The Great Intercooler Water Pump Test

Department Of Boost

Alpha Geek
Joined
May 26, 2010
Posts
8,809
Reaction score
28
Well here ya go kids, we did our intercooler (IC) water pump test. And here is some of the data for you. Not all the data we gathered, it was really a IC design test for us and the pump data was a bonus. We’re not going to be publishing data on what we found out about IC’s because that stuff is top secret and for us. No need to give the competition a leg up.

Test Rig:

Our test rig was set up like and engine compartment. IC where the IC goes, HE where the HE goes, pump where the pump goes (well, should go) and we used hose lengths pretty much identical to most of the IC systems out there. The HE we used is a stock 2007 GT radiator. We figured it would be a good middle of the road substitute. We couldn’t go and test every HE out there too. Every pump was run on the same rig, the same way, the same day. It doesn’t get anymore A/B than this.

Pumps:

-GT500/Terminator/Lightning Bosch pump. We used this as the baseline. This is what most kits come with. So everything is judged off of this.

-Lingenfelter Stewart EMP. This is the big daddy pump everyone has been talking about. It is a Steward Warner pump that Lingenfelter hot rodded to run at 18v. It’s a beast, and expensive ($570), and big, and heavy, and will be tricky to mount. Do not confuse this with the standard Stewart pump, that one runs at 12v and doesn’t perform as well.

-Davies Craig EWP150. We got this little fella from an Australian company that has been doing water pumps for years. Trick little unit. Because of how it looks we call it the “Turbo Pump”. It’s light, it’s inexpensive ($250) and looks cool.

-Meziere 55gpm

-Meziere 20gpm

-Rule 2000. This is a bilge pump. A lot of drag race guys use them in ice chests.



Stock IC Systems:

This is what most of you are going to be interested in. We replicated your average stock IC system and its restrictions. Pretty much all of the manifolds/IC’s out there have water fittings/ports that are way undersized compared to the hose diameter (3/4”). To add insult to injury most of them have a couple of hard 90’s in and out of the IC. We replicated a GT500 setup. This is a good benchmark. The Whipple HO for the 3v is about the same, the Kenne Bell’s a little worse, the Roush 3v TVS is about the same, the M90 is worse and the Coyote manifolds (Roush/Whipple which are the same) are about the same too. The Saleen and Eforce are a lot worse. So this data is going to be what most of you can expect from your current systems.

Results are in order of best to worst in gallons per minute (gpm):

Lingenfelter – 9.3gpm

Davies Craig EWP150 – 6gpm

Meziere 55gpm – 5.2gpm

Rule 2000 – 5.1gpm

Bosch/GT500 pump - 5.1gpm

Meziere 20gpm – 4.25gpm

Clearly the Lingenfelter is s a BEAST! It should be though. It’s expensive, it’s huge, it’s heavy, it’s a PITA to mount and it draws a PILE of amps. You had better have a good alternator to be running this bad boy. That said, if you want a silver bullet to solve your IAT issues and don’t want to gut your entire system and go crazy this is a great option.

Department Of Boost GT450 Intercooler:

We always thought that we made a better IC than the other stuff out there. And we were right (patting myself on back). Less restriction at the point where the water goes in/out of the IC pays off. Here is how the pumps ran with our IC on there.

Results are in order of best to worst in gallons per minute (gpm):

Lingenfelter – 13gpm

Davies Craig EWP150 – 8.5gpm

Meziere 55gpm – 7.5gpm

Rule 2000 – 7.25gpm

Meziere 20gpm – 7.25gpm

Bosch/GT500 pump – 6.5gpm

As you can see with the GT450's less restrictive setup all the pumps work better. An average of 28% better! 28%!!!!!! I don’t care what performance you are measuring, 28% is a huge increase. This is all down to our IC design, there were no other changes, none, zero, zip.

Other Stuff:

And this is just the start. We learned a lot from this test. You guys will be seeing our newer manifolds (R-Spec and Coyote) with even better IC’s than our GT450 is.

-The Coyote manifold is 43% better than the GT450 (and a whole lot better than the stock type stuff).

-The R-spec…..wait for it……is 124% better than the GT450. Yeah, 124%! We had that sucker moving 24.5gpm, and it was not with the Lingenfelter pump.

-Our GT500, Terminator, GT Supercar IC replacement kits will be 43-51% (depending on what car) better than the stock stuff that is in there now.

One more tidbit of info. We ran all the pumps at 14.2 volts (what your car is supposed to make) and 13.2 volts which is what a lot of them do make because of weak alternators/batteries/wiring issues, etc. That 1 volt in power drop killed every single pumps flow right about 20%. So if you have a weak alternator your pump performance is suffering. Check your voltage while running (you can do this with a SCT, Aeroforce, etc). You can find some quick pump performance just making sure your charging system is doing it’s thing.

So there it is, short and sweet. If you have any questions LMK. I’m going to post this and head out the door to play with the S550 for the day (working on it). I won’t be able to check this until late tonight or tomorrow morning.

Hope this helps!
 
Last edited:

eighty6gt

forum member
Joined
May 9, 2011
Posts
4,292
Reaction score
403
No 2013 GT500 pump!

I'm going to bucket test one next year I guess vs. the stock Bosch.

Good info, thanks for the tests. Maybe I'll run 3 pumps, since I am running the M90 intercooler core...
 

JeremyH

3V Fuel Guru
S197 Team Member
Joined
Jan 10, 2009
Posts
20,857
Reaction score
197
Location
Virginia Beach
No 2013 GT500 pump!

I'm going to bucket test one next year I guess vs. the stock Bosch.

Good info, thanks for the tests. Maybe I'll run 3 pumps, since I am running the M90 intercooler core...


I think what this proves is an intercooler upgrade is better than just a pump upgrade as the better intercooler will increase flow with the same pump.
 

Fullboogie

King of the Ski - OG Jr.
Joined
Oct 27, 2008
Posts
8,465
Reaction score
2,858
Location
Houston
Very interesting that the stock GT500 pump outperformed the Mezziere 20gpm pump in your first test! That just saved me $200.
 

CPRsm

forum member
Joined
Jul 11, 2011
Posts
3,043
Reaction score
1
Does the lower restriction IC drop inlets better though? Usually lower restriction is less contact, and less heat transfer. Great info though. Had been looking at the lingenfelter pump trying stretch an intercooler for all it's worth
 

Rasmus

forum member
Joined
Feb 9, 2009
Posts
2,295
Reaction score
26
Does the lower restriction IC drop inlets better though? Usually lower restriction is less contact, and less heat transfer.

I was wondering this as well. What did the actual temps look like?
 

o2sys

forum member
Joined
Jun 26, 2011
Posts
4,367
Reaction score
19
Location
NY/NJ
On the DOB Manifold:

Meziere 20gpm – 7.25gpm

Bosch/GT500 pump – 6.5gpm

Damn, my wallet hurts now...not to mention the increased noise I got from the Meziere 20GPM.
 

blownGTvert

forum member
Joined
Feb 24, 2011
Posts
775
Reaction score
0
Location
Zeeland, MI
No 2013 GT500 pump!

I'm going to bucket test one next year I guess vs. the stock Bosch.

Good info, thanks for the tests. Maybe I'll run 3 pumps, since I am running the M90 intercooler core...

The 2013/14 GT500 innercooler pump is rated @ 55gpm. Would the actual flow rate through the IC match the Meizere 55gpm pump above?
 

Department Of Boost

Alpha Geek
Joined
May 26, 2010
Posts
8,809
Reaction score
28
The 2013/14 GT500 innercooler pump is rated @ 55gpm. Would the actual flow rate through the IC match the Meizere 55gpm pump above?

The Meziere 55gpm pump only flowed 27gpm through 16' of 1 1/2 hose. No IC, no HE.

Actually NONE of the pumps ran what they were supposed to.

I don't trust ANY pump claims.
 

Department Of Boost

Alpha Geek
Joined
May 26, 2010
Posts
8,809
Reaction score
28
I was wondering this as well. What did the actual temps look like?

This was a water flow test rig, not a IAT test rig.

At the end of the day more water = more cooling. The IC doesn't cool anything. It is just a fancy container for water, which is what does the cooling.
 

JeremyH

3V Fuel Guru
S197 Team Member
Joined
Jan 10, 2009
Posts
20,857
Reaction score
197
Location
Virginia Beach
This was a water flow test rig, not a IAT test rig.

At the end of the day more water = more cooling. The IC doesn't cool anything. It is just a fancy container for water, which is what does the cooling.


Not the case, with that the core plates and aluminum fins have to take heat from the air and then the water takes that heat from the aluminum fins and plates so it does play a part. More surface area in the core is also better. The faster you can get colder water into the core the better the delta in temperature which will cool the aluminum and pull heat from the air better. The intercooler media does have a part in it, hence why they are most often made of aluminum which absorbs and transfer heat readily.


This was a water flow test rig, not a IAT test rig.

At the end of the day more water = more cooling. The IC doesn't cool anything. It is just a fancy container for water, which is what does the cooling.


I think the pump claims are with no output restriction and system to push through, since how would the pump manufacturer know what system the pump is going in, which indeed is a bit misleading.
 
Last edited:

CPRsm

forum member
Joined
Jul 11, 2011
Posts
3,043
Reaction score
1
This was a water flow test rig, not a IAT test rig.

At the end of the day more water = more cooling. The IC doesn't cool anything. It is just a fancy container for water, which is what does the cooling.

To a point more water is better. It would flow even more in the water core had 0 fins inside. But the fins are what pull the heat from the air charge side and transfer it to the water.
There was a company a few years ago that made a new cooler for the gt500's claiming boost drop was minimized. We knew that meant the fin count had to be reduced since the core was the same size. Guys did indeed pick up boost, but inlets were terrible. I think more water is better, if compared in the same core. I do not think more water is always better when the core has to be modified for more flow by reducing resistance. I'd be really interested in inlet numbers to compare to.
 

Department Of Boost

Alpha Geek
Joined
May 26, 2010
Posts
8,809
Reaction score
28
Not the case, with that the core plates and aluminum fins have to take heat from the air and then the water takes that heat from the aluminum fins and plates so it does play a part. More surface area in the core is also better. The faster you can get colder water into the core the better the delta in temperature which will cool the aluminum and pull heat from the air better. The intercooler media does have a part in it, hence why they are most often made of aluminum which absorbs and transfer heat readily.





I think the pump claims are with no output restriction and system to push through, since how would the pump manufacturer know what system the pump is going in, which indeed is a bit misleading.

To a point more water is better. It would flow even more in the water core had 0 fins inside. But the fins are what pull the heat from the air charge side and transfer it to the water.
There was a company a few years ago that made a new cooler for the gt500's claiming boost drop was minimized. We knew that meant the fin count had to be reduced since the core was the same size. Guys did indeed pick up boost, but inlets were terrible. I think more water is better, if compared in the same core. I do not think more water is always better when the core has to be modified for more flow by reducing resistance. I'd be really interested in inlet numbers to compare to.


I 100% get where you guys are coming from. That said when working with PD blowers you have certain constraints. The biggest is space. We're pretty much out of it, everyone is. Putting bigger IC cores in is not an option. So that is a dead end road. I also see what you guys are saying about fin count, tube diameter, etc. Of course of you took all of the fins out and had huge tubes it would flow better but it wouldn't transfer heat as fast. That is not what we did or what we will be doing though. We used the exact same Bell core we use in our GT450 manifolds and the first two generations of R-Spec. It has plenty of fins, tube size is right, etc. We did not make improvements by changing the core. When it comes to tube size/fin count, etc that horse has been beaten to death too. There is nothing more to be had there like size. At this point it is what it is and everyone works with the cores that are out there working.

About the core being maxed I'll give you this tidbit. GT500's will boil the water in the IC at about 16-17psi with a stock pump. Clearly they are transferring plenty of heat to the water. The problem is that the water is in the IC to long and is getting cooked. Can't put a bigger core in, there is no room. Don't need more heat transfer, it has plenty of that. It needs more water. It's the only thing that is left to go after.

As far as pushing water through to fast, basically impossible. In theory, on paper it can be done. In practice it's effectively impossible. You would need some sort of positive displacement pump and it would have to run well over 100psi. And that is nowhere near where we are at or will ever get.
 

BruceH

BBB Big Bore Boss 322
Joined
Sep 13, 2009
Posts
13,801
Reaction score
14
Location
Pacific Northwest
So when are people going to start putting baps on the water pumps? Should be an easy way to get 17 or so volts when you need it for more flow. Turn it down for normal driving. It's only going to cost someone $100 or so for a used 20a bap.
 

Latest posts

Support us!

Support Us - Become A Supporting Member Today!

Click Here For Details

Sponsor Links

Banner image
Back
Top