How quickly does coolant circulate through the system?

[CharlesInCharge]

I'm curious if there is a general consensus on how quickly a typical car or truck will circulate coolant thru the engine, radiator, and overflow tank when at operating temps? Is it a constant flow like at a faucet speed running thru? Or some other rate? When warm, would the coolant tend to circulate thru entirely in about 1 minute, or 5 minutes, or 10 minutes? A liter per minute, so a 10L system might circulate all of it in 10 minutes?

This isn't necessarily vehicle specific, but since it's easier to use a Mityvac to extract 1/3 of the coolant thru the radiator and overflow tank, than getting under and fussing with draining the block or valves, I'm just curious if doing partial coolant replacements with a Mityvac makes sense if the coolant is circulating and mixing at a relatively fast pace.

 

[Rand]

Obviously it would vary highly with rpm and engine.. but
At idle I would think every few mins.. at highway speed or couple thousand rpm once a min?

but just WAG.

 

[ekrampitzjr]

On most cars the water pump rotation speed will vary with engine speed because of direct drive to the pulley. Engine block size and design make a difference in restricting flow and amount of coolant in the block.

This is just a rough guess based on the rate of flow I've seen looking into radiators with the cap off while running. If the radiator is not plugged, I'd figure 30 seconds to 1 minute for complete circulation between block and radiator. The thermostat is a bottleneck to flow, but most cooling systems work best with some back pressure, which the thermostat is designed to provide.

If I'm wrong about my time guess, please correct me.

 

[ekrampitzjr]

Found what I was looking for online. An estimate from one source says most car cooling systems flow ~4–9 gallons per minute. If you assume the cooling system holds about 4 gallons, which is typical for a large V-6 or smaller V-8 in a stock street application, then my guess of 30 seconds–1 minute for circulation is close.

 

[Rand]

Found what I was looking for online. An estimate from one source says most car cooling systems flow ~4–9 gallons per minute. If you assume the cooling system holds about 4 gallons, which is typical for a large V-6 or smaller V-8 in a stock street application, then my guess of 30 seconds–1 minute for circulation is close.

right but what is it at idle. I think 4gpm sounds quite high for ~~600rpm. for highway driving or 1500-2500rpm probably spot on actual engine dependent.

 

[mightymousetech]

The flow rate is much higher than mentioned. A BMW electric pump is about 40 gal/min.

 

[ekrampitzjr]

The flow rate is much higher than mentioned. A BMW electric pump is about 40 gal/min.

Wow. BMW must have a reason. I might well be wrong on newer and more expensive cars. Interesting that BMW maintains the same rate of flow at idle and at, say, 6000 rpm using an electric pump.

The engines I've seen running at temperature with the radiator cap off, which were Japanese and American vehicles some years ago, could not have been flowing anything like 40 gallons/minute. The online estimate of 4–9 seems accurate for those engines. But design varies.

It's harder to see coolant flowing in today's radiators because of the location of the filler neck or using pressurized overflow tanks where the cap now is. When the radiator cap was right above where the upper hose attached to the radiator, watching flow was easy.

 

[Timmastertech]

Completely depends on the vehicle. Ford 6.0 diesel moves over 100 gallons per minute at idle.

 

[Passport1]

I have used these water pumps for years. They flow an adequate amount of coolant, what ever that is, and last eight to ten years at least until they begin weeping. A quality water pump is a mission critical part. No, not sponsored by anybody but my S.O.

 

[mightymousetech]

Wow. BMW must have a reason. I might well be wrong on newer and more expensive cars. Interesting that BMW maintains the same rate of flow at idle and at, say, 6000 rpm using an electric pump.

The engines I've seen running at temperature with the radiator cap off, which were Japanese and American vehicles some years ago, could not have been flowing anything like 40 gallons/minute. The online estimate of 4–9 seems accurate for those engines. But design varies.

It's harder to see coolant flowing in today's radiators because of the location of the filler neck or using pressurized overflow tanks where the cap now is. When the radiator cap was right above where the upper hose attached to the radiator, watching flow was easy.

Well, one benefit of an electric pump is that it is not always running. Often does not run at all if the engine is cold and the heat is turned off. The speed is also variable depending on requirements. But it can be at full speed, even at idle.

 

[ctechbob]

Completely depends on the vehicle. Ford 6.0 diesel moves over 100 gallons per minute at idle.

That seems awfully high.

 

[Timmastertech]

That seems awfully high.

That spec comes from the factory training information from the new product intro info on those engines and was backed up later in other classes. Its an absolutely massive cooling system

 

[meep]

Impressive. 100gpm is huge. Also a big energy pull. granted, that engine is designed to work.

I was feeling pretty good for 1 full circulation at idle per minute for a passenger car, though we miss some details here. The heads will circulate more frequently than the radiator loop, so eyeballing the radiator flow alone won’t be the whole story. If the radiator sees 4gpm, the head look could easily see 8 during the same timeframe. That would make 12. I’m pretty sure that pump at 650 rpm isn’t moving 12 gpm, but easily at 1500.

 

[Jetronic]

That seems awfully high.

Remember that's with very low pressure buildup by thye pump. As soon as resistance is met the flow rate plummets.

Coolant pumps have the same basic design as one of these:

and similar flow rate/ decline aswell:

With the added variable of the pump having a variable rpm of course, but that's what flow rates will look like at full rpm, with a closed thermostat to the left to fully open at the right.

 

[Skippy722]

It’s going to vary highly depending on RPM, the particular water pump, the rest of the cooling system, and the engine itself. From LSMag: “According to Meziere, most mechanical pumps tested in V8 applications flow approximately 8 GPM at idle, and around 70 GPM at 6,500 rpm.”

You also need the thermostat to be open to get flow through the radiator, otherwise it’s just circulating coolant through the block and possibly heater core.

Wow. BMW must have a reason. I might well be wrong on newer and more expensive cars. Interesting that BMW maintains the same rate of flow at idle and at, say, 6000 rpm using an electric pump.

The engines I've seen running at temperature with the radiator cap off, which were Japanese and American vehicles some years ago, could not have been flowing anything like 40 gallons/minute. The online estimate of 4–9 seems accurate for those engines. But design varies.

It's harder to see coolant flowing in today's radiators because of the location of the filler neck or using pressurized overflow tanks where the cap now is. When the radiator cap was right above where the upper hose attached to the radiator, watching flow was easy.

As Mightymousetech said, it gives them flexibility you can’t get with a mechanical pump. If you’re say pushing the car very hard on the track and then say do some low speed driving to cool it back down or let it sit and idle, a mechanically driven pump might see coolant temps rise for a bit as it can’t get that hot coolant to the radiator fast enough. An electric pump will allow them to keep the coolant flow up without elevated RPM.

 

[nthach]

I think anywhere between 15 seconds to a minute is a fair estimate for all engines. Water pump GPM, thermostat and how complex the cooling loop is are all considerations.