Dane Posted January 31, 2008 Share Posted January 31, 2008 Lately I've been wondering how much the pumping efficiency of the cooling system could be improved for the sake of reducing parasitic drag. I know that many engine modifications are made to improve the pumping efficiency of the intake and exhaust system...but considering that water has such a greater density and viscosity than air, I think it would be a good area to look into. Some little things I thought of: -Knife edge the ends of all pipes where cooling hoses attach to maintain laminar flow at pipe/hose junction. -Smooth casting bumps within water pump housing and on impeller to improve laminar flow. (I know casting bumps within water jacket increase surface area and turbulence for better cooling) -Install a higher flowing thermostat or somehow modify thermostat to increase flow. -Replace any T-shaped pipe bends with smoother radiused bends. -Install a higher flowing radiator. I also had another more radical idea that I got from looking at pictures of the OS Giken DOHC head in the How to Modify book. It has sort of a coolant plenum on the exhaust side with coolant runners going over each cylinder...with the thermostat on the end of the plenum. Maybe holes could be drilled and tapped on the spark plug side of the head (in the interest of available space) for barbed fittings...which you could then run to a similar coolant plenum via short heater hoses and fab up a flange for a thermostat housing and an external bypass line. This could help cylinder to cylinder temps in the head...but I'm not sure what effects it would have on pumping efficiency. BTW...the reason I like the idea of using barbed fittings and short heater hoses to the coolant plenum is that the system could be removed and plugged if necessary, and it would allow more tolerance in alignment between the holes drilled in the head and the holes drilled in the plenum (hoses are flexible). The shortest possible hoses would be used to minimize flexing/bending in the hoses due to the weight of the plenum. Anybody ever seen any gains from improved pumping efficiency or using a head cooling set-up similar to the OS Giken head? Quote Link to comment Share on other sites More sharing options...
Tony D Posted January 31, 2008 Share Posted January 31, 2008 Most of the serious horsepower turbo L engines run auxillary coolant lines out the top of the cylinder head on the spark plug side. And electric pump could save drag by only running at the speed required for cooling, speed variable by temperature. And of course, reverse flowing the system could pay benefits, but isn't as easy as on a SBC. The Nissan LD28 water pump is of higher capacity as well. Check the thread on that. Little known tidbit that I picked up hanging around JGK and TimZ... Quote Link to comment Share on other sites More sharing options...
ron260z Posted January 31, 2008 Share Posted January 31, 2008 if you want your car to run colder you can do two things. take the thermostate out. or drill holes in it untill you achive the temperature you want. Quote Link to comment Share on other sites More sharing options...
Tony D Posted January 31, 2008 Share Posted January 31, 2008 argh... Quote Link to comment Share on other sites More sharing options...
Dane Posted January 31, 2008 Author Share Posted January 31, 2008 Tony, I have read about the LD water pump and the larger impeller...that is very interesting. I'm kind of following the school of thought of reducing the pressure drop across the cooling system required for a given flow...as opposed to increasing total water pump output. I wasn't aware that electric water pumps used variable speed control. I was under the assumption that they worked similar to electric radiator fans with a cut-off temperature when they just turn on or off. Of course that works for a cooling fan...it can have a low duty cycle since it only kicks in when airflow through the radiator isn't sufficient for cooling (like idling at a stoplight). But it wouldn't make sense for the water pump to just shut down altogether...because then you would have these up and down spikes in engine temperature as the coolant in the block goes from stationary to moving to stationary again. Whatever...I'm just thinking aloud. Anyways, what would be used to control the water pump speed...a temperature based variable resistor or something like that? Quote Link to comment Share on other sites More sharing options...
Chemicalblue Posted January 31, 2008 Share Posted January 31, 2008 one of these would work http://www.spal-usa.com/FAN-PWM2.html Quote Link to comment Share on other sites More sharing options...
510six Posted January 31, 2008 Share Posted January 31, 2008 http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump'>http://www.emp-corp.com/media/SAEPapers/2005-01-1380.pdfhttp://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump Interesting reading, although I wouldn`t use a Davies Craig electric water pump as a primary pump one is used as a "helper" pump in my current setup.It is controlled by the AEM EMS and activates with the secondary fan control.Works great, the car doesn`t heat up sitting in traffic in the summer like it used to with the stock pump. An added side benifit is that the pump can be run as a cooldown pump between runs at the drag strip , with the electric system using a jump starter box it doesn`t drain the main battery and with a water cooled turbo bearing the car cools pretty quickly and is ready for the next run. Quote Link to comment Share on other sites More sharing options...
Mr.INSANE Posted January 31, 2008 Share Posted January 31, 2008 http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump'>http://www.emp-corp.com/media/SAEPapers/2005-01-1380.pdfhttp://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump Interesting reading, although I wouldn`t use a Davies Craig electric water pump as a primary pump one is used as a "helper" pump in my current setup.It is controlled by the AEM EMS and activates with the secondary fan control.Works great, the car doesn`t heat up sitting in traffic in the summer like it used to with the stock pump. An added side benifit is that the pump can be run as a cooldown pump between runs at the drag strip , with the electric system using a jump starter box it doesn`t drain the main battery and with a water cooled turbo bearing the car cools pretty quickly and is ready for the next run. How does the water get past the belt driven pumps impeller? Quote Link to comment Share on other sites More sharing options...
Dane Posted January 31, 2008 Author Share Posted January 31, 2008 How about underdrive water pump pulleys? I swear I've seen a photo of an L4 engine with a larger than stock water pump pulley. Any kind of parts interchangability with other cars or anything aftermarket? (I know that some aftermarket crank dampers have a smaller pulley diameter...I'm thinking specifically about the water pump pulley.) Quote Link to comment Share on other sites More sharing options...
ozconnection Posted January 31, 2008 Share Posted January 31, 2008 There is a very good book that attempts to give its readers insight into what is really going on in an engines cooling system. The book is "Four Stroke Performance Tuning" by A. Graham Bell. Chapter nine, Lubrication and Cooling discusses the do's and dont's of cooling systems. It's a bloody good read and well worth buying or borrowing from your local library to do some research. 'ARGH' is pretty much what the author had to say about removing thermostats too, don't do it!!!! It does talk about things like fans and shrouds, radiator fin counts, materials used ie aluminium pump types and mods for increasing flow etc. I have adapted many of the ideas in the book to my L28 which runs an auto trans and air con in the hot Aussie summer and it hasn't let me down yet!! Nothing too radical, just basic and solid engineering. Good luck with your cooling system, Cheers:smile: Quote Link to comment Share on other sites More sharing options...
510six Posted February 1, 2008 Share Posted February 1, 2008 How does the water get past the belt driven pumps impeller? http://www.daviescraig.com.au/main/display.asp?pid=46 http://www.emp-corp.com/media/SAEPapers/2005-01-1380.pdf http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump One thing that I have notices is after ceramic coating the pistons the in/ex ports and manifolds, the motor "sheds" heat much faster than it did previous to the ceramic coatings. Quote Link to comment Share on other sites More sharing options...
heavy85 Posted February 1, 2008 Share Posted February 1, 2008 I assume the point here is to reduce parasitic loss by reducing the pump pressure? If so impeller pumps are not fixed displacement and vary nonlinear with output pressure. This means for a given speed the flow varies as a function of pressure ... and in a non-linear fashion (this is important so remember this). Power is flow*pressure. Now remember that non-linear aspect I keep saying? For example if you cut the pressure in half you dont get double flow (constant power) you may get three or four times the flow. So the resulting power will be 1/2pressure*4flow = double the power you started off with. So reducing pressure is not necessarily the answer to saving power with these type of pumps. You really need to properly size the pump to meet the system demands. OK so real world example. Electrical motor driven fuel transfer pumps for relatively large engines (think commercial marine stuff). They have to calculate the pump discharge line losses and to make sure the pressure is HIGH enough to not burn up the electric motor. Kind of not intuitive but that's how it works. Cameron Quote Link to comment Share on other sites More sharing options...
Dane Posted February 1, 2008 Author Share Posted February 1, 2008 heavy85, Awesome...you're speaking my language. Yes...I understand that the pump isn't positive displacement...it would probably be appropriate to model it with a map as a function of pressure and flow (kind of like a turbo compressor map). I know what you mean by Power = Pressure * Flow, with the pressure actually being the pressure difference between the outlet and inlets of the pump. We're on the same page so far. Here's an equation for fluid flow through a pipe that I referenced from a college physics textbook: Poiseuille's Law: Change in Volume/Change in Time = (pi/(8*Viscosity))*(Change in Pressure/Change in Length)*(Radius^4) Plugging that into the "Power = Pressure * Flow" equation: Power = (Change in Pressure)*(Change in Volume/Change in Time) Power = (pi/(8*Viscosity))*(Change in Pressure^2/Change in Length)*(Radius^4) But with that in mind...I don't see how you can come to the conclusion that reducing the pressure difference doesn't reduce the power required. I mean...plug in a smaller number for "Change in Pressure" on the right side of the equation, and you get a smaller number for "Power" on the left side of the equation. As far as the electric pumps burning up with too light of a load...would I be correct to assume that these types of pumps are designed to work against a load? Can we be sure that this applies to an L6 water pump as well? The only way I can think of that reducing pressure would effect the operation of the water pump is that if it took the impeller out of its operating efficiency range (kind of like a turbo compressor out of its efficiency range)...so that the impeller generated more heat and moved less water. I just can't see how reducing the flow restriction through the cooling system...therefore reducing the pressure difference for that flow...wouldn't decrease the power required to force that water through the system. Quote Link to comment Share on other sites More sharing options...
MJLamberson Posted February 1, 2008 Share Posted February 1, 2008 if you want your car to run colder you can do two things. take the thermostate out. or drill holes in it untill you achive the temperature you want. thermostats like most things serve a purpose, true an engine shouldnt run t hot, but it does NEED to heat up, without the thermostat this would take to long. Quote Link to comment Share on other sites More sharing options...
Tony D Posted February 1, 2008 Share Posted February 1, 2008 A water pump IS a turbo! It has the same sort of mapping. The premise of reducing line losses is a red herring IMO. The engine needs turbulent flow to get a good heat exchange, and the 'improvements' in reduced parasitic drag by streamlining hose connectors and the like will be purely theoretical. An Engineering Problem to eat up time, and see a theoretical savings. But in the real world the effective available horsepower will be nil. IMO, the more efficient method would be to run an appropriately sized electric pump which can idle down during low-load periods and which has a battery to smooth out impact-loading of high load times so you are not drawing maximum amps during periods coinciding with maximum heat production. Decreasing your pressure drop across the engine by increasing the flow will result in decreased thermal transfer time (residence time) as well as possibly lowering the operating pressure---which will have detrimental effects in the formation of steam pockets and impingement-induced cavitation damage on castings, impellers, etc... It's a much more complex system than simply streamlinging this or decreasing that. Change one thing, and you change a LOT of other things. There was a posting recently about 'cavitation' which had a 1950's era engineering filmstrip linked within it---it is a very good 32 minutes to spend watching. You learn a lot about increasing flowrates and decreasing pressures and what happens within a fluid when you do those things. Quote Link to comment Share on other sites More sharing options...
heavy85 Posted February 1, 2008 Share Posted February 1, 2008 But with that in mind...I don't see how you can come to the conclusion that reducing the pressure difference doesn't reduce the power required. I mean...plug in a smaller number for "Change in Pressure" on the right side of the equation, and you get a smaller number for "Power" on the left side of the equation. As far as the electric pumps burning up with too light of a load...would I be correct to assume that these types of pumps are designed to work against a load? Can we be sure that this applies to an L6 water pump as well? Remember the thing I kept saying? Flow is a function of pressure in a NON-LINEAR fashion. In the example I gave it was a square function but in reality its probably somewhere between depending on the pump design. Keeping on the square function if you cut the flow in half the flow will go up by 4 times for a given speed. Power = 1/2Pressure*4Flow = 2*The original power. So by lowering the pressure you are disproportionately increasing the flow and therefore increasing the power. The electric pump is then running at a HIGHER load (power). The point is you can reduce the pressure to save power but ONLY if you resize the pump to operate in that lower pressure range. Using the same pump could hurt you. Cameron Quote Link to comment Share on other sites More sharing options...
Mr.INSANE Posted February 2, 2008 Share Posted February 2, 2008 http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump'>http://www.emp-corp.com/media/SAEPapers/2005-01-1380.pdfhttp://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump http://www.stewartcomponents.net/Merchant2/merchant.mvc?Screen=CTGY&Category_Code=ElectPump Interesting reading, although I wouldn`t use a Davies Craig electric water pump as a primary pump one is used as a "helper" pump in my current setup.It is controlled by the AEM EMS and activates with the secondary fan control.Works great, the car doesn`t heat up sitting in traffic in the summer like it used to with the stock pump. An added side benifit is that the pump can be run as a cooldown pump between runs at the drag strip , with the electric system using a jump starter box it doesn`t drain the main battery and with a water cooled turbo bearing the car cools pretty quickly and is ready for the next run. What I meant was how do you expect to cool the car after its turned off with mechanical pump not moving. You mentioned using a Helper electric pump. Unless the mechanical pump is replaced with the electric I don"t see how running an extra electric could help when the car is off Quote Link to comment Share on other sites More sharing options...
heavy85 Posted February 2, 2008 Share Posted February 2, 2008 What I meant was how do you expect to cool the car after its turned off with mechanical pump not moving. You mentioned using a Helper electric pump. Unless the mechanical pump is replaced with the electric I don"t see how running an extra electric could help when the car is off These pumps have very poor efficiency. It's not like there is a seal or something so the water would just flow through the mechanical pump. Maybe not as well as with the pump spinning but at least quite a bit would get through. Take a look at an impeller water pump (or turbo) and you will see this pretty easily. Cameron Quote Link to comment Share on other sites More sharing options...
Tony D Posted February 3, 2008 Share Posted February 3, 2008 The impeller blades in the factory pump are far enough spaced to allow plenty of flow with a pump in the lower radiator hose like a Davies-Craig. It will simply bypass the vanes, there is not any interference fit, and it's not positive displacement, it can flow backwards easy enough. Again, remember that you DON'T want to reduce the pressure at which the pump operates! The higher pressures induced by the pump at higher rpms DIRECTLY affect the incidence of spot-boiling on the surfaces of the engine that is the hottest: upper cylinder walls and head. Lower the operating pressure in the engine block (at the pump outlet), and you will start to see cylinder wall failures due to steam-impingement cavitation as it eats the cylinder walls and makes them like swiss-cheese perpindicular to the piston pin bores (ask me how I know this....) You can lower the pressure, but you will correspondingly have to RAISE the blanket pressure on the top of the water (giving more pump suction head pressure (NPSH) to the water pump. More pressure at the inlet will result in the same pressure at the outlet with less work put in...but you will be waiting a time for the inlet pressure to come up due to expansion. unless you want to run an inert gas blanket from an onboard N2 Bottle... This is theoretical. The HP you will 'save' will be at the expense of metal parts experiencing cavitation and steam pocket formation at lower loads. For something like a hit-or-miss farm engine, it's fine. But when you start talking about high horsepower engines under load, it's not a good idea IMO. Quote Link to comment Share on other sites More sharing options...
ozconnection Posted February 3, 2008 Share Posted February 3, 2008 Does anyone else feel like me on this thread? I thought I knew a little bit about cars but it seems that I have been fooling myself for far tooooooo long. (I must have missed something....buggar!) Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.