the storage side as much as on the generation side. Coordinates are more useful than a place name, because they can be resolved against irradiation and temperature datasets directly.
What the site is and what it does. An industrial process, a commercial building, a mine, a resort, a telecom installation and a remote community have different load shapes, different tolerance for interruption, and different growth expectations. The purpose of the facility determines how the rest of the information is read.
Available area and layout constraints. How much ground or roof is available for generation, and where storage can physically stand — including the clearances it has to maintain and any setback the site is required to observe. This is a hard constraint rather than a preference, and it interacts directly with the equipment format the project can use; the physical and logistical limits that decide how storage can be delivered and placed are set out in how the format of a battery storage system is matched to a site.
Access and logistics. The route in, the road class, overhead and weight limits, and the lifting equipment available on site. This is the input that most often arrives last and most often changes the answer.
Load: The Input Most Enquiries Get Wrong
Load information is where a design is won or lost, and it is where enquiries are typically thinnest. Four numbers matter, and they are not interchangeable.
Peak demand, in kilowatts. This sets the rating of the conversion equipment and the size of the generator, and it is a different quantity from average demand. Motor starting, welding plant, HVAC compressors and lifts can pull a peak that a system sized on average load will not carry.
Daily energy, in kilowatt-hours per day. Together with peak demand, this sets the capacity of the battery energy storage system and the scale of the generation. It is the number most often supplied and the one that says least on its own, because the same daily energy can arrive as a flat load or as a heavy evening peak.
The load profile. How demand varies across the day, and across the year. A 24-hour curve, or a month of half-hourly readings, is worth more than any summary statistic, because it determines how much storage is needed, when it will charge, and how often the generator has to start. Where the site has a year of metered data, the design problem becomes largely arithmetic. Where it has none, the profile has to be measured or reconstructed, and that becomes the first task rather than the first assumption.
Load composition. Which loads are essential and which can be shed, whether large motors start direct-on-line or through a soft starter, and whether there are significant non-linear loads that affect power quality. These determine how the system behaves at the worst moment, not on an average day.
If there is a single recommendation to take from this article, it is that a month of metered load data at the point of supply is the highest-value document a site can provide — higher than any specification it could write.
The Grid and the Existing Generator
The second supply, present or absent, changes the shape of the whole system.
Grid availability and reliability. Whether the grid is present, how often it fails, and for how long. Frequency and duration of outages are what determine whether storage is being sized for economic dispatch or for continuity, and the two produce very different systems.
Grid characteristics. Voltage and frequency, and how stable they are. A weak or unstable supply changes what the conversion equipment and the control system have to tolerate — the failure modes and the ride-through behaviour involved are covered in what equipment has to tolerate on a weak or unstable grid.
Operating mode. Whether the system will run grid-tied, off-grid, or in parallel with the grid, and whether it has to switch between modes. This is a control-architecture decision that follows from a factual answer, not a preference.
Tariff and metering arrangements. Import and export limits, time-of-use structure, and what the metering can actually record. These decide whether the storage is worth running for peak shifting, and how it should be scheduled.
The existing generator installation. For each set: capacity in kVA or kW, age and condition, running hours, fuel type and consumption, and — critically — its minimum loading, the point below which running it does more harm than good. A hybrid system that includes an existing generator is designed around that generator’s real operating envelope, and a capacity figure on its own does not describe one.
How these supplies are combined into a single system — the arrangement described as a solar, battery and diesel hybrid microgrid — follows from the answers above rather than preceding them. The architecture is the output of the information, not a substitute for it.
Backup Duration, Critical Loads and Interconnection
Required autonomy. How long the system must carry the load with no solar input and no grid — expressed in hours or days, and stated for a defined load, not for the whole site. Autonomy is the input that most directly sets battery capacity, and it is the one where the gap between what a site asks for and what it needs is usually widest. The reasoning that connects a required autonomy figure to an actual generator and storage combination is worked through in how much diesel backup a hybrid microgrid actually needs.
The critical load list. The specific loads that must remain energised, with their ratings. This is a shorter list than the total load and a more useful one, because it is what the system is designed to protect.
Staged shedding. Whether the site can accept a sequence in which non-essential loads are dropped before essential ones, and in what order. Where a site can shed, the storage requirement falls; where it cannot, the system has to carry everything.
Interconnection and compliance. The grid code or interconnection standard that applies, the protection and control requirements that follow from it, the metering arrangement, and any import or export limit imposed by the utility. Where the project is off-grid, the applicable standards come from the equipment and safety side instead. This group is the one most often discovered late, and it has the longest lead time, because approval is not something a supplier can shorten.
Site Conditions and Environmental Data
The environment decides how equipment is specified, positioned and protected, and it is worth collecting as hard data rather than regional averages.
Solar resource. Irradiation data for the location, ideally as a monthly profile rather than an annual total, since the design has to work in the worst month.
Temperature. The minimum and maximum ambient, and the range the equipment will actually see where it is placed. Temperature drives both generation output and usable storage capacity, and the mechanism by which heat reduces what a battery can deliver is set out in how temperature affects usable storage capacity.
Altitude. It affects cooling, insulation and the derating of several classes of equipment, and it is routinely omitted from enquiries.
Dust, humidity, salt and corrosion. Conditions that determine enclosure and corrosion requirements, and that have to be recorded for the position the equipment will occupy rather than for the site as a whole — the distinction that decides what a protection rating actually delivers is covered in how IP ratings and corrosion classes relate to where equipment is placed.
The Documents That May Already Exist
Most sites hold more of the required information than they realise, in documents written for other purposes.
Single-line diagram. Even an outdated one shows the topology, the voltage levels, the protection arrangement and the switching — and can be marked up rather than rebuilt.
Historical load data. Meter readings, utility bills, or the logs of a building management system. Twelve months of bills will not give a load profile, but they will bound the energy and reveal the seasonality.
Equipment schedules and nameplate data. Ratings, ages and duty of the existing generator, transformers and switchgear.
Site layout and drawings. Available area, existing services, access routes and any planning constraint already recorded.
Studies and reports. Any earlier energy audit, feasibility study, grid study or interconnection application. These often answer questions that would otherwise be asked again from the beginning.
Applicable standards. The code or standard the installation has to comply with, which determines much of the protection and metering design.
What Each Missing Item Changes
The value of the list is not in completing it but in knowing what a gap costs.
No measured peak demand means the conversion and generator ratings are sized on an estimate, and the first hot afternoon with full motor load will test that estimate.
No load profile means storage is sized on daily energy alone, and the system is likely to be either oversized for the average day or unable to cover the peak hour.
No minimum loading figure for the existing generator means the hybrid control logic is designed blind, and the common failure — a generator that is cycled below the load it can tolerate — is built in from the start.
No autonomy definition means battery capacity is whatever the budget suggested, rather than what the site actually requires, and the two are rarely the same.
No interconnection requirement means the design reaches the approval stage before the constraint is known, which is the slowest possible order in which to discover it.
No environmental data for the equipment position means protection and thermal design are based on an average that no individual piece of equipment will experience.
Preparing the Pack
The information does not have to be complete before the first conversation — it has to be classified. What is measured and what is estimated should be kept apart, because the second can be replaced by the first and the first cannot be quietly substituted by the second.
A workable sequence: collect the load data first, since almost everything else depends on it; establish the site’s physical and access constraints second, because they eliminate options; document the existing supply third, including the generator’s real operating limits; and gather the compliance and site-condition data last, since those refine a design rather than define it.
Sites that arrive at the design conversation with measured load data, a site plan, an up-to-date single-line diagram and a clear statement of the required autonomy tend to move from enquiry to a workable system concept in a single exchange. Sites that arrive with a capacity figure and a budget tend to spend that exchange establishing what was actually being asked for. The difference is not engineering effort. It is preparation, and it is the party with the site who can do it.

FAQ
Q: We do not have measured load data. Can a design still be produced?
A: Yes, but it will be a design based on an estimate, and it should be labelled as one at every stage. The practical alternative is to measure — a week of half-hourly readings at the point of supply, taken across a representative period, replaces most of the guesswork and is far cheaper than resizing equipment later. Where measurement is impossible, the design should state the assumed profile explicitly so that it can be checked when the system is commissioned.
Q: How detailed does the single-line diagram need to be?
A: Detailed enough to show the topology, the voltage levels, the main protection devices and the switching arrangement. It does not have to be current or complete. An outdated diagram that shows the original layout can be marked up in a conversation; a missing one means the arrangement has to be surveyed from scratch, which is slower and less reliable.
Q: We have three quotations with very different sizes. How do we tell which is right?
A: Compare the assumptions behind them, not the headline capacity. Offers that differ significantly are usually built on different load profiles, different autonomy requirements or different assumptions about the existing generator. Ask each supplier to state the load profile, autonomy and generator duty its design was calculated from. The offer whose assumptions match your site is the one worth pursuing, regardless of which is largest.
Q: How long does it take to gather this information?
A: The load data and the existing documents usually take the longest, because they involve retrieving records the site already holds rather than producing anything new. Site conditions and compliance requirements can often be resolved quickly for sites in a well-documented jurisdiction, and slowly where the applicable code has to be established first. Starting with the load data is the single decision that shortens the rest of the process most.
Q: What if the site plans to expand later?
A: Then the growth expectation belongs in the same pack as the current figures, because it changes the design now — the rating of the conversion equipment, the capacity of the switchboard, and the space and connection reserved for a second stage. Expansion is far cheaper to accommodate at the design stage than to retrofit; the constraints that expansion has to respect are the same ones described in how an off-grid system is sized, applied with a longer horizon.
Q: Do we need to know the exact solar resource before we start?
A: A good estimate for the location is enough to begin, because the design has to work in the worst month rather than the annual average, and that figure is what drives the sizing. Precise resource data matters later, when the yield has to be confirmed, but it should not hold up the load and storage work, which is where most of the design risk sits.
Q: Our existing generator is in poor condition. Does that change what we should provide?
A: It changes it substantially, and it should be stated up front rather than discovered during commissioning. Age, running hours and condition determine whether the generator can serve as a reliable backup, whether it should be retired and replaced, and what duty the hybrid control should assign it. A design that assumes a healthy generator and is built on a marginal one will not deliver the autonomy it promised.

The information in this list is not difficult to assemble for a site that already has it, and not difficult to obtain for one that does not. What it requires is a decision to treat the facts about the site as the first deliverable of the project rather than as an afterthought to the equipment. A system designed from measured peak demand, a real load profile, a known autonomy requirement and a documented generator duty is one that can be built as drawn. A system designed from a capacity figure and an intention is one that will be redesigned on site, at the point where redesign is most expensive.
YURI POWER Engineering Team · October 9, 2026
