SL-C electric water pump questions

[Scott]

Rich, I bet you only wait long enough for the edge to cool My understanding is that the "flowing through the radiator too fast reduces heat transfer efficiency" argument is wrong. It's true that the longer the water is in the radiator, the more it cools. However, as is cools the delta T decreases and heat transfer efficiency therefore decreases. Assuming that you have a properly sized and designed radiator you'd rather flow faster so that the delta T is higher which results in higher efficiency. Increased flow also increases turbulence which is a good thing when you're trying to reject heat (my point in the prior post). There is certainly a point where too much flow is detrimental, but I think that manifests itself elsewhere (i.e., pump cavitation) before it effects heat transfer efficiency in the radiator.

A couple of more points on the Pierburg. If I recall correctly, the race SL-C never had to run the pump greater than 70% duty cycle. Also, there are lots of crap Chinese fakes out there so you need to be careful. The picture below shows how compact the pump is (ignore orientation) and what a legit box looks like.

 

[Howard Jones]

S2 is correct. "The water needs to stay in the radiator long enough to cool" or "the water flows through the radiator too fast" is a wife's tail. The physics simply don't support it. The water can compressed too much "too fast" and the result will be cavitation in the water pump. This is a result of pump speed being so high that the water looses laminar flow contact with the impeller elements. Think bubble forming behind the impeller blade and then collapsing causing the water to flow back into the void left by the collapsing bubble. This reverse flow is in the opposite direction than the impeller rotation. These continuing little shocks can add up to enough force to damage the impeller. In fact I have had to weld the steel impeller onto the pump shaft to prevent it from spinning on it in my Eldenbrock RPM performer water pump that I have used in my GT40. Other restrictions, tight turns, or roughness in the system piping does a very similar thing and causes loss of laminar flow in that region.

S2 is correct again when he says that the goal is to maintain laminar flow everywhere the water is in motion, But I contend that laminar flow is continued in the radiator as well. The pressure drop across the radiator causes the water to slow down dramatically because of the huge increase in cross section when the coolant leaves the piping and flows into the huge volume in the radiator. Remember higher flow speed increases the difficulty of maintaining laminar flow. So as the water slows down in the radiator it is relatively more contusive to efficient laminar flow.

This is why I contend that the more efficiency the complete system has, as that relates to reducing turbulence through it, the better.

Again: the more coolant volume that can be passed across the heat source and then passed across the heat exchanger in a given period of time will cool the most water. I.E. exchange the most heat. Every thing being discussed here is to do that task with the most efficiency. Pumps can only be made so big, radiators have to fit in the location designed for them, and the distance the coolant needs to transit is fixed. As well as more and more powerful engines producing more and more heat.

S2, that is a really nice looking piece. What do they cost?

 

[Neil]

Coolant flow through the system should be laminar flow except through the radiator, block and heads. Turbulent flow transfers more heat from the coolant to the air in the case of a radiator and transfers more heat from the coolant passage walls to the coolant in the case of the heads and block. Laminar flow in a heat exchanger is not as efficient as turbulent flow.

 

[Howard Jones]

Well that's two to one. I guess I need to do some more reading. Thanks guys. Did the reading and you guys are right. Thanks again.

https://www.deppmann.com/blog/monday-morning-minutes/minimum-velocity-heat-exchangers-part-2/

 

[JordonMusser]

I would be VERY surprised if you can cool the car if driven hard on a race track with a single electric pump. Two as howard mentioned will probably be required. I had planned to go that route, but was able to make a custom accessory drive to fit a big daddy mechanical pump from Meziere (bottom right). Supposedly this is what the baja trucks run. Motor should make 1500hp on kill, and 900hp in 'track' mode. Note this is for a different platform, but mid engine.

View this content on Instagram

View this content on Instagram

 

[Howard Jones]

Here's a forum that you can ask anything you want about Stewart pumps. I would think that any generic technical question would also be welcomed.

https://www.stewartcomponents.com/forum/

 

[Neil]

The Stewart website addresses a seldom- considered problem with electric fuel pumps- coolant pressure in the cylinder heads. One of the functions of a water pump is to circulate coolant through the engine to the radiator and return it to the engine. The other function is to create additional pressure in the coolant passages in the cylinder heads, over & above the pressure generated by the coolant temperature. The additional pressure minimizes nucleate boiling around the hot spots in the heads and maintains good heat transfer from the heads to the coolant. A high flow/high pressure pump requires quite a few horsepower to generate that flow & pressure.

Since 1 horsepower requires 746 watts to a 100% efficient electric pump, and @ 13.7 volts, this draws 54.5 amps... per horsepower! Electric water pumps simply do not have enough power to cool a high horsepower engine efficiently.

Take what I've said with a grain of salt (no pun intended)- in applications where an engine is producing lots of horsepower for an extended period of time, this becomes a serious consideration. These are not street or road racing applications where the output power is either low or only high for limited periods of time. The critical applications are for high power outputs where the throttle is fully open for for many miles, such as boat racing, NASCAR, dynamometer testing, or land speed racing.

 

[Joel K]

The Stewart website addresses a seldom- considered problem with electric fuel pumps- coolant pressure in the cylinder heads. One of the functions of a water pump is to circulate coolant through the engine to the radiator and return it to the engine. The other function is to create additional pressure in the coolant passages in the cylinder heads, over & above the pressure generated by the coolant temperature. The additional pressure minimizes nucleate boiling around the hot spots in the heads and maintains good heat transfer from the heads to the coolant. A high flow/high pressure pump requires quite a few horsepower to generate that flow & pressure.

Since 1 horsepower requires 746 watts to a 100% efficient electric pump, and @ 13.7 volts, this draws 54.5 amps... per horsepower! Electric water pumps simply do not have enough power to cool a high horsepower engine efficiently.

Take what I've said with a grain of salt (no pun intended)- in applications where an engine is producing lots of horsepower for an extended period of time, this becomes a serious consideration. These are not street or road racing applications where the output power is either low or only high for limited periods of time. The critical applications are for high power outputs where the throttle is fully open for for many miles, such as boat racing, NASCAR, dynamometer testing, or land speed racing.

Neil, you continue to amaze me with all the knowledge you have across so many topics. I really enjoy reading your posts.

 

[Neil]

Thank you, Joel. I guess in the last 80 years I've learned a few things so I like to pass that knowledge along to others when I can.

 

[Howard Jones]

Neil. you got me thinking again so I did a little checking. DC says that the DC150 I am using draws as little as 2 amps and a maximum of 10 amps. I have protected that circuit with 15 amp ckt breaker. That would infer a power consumption of some where in the neighborhood of 24w-120w or .03hp - .16hp @ 12vdc.

I agree with you. When the rpm band of the engine is very narrow such as NASCAR speedway ovals, boats, landspeed attemps once in top gear and at fullpower a belt driven mechanical pump can be used at the most efficient point of it's power consumption/ volume pumped curve
AND kept out of cavitation. For our cars Street/ road racing however the ability to run the electric pump at a constant speed is most beneficial.

Jordon has got me reading again and DC does recommend that in all out racing applications two pumps can be used in series with a controller. One being controlled by the controller and the other driven off the fan circuit via a relay. If a relay is used in this manner I would recommend finding one that is overrated for both current and duty cycle due to the constant on off cycling of the second pump. Should I ever see the need it would be very easy to add a second pump up in the nose.

That two pump configuration is exactly the one recommended by Stewart when I talked to them. Two pumps at the two opposite ends of the system. At the engine cool side return and the radiator hot side input. They also said to use the smaller of the two electric pumps if I did it that way and run them all on all the time Then control water temp with a thermostat or a tapered restrictor.