Leccy water pump question
Discussion
Hi all,
I'm just looking at sizing up the water pump setup for my race engine (4.7 TVR AJPV8). I'm hoping to achieve somewhere around 500bhp, which is above the recommended figure for a Davies Craig EWP115. I have the option of a couple of 110L/min DC pumps which I could run in parallel to achieve a couple of hundred L/min at a pressure of less than 0.5bar. Is this going to be enough? Is it way too much? How do I work out the requirement? Obviously this is max flowrate we're talking about. The nominal flowrate will be controlled by voltage drop via the ECU.
I'm just looking at sizing up the water pump setup for my race engine (4.7 TVR AJPV8). I'm hoping to achieve somewhere around 500bhp, which is above the recommended figure for a Davies Craig EWP115. I have the option of a couple of 110L/min DC pumps which I could run in parallel to achieve a couple of hundred L/min at a pressure of less than 0.5bar. Is this going to be enough? Is it way too much? How do I work out the requirement? Obviously this is max flowrate we're talking about. The nominal flowrate will be controlled by voltage drop via the ECU.
Conventional cooling needs on smaller engines with mechanically driven water pumps varybetween 2.0 to 2.6 L/min/kW.
(I borrowed that from an SAE paper)
500bhp is 370Kw so based on that you need a flow rate of around 850 litres per minute. I note the words "on a smaller engine" but not sure if a larger engine needs proportionally more or less cooling but either way the biggest Davies Craig pump is 115 Litres per minute so it's going to be very undersized.
I suppose the difference between a smaller engine and a larger one is the thermal effciency and larger ones generally are less efficient. Personally I would (have) stick with the original pumps.
Don't forget the AJP has 2 water pumps too. I have heard a lot of tuners state that "electric water pumps are totally useless" and others swear by them. Going strictly by the maths I'm in the former camp unless you are using a big EWP on a quite small engine. Certainly on a race engine which will be near it's peak output most of the time an EWP may be a bad idea.
(I borrowed that from an SAE paper)
500bhp is 370Kw so based on that you need a flow rate of around 850 litres per minute. I note the words "on a smaller engine" but not sure if a larger engine needs proportionally more or less cooling but either way the biggest Davies Craig pump is 115 Litres per minute so it's going to be very undersized.
I suppose the difference between a smaller engine and a larger one is the thermal effciency and larger ones generally are less efficient. Personally I would (have) stick with the original pumps.
Don't forget the AJP has 2 water pumps too. I have heard a lot of tuners state that "electric water pumps are totally useless" and others swear by them. Going strictly by the maths I'm in the former camp unless you are using a big EWP on a quite small engine. Certainly on a race engine which will be near it's peak output most of the time an EWP may be a bad idea.
I have a couple of reasons for going with the EWP setup, firstly I do not have the mechanical ones as the car was running an EWP and belt driven Pace oil pump. Secondly it enables be to bleed the system without running the motor, and allows me to leave the water circulating after the shutdown to better cool the engine after a race. I'm fairly sure the engine won't need 850L/min as the Davies Craig website describes its EWP115 as being 'suitible for V8 Engines or engines over 400bhp'. So I'm guessing that the requirement is somewhere slightly north of 115L/min but not hugely so. I've also read somewhere that there is a peak flowrate through an engine at which maximum heat transfer can take place. Above this point you're just losing power trying to push the water faster and thats what a lot of mechanical pumps do, hence why EWPs ofter give a slight performance gain at high engine speeds.
Unfortunately I don't know what this flow rate is!
Unfortunately I don't know what this flow rate is!
Edited by 450Nick on Thursday 28th October 15:14
Stewart Warner make some decent looking pumps. Much better than that plastic DC crap. On a recent visit to Prodrive, seen one of their pumps fitted to a LeMans engine. I'd take that as a very good sign.
http://www.stewartcomponents.com/
Although still hard to beat a mechanical setup
http://www.stewartcomponents.com/
Although still hard to beat a mechanical setup
Edited by stevieturbo on Thursday 28th October 17:35
Typical l/min per kw for a "performance" engine designed with "precision" cooled heads (i.e. the flow is actually targeting to have a high velocity where it is needed) is about 1 l/min per kw.
The craig david ;-) water pumps are completely useless** unfortunately, as normally you have approx 1 bar pressure head across the head gasket alone! (at which figure the EWP flows precisely zero l/min !!) (fail to compare various pump flows with at least 1bar head at your peril!!)
The AMR/prodrive V12's do indeed run with the Stewart components pumps, which are ok (2x wattage of EWP's!) and the only other decent pump is the Siemens OEM elec pump fitted to several std cars (BMW /audi etc)but it is er, reasuringly expensive to buy and requires CAN to control it.
Heat transfer from head to coolant is driven by a few factors:
1) heat transfer co-efficient between alluminum and coolant (use min glycol % as it has less specific heat capacity than pure water!!) Otherwise pretty much fixed (you have an ally head, and you need to put water in the engine, so unless your gonna go for a liquid sodium system thats your lot!!.
2) velocity across the heat transfer boundary - this one is very important, as the total heat transfered is pretty much proportional to the velocity gradient. Faster = better
3) Delta T across the heat transfer boundary - again, important as the effectively "drives" the heat flow. Higher delta T = higher heat transfer.
So if you look at what happens with a "Low flow" system, where the velocity is low and the net result is that the delta T must increase to push the heat across the poor conduction boundary. Now, if the delta T must go up, then the temperature of the metal must go up. So what effectively happens is the material of the cylinder head increases in temp, which is bad (more chance of detonation, less volumetric efficiency because of air charge upheat in the runners, and less mechanical strength = more deflection under load etc)
Now unless you are measuring cylinder head metal temps in critical areas (which is how the oem etc arrive at the required water pump performance in the first place) you will never know that your system is "non optimal" unless you get so bad you have a mechanical failure (cracking of the inter-valve bridge area typically)
The only saving grace, and one which will "save" many a poor cooling system is an effect called nucleat boiling, where very localised boiling occurs of the coolant, and luckily boiling is capable of massive cooling (check out the latent heat of steam!!!) so you get to a point where the system sudenly has a massive gain in localised cooling, which prevents metal temps going too high. Some cooling system are now designed to operate in this zone all the time
The craig david ;-) water pumps are completely useless** unfortunately, as normally you have approx 1 bar pressure head across the head gasket alone! (at which figure the EWP flows precisely zero l/min !!) (fail to compare various pump flows with at least 1bar head at your peril!!)
The AMR/prodrive V12's do indeed run with the Stewart components pumps, which are ok (2x wattage of EWP's!) and the only other decent pump is the Siemens OEM elec pump fitted to several std cars (BMW /audi etc)but it is er, reasuringly expensive to buy and requires CAN to control it.
- I should note that "useless" does not mean that the car won't "cool", many people fit EWP's and say "it cools fine, the temp gauge never gets too hot. Unfortunately the temperature of the water has little bearing on the actually perforance of the cooling system, and most critically the cylinder head metal temps that result.
Heat transfer from head to coolant is driven by a few factors:
1) heat transfer co-efficient between alluminum and coolant (use min glycol % as it has less specific heat capacity than pure water!!) Otherwise pretty much fixed (you have an ally head, and you need to put water in the engine, so unless your gonna go for a liquid sodium system thats your lot!!.
2) velocity across the heat transfer boundary - this one is very important, as the total heat transfered is pretty much proportional to the velocity gradient. Faster = better
3) Delta T across the heat transfer boundary - again, important as the effectively "drives" the heat flow. Higher delta T = higher heat transfer.
So if you look at what happens with a "Low flow" system, where the velocity is low and the net result is that the delta T must increase to push the heat across the poor conduction boundary. Now, if the delta T must go up, then the temperature of the metal must go up. So what effectively happens is the material of the cylinder head increases in temp, which is bad (more chance of detonation, less volumetric efficiency because of air charge upheat in the runners, and less mechanical strength = more deflection under load etc)
Now unless you are measuring cylinder head metal temps in critical areas (which is how the oem etc arrive at the required water pump performance in the first place) you will never know that your system is "non optimal" unless you get so bad you have a mechanical failure (cracking of the inter-valve bridge area typically)
The only saving grace, and one which will "save" many a poor cooling system is an effect called nucleat boiling, where very localised boiling occurs of the coolant, and luckily boiling is capable of massive cooling (check out the latent heat of steam!!!) so you get to a point where the system sudenly has a massive gain in localised cooling, which prevents metal temps going too high. Some cooling system are now designed to operate in this zone all the time
Personally I would stick with a tried and proven solution on a racing engine than try to engineer / bodge a system few people know how to do well.
Just remember, to do this properly this is more than just sizing a pump or two, you also have to work on the alternator and battery, because you want to not drive the alternator on acceleration to get the engine power pushing the car forward, and then drive the alternator on decelleration, putting enough energy into the batteries to drive the pump on acceleration again. If you are sizing the alternator to drive the pumps all the time that is a pretty inefficient way to drive them - a belt is better. So we are back to square one.
As Max Torque says ethylene glycol will reduce heat transfer but it also helps stop the engine from freezing, not that big an issue if you are laying the car up over winter with no coolant in it, but you need to make sure that you use a reasonable amount of coolant to protect the engine from corrosion. I can't emphasise this strongly enough - without corrosion protection water will find it's way through high heat areas like the exhaust ports in no time at all.
I would try to go for more flow than less in a cooling system, while nucleate boiling is good, film boiling is the next step and it is the opposite of good. Avoid like the plague.
Just remember, to do this properly this is more than just sizing a pump or two, you also have to work on the alternator and battery, because you want to not drive the alternator on acceleration to get the engine power pushing the car forward, and then drive the alternator on decelleration, putting enough energy into the batteries to drive the pump on acceleration again. If you are sizing the alternator to drive the pumps all the time that is a pretty inefficient way to drive them - a belt is better. So we are back to square one.
As Max Torque says ethylene glycol will reduce heat transfer but it also helps stop the engine from freezing, not that big an issue if you are laying the car up over winter with no coolant in it, but you need to make sure that you use a reasonable amount of coolant to protect the engine from corrosion. I can't emphasise this strongly enough - without corrosion protection water will find it's way through high heat areas like the exhaust ports in no time at all.
I would try to go for more flow than less in a cooling system, while nucleate boiling is good, film boiling is the next step and it is the opposite of good. Avoid like the plague.
Thanks for the input guys, proper backed up opinions are always good to hear! So it sounds like I'd be better off going for a standard mechanical one? as I understand it, the road v8s (I have a road block) ran with one water pump instead of two in the race variant. Would I be better of sticking with one mechanical one, a larger belt driven one, or a couple of bigger ewps? Or run with the standard mech one and have the dc 80l/min one I already have in there as a booster/cooling/bleeding pump?
http://www.daviescraig.com.au/Electric_Water_Pumps...
Questions 21 & 22 seem to disagree with the statements here. Who's right? Just looking for the proper solution :-)
Questions 21 & 22 seem to disagree with the statements here. Who's right? Just looking for the proper solution :-)
?Right, just done a bit more digging into this and spoken to Docking Engineering, Steve Gugliemi and a couple of other race/cooling bods, and the general concencus seems to be that you want nowhere near 1 bar of pressure in the head (more like 6-7psi), and you do not want too much velocity through the block as this also means that the flow is too fast through the rad so the rad does not have enough time to cool the fluid.
Furthermore, in a road car it seems you want an EWP every time for quick warmup and emissions control, though with a racer it matters less and either is good, though EWPs do allow you to properly cool the engine when you come off track, and you may pick up a couple of horsepower as you no-longer have a water-braked dyno attached to the engine. In short, as long as a flow rate of around 200L/min (based on a 600bhp V8) is achieved at about 6 psi, there should be no problems at all using an EWP.
God this subject is making me change my mind an awful lot!
Furthermore, in a road car it seems you want an EWP every time for quick warmup and emissions control, though with a racer it matters less and either is good, though EWPs do allow you to properly cool the engine when you come off track, and you may pick up a couple of horsepower as you no-longer have a water-braked dyno attached to the engine. In short, as long as a flow rate of around 200L/min (based on a 600bhp V8) is achieved at about 6 psi, there should be no problems at all using an EWP.
God this subject is making me change my mind an awful lot!
450Nick said:
...God this subject is making me change my mind an awful lot!
I found this a while back :-http://www.vetteweb.com/tech/vemp_0808_corvette_el...
11-13 bhp more on a LS engine potentially.
vetteweb said:
"Typically, our electric pump will outflow a stock mechanical pump by about 3:1 at idle, and the flow advantage is there all the way into the middle-rpm range. A well-designed and properly-driven mechanical pump will outflow our electric past about 3,500 rpm, but the flow of the electric pump in this range is still quite adequate for street/strip vehicles. The case would be different during the sustained high-rpm operation of many road-race situations."
That doesn't help does it !! 
Who is right? Good question.
Max Torque actually does this stuff for a living, he has worked for a number of big name consultants who are approached by enormous car companies asking for guys like him to design / fix their vehicles.
On the other hand we have some guys that race these vehicles. Their record is what it is.
Personally, as I have been in the position of managing projects done by people like Max Torque, I'd say his advice was very good.
There's a simple answer - buy the race guys' set up as a complete working system and see how it goes. Maybe it works, maybe it doesn't. Good luck.
Max Torque actually does this stuff for a living, he has worked for a number of big name consultants who are approached by enormous car companies asking for guys like him to design / fix their vehicles.
On the other hand we have some guys that race these vehicles. Their record is what it is.
Personally, as I have been in the position of managing projects done by people like Max Torque, I'd say his advice was very good.
There's a simple answer - buy the race guys' set up as a complete working system and see how it goes. Maybe it works, maybe it doesn't. Good luck.
Firstly and most importantly, I am not trying to sell you a water pump, my advice is simply based upon the physical properties required for a cooling system!!
1) "Craig David" points 21 & 22: unfortunately both points are, how do we say "bo**ocks" no OEM has ever spec'd a water pump by the idle operating point!! Why?? well the heat generated by an IC engine is proportional to the fuel burnt (typicaly heat to all losses (coolant, oil, convection, conduction) is about the same as the power produced by that engine (approx speaking, 100% fuel goes to 30% cooling, 30% useful power, 30% exhaust energy, and a bit to sound). Now at idle, fuel consumption is typically 1 to 3 l/hr, whereas at peak power that figure can easily be 150+kg/hr for a big power engine. Therefore the peak idle heat rejection value is tiny!! Added to which, at idle, the vehicle speed is very low (possiby stationary), so why would you want to push lots and lots of water to a radiator that can't er, radiate??. No, water pump size IS set by peak flow capability at peak power rpm, and sizing the pump to create critical zone cylinder heat metal temperatures that are within the limits set by the engine designers
The "flow above 80/l/min increases heat loss very little" point is also i'm afraid a red hearing. The engine they were testing, generates a fixed amount of heat, hence, as you increase the flow rate, that heat can be transfered by a lower and lower delta T between metal and coolant. SO, if you take a std engine, and keep upping the flow rate then obviously that engine will continue to reject the amount of heat it is producing, and not "magically" suddenly produce a lot more heat. The simply physics of the matter are that the heat transfer is driven by the delta T and heat transfer co-efficient, which is proportinal to the velocity gradient across the heat transfer boundary. More velocity = lower delta T
Regarding the 6/7psi head pressure figure from Docking. The "pressure" in a cooling system is a result of the static pressure required to avoid boiling, and the dynamic pressures created by the flow around that system. The point of minimum pressure is the water pump intake, and cooling systems typically have the header tank feed attached at this point to prevent the minimum pressure dropping at high flows and causing boiling (or worse cavitation). To move the required water around the system, the pump must generate a pressure head, which is then "lost" across various "obstructions" in the circuit. (closed loop system so pressures must sum to zero).
Typical system, to move 1l/min per kW: (all pressures are gauge)
Header tank cap pressure: 1 bar min, (modern trend is now to 1.5bar) which sets water pump intake prssure to 1 bar.
Post pump pressure at peak power rpm: 3 bar
Head gasket head loss: 1 bar (majority of loss as the heat gasklet "holes" are designed to do the "flow control" and steer the water around the systems. (Homework: take a close look at a HG, and compare diameter of the coolant passage holes and their position compared to the water inlet in the block;-)
Thermostat head: 0.5bar
Radiator head: 0.5 bar
Most cooling systems will operate with a total system delta T (temp diff between hotest and coolest fluid, top hose to bottom hose usually) of approx 5 to 9 degC, anymore than this is an indication of "underflow". Conversely to what would appear to be normal, the less the system delta T, then better the system performance.
1) "Craig David" points 21 & 22: unfortunately both points are, how do we say "bo**ocks" no OEM has ever spec'd a water pump by the idle operating point!! Why?? well the heat generated by an IC engine is proportional to the fuel burnt (typicaly heat to all losses (coolant, oil, convection, conduction) is about the same as the power produced by that engine (approx speaking, 100% fuel goes to 30% cooling, 30% useful power, 30% exhaust energy, and a bit to sound). Now at idle, fuel consumption is typically 1 to 3 l/hr, whereas at peak power that figure can easily be 150+kg/hr for a big power engine. Therefore the peak idle heat rejection value is tiny!! Added to which, at idle, the vehicle speed is very low (possiby stationary), so why would you want to push lots and lots of water to a radiator that can't er, radiate??. No, water pump size IS set by peak flow capability at peak power rpm, and sizing the pump to create critical zone cylinder heat metal temperatures that are within the limits set by the engine designers
The "flow above 80/l/min increases heat loss very little" point is also i'm afraid a red hearing. The engine they were testing, generates a fixed amount of heat, hence, as you increase the flow rate, that heat can be transfered by a lower and lower delta T between metal and coolant. SO, if you take a std engine, and keep upping the flow rate then obviously that engine will continue to reject the amount of heat it is producing, and not "magically" suddenly produce a lot more heat. The simply physics of the matter are that the heat transfer is driven by the delta T and heat transfer co-efficient, which is proportinal to the velocity gradient across the heat transfer boundary. More velocity = lower delta T
Regarding the 6/7psi head pressure figure from Docking. The "pressure" in a cooling system is a result of the static pressure required to avoid boiling, and the dynamic pressures created by the flow around that system. The point of minimum pressure is the water pump intake, and cooling systems typically have the header tank feed attached at this point to prevent the minimum pressure dropping at high flows and causing boiling (or worse cavitation). To move the required water around the system, the pump must generate a pressure head, which is then "lost" across various "obstructions" in the circuit. (closed loop system so pressures must sum to zero).
Typical system, to move 1l/min per kW: (all pressures are gauge)
Header tank cap pressure: 1 bar min, (modern trend is now to 1.5bar) which sets water pump intake prssure to 1 bar.
Post pump pressure at peak power rpm: 3 bar
Head gasket head loss: 1 bar (majority of loss as the heat gasklet "holes" are designed to do the "flow control" and steer the water around the systems. (Homework: take a close look at a HG, and compare diameter of the coolant passage holes and their position compared to the water inlet in the block;-)
Thermostat head: 0.5bar
Radiator head: 0.5 bar
Most cooling systems will operate with a total system delta T (temp diff between hotest and coolest fluid, top hose to bottom hose usually) of approx 5 to 9 degC, anymore than this is an indication of "underflow". Conversely to what would appear to be normal, the less the system delta T, then better the system performance.
Edited by anonymous-user on Friday 29th October 13:47
I feel a dsicusion on "total system efficiency" would also be useful:
Lets assume we require a nice round 1kW to move the "optimum" amount of water around our cooling system: And, lets also assume that the pump impeller is equally efficient in each design. (note, "old" mechanical pumps typically had very poor efficiency as they were just simple flat "stamped" metal impellers, these days (since late 1990's) all oems have gone over to precision impellers to minimise power losses. Lots of people fit EWP's etc to old engines like say Rover V8, in which the basic std design is so poor that the ewp can actually help!!)
Mechanical pump:
impeller efficiency, (shaft power to fluid power) 65%
Drive loss from crankshaft (belt) 93% (at peak load)
total power from crank 1kW / 65% / 93% = 1.65kW
Electrical system of same power
impeller efficiency, (shaft power to fluid power) 65%
Electrical power to shaft power: (assume brushed low voltage DC motora-la EWP) 80%
Shaft power to electrical power (via conventional claw pole alternator) 73%
crankpower to alternator shaft power (belt) 93%
total power from crank 1kW / 65% / 80 / 73 / 93% = 2.83kW
This "loss" of efficency (as a result of the requirement to transfer power between more "states") can be offset however by:
1) only running at full pump power when required (mech pump is fixed rpm vs engine rpm)
2) clever "regen" strategys that only charge alternator when braking for example
3) for short use (drag runs etc) just run from battery without alterntor, so power does not directly come from engine during run etc.
4) "downsizing" the power delieved (only deliver 500w of fluid power for example
Now, depending upon what your engine is doing, some of these factors will be more important than others: As mentioned, for a pure road car, which most of the time is no were near peak power, and has to meet all sorts of emissions, fuel economy, warmup, cabin heater performance requirements, then an electric system with "regen" alternator makes a lot of sense. But the simple fact of the mater is, even with all these "plus" points, the cost of such systems has uptil now limited there introduction to "premium" platforms (bmw has a nice "switchable mechanical pump on the PSA mini engine btw - clever!!)
For a race engine, the ONLY reason i would run an electric pump is to optimise packaging (no room to fit belt system etc), and then, it would have to be shown that the electric system does not limit engine power or durability(back to those pesky metal temps again)
If you are really concerned about being able to "post cool" or probably more importantly "pre warm" your engine, 2 simple quickfit points in top and bottom hose (or the original cabin heater tappings)and a small remote 12v pump (a-la BOSCH 12v 20l/min pumps) makes a lot of sense
(It should also be noted that unfortunately 12V is a terrible potential from which to run high power devices, 1kW at 12v is 83amps. So say you have a dc motor or alternator winding resistance of 50mohm (typical value) then you will be loosing 347watts to I2R losses alone in each of your motor/alternator!!)
Lets assume we require a nice round 1kW to move the "optimum" amount of water around our cooling system: And, lets also assume that the pump impeller is equally efficient in each design. (note, "old" mechanical pumps typically had very poor efficiency as they were just simple flat "stamped" metal impellers, these days (since late 1990's) all oems have gone over to precision impellers to minimise power losses. Lots of people fit EWP's etc to old engines like say Rover V8, in which the basic std design is so poor that the ewp can actually help!!)
Mechanical pump:
impeller efficiency, (shaft power to fluid power) 65%
Drive loss from crankshaft (belt) 93% (at peak load)
total power from crank 1kW / 65% / 93% = 1.65kW
Electrical system of same power
impeller efficiency, (shaft power to fluid power) 65%
Electrical power to shaft power: (assume brushed low voltage DC motora-la EWP) 80%
Shaft power to electrical power (via conventional claw pole alternator) 73%
crankpower to alternator shaft power (belt) 93%
total power from crank 1kW / 65% / 80 / 73 / 93% = 2.83kW
This "loss" of efficency (as a result of the requirement to transfer power between more "states") can be offset however by:
1) only running at full pump power when required (mech pump is fixed rpm vs engine rpm)
2) clever "regen" strategys that only charge alternator when braking for example
3) for short use (drag runs etc) just run from battery without alterntor, so power does not directly come from engine during run etc.
4) "downsizing" the power delieved (only deliver 500w of fluid power for example
Now, depending upon what your engine is doing, some of these factors will be more important than others: As mentioned, for a pure road car, which most of the time is no were near peak power, and has to meet all sorts of emissions, fuel economy, warmup, cabin heater performance requirements, then an electric system with "regen" alternator makes a lot of sense. But the simple fact of the mater is, even with all these "plus" points, the cost of such systems has uptil now limited there introduction to "premium" platforms (bmw has a nice "switchable mechanical pump on the PSA mini engine btw - clever!!)
For a race engine, the ONLY reason i would run an electric pump is to optimise packaging (no room to fit belt system etc), and then, it would have to be shown that the electric system does not limit engine power or durability(back to those pesky metal temps again)
If you are really concerned about being able to "post cool" or probably more importantly "pre warm" your engine, 2 simple quickfit points in top and bottom hose (or the original cabin heater tappings)and a small remote 12v pump (a-la BOSCH 12v 20l/min pumps) makes a lot of sense
(It should also be noted that unfortunately 12V is a terrible potential from which to run high power devices, 1kW at 12v is 83amps. So say you have a dc motor or alternator winding resistance of 50mohm (typical value) then you will be loosing 347watts to I2R losses alone in each of your motor/alternator!!)
450Nick said:
Right, just done a bit more digging into this and spoken to Docking Engineering, Steve Gugliemi and a couple of other race/cooling bods, and the general concencus seems to be that ..... you do not want too much velocity through the block as this also means that the flow is too fast through the rad so the rad does not have enough time to cool the fluid.
Oh dear oh dear. Anyone who says this has no understanding of thermodynamics or heat transfer. Total b****cks I'm afraid. All other things being equal heat transfer is proportional to momentum transfer and hence velocity and pressure drop. Ok, i've given the benefit of my knowledge, so i guess it's time to lay my cards on the table, and say what i'd do if it were my race engine:
(in order of preference and assuming a limited budget):
1) Fit the std mechanical water pump to the engine (or you could use the "remote" mech pump off the jag V6??) and use a small "run-on" 12v pump if worried about cool down (or just leave engine idling for 1min afterwards??)
2) fit a Stewart components EWP if package problems or difficult to track down / rehouse a mech pump
3) as an absolutely last resort, and probably only if someone was holding a gun to my head, then fit about 3 craig david ewps in series and cross my fingers !!!
(Actually what i would probably do, is get the siemens OEM brushless elec pump from a bmw 335i, hack the CAN control and use that with a battery chargecontroller for regen alternator loading etc........... but i do like to make things difficult for myself ;-)
(in order of preference and assuming a limited budget):
1) Fit the std mechanical water pump to the engine (or you could use the "remote" mech pump off the jag V6??) and use a small "run-on" 12v pump if worried about cool down (or just leave engine idling for 1min afterwards??)
2) fit a Stewart components EWP if package problems or difficult to track down / rehouse a mech pump
3) as an absolutely last resort, and probably only if someone was holding a gun to my head, then fit about 3 craig david ewps in series and cross my fingers !!!
(Actually what i would probably do, is get the siemens OEM brushless elec pump from a bmw 335i, hack the CAN control and use that with a battery chargecontroller for regen alternator loading etc........... but i do like to make things difficult for myself ;-)
450Nick said:
It's not that it aint broke, its that it don't exist! I have no water pumps at the moment so I need something one way or another, though it will be significantly easier to put in a pair of EWPs than source an original (and probably underpowered) TVR water pump...
Just remember, you can always change the pulley ratios to get the pump to work a little bit harder if it lacks delivery. The cost of the machining may seem expensive but will be cheap compared to engineering an electric set up.Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff