Size a charge controller
Calculates the PWM loss from your module data instead of a rule of thumb — an 18 V module on a 12.5 V battery loses 31 %. Checks the voltage window at the design minimum temperature and cross-checks your figure against your site’s typical annual minimum instead of accepting it unverified. And honestly says when MPPT is NOT worth it: payback uses the annual mean of sun hours at your location — for small systems PWM is the right recommendation.
The formulas behind the calculator
Every number above can be recomputed: the full calculation path, all assumptions and the data source with retrieval date — plus cross-validation against independent references. Disclosed, not claimed.
An estimate based on the stated assumptions. The final design must be checked by a qualified professional against the rules that apply where you are.
| Step | Formula | Value | Provenance |
|---|---|---|---|
| Temperature-corrected open-circuit voltage | V_oc * (1 + beta/100 * (T_min - 25)) | 45.982 V | exact |
| String open-circuit voltage | n_series * V_oc,corr | 183.93 V | exact |
| MPP voltage when hot | n * V_mp * (1 + beta_Vmp/100 * (T_max - 25)) | 116.27 V | exact |
| Usable power with PWM | U_bat * I_mp,array | 243.6 W | exact |
| Usable power with MPPT | P_array * eta_MPPT | 1,358.7 W | exact |
| Charge current | P_array * eta_MPPT / U_absorb | 47.176 A | exact |
- Formula
I_MPPT = P_array · η / U_absorb · P_PWM = U_bat · I_mp- Valid for
- Voltage window from the Voc core shared with calculator 11 (cold-corrected open-circuit voltage, controller limit, minimum headroom of hot V_mp above battery voltage); the entered design temperature is checked against the site TMY minimum but NOT set automatically — for a safety-critical quantity the designer sets the number. The PWM loss follows from the actual module data, not a rule of thumb. MPPT payback uses the ANNUAL MEAN of sun hours at the site (Berlin, 35 degrees south: 3.58 h/d), because it is an economic and not a sizing question.
- Not covered
- Extreme years below the TMY minimum (Annex B1 remains open — the user value governs), partial shading and its effect on the MPP tracker, multi-tracker units, the controller’s efficiency curve across the load range, the controller’s own consumption.
Frequently asked questions
MPPT or PWM — which charge controller is worth it for my system?
The calculator derives the PWM loss from your actual module data instead of a rule of thumb: a 100 W module with V_mp 18 V and I_mp 5.55 A delivers only 69.4 W into a 12.5 V battery — a 31 % loss, because PWM pulls the module down to battery voltage. It then weighs the extra yield against the price difference, and if MPPT does not pay back within five years it honestly recommends PWM.
How many amps does my MPPT controller need to be rated for?
The maximum charge current is I = P_array · η_MPPT / V_charge-end, and the controller should be rated at 1.25 times that. The calculation deliberately uses the charge-end voltage rather than the nominal voltage: using 24 V instead of 28.8 V on a 24 V bank overstates the current by about 20 % and leads to an oversized controller. In the reference case, 1200 Wp on a 24 V bank gives 40.4 A.
Why does the string open-circuit voltage have to be checked at the coldest design temperature?
Open-circuit voltage rises in the cold: V_oc,corr = V_oc · (1 + β/100 · (T_min − 25)). The string voltage must stay below the controller's maximum input voltage even on the coldest design day — this check is safety-critical. The calculator also verifies that the hot-day MPP voltage stays above battery voltage plus headroom; partial shading and controllers with several trackers are outside its scope.
How large is the PWM loss really — and where does it come from?
A PWM controller drags the module down to battery voltage instead of operating it at the maximum power point. The spec reference example: a 100 W module with V_mp = 18 V and I_mp = 5.55 A delivers only 12.5 × 5.55 = 69.4 W at a 12.5 V battery — a 30.6% loss. MPPT harvests ~97 W. Competitors assert a blanket “20–30% better”; this calculator computes it from your specific module data, and the curve chart shows where the power is lost.
Why is the cold-corrected open-circuit voltage safety-critical?
Because module voltage rises as temperature falls (β ≈ −0.27 %/°C): four modules at 41.5 V Voc reach 183.9 V together at −15 °C — a 150 V controller would be impermissible and dies on the first cold, sunny winter morning; the 250 V class is needed. Exactly this case (cold AND sunny) is the design case, and the Voc line in the chart shows where your string crosses the limit.
When is PWM still the right choice?
When the MPPT premium does not pay back: the calculator converts the extra yield in Wh/day via your sun hours into euros and divides the price difference by it. For small systems (one 100 W module at a garden shed) the honest result is “does not pay back” — then PWM is right, provided the module voltage matches the battery (36-cell module at 12 V). A calculator that sometimes recommends the cheaper part is one you can trust.
Does the calculator check my design temperature against the site?
Yes, since 5 August 2026 — with the same check as the open-circuit voltage calculator, whose core this calculator shares. If you enter a temperature warmer than your site's typical annual minimum, the calculator says so clearly: the Voc calculation then sits on the unsafe side. For Berlin that minimum is −7.9 °C; the worked example deliberately designs colder at −15 °C.
Why doesn't the calculator just set the temperature itself?
Because string voltage is safety-critical. For such a quantity the designer sets the number and the tool checks it — a silently inserted value would shift responsibility without anyone noticing. Also, a TMY year is a typical year, not an extreme one. Designing to standard requires the coldest expected temperature, which lies below the TMY minimum.
Which sun hours drive the MPPT payback?
The annual mean at your location, not the design month. That distinction matters: “when has the premium paid for itself” is an economic question spanning the whole year, while “how many modules do I need” depends on the weakest month. For Berlin at 35 degrees south it is 3.58 sun hours as an annual mean against 1.03 in December.
What would the December value have changed?
It would have stretched payback more than threefold — from 3.2 to 11.0 months in the reference case. For this system MPPT remains the right choice either way, but for a smaller system near the five-year threshold the wrong reference would have flipped the recommendation. That is exactly why it now appears as its own note on the page.
Where does the electricity price for the comparison come from?
It is now an input field — previously it was fixed at 30 ct and unreachable through the form, as was the MPPT premium. The default of 31.1 ct matches the existing-customer average from the BNetzA/SMARD price analysis. If you export rather than self-consume, enter the feed-in tariff here instead: the extra yield is then worth considerably less.