Water Pump Flow Rates

live4theking

Old Man with a Hat
Joined
Nov 24, 2014
Messages
12,419
Reaction score
8,469
Location
Rockland; Venango County Pennsylvania
In another post I was asking some water pump questions and Big_John linked to this thread, start reading at post 191.

I took some of the information from this thread and put it in a chart to make it easier for me to compare. Here it is.

Water Pump Flow.jpg
 
Thanks for posting. I would point out that chasing flow is not necessary a cure all. Higher flow rates are only effective if a high flow t-stat is also used. Robert Shaw brand were the best in the high flow t-stats. The factory did a pretty good job of mating pumps, t-stats, shrouds and fans to get sufficient flow with the appropriate air movement thru the radiator to get adequate cooling.

Dave
 
I posted about this on the C-body Dry Dock years ago. I had a pump like #2 on my 451 stroker. Always had overheat problems at idle and stop-and-go traffic. 3-core 22" rad, Flex-a-lite s/s flex fan. Added a fan shroud which did not help. With the fan shroud it took longer to get hot, but it still did get hot, and once I got moving it took longer to cool down.

Then I found the same water pump flow rate study that you did, and saw that my pump style was WORST performer of the bunch, especially at low RPM. So I switched to a Milodon high-flow pump with anti-cavitation plate and high-flow thermostat which they recommended. That completely solved my overheat problems.
 
Thanks for posting. I would point out that chasing flow is not necessary a cure all. Higher flow rates are only effective if a high flow t-stat is also used. Robert Shaw brand were the best in the high flow t-stats. The factory did a pretty good job of mating pumps, t-stats, shrouds and fans to get sufficient flow with the appropriate air movement thru the radiator to get adequate cooling.

Dave
...with a stock engine and cooling system. With recored or aftermarket radiators, rebuilt or upgraded engines and altered ignition timing and all bets are off.

My first high-flow t-stat was a Moroso because it was cheaper than the Mildon. That t-stat failed open and I wound-up replacing it with a Milodon. Search online for Moroso thermostat failure and you'll get lots of hits.
 
Any opinions on
Milodon 16260 Performance Aluminum High Volume Water Pump for Mopar Big Block
 
Agreed. Built engines and altered cooling system components are always going to be a crap shoot. More horsepower always means that more heat is generated as a by product. Re-cored radiators do not necessarily have the same number of tubes or the same tube diameter and that can effect cooling. Aluminum radiators have different flow characteristics which also can effect cooling. Often, one just needs to experiment to find what works.

Dave
 
Also remember there have been issues with some aftermarket aluminum water pump housings. The inlet connection on some of them are restricted, which has caused overheat problems for people. I am running an OEM cast iron housing, so I knew that wasn't part of my overheat problem.
 
I should be Ok as my engine has the factory cast housing. I'll make sure to get a high flow thermostat as well. Then I'll get my 22" rad rebuilt for max cooling. Hopefully all my over heating problems will go away!
 
Any opinions on
Milodon 16260 Performance Aluminum High Volume Water Pump for Mopar Big Block
It's a good one. Milodon's aluminum housing is good also.
There are 440 source ones that need some blending in the interior to flow better.
If I remember correctly the A/C water pump slows the coolant flow allowing it time to reject heat in the larger but hotter running radiator behind the condenser that is heating the air first. Ross is correct the less vanes in A/C pump. I do think in most cases it spins faster due to pulley sizing, also providing more rpm potential for the fan clutch to be able to get to if needed.
 
Mopar Performance aluminum water pump and housing.

20180817_141218.jpg
 
Last edited:
If I remember correctly the A/C water pump slows the coolant flow allowing it time to reject heat in the larger but hotter running radiator behind the condenser that is heating the air first. Ross is correct the less vanes in A/C pump. I do think in most cases it spins faster due to pulley sizing, also providing more rpm potential for the fan clutch to be able to get to if needed.

There has to be some sort of compromise between enough flow to move the fluid through the block and radiator and slow enough to get maximize heat transfer. It would be great to know what that rate of flow is and build around it.
 
There has to be some sort of compromise between enough flow to move the fluid through the block and radiator and slow enough to get maximize heat transfer. It would be great to know what that rate of flow is and build around it.
Lot of variables in there. Then throw in different materials like aluminum instead of brass radiator, fin count, fan CFM, fan clutch normal %, rich or lean mixture.
 
Lot of variables in there. Then throw in different materials like aluminum instead of brass radiator, fin count, fan CFM, fan clutch normal %, rich or lean mixture.
True, but I think the common factor is the cast iron block. After all, that is what you are trying to pull the heat out of. If you had a large enough capacity radiator to drop the coolant temperature 100 degrees, what would the ideal speed through the block be?

But, that may not be the limiting factor. As you say, there's all sorts of other materials too.

It's just some food for thought.... I'm sure that in each of the car companies, there's some squirrel little guy that has the answers, but has to fight against the stylists and bean counters for the ideal cooling solution.
 
It's just some food for thought.... I'm sure that in each of the car
I just put a radiator in my wife's Jeep GC. Aluminum with plastic tanks, about as wide and tall as A/C classic Mopar barely a inch thick front to back, sits behind a giant condenser(which now contains the trans cooler), a power steering cooler that would keep my 3800 stall converter cool in traffic. So there has to be some formula, because they are not over cooling the coolant in it with a 5.7 behind it.
 
One coolant pump innovation I would like to experiment with a little would be an electric pump with some temperature feedback controlling its behavior. I like the idea of being able to run both pump and fan when parked on hot days. At present this year, Mathilda still hasn't gone over ~210F at the hottest and never for any time over 3 minutes. I run a stock Gates pump in an aluminum housing at present with the Cold Case MOP753A radiator and a pusher electric fan in addition to an 18" 6 blade fan with a Hayden 2747 clutch. The pusher only gets run when I see temperatures over 200F in the coolant.

I like the idea of driving both pump and mech/clutched fan off a sturdy DC motor instead of the crank pulley. Have any of you experience with such a setup for normal driving?
 
I run an electric water pump on my 35 Plymouth 340 4 speed hot rod and an electric fan. the water pump runs at 1600 rpm whether you are at a stop light or cruising. as you suggested you can stop and leave both running. it cools everything down nicely.
 
I run an electric water pump on my 35 Plymouth 340 4 speed hot rod and an electric fan. the water pump runs at 1600 rpm whether you are at a stop light or cruising. as you suggested you can stop and leave both running. it cools everything down nicely.

That's one of the big advantages to such a system; post drive cooling. The other I seek is to CONTROL the flow rate electrically as well. Thus one may vary the angular speed of both pump and fan as needed. 1600 rpm is a good angular speed for a pump. You run a dc motor bracketed on the side and belted to the pump or w the pump motor integral to the pump as most electric pumps are designed and the electric fan separate w its own motor and control circuit?

How much power (current) does your cooling system draw? I suspect I would want a ~90-100 amp alternator to supply such on a regular basis.
 
True, but I think the common factor is the cast iron block. After all, that is what you are trying to pull the heat out of. If you had a large enough capacity radiator to drop the coolant temperature 100 degrees, what would the ideal speed through the block be?

But, that may not be the limiting factor. As you say, there's all sorts of other materials too.

It's just some food for thought.... I'm sure that in each of the car companies, there's some squirrel little guy that has the answers, but has to fight against the stylists and bean counters for the ideal cooling solution.

You all should do a little thermodynamic research to really optimize your numbers for cooling engines. I dug this gem up first:
conductionequation.png


where:

Q over t is the rate of heat transfer - the amount of heat transferred per second, measured in Joules per second, or Watts. k is the thermal conductivity of the material - for example, copper has a thermal conductivity of 390, but wool has a thermal conductivity of just 0.04. T1 is the temperature of one object, and T2 is the temperature of the other. Since it's a temperature difference, you can actually use Celsius or Kelvin, whichever is most convenient. And d is the thickness of the material we're interested in. [ Heat Transfer Through Conduction: Equation & Examples - Video & Lesson Transcript | Study.com ]

Calculate your Q based on the heat of enthalpy of gasoline/air in the cylinder for the initial heat, and the initial t would be how long it takes the power stroke to complete. Use the thickness of the cooling jacket for d. The delta T would be the temperature of the flamefront minus the desired temperature of the coolant, say, 373 Kelvin, the boiling point of water at one atmosphere pressure.

I then would solve for t, the time and look at the flow rate for the coolant in the neighborhood of the cylinder in this case, and then work with the capacity of the two hgeads for total volume of coolant, to puke out the desired capacity of the pump. Since my tiime is short for now, I can't do these sums but maybe later it could be FUN to work ths out, then puke it into a spreadsheet.
 
The actual flow rate is throttled by the thermostat. This both regulates flow to try to maintain a constant temp in the block and also provides a restriction which creates high pressure in the block, which are both good things. The only downside is that the water pump pushing against a restriction can lead to cavitation (bubbles) which impedes cooling. The anti-cavitation plate on the Milodon pump is supposed to reduce cavitation by reducing coolant shearing between the moving impeller blades and the fixed housing wall.

The argument of "too much flow" boils down to the idea where a condition could exist where the thermostat is wide open and coolant is circulating so quickly that it cannot extract enough heat from the engine before exiting to the radiator, leading to a runaway overheating condition. I have doubts that this actually happens in practice, or that the real problem is something else, such as scale buildup in the block causing hot spots and/or impeding heat transfer, scale buildup or blocked tubes in the radiator, or insufficient airflow across the radiator due to several possible reasons.

The rate of heat transfer is proportional to the difference in temperature. (This can be shown from @Gerald Morris formula above. All other variables held constant, the rate of heat transfer Q is proportional to the temperature difference T2-T1.) That means that the hotter the coolant is when it enters the radiator, the faster heat energy will be dissipated from it. In other words, hotter coolant will not need as much dwell time in the radiator to cool. This will, of course, be affected by airflow across the radiator.

Higher coolant flow rates will improve coolant circulation, which reduces hotspots in the engine. Higher flow rate also implies higher pressure in the block. This reduces localized coolant boiling, called "film boiling", on the surface of the cylinder walls. Film boiling is bad because the bubbles insulate the liquid coolant from contacting the cylinder walls, which can greatly reduce heat transfer. So even if the temp on the gauge does read a bit hotter, the engine may not be any more prone to pinging/detonation.
 
Back
Top