The Great Intercooler Water Pump Test

Discussion in 'Mustang Chit Chat' started by Department Of Boost, Oct 22, 2014.

  1. DiMora

    DiMora More Is Better

    828
    9
    If somebody sends me one I’ll test it and return it.
     
  2. Pentalab

    Pentalab forum member

    2,541
    104
    I would have thought that a flow meter in a closed loop would be the real deal, vs a return line into a calibrated bucket ? It would be interesting to compare the 2 methods. IF the calibrated bucket method resulted in slightly more flow, that could be factored into the final equation. IE: '.92 x 6 gpm = 5.52 gpm'.
     
  3. DiMora

    DiMora More Is Better

    828
    9
    It works fine...and you can easily see flow rate differences:

    Afco dual-pass:

    [​IMG]

    Steeda triple-pass:

    [​IMG]
     
    Last edited: Feb 8, 2019
  4. DiMora

    DiMora More Is Better

    828
    9
    Jason / DOB has correctly identified that intercooler system plumbing diameter and inlet and outlet size is absolutely CRITICAL along with choosing a good pump in order to get the best flow rates. I love that they are moving to 1.25" inlets and outlets in their intercooler / manifold. That is HUGE...here is why:

    I had an amazing physics teacher in high school - Mr. Byron Anderson. I hope he is doing well. Anyway, he had a talent for taking physics laws and having us apply them to real life. Remembering "Byrophysics" as we called it, I started studying up on this topic last night, and here is what I found:

    I need to choose an amazing pump and get all the plumbing that I can control up to the largest diameter I can. Even if there is a restriction in the system (the intercooler itself with 3/4" plumbing), diameter is king!

    You could stop reading right there, but here is why if you want to go deep with me...and I'll summarize my plan at the end.

    Poiseuille's law (Hagen-Poiseuille equation) is what governs what is going on with our supercharger intercooler systems...it governs fluid flow in a pipe. Don't quit reading when you see the formula below...it is quite easy to understand.

    ΔP = 8μLQ / πR4


    or

    Q = ΔP πR4 / 8Lμ

    Where:

    ΔP is the pressure difference between the two ends, L is the length of pipe, μ is the dynamic viscosity, Q is the volumetric flow rate, R is the pipe radius.

    Here is what it distills down to as it applies to our systems:

    • ΔP (pressure) is going to be directly impacted by what fluid pump we choose and the inlet / outlet diameter of that pump as well as how that pump performs with flow resistance in the rest of the system
    • L (pipe length) is more-or-less "fixed" in our cars; the only impact we can have on that is by routing our hoses "smartly" to avoid excess hose length when not necessary. Perhaps mount the pump as close to the heat exchanger inlet as possible...or as close to the intercooler inlet as possible...the bottom line is avoid excessive plumbing length if not necessary.
    • μ (dynamic viscosity) we have very little control over...we are using water / glycol (antifreeze) and perhaps water-wetter. The higher the viscosity (thicker the fluid) the worse flow rate. Water is thinner than glycol...so using the least amount of glycol for our environmental conditions that we can get away with is best (don't use so little that your intercooling system freezes in winter and you destroy things)
    • Q (volumetric flow rate) is what we are really trying to "solve" for...so all the other parameters in the formula impact Q
    • R (Pipe radius) we *can* control up to some extent...more on this in a second, as you will soon see why that item is the most critical along with pump selection. Here is the most important part of R. Notice in the formula that it is to the 4th power. That means that compared to the other variables, radius of the pipe has the absolute largest impact on our flow rate. Layman's terms points concerning radius:
    1. Fluid in a pipe moves slower near the walls of the pipe and faster in the middle.
    2. For a given original flow rate of a given fluid...the following are true:
    • Double the length of the pipe and the flow rate is cut in half
    • Double the viscosity of the fluid and the flow rate is cut in half
    • Double the system pressure and the flow rate doubles (Pump selection!)
    • Double the pipe radius and the flow rate increases by 16 (Plumbing diameter!)
      • A 19% increase in pipe radius doubles flow rate
    I found this demo online that summarizes this nicely here. Check this out:

    [​IMG]

    See the two tanks?

    The top one with the 6MM drain capillary will empty at the exact same time as the lower one that has 16 3MM bore capillary tubes!

    Summary:
    • Choosing the largest diameter plumbing in the system that is possible can increase flow rate. Even if there is a restriction in part of the system that we cannot overcome without replacing the intercooler (3/4" inlet / outlet) decreasing the negative impact of small pipe diameter in the rest of the system should increase overall flow rate by mitigating flow resistance of the coolant at the pipe / hose walls
    • Choosing a pump that flows a high volume and works with with a restriction in the system is critical. As the experiments of others have proven, a pump that flows well with no restriction (Meziere) does not do so well with resistance (20 GPM unrestricted vs. 3 - 3.5 GPM in a restricted 3/4" system)
    • Use the shortest plumbing you can get away with...route hoses wisely
    • Use the least amount of glycol you can get away with based on how cold it gets where you live (impact of this is minor compared to plumbing hose diameter and pump selection)
    My long-term plan:

    • Install Stewart-EMP 25A pump because it is current king-of-the-hill for a car with a 3/4" plumbed intercooler
    • I'm considering 1.25" inlets and outlets using -20AN fittings and Earl's elbows to 1.25" lines...this would entail finding a local TIG welder and adding -20AN bungs to my Canton ice-box inlet and outlet and my Steeda triple-pass heat exchanger. I'd use 1.25" to 1" reducers at the inlet and outlet of the Stewart EMP pump, or I could also have my welder add -20AN fittings to the Stewart, which would probably be best.

    In the short term, I'll get the Stewart-EMP installed and report back my flow rate with no other changes.

    Shane
     
    Last edited: Feb 10, 2019
    Department Of Boost likes this.
  5. 06 T-RED S/C GT

    06 T-RED S/C GT forum member

    1,222
    57
    This isn't related to the thread post, however, I submitted a PM to your inbox 3 days ago, but have not received a response back. Therefore I'm re-submitting what was posted in the PM on this thread, just in case for some reason you didn't receive it.


    I came across a "cut n clamp" ProChamber installation you did from nearly 5 years ago and wanted to get some feedback from you. I have a 2006 Mustang GT with a Saleen series VI supercharger and have been running the Mac off/road ProChamber with Borla S-Type axle backs (stingers) along with the factory exhaust manifolds. Although I really like the overall sound of the Borla's, I cannot stand the interior resonance drone I get from them between 1500-2500 RPM while at cruising and also when accelerating. Anyhow when I read your ProChamber (How-To) install, I became very curious and began thinking that perhaps by adding the factory cats to the ProChamber just might be the solution for either significantly reducing the interior resonance drone or possibly even eliminate it altogether? However, on the flip side, I also have this concern that if adding the factory cats to the ProChamber? that it may end up causing the overall sound of the exhaust to be too quiet, like similar to the factory exhaust. At any rate, I'd be very interested in getting your feedback concerning both the pros and cons along with any opinions/advice that you might be able to provide as to whether or not I could benefit from adding the factory cats to the Mac ProChamber. Thanks in advance.
     
  6. Department Of Boost

    Department Of Boost Alpha Geek

    8,780
    6
    If you run all 1.25" line you will get about 12gpm with the Stewart.
     
  7. DiMora

    DiMora More Is Better

    828
    9
    That’s over 3x what I’m getting now.

    Do you guys cut off the EMP necks and weld on 1.25” (-20AN) fittings - or do you just run a tapered 1” to 1.25” silicone hose then a 1.25” coupler to plumb the pump to the rest of the lines?
     
  8. Department Of Boost

    Department Of Boost Alpha Geek

    8,780
    6
    If you were to run one of my new manifolds (have you seen the new 3v manifold?) with 1.25" lines you would get 7x the water flow you're getting now (28gpm).:eek: My new manifolds are seeing IAT's 8-12deg over ambient in traffic/surface streets and only gain 15-20deg over a 15sec WOT run. Then they recover instantly. They're good enough to run at the road course with zero IAT issues.

    The new manifold would directly replace your Roush one.

    I've done it both ways.
     
  9. 1950StangJump$

    1950StangJump$ forum member

    364
    18
    So . . . I have this Ford Performance pump on my KB Stage 1:

    https://www.americanmuscle.com/frpp-intercooler-pump-cobra.html

    It's running through a VMP Triple Pass. I realize there is a lot that goes into flow rates (and thus cooling abilities), including hose size, bends, and the smaller KB IC inlet/outlets.

    But, will the 13/14 GT500 or the Stewart-EMP simply plug and play in place of my current pump to increase flow substantially? I thought I read where the GT500 flowed so much, it needed a bypass of sorts? Even saw where some argue that flow too fast could actually hamper cooling ability. Perhaps that was speculative info when the pump first became available?

    I'm just now starting to put some thought into this upgrade, so forgive me if the question is stupid.
     
  10. Department Of Boost

    Department Of Boost Alpha Geek

    8,780
    6


    The KB has the absolute highest flow restriction of any intercooler. An optimistic flow number with that pump is 4gpm.

    Plug and play, no. You will need to do some work on mounting and wiring them in.

    The 13' GT500 pump will get you about 7gpm. The Stewart 8gpm. They increase substantially by percentage. But where you end up is still dismally slow for water speed. You won't need a partial bypass at flow rates that low.

    It needs one, but barely.

    That person also believes the Earth is flat. You can't move the water too fast. Period, end of story. Anyone says that you can is talking from a position of complete ignorance.

    Every time we increase flow rate the performance increases. We have tested all the way up to 56gpm.

    The Bugatti Chiron moves 200gpm. That car will go WOT for 9min without having temp issues (it runs out of gas at 9min). It's the only car on the planet that can do that. It's safe to say the Bugatti engineers didn't overcome the massive design hurdles to run 200gpm because they had extra time/money and did it because it would be fun.

    Your questions aren't stupid. But, you can ask questions of stupid people. And get the wrong answers.:D
     
  11. DiMora

    DiMora More Is Better

    828
    9
    Yes, I've seen it. I think that will make a great future mod for me. For now, I need to get my 5.0 stroker short-block installed, then when I turn up the boost it will really matter. I'm only giving it 9 PSI right now...my real desire here with the intercooling mods is to get the "supporting" mods in place...plus it's all pretty cheap...I'll drop under a grand for the pump and all the 1.25" plumbing and welding.

    I've installed a DOB setup before. Good product, I thought it was well built and it performed nicely. It howled like blowing air across a coke bottle, but i understand a later design revision changed that. I wouldn't hesitate to buy one of the R-Specs in the future with the giant intercooler and 1.25" lines.

    Pertaining to hampering cooling by flowing the fluid too quickly - you simply cannot flow the water too quickly in our systems. I suppose there is some velocity that is so great that it causes too much water turbulence and the pump doesn't run right due to trapped air bubbles and cavitates, but the guys that post about "dwell time" and "soak time" for the coolant to absorb heat have forgotten what they learned in physics. Remember that per Isaac Newton, the rate of temp change is proportional to temp differential between an object and its surroundings. So...the cooler we can keep the "surroundings" of the intercooler fins (the coolant), the faster rate change (cooling) we experience.

    Pertaining to the GT500 degas bottle, it has inlets and outlets that are less than 3/4"...they take 3/4" hose, but they are thick-walled plastic and they adversely impact flow rate. The depth of the tank and orientation of the inlets and outlets also can cause the coolant to "froth up" at high velocities...the purpose of the "degas" bottle is to do just that...let any bubbles in the coolant escape... When you put a big pump on one of those tanks with no bypass, the flow rate is so high that bubbles are actually added into the circulating coolant and the trapped bubbles get into the coolant pump and cause cavitation (and bubbles in the heat exchanger and intercooler are also bad, as coolant is not in contact with the internal fins where the bubbles are). There are two solutions: Get a bigger tank with bigger inlets and outlets (like a Canton, Moroso, or trunk ice box) that degasses well at high flow rates, or run a partial bypass line so some of the coolant goes around the degas tank...this allows the degas tank not to froth up and add air bubbles...but it allows any air bubbles that do pass through to percolate up and escape.

    Lastly, your existing pump is a standard Bosch-type pump that flows 2.5 to 3.5 GPM. You definitely would see a performance enhancement by upgrading. Plug and play? I would add a relay and a fused direct power line to the fusebox instead of trying to power it off the existing wire if I went with the Stewart EMP (that's what I will be doing). You might get by with plug-and-play on the GT500 pump, but the Stewart EMP draws ~17 amps nominal and is rated up to 25, so I'd give it some good power with a relay. If you don't know how to add a relay, I can draw you a picture and post it. You can buy a relay at any auto parts store.
     
    Last edited: Feb 11, 2019
  12. 1950StangJump$

    1950StangJump$ forum member

    364
    18
    Hey, thanks. I knew the mount would be different, and I assume the connector is different, so I'll just have to splice and solder in a compatible connection?

    So, once that's done, adding the GT500 pump alone and keeping all other things equal (3/4" line, KB IC, et), I'll see roughly a 75% increase in flow above the Ford Performance pump I linked?
     
  13. DiMora

    DiMora More Is Better

    828
    9
    Yes (splice and solder) on the connection. You may have to fabricate a bracket for mounting or if it comes with a bracket, mount it in a new location. DOB will have to answer about flow rate; I've not tested a GT500 pump. The Stewart EMP flows more than the GT500 pump per a PM DOB sent me, but it is also a lot more expensive. I paid the extra $ because I'm trying to buy things once...I don't want to have the itch to upgrade my pump again later to get an extra GPM or two out of it.
     
  14. 1950StangJump$

    1950StangJump$ forum member

    364
    18
    Thanks for the responses. I would probably use this opportunity to also upgrade to a Canton overflow tank from the plastic KB tank (which is an eyesore anyway). I guess the trick would be to monitor for "frothing" that could jam up the pump. It would suck to do this upgrade, hoping to double flow rate, only to get no flow rate because of cavitation.
     
  15. Department Of Boost

    Department Of Boost Alpha Geek

    8,780
    6
    The standard GT450 (the new one I mentioned) has the giant intercooler and 1.25" lines now.



    There aren't any pump available to move water so fast as to create undesirable turbulence and cavitation. On paper you can move water too fat. In reality you can't get close to that.


    I'm guessing most of them never had physics.


    At 20gpm+ you have one-ish choice. Partial bypass.

    If someone has a tank it needs at least 5gal of water volume WITH VERY elaborate baffling. Not many people run trunk tanks.
     
  16. 1950StangJump$

    1950StangJump$ forum member

    364
    18
    I'm confused. If this is the case, why does a stock 2013 GT500 run a bypass?
     
  17. Department Of Boost

    Department Of Boost Alpha Geek

    8,780
    6
    He was talking about cavitation/turbulence in the system as a whole.

    The partial bypass is for cavitation/turbulence in just the de-gas tank.
     
  18. 1950StangJump$

    1950StangJump$ forum member

    364
    18
    Gotcha. Not to make it about me ... but if I added the GT500 pump, given the flow rate would likely only go from 3ish to 7ish gpm anyway, I wouldn't need a bypass, right?
     
  19. Department Of Boost

    Department Of Boost Alpha Geek

    8,780
    6
  20. 1950StangJump$

    1950StangJump$ forum member

    364
    18