TT_Rex_7

Alright, i've been thinking about running an EWP (Electric Water Pump) for quite some time now, but have tried to avoid it due to the possibilities of it failing, leaving my engine to overheat. However, I'd like to avoid these remarks, and try and focus on a possible "fail safe system."

Just to get it out of the way before anyone asks, I'd like to run an EWP so that I can mount my GT40R in the location of where the stock air pump used to be. Since the turbo is so large, it effects me trying to run hoses from the water pump to the radiator due to how I have my setup. (Turbo at an angle towards the drive side) Also, because the turbo is at an angle, the compressor housing is in the way of the filler neck. The reason I want the turbo in this location is to clear my custom intake manifold, and still have room to run a 4" DP/exhaust. I also like the idea of being able to run the water pump after the motor has been shut down. And of course, there's the "bling" factor!

FWIW, I plan on running Meziere's 55GPM EWP.

So far the best solution i've seen is lights/buzzers that are wired up to the EWP, and go off if the pump stops flowing/fails. So lets say your just cruising at an avg. speed, (50-70 mph) how long do you think you'd have to pull over after hearing the buzzer before the engine was to overheat?

Although a light/buzzer is somewhat of a good idea, I still don't think it's all that much of a "fail safe system." So after a little bit of thinking, I was considering running two EWP's, while only having 1 running continuous. Then I'd have the second wired up to my ECU. I could then set the ECU to turn the second EWP on at a specified temp. Therefore, if the first pump was to fail, the ECU would turn the second one on after the motor began to heat up, allowing coolant to start circulating again. In conjunction with this, I could also wire up a light/buzzer so that I know wether or not it was the EWP that failed, or if something else happened.

So does anyone thing that this could/would actually work?!

I also came across this thread by GuitarJunkie28 ... https://www.rx7club.com/haltech-forum-62/its-impossible-overheat-my-motor-506001/

So who thinks that by tuning the ECU to do what's linked above, running a second EWP that's wired to the ECU, AND running a light/buzzer to warn you if the pump stops flowing would be considered a "fail safe system?"

-Alex

Edit: This is also posted in the Rotary Performance section, which is probably where this belongs. However, there's not nearly as much "traffic" in that section as there is here!

NCMontegott

iTrader: (5)

i think it would work. but what are the possibilites of a ewp actually croaking on you.
do you remember that guy with the 4 rotor in australia. might have to search , but he had converted to a ewp. it would not be hard to install 2 pumps in parallel and one worked all the time and the other could be turned on with a switch like you said. very feasible.

what about running 2 pumps at the same time during hard pulls?

keep us updated.

Robertio

I run a Davies Craig EWP, something unrelated blew a fuse for it and at 70mph I had 1-2 mins before the stock temp guage went up to where it sits on a standard car (normally runs so cool on the EWP that the dial sits at the bottom of the guage) then a further 30 seconds (as I pulled off the road) before the buzzer sounded and I killed the ignition.

The pump has a 2 year warranty, so my plan is to just change it every second year. There are a few FDs in the UK running this pump and AFAIK none have failed to date.

TT_Rex_7

Although the factory gauge is almost TOTALLY useless, this is the type of responce that I wanted to hear.

I was hoping for 1-2 minutes of a "safe zone" to either get the second pump running, or the car shut down. IMO, 1-2 minutes is enough time to take action to prevent the car from overheating.

The main thing that i'm still curious about is how long it'll take the second pump to start getting the temp. back down. In other words, lets say the pump fails, and I have 2 minutes till the car will overheat. So lets say it takes 1 minute for the temps to get high enough to kick the second water pump on, then it takes 2 additional minutes for the second pump to START bringing the temps back down, that's a minute to late.

-Alex

dubulup

plastic parts will fail...it is only a matter of time PERIOD

Plastic parts and heat cycling do not get along. (my profession deals with qual testing electronic components for a highly advanced aircraft, and as a side note: we do not allow plastic electronic parts on this aircraft)

First fail safe is to get a pump that is not made of plastic...I don't know if the pump you specified is plastic or not, but I know plastic pumps are out there.

Linerize the stock temperature gauge, its cheap and easy...and can be calibrated to the PFC, then you can really watch the pump work and get use to the time it takes to bring the temps down.

The EWP should have its own circuit...nothing else can blow the fuse.

If you have the capability, set an alarm at a HIGH safe temp...if the pump isn't bringing the car back into the normal operating temp, shut down the car.

As the last resort do what GUITARJUNKIE suggests, at the highest "safe" temp you would ever run your car in the desert, dump so much fuel the car stalls.


I don't see the point of dual pumps if the first fails...the second could fail too...its in the same enviroment as the first.

Robertio

Can't comment on how long you have as I spent half an hour or so trying to find the blown fuse before leaving the ignition on to get everything nice and cool before starting the car. Based purely on how long it takes to get from the rad fans being on (stuck in traffic) to normal temps you are looking at around 2 minutes of driving (it does start dropping in just a few seconds) - so I imagine it will be tight, but without experimenting can't say for sure.

Factory dial works fairly well with the EWP as it is pretty linear at lower water temps, it's only in stationary traffic or if something goes wrong it makes it into the dead zone.

TT_Rex_7

This is one of the reasons why I chose NOT to go with the Davies Craig EWP. That, along with the fact that it flows less, and has a shorter life expectancy.

Meziere EWP's are made from 6061. (Aluminum for those that don't know)

I run an aftermarket water temp. gauge, along with a Microtech. I do plan on logging water temps. to see how everything reacts.

Bingo! I plan on running a seperate circuit for each of the two pumps. This will eliminate the possibility of anything else blowing the fuse, as you've already stated.

I plan on setting an alarm/light on both my Microtech hand unit (Which will be mounted in the car) along with my electric water temp. gauge.

As long as my tuner feels comfortable doing this, I'm going to have my car setup this way regardless!

Your right, the second could fail as well. However, I plan on replacing/rebuilding the first pump as soon as it fails, so I can continue to have a "back up" pump. Since the second pump only runs at a specified temp., there's no wear and tear on it. I just highly doubt that the second pump would fail when the first pump goes out.

There's two main reasons for running two pumps. First reason being to help prevent the motor from overheating if the first pump goes out. Second reason is so your not stranded somewhere due to a failed water pump. Once the first pump goes out, and the second is running, you can pull over, and simply switch the connections on the pumps so that the second pump is now running continuously, so that you can get to where your going.

-Alex

TT_Rex_7

Bump for more thoughts!

-Alex

fdrx-7man

I think 2 ewps is over kill .... unless you need that security!
A pully driven water pump can fail as well and you would be in the same boat as if it were a ewp.
I think a buzzer for the ewp alone would be a good enough fail safe.

But I do see where it would be nice to have 2 water pumps!

Just my 2 cents.

Black91n/a

Just a thought, but if you have 2 pumps in parallel, with only 1 on at a time, then there's the possibility of recirculation through the pump that's off, thus forming a seperate coolant loop that excludes the engine. This would be bad, as only some of the coolant would be flowing through the engine and radiator.

I have no experiance with EWP's but that's something I'd watch out for as a potential problem with the setup you proposed.

TT_Rex_7

I'm worried about making it home more so than over heating the engine. With only one pump, I have no choice but to call a tow truck, which isn't my idea of fun if I'm on a long distance trip with the FD.

From my experience, you can typically "limp" home if a pully driven water pump fails, which is one of the reasons why many choose not to switch to an EWP. I'm trying to eliminate as many downfalls to an EWP as possible.

-Alex

TT_Rex_7

The pumps will be in series. It'll flow from one pump, rither into the second.

The only concern I have with this is putting wear on the pump that's off. Even though the pump isn't actually running, all the parts are still moving as the coolant flows through it. However, I don't think it'll have near as much wear as the pump that is on. Can't really say until I put the system into action.

-Alex

Black91n/a

With pumps in series you'll need to worry about the increase of backpressure and the associated reduction in flow from pushing the fluid through the off pump, which could end up being fairly significant, as they're probably not designed to be run that way. Also, as you said, wear would also be a concern.

IMHO a better setup would be to have parallel pumps with some check valves (only 2 are needed), or better yet, some full port ball valves. Check valves would be foolproof once installed correctly, but could end up being a flow restiction themselves, whereas full port ball valves wouln't be a restriction at all. Check valves would be the way to go if you're able to rig up some sort of system to detect a fault and engage the other pump, as you could just keep driving. A parallel system shouldn't take up much more space than a series one, it'll save the second pump from the wear and tear, and it'll cause a much smaller flow restriction. Again, I have no direct experiance, this is just speaking from a engineering/design point of view.

Deals Gap Rotary Rally

iTrader: (6)

Very interesting topic. I've been thinking about doing EWP for my future 20B project and thought about going with 2 EWPs just as Alex have mentioned, one running continuous and other for back up and high load (by the way, this is Herblenny).

I'm interested in seeing what Black91n/a is proposing. Is such check valves available in the size we need?? I think it would be a great thing to put some sort of a check valve on the secondary unit, since its not running all the time and its for emergency or extra flow only. Some restriction should be really a problem.

Also, Alex, I would like to check out your car sometimes.. I visit nashville frequently.

TT_Rex_7

So let me see if I understand you here...

Your saying to place a check valve in front of each pump, so two in total. As the coolant flows through one of the pumps, it closes the check valve that's in front of the pump that's normally off. Then you simply place a "bypass" line right there. If the pump that's running fails, the second pump will begin flowing, but in the opposite direction, causing the check valve in front of the pump that is normally running to close. Then again, you have a "bypass" line at that location as well.

If this is correct, will it cause a problem with the coolant running "backwards?"

-Alex

Edit: A picture in paint, or something of the sort, would be nice!

TT_Rex_7

I know if you run them in series, you can run both of them for a high load situation. However, If I'm understanding Black91n/a correctly, you don't have that option when you run them in parallel. If you tried, they'd be flowing against each other.

PM me before your next visit, and I'll see what I can do. So far, I've only been sucessful at taking everything apart! Cobra cradle, custom gas tank, and custom intake manifold *should* be going in soon though! If there's a DGRR in 2007, I plan on being there!

-Alex

Black91n/a

I've attached a sketch of what I'd do if I was going to be doing it myself. If there's a check valve in front of each pump, then you can run either one, or both at the same time with no backflow or internal circulation. I hope that clears up what I was suggesting. I have no idea if there is such thing as a check vlave large enough, but a ball valve of that size could be sourced from a plumbing or irrigation store, and the PVC ones are reasonably compact and quite light. If there's a commercial/industrial type irrigation store nearby then that'd be a good place check there first, they may at least know if a ball valve big enough exists in their industry.

As for running 2 pumps at once for high load situations, that's of debatable use (I've read others saying that it's of no use). There is more water flow through the radiator, but it's going faster and spends less time there, meaning it doesn't get cooled down as much, but at the same time the engine's not heating the fluid up as much. The real limiting factor to cooling is most likely airflow and radiator surface area in most of the situations you're likely to encounter, so increasing the water flow rate won't do much good if any (beyond a point).

One of the key selling points of EWP's that I've heard is that it can pump at a constant rate regardless of what speed the engine is running at, thus you can keep the pump running at the point of maximum cooling efficiency at all times.

Herblenny

iTrader: (13)

I agree.. Excessive flow will not help with actual cooling of the engine. I'm hoping single pump will be enough and what I've read, it should be. But just incase, its nice to know 2 pumps could be ran for higher load.

I'm really interested in this ball valve thing. I've never seen one and I need to see if its possible for this application.

Alex,

I go over to Garfinkles quite often. I think it should be about 20 minute drive from Gallatin to Murfreesboro. We'll meet up sometime.

PHIL

Black91n/a

I did a quick search on Google and came up with some sources and information for large PVC check valves. So we at least know that they exist, but finding a local source could be more difficult.

http://www.plastomatic.com/ckscatalog.pdf
http://www.plumbingstore.com/pvccheck.html

The first link includes some technical specs, and for a flow rate of about 55GPM there's a pressure drop of only about 2.1psi for the 2" valve. For a 1.5" valve the pressure drop would be about 2.8psi.

As for fitting them it should be relatively easy, just insert a short length of PVC pipe and clamp a hose to it. One potential problem I see is the size of all the extra plumbing required to run a parallel twin pump setup like I've proposed. There may not be room for all of it.

scotty305

iTrader: (2)

This is an interesting idea, but have you seen a Meziere pump in person? They're a bit bulky, and I'm not sure you'll be able to fit two in your engine bay...


-s-

TT_Rex_7

Luckily for me, I do! I'll be running a FMIC, with the radiator in a v-mount position. This allows me to run an 1/8" thick piece of aluminum across the frame rails. It runs from the front of the car, all the way to the bar that crosses the frame rails. (Right in front of the main pulley) Doing this allows for two things...1, it fully seals the radiator, allowing all the air that enters the FMIC to pass right through the radiator...2, It allows me to keep my battery up front, and still have enough room to hold two EWP's, run the IC piping through each side, and still have some room to spare!

-Alex

scotty305

iTrader: (2)

Sounds like an interesting setup, although I still don't agree with the second waterpump. The extra lines and flow-control valves are all additional points of failure, and it might not be safe to assume that your system is any more reliable than the Meziere pump in the first place.

Personally, I'd just use one pump, install it correctly, use high-quality fittings and mounting hardware, and mount the pump so that it's easy to remove it if you ever need to. If you're really worried about failures, carry a backup with you, as well as the tools to replace it with.

TT_Rex_7

Picture helped out alot!

So pretty much, your suggesting to run two EWP's the same way people run two fuel pumps. The only thing I'm not following is the check valves?! I don't see why backflow would be an issue without the check valves. As I mentioned above, your drawing is the same setup in which most people run two fuel pumps, and i've never seen them run check valves before.

Your 100% correct on the issues you mentioned with running both EWP's at the same time. However, an EWP typically flows less than the factory water pump. Only during the RPM range in which the factory water pump is most efficent though. When running two pumps, you can atleast maintain the same flow as the factory water pump, which is why alot of people do this. I'd rather run two EWP's during a high load situation, rather than one, and be flowing less than the factory water pump. I'm sure both EWP, and pulley driven water pump manufacturers have done tests to determine if it's better to have the coolant flow faster, or slower. However, IMO, it probably boils down to your application. They'd probably tell you to have it flowing faster in one application, and slower in another. Everything has it's advantages/disadvantages.

Bingo! Not only that, but you can keep the coolant flowing after shutting the car off as well. Another reason, although debated alot, is freeing up some HP. Although electrical load eats up HP, There's typically less HP being ate up than a pulley driven water pump.

-Alex

TT_Rex_7

At this point, I don't agree with the valves either. However, a check valve really can't fail. It's just a ball that's placed in a pipe, with holes on one end. As for extra lines and fittings, it's really just a matter of two extra fittings, and a few inches of extra lines.

Well this idea is pretty much the same as carrying an extra pump in the car. Only diffrence is you don't have to uninstall the bad one, and install a new one. It's already hooked up for you, and all you have to do is flick a switch, or something of the sort.

As Black91n/a has pointed out, running the system in parallel keeps coolant flow to the pump that's off to a minimum. This in return keeps wear and tear off the second pump. The chances of the second pump being dead when the first one goes out is very slim.

-Alex

TT_Rex_7

Picture below is 4 check valves that i'll be using in my custom gas tank. (FWIW, the other two items are trap doors that'll also be used.) Pretty much, a check valve is a pipe with a ball inside. On one side of the pipe, there's holes drilled to allow the fluid to flow through them. You place the check valve in a way that allows fluid to flow into one side of the pipe, and out the drilled holes. If it trys to flow backwards, the substance will push the ball to the other side of the pipe, closing off the drilled holes. This in return, keeps the substance from making it past the check valve.



There's diffrent designs from the ones pictured above, but they all work in the same way.

I can't wait for my car to get down so I can finally get out to meet some of the local Rx-7 guys. I've heard alot about Garfinkle, but have yet to meet him.

-Alex