trainwreck517

iTrader: (1)

I am not sure if this should be posted here or Race Car Tech section, considering this is done on a 3rd gen I decided to post it here but same thing can be done to all three generations.

So in the process trying to get a solid cooling system for my NASA time trial car, I decided to go the electric water pump route.

Car engine setup is: street ported 13B-REW, FMIC (properly ducted, with fresh air passage to the radiator), two 32-row oil coolers (ducted), CSF racing radiator (2.75” core, ducted), 60-1 single turbo, 100% water injection, under drive pully, tweak it racing idler pulley.

I feel the stock water pump doesn’t flow enough for hard road racing / time trial applications with a high hp engine. I feel this is especially true when using a thicker core with higher flow resistance. Unfortunately there isn’t an aftermarket high flow mechanical pump we can just bolt up similar to that as the V8 guys (Mazmart has a modified FD pump with better designed impeller, but no flow rating / test data provided).

I began having trouble keeping water temperature down on hot days (95-100°F track temps). I would see 110°C after about 10 minutes of hard racing when shifting at ~8100-8300rpms when running anything greater than 12psi. When racing on hot days I was forced to keep my rpms low by shifting at ~6800rpms to keep my water temps under control. I hypothesized this is probably do more to the stock water pump cavitations than the radiator not keeping up. So I decided to throw on a high flow (55GPH) ewp and see what happens.

Most setup I have seen involved an in/out port on the front iron, which remove stock thermostat housing and involve custom brackets for the alternator. This also makes keeping a stable temperature difficult do to lack of a thermostat (running too cold).

An engine builder I talked to said he has had good results running a ewp with a modified thermostat housing on his 20B engines. So I have decided do a write up on my setup.

This is the ewp I went with: http://www.stewartcomponents.net/Mer...Code=ElectPump

This pump seems to be working great on the 5.4L V8 Ford GT guys, so I decided to use the same pump. One good thing about this pump is that it does not add much restriction to flow when not operating. So incase the ewp pump fails, I can swap in a mechanical pump and continue racing.

So here’s what the stock setup looks like:


First we have to take of off the thermostat housing:


Then the water pump housing:


With everything removed we must first modify the water pump by removing the impeller and sealing it. I have used JB weld to seal the impeller shaft hole:



After which we must modify the stock water pump housing by blocking the bypass ports:



I also decided to grind down a portion of the water pump housing which would be the output of the stock water pump to help increase flow with the ewp. I left most of it intact so that I can easily just swap in stock water pump at the track if I were to have the ewp fail:



I then decided to crimp down portion of the turbo water feed to help limit flow as I am now using a ewp on the bottom radiator hose. I feel an unrestricted turbo water feed would allow too much water flow to bypass the engine and feed through the turbo instead (turbo may pose less resistance than the engine, this was not a problem with the stock setup as the water pump was in between the turbo and not before it):


We must also modify the thermostat by removing the bypass plunger and drilling bypass holes:

trainwreck517

iTrader: (1)

Do to the pump being a bit longer than the original hose I needed to add a u-bend to the output of the ewp. Here’s what the ewp looks like with pluming.



Now we need the water pump housing inlet to flow in from the passenger side rather than the stock driver side. The water pump housing must be modified a by cutting the return a bit shorter. I than used a hose with a 90°bend to point the inlet to the driver side. Here is what it looks like with the hose attached:


Now with the ewp installed:




Final installation:


I wired up the pump using the existing AC relay (I no longer have an AC). I have it setup to turn on with the ignition in the ON position. This way the pump is on with the car is on, and also with the key in the on position with the engine off for cool down after a track session.



My next track day won’t be until next month, so I will update the thread then with my results.

Please note this is for my track car that sees very little street time. It is usually only driven on the street to go to local meets and to drive to and from the track. This is one of the main reasons why I wanted a functional thermostat. This is definitely an overkill for street driven car. The ewp pump I used is rated for 10,000 hours and has been used on high HP 20Bs and V8s with success, so I feel it should work fine in my application.

Thanks,
Alvaro

turboIIrotary

iTrader: (9)

Would it be possible to use just a flat plate where the water pump went instead of using the jb weld?

allrotor93

iTrader: (126)

Let us know how it works!

trainwreck517

iTrader: (1)

Yes, that would be a best looking way but as I don't have access to a plasma cutter to make a cover, I just modified the existing water pump with JB weld.

scotty305

iTrader: (2)

I think an electric water pump is a great idea, and I also like the idea of keeping the OEM waterpump housing since the alternator mounts to it. I'm not so sure about using JB Weld inside the cooling system, is this something you've done in the past or will this be the trial run?

gracer7-rx7

iTrader: (16)

Most people usually replace the stock water pump with an aluminum adaptor that has an inlet outlet. Read through this thread as there is a pic in there somewhere or search on old threads on the topic.
https://www.rx7club.com/showthread.p...&highlight=Ewp

With regards to your coolant temps, if that is a greddy fmic, then that is the source of your poor coolant temps. That ic blocks the entire opening in the stock front bumper severely restricting airflow to the radiator. You either need better ducting and try to air to the rad directly or you'll never get manageable temps at hot desert so cal tracks. Btdt and I now rUn a vmount.

GoodfellaFD3S

iTrader: (213)

Did you consider making the switch to Evans NPG+?

trainwreck517

iTrader: (1)

I have used JB weld in the past to block the bypass port on the water pump housing with no problems before. GB weld is rated at 580°F, so should work fine in a 220°F environment. Best solution would be to weld it though.

Yes, I know most people use an in/out port on the front iron. But as I said this makes it difficult to keep temperatures warm enough on street driving and requires a custom bracket for the alternator.

I am sure the Greddy FMIC doesn't help with cooling either, but I still feel that the radiator gets enough air flow. I think there is some cavitation going on at higher rpms and the flow restriction of a 2.75" core aluminum doesn't help either. I have a fresh air duct to radiator that by passes the FMIC.



I guess I will find out for sure at the next track event whether my theory was correct on the pump cavitating. Perhaps you are right and there isn't enough flow to the radiator. If that is the case, I will go with a Vmount setup in the future.

I did, but felt if the pump is cavitating then it may not help to much. I also didn't like the idea that in case of a coolant leak I would have that stuff flying at my tires.

GooRoo

iTrader: (4)

You might be right, that the cavitation isn't helping - but based on my experience there's no way to make that setup work on a road course. You don't have anywhere near enough air being forced through the radiator with a setup similar to that picture. Damian and I are using GTC noses, with ASP ICs (which are similar to a v-mount in that we select how much air goes into each opening) and we are directing 75%+ of that huge opening to the radiator and we can still get the cars hot in 90+ degree ambient heat after a 20-30 minute session on track. You need triple the air going to the rad (depending on HP), and a good solid sealed duct to make it work long term on the track.

We use Evans (at no pressure) as a backup - if anything is wrong with the cooling system or we miss a temperature issue - it doesn't matter, it won't boil. In addition, any small leak will stay small since we don't have any pressure in the lines blowing out hoses. We've had multiple cracked hoses on track and we don't even notice until you open the hose and you can smell it. If you really blow a hose then water or evans it doesn't matter, you are going off.


With regard to having to move the Alt I understand your desire to keep the mounting location the same - but with respect to keeping the temps in check (both too low and too high) once you remove the thermostat did you consider something like this?

http://www.hrpworld.com/index.cfm?tp...action=product

It actively maintains the temperatures by controlling the flow. Though it looked interesting when I was considering a electric water pump.

trainwreck517

iTrader: (1)

Thanks, I will keep that mind. Are still running the stock pump?

Yea, I have had coolant hose blow with a 30/70 mix of coolant/water which caused me to fly off the track as I locked up the rears trying to brake into sunrise at buttonwillow. I guess that's one of the main reasons why I feel bit hesitant to run sometime like evans. But you bring up a very good point on running it with no pressure.

I was close to running that exact controller, however after doing some research on Ausrotary.com I found many people were having water pump failures on the Davies Craig Water pump (the pump that controller is intended for) so I decided against it. Also the Davies Craig EWP is only rated for 2,000 hours. Much less than the one I used.

the_glass_man

iTrader: (2)

Uploaded with ImageShack.us
I completely removed my water pump housing and switched to - 20 AN fittings all the way around.

trainwreck517

iTrader: (1)

As I already mentioned twice, including the original post. I wanted to avoid that type of setup.

the_glass_man

iTrader: (2)

Why? So you can run a thermostat? Why don't you run a proper ECU that controls your pump via PWM? I don't see the point of going through all that trouble just to run an electric water pump in place of the regular pump.

trainwreck517

iTrader: (1)

Yes, so I can run a thermostat and keep the stock alternator mounting location.

This is not a simple brush type DC motor.. running PWM is not recommended by the manufacture. So that would not work as a means of controlling flow. As far as I am concerned, a Power-FC is a proper ECU.

I think this is a much simpler solution, then going through all the trouble of making a custom bracket for the alternator and finding a reliable method of controlling temperature. I would also have the problem that in case of a failure of the EWP, I would have to drop out of the race. At least with this method, I can swap in a mechanical pump and be out on the track again or at the very least be able to drive the car home.

trainwreck517

iTrader: (1)

Alright wanted to give a quick update of the results at the track.

I was unable to complete my duct work for the radiator so had some minor air gaps present (air would bypass the radiator). I will diffidently have that fixed next time out.

With that being said here are my results:

Late morning session: Ambient temperature 78°F peak water temperature 87°C ( my fans are set to come on at 86°C on the PFC) running 10psi of boost.

Late after noon session: Ambient temperature 89°F peak water temperature 101°C running 12psi of boost.

This was at Buttonwillow #13CW. My water temperature would fluctuate between 98°C -101°C during a single lap (hotter during slower corners, cooler during faster corners). So I think with perfect ducting, I can get temps down another degree or so. My goal is to keep water temps below 112°C during peak summer temps.

So doing a bit of temperature adjustment if running same track, I calculate I can run up to 108°F ambient temps without changing a thing on the car and would only have a peak water temp of 112°C.

I feel pretty confident that if I would have ran the stock water pump, my water temps would have began creeping out of control. So I diffidently feel that the ewp helped keep my water temps under control.

I'll try to get the ducting 100% sealed before my next event. Best lap time was 2:12.55 with some traffic. No areo on the car with street tires.

Video of my fastest lap of the day:
http://www.youtube.com/watch?v=rWMIRGeiQeQ

P.S, I lost in the sweeper due to someone going off and leaving a bunch of a dirt on the track.. corner worker did a stand up job of not putting up the yellow/red flag to warn me... That was my fastest lap attempt of the day.. oh well.
Thanks,
Alvaro Garcia

GooRoo

iTrader: (4)

Sorry, I didn't answer you before, but I have run a stock water pump until just recently when I switched to the Remedy pump to try to get more flow at high RPM.

Your results seem pretty reasonable so far... I have a few questions to try to frame them.

How much HP are you running at the 10psi? Do you know what the track temp was? Often when it's 80-90 it can be 120-140 at track level where the air is being ingested into the nose, but if there's enough wind then it can be much lower.

How many laps did you do at once? How was your longest continuous run of HOT laps?

Do you have any Oil temps to look at?

Between Damian and I what we have found is that you can do TT or single laps with alot of different setups, but if you want to race and do 8/10ths or 9/10ths laps over and over (and in traffic) for an hour+ then the thermal requirements go up alot so that's why I'm asking to see if your data supports that or refutes it.

trainwreck517

iTrader: (1)

I don't know how much HP. Should be low 300s at 10psi, but I honestly don't have data. Setup is: Street port, 60-1 w/ T4 0.84 A/R, high flow cat, RB Cat back, Water Injection, Intakes from my data logging showed 24°C for most of my run time until I ran out of water and they creeped up to 36°C.

I don't know what track temps were at, just ambient. It was clear day, so I am sure the asphalt was hot resulting in pretty high tack temps.

I get about 11 laps per session at this track.. ~ 25 minutes.

First lap I am driving about %80 to get tires and brakes warmed up.. after which I go 100%.. and if I feel great.. I'll go for the 110%.. but I I'll usually get a bit sloppy as you can see my corrections at 2:50min mark from giving it to much throttle.. so about 9-10 continuous HOT laps (20 min or so).

Yes, taken from the the oil pan. Hottest I seen was about 215°F.

I agree, I usually get very little to no traffic. So there is a lot of clean air going to the radiator / oil coolers (track organizers pre-griding by lap times).

Thanks,
Alvaro G.

Type_P10

Question! How does this work without a t-stat? the engine takes a lot longer to get hot, and would it not have diffeculty having a steady temp during driving??

the_glass_man

iTrader: (2)

If you use the Davies Craig controller or an ECU that supports an auxiliary output with pulse width modulation you can have precise control over the flow rate that the pump is flowing in regards to the temperature. This way you can delay the start or slowly pulse the pump motor until the block is up to operating temperature. It doesn't take any longer for the to reach operating temperature than without a thermostat.

I did try running the pump at full on without the thermostat or any controlling method and with my current setup the car would never get up to optimal operating temperature. It didn't matter how long I left it running or how hard I hammered it even with a huge FMIC and the undertray uninstalled.

Type_P10

Ok. So i am guessing the pfc does not support this puls width modulation.... I am remoging the entire WP t-stat block and using an adapterplate... This is using an stewart ewp and org fd fans running on the org fan relays ( tryd using a aftermarked fan controller but it did not handle the fd fans 2 speed fans). Would it be better just using the kit from davies craig insted of an stewart ewp and davies controller?

the_glass_man

iTrader: (2)

I don't know enough about the Stewart pumps to know if they would work well with the Davies Craig Controller. Do the Stewart pumps have built in controllers or mention anything about being driven off of PWM? The PFC is great, but once you start getting into serious modifications it begins to show its age. An ECU like a MoTeC or Link G4 would really benefit an installation such as this. However the Davies Craig Digital Controller would do a fine job regulating the temperature and optionally can control the cooling fans as well as the pump. You can even have it run after shut down to help prevent heat soak.

The problem you are going to run into is where are you going to mount all of your sensors without the thermostat housing? If you run the controller you'll need a place to put the probe and you'll need one for the sensor the PFC gets it temperature readings from. Not to mention the coolant level sensor and where will the fill cap go? Placement of all of these and tuning will be key.

Type_P10

So to make it easy for myself maybe use this plate (welded up) as a block off for the WP and using the WP t-stat assembly as normal might be an idea.. not to keen on having "instant weld" in there as in this thread, surly a nice way to do it, guess im just paranoid...

http://www.meziere.com/ps-1300-1266-wp91.aspx