The Other Datasheet: Reading a Solar Panel Label for a Pump System

The short answer: a pump system is a marriage between two datasheets, and most buyers read only one. The pump side gets plenty of attention — our watts guide shows what the nameplate hides and the curve guide teaches the duty point. The panel side is usually chosen with one number, the wattage, because wattage is the one thing printed in giant type. But the four small numbers under it — Vmp, Imp, Voc, Isc — decide whether your string wakes up inside the controller’s window on a frosty dawn or cooks off it at noon. This guide covers what each of the four numbers measures, what standard test conditions quietly assume, how the two temperature coefficients move your string through the day, and three tests that turn any datasheet into a panel count. Eight models cited here are verified against the live product database.

Why the second datasheet deserves the same respect

The pump curve and the panel datasheet are the same kind of document: a measured promise under stated conditions. Nobody would order a 750 W pump because “750” sounds big without checking that it delivers the flow at their head — yet the same buyer will order panels on wattage alone and discover at commissioning that the string voltage sits outside the controller’s window, or that a cold morning pushes the open-circuit voltage past the ceiling.

The failure is silent and one-directional. A string that is a little too small wastes noon hours; a string that is a little too high kills the controller the first hard frost. Panels do not protect their buyer the way a pump’s dry-run protection does. The datasheet is the protection — if it gets read.

The four numbers, and what each one measures

Every crystalline module datasheet carries the same quartet, measured under standard test conditions:

• Vmp — voltage at maximum power. Where the module actually operates when working: the voltage the controller sees for most of the sunny day. Series strings add Vmp.
• Imp — current at maximum power. What flows at that operating point. Parallel strings add Imp; series strings do not.
• Voc — open-circuit voltage. The ceiling: what the module shows with nothing connected, first light, before any current flows. It is the highest voltage your equipment will ever see from that string, and every controller max-Voc rating exists to be checked against it.
• Isc — short-circuit current. The worst-case current, used to size fuses and DC breaker ratings.

For the common 550 W class the numbers land near Vmp 42 V, Imp 13 A, Voc 50 V, Isc 14 A. For legacy 100 W 12-volt-class modules: Vmp 18 V, Imp 5.6 A, Voc 22 V, Isc 6.1 A. Memorise your own class from the datasheet, not from this paragraph — the arithmetic below is the transferable part.

What standard test conditions quietly assume

STC is 1,000 W per square metre of irradiance, a cell temperature of exactly 25 °C, and a standardised spectrum. The fine print is the temperature. A cell under full noon sun runs 20-30 °C above the air around it, so a 25 °C cell usually means the air was near single digits. A realistic summer noon puts the same cell at 55-65 °C, and crystalline output falls by very roughly 0.4 % per degree — 8-12 % gone to heat alone before dust, wiring and tolerance touch anything.

This is the honest derivation of the 1.3x rule our catalogue prints on every model: a solar pump is expected to run at full load for hours, real panels in real heat deliver about 80 % of label, and 1.3 divided by 0.8 leaves roughly 1.04 — the array can feed the pump at noon and little more. The rule is a floor, not a target, and it is derived in full in the watts guide.

The two coefficients that move your string through the day

The datasheet lists temperature coefficients near the bottom, and two of them matter for pumping:

• Voc rises as the cell cools — about 0.3 % per degree below 25 °C. A winter sunrise at −13 °C puts the string about 15 % above its nameplate Voc; a mild winter morning adds 7-10 %. The check that matters is always the coldest dawn your site produces, not the average.
• Vmp sags as the cell heats. The hot-noon operating voltage of the 550 W class lands near 37 V against the nameplate 42 V — and the MPPT window’s floor is checked against the sagging number, not the nameplate.

So the string lives between two moving walls: the cold-dawn Voc against the controller ceiling, and the hot-noon Vmp against the MPPT floor. Both walls are printed on two different documents — the ceiling in the controller’s table, the floor too — and the datasheet gives you the numbers to test against both. The same physics is why altitude and cold sites need their own arithmetic.

Three tests, in order, turn a datasheet into a panel count

Take any pump and any module datasheet and run the tests in this order — cheap checks first:

1. Watts: array ≥ 1.3 × pump power. This sizes the total, nothing more.
2. Cold Voc: string Voc × 1.15 < controller max Voc. Count how many modules in series this allows. Never skip the 1.15 unless your site truly never frosts.
3. Hot Vmp: string Vmp sagged > MPPT floor. For the 550 W class use 37 V per module; for others, Vmp × 0.9 is a fair hot estimate.

The controller windows for our ladder, reprinted from the controller guide:

The 72 V row is the lesson the table teaches for free: with the wrong module class, no string count works — which is why the module is chosen against the window first and the wattage second, not the other way round. Above the fixed-voltage classes, the wide-window and A/D machines turn string layout into architecture, and that is a different guide.

Worked example A: the 200 W pump where watts and voltage disagree

The 2DPC1.5-35-24-200 wants 260 W of array by the 1.3x rule. With legacy 100 W modules the watt math says three: 300 W clears the bar. The voltage math says two in series is the ceiling — three in series reaches 66 V of cold Voc, over the 60 V limit, and the very first frosty dawn would find the controller’s limit for you, permanently.

The resolution is parallel: two strings of two, four modules, 400 W. Voltage check: 2 × 22 V = 44 V Voc, comfortably under 60; operating Vmp 36 V, inside 30-48 V. Current check: two strings in parallel double the Imp to about 11 A — under the controller’s 17 A input budget, with Isc-based fuse sizing left to the protection guide (14 A class Isc × 1.56 → a 25 A gPV fuse). The watt overshoot is not waste: it is morning and afternoon hours, which is exactly what a small pump starved of noon power needs most.

The modern shortcut, as noted in the string-design guide: one 550 W module now doubles a 200 W pump’s target on its own, with Vmp 42 V and Voc 50 V both inside the same window. The datasheet tests are identical either way.

Worked example B: two modules and a 4 % frost margin

The 3DSC6-60-48-750, 750 W, wants at least 975 W of array. Two 550 W modules in series: 1,100 W clears the bar. Voltage: Vmp 84 V sits inside the 60-90 V window; hot-noon sag to about 74 V stays above the floor. Cold check: 2 × 49.9 V = 99.8 V of nameplate Voc, and at −13 °C it climbs to about 115 V against the 120 V ceiling.

That is a pass — with 4 % of margin. It is exactly the kind of answer that looks identical to a fail on a supplier’s optimistic quote where “Voc ≈ 48 V” rounds the arithmetic to 110 V, and identical to a disaster on a datasheet with a 51.5 V module. The number did not change; the datasheet did. This is why the check happens with the manufacturer’s PDF open, before ordering, and gets re-verified with a multimeter at commissioning.

Worked example C: the tempting extra panel, and the class ceiling

The 4DSC3.5-140-110-1100, 1,100 W, wants 1,430 W. Three 550 W modules: 1,650 W clears it, Vmp 126 V inside 110-150 V, cold Voc about 172 V against a 220 V ceiling — comfortable in both directions. The tempting fourth panel breaks two tests at once: nominal Vmp 168 V leaves the window’s top, and the frost Voc reaches about 230 V against 220 V. Both failures are invisible on a warm afternoon at the counter where the extra panel gets bought.

One rung up, the ceiling of the whole class appears. The 4DSC16-70-110-1500 wants 1,950 W; three modules give only 1.1x and four are barred by the same two tests. The honest answer is not more maths — it is a different machine whose window gives the modules room. The 3DSC5-135-150-1500-HV accepts 80-430 V, where four modules of the same 550 W class fit with ease (frost Voc about 230 V against a 430 V ceiling) and the same three tests pass with real margins. At the farm tier the pattern repeats at a higher voltage — the 6DSC65-85-380/520-7500-A/D prints its optimum DC input as 520-650 V on the spec sheet, and an 18.5 kW 8DSC150-37-380/520-11000-A/D turns the same arithmetic into a 26-module ground array, which is a construction project, not a shopping list.

The arc is the lesson: the three tests never change — watts, cold Voc, hot Vmp. What changes at each rung of the model ladder is how much room the window leaves for the modules you can actually buy. Choosing the window is as much a part of pump selection as choosing the pump.

Reading the label in the yard, not just the PDF

Two field notes close the loop. First, the rear label always prints Voc and Isc — those set the DC safety ratings — while Vmp and Imp sometimes hide in the datasheet PDF. Estimates (Vmp ≈ 0.8 × Voc, Imp ≈ 0.93 × Isc) are fine for arithmetic, but a ten-minute multimeter check of one module at commissioning turns every estimate in this guide into a measurement. Second, read the tolerance line: “550 W” may carry ±3 % tolerance or a “up to 550 W” bin, and the bottom of that range is the number your cold morning will meet.

The pre-order checklist:

1. ☐ Pump model fixed, 1.3x wattage target written down.
2. ☐ Module datasheet PDF in hand — label numbers alone are not enough.
3. ☐ String Voc × 1.15 (your site’s real frost) < controller max Voc.
4. ☐ String hot-noon Vmp > MPPT floor (37 V per 550 W-class module).
5. ☐ Parallel strings: identical, and total Imp within the controller’s input current.
6. ☐ Fuse rating sized from Isc × 1.56, gPV rated.
7. ☐ One module’s Voc and Isc measured with a multimeter before the full array is wired.

The pump gets chosen with its curve. The array deserves the same two minutes with its datasheet — and the frost dawn will never know the difference.