What Size Solar Charge Controller? The Cold-Morning Voltage Nobody Checks

Size a charge controller on open-circuit voltage on the coldest morning, not on watts. Victron’s datasheets say too much PV power is simply throttled — but too much voltage destroys the controller, and panel voltage rises as temperature falls. Four 150 W panels in series read 89.2 V in the spec table and 100.1 V at 14°F, which is over the limit of every 100 V controller made.

The figure that actually breaks controllers

Panels are rated at Standard Test Conditions, which Victron defines as 1000 W/m² at 25°C. Almost nobody charges a battery at 25°C in the morning.

Open-circuit voltage has a negative temperature coefficient: as the panel gets colder, Voc goes up. Victron publishes −0.35% per °C across its whole monocrystalline range. So:

Temperature	Voc multiplier	A 22.3 V panel reads
40°F (4.4°C)	×1.072	23.9 V
32°F (0°C)	×1.088	24.3 V
20°F (−6.7°C)	×1.111	24.8 V
14°F (−10°C)	×1.123	25.0 V
0°F (−17.8°C)	×1.150	25.6 V
−20°F (−28.9°C)	×1.189	26.5 V

Multiply by the number of panels in series and the problem appears fast. Four 150 W panels in series: 89.2 V at STC, 100.1 V at 14°F. A Victron 100/30 is rated to 100 V. You are 0.1 V over, on the first cold morning, and the controller is finished.

Watts are the number that does not decide it

This is the part that makes every sizing guide wrong in the same way. They all give you the same formula — array watts ÷ battery volts × 1.25 — and Victron’s own datasheet footnotes say what each excess actually does:

1a) If more PV power is connected, the controller will limit input power.

2. A PV array with a higher short circuit current may damage the controller.

Read those together. Too many watts: throttled. Too much short-circuit current: may damage. Too much open-circuit voltage: destroyed — which is why Victron puts the voltage limit in the product name. A 100/30 is 100 V maximum PV and 30 A charge current. The first number is the one the guides never check.

Oversizing an array in watts is a money problem, not a safety one. You lose the difference between your array and the controller’s nominal PV power, and nothing breaks.

How to read a controller’s name

Victron’s model numbers are the specification, which makes them unusually easy to check: 100/30 is 100 V maximum PV and 30 A charge current. 75/15, 150/35 and so on read the same way.

The second number — the charge current — is the one the usual formula sizes, and it is also the one Victron expresses a second way on the datasheet, as nominal PV power: 440 W at 12 V for the 100/30, 700 W for the 100/50. Those are the same statement. A 30 A controller on a 12 V bank cannot pass much more than 440 W, so exceeding it is where the throttling happens.

This is why “round up to the next size” is not wrong so much as incomplete. Rounding up the current buys you headroom you may not need. Rounding up the voltage — 75 to 100, or 100 to 150 — is what stops a cold morning ending the controller, and the two do not move together: the 150/35 in our records has a lower current limit than the 100/50.

Series or parallel — and why rewiring is free

Series adds voltage. Parallel adds current. The panels do not care; only the controller does, and that makes wiring the cheapest fix available.

Those four 150 W panels:

• All four in series: 100.1 V at 14°F, 8.69 A. Over every 100 V controller.
• Two series × two strings: 50.1 V, 17.38 A. A 100/50 takes it whole.
• All four in parallel: 25.0 V, 34.76 A. Needs a controller rated past 34.76 A of short-circuit current.

Same panels, same 600 W, three completely different controller requirements. The calculator above says when a rewire rescues an array, and it says it before recommending anything, because moving two MC4 connectors costs nothing and a bigger controller does not.

The trade-off going to parallel is cable: four times the current needs conductors sized for it. Our 12V wire size calculator sizes the run from array to controller.

The current side, and the margin that is not a margin

Short-circuit current is the second limit, and on parallel arrays it is usually the binding one. Four panels in parallel at 8.69 A each is 34.76 A, against a Victron 100/30’s 35 A ceiling.

That clears it — by 0.24 A, or 0.7%.

This page treats that as no margin at all. Isc has a temperature coefficient too, +0.04% per °C by Victron’s own figure, so a panel hotter than 25°C draws more than the number being checked. A margin thinner than a few percent is inside the measurement, not outside it, and the calculator flags anything within 5% rather than calling it a pass.

Notice what that does to the recommendation: on a six-panel parallel array the 150/35 fails on current while the cheaper 100/50 passes. The dearest controller is not the answer, because its 40 A limit is lower than the 100/50’s 60 A. Sizing on price, or on the first number in the model name alone, gets this backwards.

MPPT against PWM, in watts

PWM controllers are cheaper, and the usual explanation — “MPPT is up to 30% more efficient” — understates the gap badly for series arrays.

A PWM controller connects the array more or less directly to the battery, so the panels are dragged down to battery voltage. It harvests roughly battery volts × Impp instead of Vmpp × Impp. On Victron’s 150 W panel, Vmpp is 18.2 V and Impp is 8.25 A:

• Four in parallel on 12 V: 13.8 × 33.0 ≈ 455 W of 600 W. A 24% loss.
• Four in series on 12 V: 13.8 × 8.25 ≈ 114 W of 600 W. An 81% loss.

Series wiring and PWM are close to incompatible, because everything series wiring adds is voltage and PWM throws voltage away. If you have a PWM controller and a series string, that is the whole problem.

Model	Max PV voltage	Max PV short circuit	Charge current	Nominal PV power	Battery	Buy
SmartSolar 75/15	75 V	15 A	15 A	220 W at 12 V, 440 W at 24 V	12/24 V auto select	Check price(opens in a new tab)
SmartSolar 100/20	100 V	20 A	20 A	290 W at 12 V, 580 W at 24 V	12/24/48 V auto select	Check price(opens in a new tab)
SmartSolar 100/30	100 V	35 A	30 A	440 W at 12 V, 880 W at 24 V	12/24 V auto select	Check price(opens in a new tab)
SmartSolar 100/50	100 V	60 A	50 A	700 W at 12 V, 1400 W at 24 V	12/24 V auto select	Check price(opens in a new tab)
SmartSolar 150/35	150 V absolute maximum coldest conditions, 145 V start-up and operating maximum	40 A	35 A	500 W at 12 V, 1000 W at 24 V	12/24/48 V auto select	Check price(opens in a new tab)

Where this goes wrong

• Sizing on watts. It is the one excess the manufacturer says is handled.
• Reading Voc off the spec table. That figure is at 25°C. Your controller meets the cold one.
• Assuming the 1.25 factor covers it. It is applied to current. Voltage is a separate limit on a separate line.
• Treating a 1% current margin as a pass. Isc rises with heat; a margin that thin is inside the coefficient’s own swing.
• Buying a bigger controller to fix a voltage problem. Rewiring is free and often enough, and a “bigger” controller can have a lower current limit.
• Putting a PWM controller on a series string. On four panels in series that discards about 81% of the array.
• Forgetting the cable when going parallel. Four times the current needs the conductors to match — see the 12V wire size calculator.
• Sizing the controller before the array. This page takes the array as given. If you have not settled how many panels you need yet, how much solar an RV needs works it out from the month you actually camp in — which changes the array, and therefore this answer.

Questions worth answering