Reverse Flow/Electric Water Pump Cooling System
06-01-2009, 06:27 PM
#41
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Tony I must confess this is way over my head but I need to ask: do you have pics of your install and can you point me to where I can buy the water pump you used?
Thanks
Thanks
Reply
0
0
06-01-2009, 07:24 PM
#42
Junior Member
Thread Starter
iTrader: (1)
Join Date: Mar 2007
Location: South Florida
Posts: 483
Total Cats: -2
Here it is on ebay:
DAVIES CRAIG ELECTRIC WATER PUMP EWP 80 - eBay (item 200348123739 end time Jun-10-09 17:11:43 PDT)
That ebay price is a good deal. I remember last year when I was looking for mine, they were going for about $210 or so dollars new.
If you mean pics as I went through the install, no sorry. The only photos I took were the other day for this post.
Reply
0
0
06-01-2009, 08:20 PM
#43
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Rafa, the manufacturer' link is: Davies Craig
Here it is on ebay:
DAVIES CRAIG ELECTRIC WATER PUMP EWP 80 - eBay (item 200348123739 end time Jun-10-09 17:11:43 PDT)
That ebay price is a good deal. I remember last year when I was looking for mine, they were going for about $210 or so dollars new.
If you mean pics as I went through the install, no sorry. The only photos I took were the other day for this post.
Here it is on ebay:
DAVIES CRAIG ELECTRIC WATER PUMP EWP 80 - eBay (item 200348123739 end time Jun-10-09 17:11:43 PDT)
That ebay price is a good deal. I remember last year when I was looking for mine, they were going for about $210 or so dollars new.
If you mean pics as I went through the install, no sorry. The only photos I took were the other day for this post.
Reply
0
0
06-02-2009, 03:33 PM
#44
Elite Member
Join Date: Jul 2005
Posts: 6,420
Total Cats: 84
Tonyg,
Another very useful datapoint would be if you could measure the pressure at the intake, and at the outlet, of the DC pump, with the t-stat fully open (disable the fans and idle til coolant is 10*C above your t-stat temp). From this and the pump curves, one can deduce the flowrate.
Rafa, at the time I researched this, the DC pump was a more reliable design than the Meziere, IIRC.
Anyone with a reroute can test an electric pump by plumbing it in, then removing the fan/alternator belt - do this after idling and the car is warm so you don't discharge the battery while warming up- it would help to use jumper cables and jumper your miata to another car that's idling so the system voltage is up at 14V and the pump sees 14V. The water pump shouldn't be as much a restriction as the open t-stat is.
Another very useful datapoint would be if you could measure the pressure at the intake, and at the outlet, of the DC pump, with the t-stat fully open (disable the fans and idle til coolant is 10*C above your t-stat temp). From this and the pump curves, one can deduce the flowrate.
Rafa, at the time I researched this, the DC pump was a more reliable design than the Meziere, IIRC.
Anyone with a reroute can test an electric pump by plumbing it in, then removing the fan/alternator belt - do this after idling and the car is warm so you don't discharge the battery while warming up- it would help to use jumper cables and jumper your miata to another car that's idling so the system voltage is up at 14V and the pump sees 14V. The water pump shouldn't be as much a restriction as the open t-stat is.
Reply
0
0
06-05-2009, 03:05 PM
#45
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
I decided to start some additional research of this subject. I must say that I've never been pleased with the Miata's engine cooling system. Even after my latest reroute I'm still searching for what I feel is the "sweet spot" of the engine's temp.
I wanted to thank you Tony for this thread and also thank those who've contributed to it so far. As I mentioned in my pm to you; I really don't want to crap on your thread but since you were kind enough to allow me to keep asking dumb questions here; that's precisely what I'll do.
I had driven around for about 1 month with coolant temps under control until both my fan spals broke. I just got the 2 new ones from Summit so I think next thing to do is follow your path Tony. I already have a coolant pipe coming from the back of my engine to the radiator so it won't take me much to make a "Y" and do what you've done.
Jason insisted on the DC pump. I read all he posted and I see it has merit but, on the other hand; the pump you're using has been installed for 1 1/2 years without issues and we basically have the same climate conditions where we both live. I also blocked the heater core portion of my cooling system.
Now on to the questions:
1) Would there be any advantage in buying the higher flowing pump?
2) The pump manufacturer you suggested has the option for controlling the flow in different conditions; any potential gains to be had there?
BTW, I found this pdf very interesting:
I wanted to thank you Tony for this thread and also thank those who've contributed to it so far. As I mentioned in my pm to you; I really don't want to crap on your thread but since you were kind enough to allow me to keep asking dumb questions here; that's precisely what I'll do.
I had driven around for about 1 month with coolant temps under control until both my fan spals broke. I just got the 2 new ones from Summit so I think next thing to do is follow your path Tony. I already have a coolant pipe coming from the back of my engine to the radiator so it won't take me much to make a "Y" and do what you've done.
Jason insisted on the DC pump. I read all he posted and I see it has merit but, on the other hand; the pump you're using has been installed for 1 1/2 years without issues and we basically have the same climate conditions where we both live. I also blocked the heater core portion of my cooling system.
Now on to the questions:
1) Would there be any advantage in buying the higher flowing pump?
2) The pump manufacturer you suggested has the option for controlling the flow in different conditions; any potential gains to be had there?
BTW, I found this pdf very interesting:
Reply
0
0
06-05-2009, 03:33 PM
#46
Junior Member
Thread Starter
iTrader: (1)
Join Date: Mar 2007
Location: South Florida
Posts: 483
Total Cats: -2
Rafa,
It's ok you are not crapping on my thread. In fact, the opposite is true, it invites people to further develop the system.
One point that needs to be made on my car, is that yes, the system has been installed for 1 1/2 years, but this is not my primary car. It sees very limited use and I would say that it has seen little bumper to bumper traffic or heavy track use. That said, I feel confident that it should handle these situation with no problem, but the fact remains that it has not been fully tested.
Is your Miata your primary car?
Q&A:
1- I used the smaller pump, but using their larger pump can only help a high performance car.
2- I am not using their controller, but I can see where having a controller that can vary the pump input voltage assist in both warm up and heavy duty driving (lower voltage during warm up/higher voltage if running too hot).
Good luck!
It's ok you are not crapping on my thread. In fact, the opposite is true, it invites people to further develop the system.
One point that needs to be made on my car, is that yes, the system has been installed for 1 1/2 years, but this is not my primary car. It sees very limited use and I would say that it has seen little bumper to bumper traffic or heavy track use. That said, I feel confident that it should handle these situation with no problem, but the fact remains that it has not been fully tested.
Is your Miata your primary car?
Q&A:
1- I used the smaller pump, but using their larger pump can only help a high performance car.
2- I am not using their controller, but I can see where having a controller that can vary the pump input voltage assist in both warm up and heavy duty driving (lower voltage during warm up/higher voltage if running too hot).
Good luck!
Reply
0
0
06-05-2009, 07:26 PM
#47
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Rafa,
It's ok you are not crapping on my thread. In fact, the opposite is true, it invites people to further develop the system.
One point that needs to be made on my car, is that yes, the system has been installed for 1 1/2 years, but this is not my primary car. It sees very limited use and I would say that it has seen little bumper to bumper traffic or heavy track use. That said, I feel confident that it should handle these situation with no problem, but the fact remains that it has not been fully tested.
Is your Miata your primary car?
Q&A:
1- I used the smaller pump, but using their larger pump can only help a high performance car.
2- I am not using their controller, but I can see where having a controller that can vary the pump input voltage assist in both warm up and heavy duty driving (lower voltage during warm up/higher voltage if running too hot).
Good luck!
It's ok you are not crapping on my thread. In fact, the opposite is true, it invites people to further develop the system.
One point that needs to be made on my car, is that yes, the system has been installed for 1 1/2 years, but this is not my primary car. It sees very limited use and I would say that it has seen little bumper to bumper traffic or heavy track use. That said, I feel confident that it should handle these situation with no problem, but the fact remains that it has not been fully tested.
Is your Miata your primary car?
Q&A:
1- I used the smaller pump, but using their larger pump can only help a high performance car.
2- I am not using their controller, but I can see where having a controller that can vary the pump input voltage assist in both warm up and heavy duty driving (lower voltage during warm up/higher voltage if running too hot).
Good luck!
This is not my primary car (which doesn't mean that I don't love to invent excuses to use it anyways!
). My office is 2 blocks away from my home and ever since I started to mess with turboing the car, there have been many days of downtime anyways. On the other hand, I love to drive cars I can take anywhere whenever I feel like doing so. I guess I like to have reliable cars. My next project entitles driving the car about 200 miles to a friend's hotel at the beach and back and that takes a well sorted car.
My youngest son (the one leaving for Texas in mid July) works outside the city (about a 40 mile trip back and forth) and since he's going to be leaving soon I tend to let him take my car 2 or 3 times a week to work.
Finally; this project is taking shape because I've already had the car for 2 years and summer has just started. Temps really climb these days and that makes it the best time for testing new things.
Reply
0
0
06-06-2009, 02:59 PM
#48
Elite Member
Join Date: Jul 2005
Posts: 6,420
Total Cats: 84
Rafa, as I mentioned the big question is if the flowrate of the DC is enough for track duty. I think probably not with the t-stat in place, and probably enough without. Insufficient flow will obviously show as overheating; if one overheats, the way to determine if it's lack of flow, is to look at the coolant temperature delta across the engine or the radiator (same thing) - if the coolant coming out of the radiator is very cool and the coolant coming out of the engine is very hot, the coolant flow is too slow.
Reply
0
0
06-06-2009, 03:54 PM
#49
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Rafa, as I mentioned the big question is if the flowrate of the DC is enough for track duty. I think probably not with the t-stat in place, and probably enough without. Insufficient flow will obviously show as overheating; if one overheats, the way to determine if it's lack of flow, is to look at the coolant temperature delta across the engine or the radiator (same thing) - if the coolant coming out of the radiator is very cool and the coolant coming out of the engine is very hot, the coolant flow is too slow.
One thing I really would like to try whenever I test this would be the controller to increase or decrease flow offered by the company Tony linked to.
Reply
0
0
06-06-2009, 04:31 PM
#50
Elite Member
Join Date: Jul 2005
Posts: 6,420
Total Cats: 84
The problem is not the flow when the t-stat is closed; the DC pump will just flail. The problem is the flow resistance it presents even when fully open. If you don't have a t-stat then you definitely need to control the pump based on temperature. You could probably get away with 2 thermoswitch controlled relays - a first one that turns on the pump at say 85*C but with a blower resistor in series. A second one that shorts out the resistor so the pump runs at full power, at 90*C. Then if both fans come on at 95*C you have a nice staged system.
Reply
0
0
06-06-2009, 04:40 PM
#51
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,020
Total Cats: 6,588
Since I haven't seen any actual facts posted so far, here is the flow data for the pump in question:
To interpret this data, understand that pressure, as indicated on the chart, is not the actual pressure in the cooling system, but the pressure differential which the pump is achieving between its inlet and outlet. This represents the restriction it is working against, which is principally the thermostat. As the volume of water pumped through the restriction increases, so does the pressure differential.
Now, clearly this pump is capable of working against a restriction. The question then is how much of a restriction is it capable of flowing through while still maintaining enough volume to cool the engine. Let's do a very quick-n-dirty estimation, for which I shall use this calculator.
For pipe diameter, I assume 1.25", which is the diameter of a stock Miata coolant hose. For orifice diameter, I'm going to ballpark the thermostat at 0.5". I know that the actual opening is much larger than that, but it's partially obstructed and never fully open.
For the pressure differential, we have to pick a spot on the chart to test. I'll use 0.3 bar, since it's kinda in the middle. 0.3 bar = 30,000 Pa. Fluid density is easy: water at 95°C is 960kg/m^3. Lastly, it wants to know the Cf of the orifice. This is pure guesswork since we don't actually know the Cf across an open thermostat, so I'm going to leave their default of 0.7 there.
We hit Calculate, and it gives us a volume flow rate of 0.7 l/sec, which is 42 l/min.
On the chart, the pump is capable of of 50 l/min at 0.3 bar. So in reality, it's going to have no problem moving this quantity of fluid through the hypothetical restriction that we just calculated. (and don't forget, you need some bypass circulation anyway, so that eases the load slightly.)
Now, if you're not satisfied with all this, let me toss out a random suggestion: run two thermostats. In parallel. Take the water outlet at the bottom of the block, split it, and run it into two thermostat housings. Combine the returns into the upper radiator inlet.
To interpret this data, understand that pressure, as indicated on the chart, is not the actual pressure in the cooling system, but the pressure differential which the pump is achieving between its inlet and outlet. This represents the restriction it is working against, which is principally the thermostat. As the volume of water pumped through the restriction increases, so does the pressure differential.
Now, clearly this pump is capable of working against a restriction. The question then is how much of a restriction is it capable of flowing through while still maintaining enough volume to cool the engine. Let's do a very quick-n-dirty estimation, for which I shall use this calculator.
For pipe diameter, I assume 1.25", which is the diameter of a stock Miata coolant hose. For orifice diameter, I'm going to ballpark the thermostat at 0.5". I know that the actual opening is much larger than that, but it's partially obstructed and never fully open.
For the pressure differential, we have to pick a spot on the chart to test. I'll use 0.3 bar, since it's kinda in the middle. 0.3 bar = 30,000 Pa. Fluid density is easy: water at 95°C is 960kg/m^3. Lastly, it wants to know the Cf of the orifice. This is pure guesswork since we don't actually know the Cf across an open thermostat, so I'm going to leave their default of 0.7 there.
We hit Calculate, and it gives us a volume flow rate of 0.7 l/sec, which is 42 l/min.
On the chart, the pump is capable of of 50 l/min at 0.3 bar. So in reality, it's going to have no problem moving this quantity of fluid through the hypothetical restriction that we just calculated. (and don't forget, you need some bypass circulation anyway, so that eases the load slightly.)
Now, if you're not satisfied with all this, let me toss out a random suggestion: run two thermostats. In parallel. Take the water outlet at the bottom of the block, split it, and run it into two thermostat housings. Combine the returns into the upper radiator inlet.
Reply
0
0
06-06-2009, 06:11 PM
#52
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Since I haven't seen any actual facts posted so far, here is the flow data for the pump in question:
To interpret this data, understand that pressure, as indicated on the chart, is not the actual pressure in the cooling system, but the pressure differential which the pump is achieving between its inlet and outlet. This represents the restriction it is working against, which is principally the thermostat. As the volume of water pumped through the restriction increases, so does the pressure differential.
Now, clearly this pump is capable of working against a restriction. The question then is how much of a restriction is it capable of flowing through while still maintaining enough volume to cool the engine. Let's do a very quick-n-dirty estimation, for which I shall use this calculator.
For pipe diameter, I assume 1.25", which is the diameter of a stock Miata coolant hose. For orifice diameter, I'm going to ballpark the thermostat at 0.5". I know that the actual opening is much larger than that, but it's partially obstructed and never fully open.
For the pressure differential, we have to pick a spot on the chart to test. I'll use 0.3 bar, since it's kinda in the middle. 0.3 bar = 30,000 Pa. Fluid density is easy: water at 95°C is 960kg/m^3. Lastly, it wants to know the Cf of the orifice. This is pure guesswork since we don't actually know the Cf across an open thermostat, so I'm going to leave their default of 0.7 there.
We hit Calculate, and it gives us a volume flow rate of 0.7 l/sec, which is 42 l/min.
On the chart, the pump is capable of of 50 l/min at 0.3 bar. So in reality, it's going to have no problem moving this quantity of fluid through the hypothetical restriction that we just calculated. (and don't forget, you need some bypass circulation anyway, so that eases the load slightly.)
Now, if you're not satisfied with all this, let me toss out a random suggestion: run two thermostats. In parallel. Take the water outlet at the bottom of the block, split it, and run it into two thermostat housings. Combine the returns into the upper radiator inlet.
To interpret this data, understand that pressure, as indicated on the chart, is not the actual pressure in the cooling system, but the pressure differential which the pump is achieving between its inlet and outlet. This represents the restriction it is working against, which is principally the thermostat. As the volume of water pumped through the restriction increases, so does the pressure differential.
Now, clearly this pump is capable of working against a restriction. The question then is how much of a restriction is it capable of flowing through while still maintaining enough volume to cool the engine. Let's do a very quick-n-dirty estimation, for which I shall use this calculator.
For pipe diameter, I assume 1.25", which is the diameter of a stock Miata coolant hose. For orifice diameter, I'm going to ballpark the thermostat at 0.5". I know that the actual opening is much larger than that, but it's partially obstructed and never fully open.
For the pressure differential, we have to pick a spot on the chart to test. I'll use 0.3 bar, since it's kinda in the middle. 0.3 bar = 30,000 Pa. Fluid density is easy: water at 95°C is 960kg/m^3. Lastly, it wants to know the Cf of the orifice. This is pure guesswork since we don't actually know the Cf across an open thermostat, so I'm going to leave their default of 0.7 there.
We hit Calculate, and it gives us a volume flow rate of 0.7 l/sec, which is 42 l/min.
On the chart, the pump is capable of of 50 l/min at 0.3 bar. So in reality, it's going to have no problem moving this quantity of fluid through the hypothetical restriction that we just calculated. (and don't forget, you need some bypass circulation anyway, so that eases the load slightly.)
Now, if you're not satisfied with all this, let me toss out a random suggestion: run two thermostats. In parallel. Take the water outlet at the bottom of the block, split it, and run it into two thermostat housings. Combine the returns into the upper radiator inlet.
I have a question about the 2 thermostats in parallel: what's your reasoning behind that suggestion and what are you trying to get as a result?
Reply
0
0
06-06-2009, 06:19 PM
#53
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
The problem is not the flow when the t-stat is closed; the DC pump will just flail. The problem is the flow resistance it presents even when fully open. If you don't have a t-stat then you definitely need to control the pump based on temperature. You could probably get away with 2 thermoswitch controlled relays - a first one that turns on the pump at say 85*C but with a blower resistor in series. A second one that shorts out the resistor so the pump runs at full power, at 90*C. Then if both fans come on at 95*C you have a nice staged system.
If I could by some miracle follow your instructions to the letter; wouldn't I end up with coolant flowing too fast? What could be the consequences of such a thing (if any)?
Reply
0
0
06-06-2009, 07:18 PM
#56
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,020
Total Cats: 6,588
The amount of pressure drop which the pump experiences is basically proportional to the amount of water it is trying to move, relative to the size of the restriction. If you increase the size of the restrictive orifice (the hole in the middle of the thermostat) then you decrease the restriction and increase the flow.
We can't make the hole in the thermostat any bigger, however we can add a second thermostat. Two thermostats in parallel will be equivalent to one thermostat whose opening is twice as large as normal.
2: Clicked "Electric Water Pumps."
3: Clicked EWP80.
4: Clicked button which said "Technical Guide", which opened up a .PDF document containing all the tech data for the EBP, EWP80 and EWP115 pumps, plus the EWP controller.
Reply
0
0
06-06-2009, 08:29 PM
#57
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Well, Jason's argument here is predicated on the idea that because the opening in a thermostat, even when "fully open", is relatively small, that this restriction will impede the pump's ability to move a sufficiently large volume of water to properly cool the engine.
The amount of pressure drop which the pump experiences is basically proportional to the amount of water it is trying to move, relative to the size of the restriction. If you increase the size of the restrictive orifice (the hole in the middle of the thermostat) then you decrease the restriction and increase the flow.
We can't make the hole in the thermostat any bigger, however we can add a second thermostat. Two thermostats in parallel will be equivalent to one thermostat whose opening is twice as large as normal.
Too fast? I don't believe such a thing is possible.
The amount of pressure drop which the pump experiences is basically proportional to the amount of water it is trying to move, relative to the size of the restriction. If you increase the size of the restrictive orifice (the hole in the middle of the thermostat) then you decrease the restriction and increase the flow.
We can't make the hole in the thermostat any bigger, however we can add a second thermostat. Two thermostats in parallel will be equivalent to one thermostat whose opening is twice as large as normal.
Too fast? I don't believe such a thing is possible.
Joe; many thanks for that explanation. That makes all the sense in the world. I can do that.
Reply
0
0
06-06-2009, 08:33 PM
#58
Elite Member
iTrader: (22)
Join Date: Dec 2006
Location: Sunny Spanish speaking Non US Caribbean
Posts: 3,224
Total Cats: 3
Actually Jason, I was basing my question on your previous explanation about not using a t-stat whatsoever.
Sorry I didn't explain myself too clearly (as you know; English is my second language
)
Reply
0
0
06-06-2009, 08:52 PM
#59
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,020
Total Cats: 6,588
I should add another note here, regarding operation with no thermostat. There is some evidence which suggests that having some degree of restriction at the point where coolant exits the engine is beneficial to proper engine cooling. The idea is that having this restriction between the outlet of the engine and the inlet to the radiator effectively increases the pressure within the engine itself (by an amount equal to the pressure drop across the restriction) to a level even higher than the average system pressure, such as the pressure measured at the radiator cap. This raises the boiling point of the water within the head, which further improves its ability to resist boiling, particularly non-nucleatic boiling, which is the formation of a thin but coherent layer of steam over the surface of the coolant passage.
Sorry I didn't explain myself too clearly (as you know; English is my second language )
)
Reply
0
0
06-06-2009, 10:13 PM
#60
Junior Member
Join Date: Jan 2006
Location: Toronto
Posts: 181
Total Cats: 0
There is some evidence which suggests that having some degree of restriction at the point where coolant exits the engine is beneficial to proper engine cooling. The idea is that having this restriction between the outlet of the engine and the inlet to the radiator effectively increases the pressure within the engine itself (by an amount equal to the pressure drop across the restriction) to a level even higher than the average system pressure, such as the pressure measured at the radiator cap. This raises the boiling point of the water within the head, which further improves its ability to resist boiling, particularly non-nucleatic boiling, which is the formation of a thin but coherent layer of steam over the surface of the coolant passage.
Reply
0
0
