Capacity is a sum of parts nobody wrote down
On a factory engine the cooling capacity is in the manual. On anything built out of parts it is not, because the block came from one place, the radiator from another and the hoses were cut on the bench. The only way to it is to add the pieces: 11 quarts in the block and heads, 4 in the radiator, 2.5 in hoses and the tank, 1 in the heater core. That is 18.5 quarts, 4.63 gallons, 17.5 litres, and about 40 lb of coolant that the engine has to warm through before the thermostat opens.
The block is usually the biggest share and it is the one people guess at. Filling the dry assembly through a measured jug takes twenty minutes and settles it permanently.
Heat divided by temperature drop is the whole of the flow calculation
One horsepower is 2,544 BTU per hour. At 450 hp with 70 percent of it going into the coolant, that is 801,497 BTU an hour to shift. Divide by the specific heat of the mix and by the temperature drop you want across the radiator — 0.85 and 25 F — and you get 37,717 lb of coolant an hour. At 8.6 lb per gallon that is 4,386 gallons an hour, or 73.1 gallons a minute.
Seventy-three gallons a minute through an 18.5 quart system is 15.8 turnovers a minute: any given cupful of coolant goes round the whole loop once every 3.8 seconds. That number is worth having in your head, because it explains why a cooling problem shows up so fast.
Flow and temperature drop trade directly
They are inversely proportional at a fixed heat load, and the table on the page shows it. Ask for a 12.5 F drop instead of 25 and the flow needed doubles to 146 GPM. Accept a 50 F drop and it halves to 36.6. There is no free choice here — a small drop means an even block temperature and a large pump, a big drop means a modest pump and a bigger difference between the top of the engine and the bottom.
The old argument that coolant can move through a radiator too fast to cool does not survive this arithmetic. Faster flow means a smaller drop per pass and more passes; the heat rejected is the product of the two and it does not fall.
Specific heat is why the mix ratio keeps coming up
Water carries 1.0 BTU per pound per degree. A glycol mix carries less, and at 0.85 the same heat at the same temperature drop needs 18 percent more mass moving. That is a real cost and it is why racing systems that do not need freeze protection often run something closer to water. It is not an argument against glycol, which is in there for freeze point, boiling point and corrosion inhibition rather than for heat capacity, and none of those are things this page weighs up.
What the flow figure does not settle
It says nothing about whether the radiator can actually reject the heat. That is the air side — core area, fin density, air mass through the core, the temperature of the air arriving — and it is a separate calculation with its own set of numbers that this page does not ask for. A pump that moves the required gallons through a radiator too small to give the heat up will hold a steady temperature drop across a rising coolant temperature, which is exactly the failure that looks like a flow problem and is not.
The genuinely useful thing to do with the figure is the reverse check: measure the top and bottom hose temperatures on the running engine, put your measured drop in the field, and see what flow that implies. If it comes out at a third of what the pump curve says, something in the loop is short-circuiting.
Questions people ask
How much heat goes into the coolant?
It depends on the engine and the condition, and this page takes the figure from you rather than supplying one. Heat to coolant is commonly of the same order as the crank output, which is where the percentage field starts, but the real number for a specific engine at a specific load comes from the engine builder or from manufacturer heat rejection data. If you have a BTU figure off a data sheet, switch the first field and enter it directly.
Can coolant flow through a radiator too fast to cool properly?
The arithmetic does not support it. Doubling the flow halves the temperature drop per pass and doubles the number of passes, and the heat carried is the product of the two, so it does not fall. What changes is that a fast-flowing system runs a smaller difference between the top hose and the bottom, which is generally what you want across a block. The page shows the trade directly in the flow against drop table.
Why does the specific heat of the mix matter?
Because heat carried is mass times specific heat times temperature change. Water is 1.0 BTU per lb per F and a glycol mix is below it, so at 0.85 you have to move about 18 percent more mass for the same heat at the same drop. The glycol is in the system for freeze protection, boiling point and corrosion inhibition rather than for heat capacity, and this page takes no position on the mix — it just uses the number off your datasheet.
How do I find the volume of my block?
Fill the dry assembly through a measured container and count. On a built engine there is no published figure that means anything, because the block, the heads, the gaskets and whatever is plumbed into the water side all came from different places. It is the largest single volume in most systems and the one that makes the capacity figure wrong when it is guessed at.
Does this tell me if my radiator is big enough?
No, and it is not close to being able to. Radiator capability is an air-side calculation — core area, fin density, the mass of air going through it and the temperature of that air — and none of those are inputs here. This page works out how fast coolant has to move to carry a stated heat load at a stated temperature drop, which is a pump and plumbing question rather than a radiator one.