Stock Intercooler - putting some numbers on it

Discussion in 'Car Builds, Projects, Idea's Experiments' started by Mike, Sep 27, 2009.

  1. Mike

    Mike New Member

    May 4, 2009
    403
    3
    0
    Ratings:
    +3 / 0 / -0
    #1 Mike, Sep 27, 2009
    Last edited: Sep 27, 2009
    After reading many testimonials about intercooler upgrades (including a few good back-to-back comparisons), I wondered what I could do to investigate the subject. I'd like to try a bunch of intercoolers and pick the right one for me - or better yet, have a custom one made. However, I only had $25.00. So much for those ideas.

    Still, I thought perhaps I could do something useful. I work at a small aerospace company, and we have access to some equipment (as long as we're considerate about it!). So I decided to instrument a stock intercooler and see what kinds of temperatures and pressures I could generate.

    My original idea was to put together a DAQ (data acquisition system), but all of the company DAQs were tied up in real work. So, I put some pressure taps in a stock intercooler (the used intercooler was purchased for $25.00) and also added two very thin, sheathed thermocouples in the airflow at either end. This way I could read the pressures and temperatures at the inlet and outlet of the intercooler (for the engine intake air) using a craptastically installed set of gauges.

    Here are two pictures of the setup:
    SCsensors1.jpg scgauges1.jpg

    The procedure was as follows:

    1. Warm up the car by driving 15 miles out to the highway
    2. Get on the highway and wait for a "bubble" free from traffic
    3. Perform 10 to 15 50-70 mph runs in second gear (up to redline without speeding).
    4. Record the readings.

    The ambient conditions were:

    Altitude: 4000 ft
    Air temperature: 100 F
    Humidity: 8%
    Winds: very low

    A summary of readings:

    Max inlet (to intercooler) temperature: 284 F
    Max delta temperature (inlet to outlet): 90 F
    Max outlet temperature: 194 F

    Max inlet pressure: 14.0 PSI
    Max delta pressure (loss): 1.1 PSI

    What does this mean? Well, the pressure drop is very low, which is a good thing. We can expect that, though, because the intercooler is so short that there is not much length for the internal fins to cause a large pressure drop.

    Unfortunately, the short intercooler also means that the temperature at the outlet is very high. One way to look at this is to take the difference between the intercooler outlet temperature and the ambient air temperature, a measurement called "approach." On my MCS (a 2003 model), I measured an approach of 94 F. To put this in perspective, large stationary commercial intercoolers typically have an approach of about 20F, and turbo cars with big intercoolers typically have an approach of 40 to 60 F.

    Now, I gamed the system a bit by sticking with 50-70 mph runs, which were the fastest I could do legally. Higher speeds will flow more air through the intercooler, increasing heat transfer. Why, I'd venture to guess that a third gear run might produce an outlet temperature 15 degrees F cooler. It's just a guess, though. Purely hypothetical.

    So, it's no surprise that the stock intercooler is too small for efficient cooling at max power (for the record, my engine has a 15% pulley, a CAI, and an old version of Cartune software installed). None of this indicates specifically what different intercooler or air scoop design would be better, but it definitely does show that the R53 should benefit from a better intake air heat exchanger.

    BTW, here's a brief video showing the ghetto-tastic pressure gauges during the test:

    [ame=""]YouTube - R53 MINI Cooper S stock intercooler delta P[/ame]
     
  2. drPhilGandini

    drPhilGandini MINI of the Month May 2009

    May 4, 2009
    113
    0
    0
    Ratings:
    +0 / 0 / -0
    Very interesting experiment, Dr Mike. and thanks, as all data relating to the R53 intercooler are welcome.
    There are many threads on NAM that report data acquired over the last 5 or so years, starting with Dr Obnxs, and being provided by many of us over the years who have found this question interesting to say the least.
    Much of your reported data "fits" with that which I have reported elsewhere, but I am surprised that you were not able to achieve lower approaches. I never gathered data with my OEM IC, but I regularly observe approaches below 20F, in fact I'm pretty upset when I don't see them this low under steady state conditions. I have seen very high ones under severe conditions, both on the street in 100F ambient temps and stop-start commuter traffic, or on my local track.
    That said, I believe your observations are generally quite consistent with those that have been produced in the past. Glad to see the Central Limit Theorem is alive and well...
     
  3. Mike

    Mike New Member

    May 4, 2009
    403
    3
    0
    Ratings:
    +3 / 0 / -0
    Ah, so this falls in line with other data - great!

    Anyway, I only reported the data for hard accelerations. I've been thinking about track applications, though I was surprised that ten run-ups on the highway with such a mildly-tuned engine would bring 90+ degrees approaches.

    Let's see, I have here an approach for 70 mph steady running, same ambient, slightly uphill. I got 11 degree F delta, even after the heat soaking runs - as you mentioned, below 20 degrees.
     
  4. Dr Obnxs

    Dr Obnxs New Member

    Jun 11, 2009
    1,158
    3
    0
    A Man of Wit and Charm! (Just ask my wife!)
    Woodside, CA, up in the hills and trees.
    Ratings:
    +3 / 0 / -0
    Welcome to the world of IC testing!

    One of the really, really good design features of the stock IC is the pinched tube terminations. Close to every other IC has some form of flat bar facing the air flow, and this creates un-needed pressure drops that hurt the aftermarket offerings.

    You're right that third gear does better than first, but it's hard to test on the street for speed limit issues.

    I'm not a fan of steady state tests. The car isn't really on boost so it's close to a meaningless number. The peak numbers near red-line on power runs are more indicitive of how the IC really is working.

    And yes, it's a toasty environement for sure. Depending on the set up, pre-IC temps of around 300 (sometimes higher) are pretty easy to come by. I think the high post IC temps have a lot to do with the crappy airflow. Improved scoops help with this, but the increased drag means they may be a wash on a fast track.

    Matt
     
  5. TGS91

    TGS91 New Member

    May 8, 2009
    1,593
    18
    0
    Sales Dude
    St. Louis, MO
    Ratings:
    +18 / 0 / -0
    Kind of related. Aftermarket Air/Air IC's are certainly high on allot of peoples mod list.

    Question I have is what real WHP do you see using a Air/Air? 5, maybe 10WHP?

    I know some folks are developing a Water/Air for the R56 which while kinda expensive sounds like a winner with a 20-30WHP boost

    Thoughts?
     
  6. Rixter

    Rixter Well-Known Member

    Jun 14, 2009
    1,230
    79
    48
    Technology Architect
    North of the 49th
    Ratings:
    +79 / 0 / -0
    Hang on a minute here, is TSG91 qualitified to post in this thread, he's not a doctor of anything, unlike Dr Mike, Dr Phil, and Dr Obnxs :D
     
  7. Mike

    Mike New Member

    May 4, 2009
    403
    3
    0
    Ratings:
    +3 / 0 / -0
    Sure he is, everyone's welcome. I'm not a doctor either (it's just a nickname).
     
  8. TGS91

    TGS91 New Member

    May 8, 2009
    1,593
    18
    0
    Sales Dude
    St. Louis, MO
    Ratings:
    +18 / 0 / -0
    well.... I have been told I'm obnoxious! :p
     
  9. phydeaux

    phydeaux New Member

    May 5, 2009
    185
    1
    0
    Maple Valley, WA
    Ratings:
    +1 / 0 / -0
    I'm not a doctor, but I like to play doctor...

    Interesting information, though it's probably going to take me a little bit to really understand what it's telling me. Also good timing, as I was talking to one of our local members at our car show yesterday about how to make improvements to the airflow and cooling. Heat soak was mentioned above, which leads to my question. Would something like aerogel help at all? Or is it possible that there isn't enough benefit to be gained to justify cost in the "bang for the buck" category? I'm also thinking that the heat would have to go SOMEWHERE if aerogel is so good at keeping heat from the IC, but that's just my uneducated opinion on the subjuct.
     
  10. k-huevo

    k-huevo Club Coordinator

    May 6, 2009
    683
    159
    0
    Pipe Creek, Texas
    Ratings:
    +159 / 0 / -0
    First off, thank you Mike, and those analog gauges are way cool!

    This philosophical laden paragraph is not necessarily connected to Mike’s test, so be forewarned this is slightly off topic and may invite more off topic comments. I’ve had a few discussions with Gerhard Schruf at Bell Intercoolers About Bell Intercoolers, about improving the stock MCS intercooler, it is not the size of the OE intercooler that creates its inefficiency, its location is the limiting factor. Gerhard has looked at the R53 up, down, and sideways, if anyone could substantially improve upon the stock IC he could. He concedes improved airflow-capture and routing could be beneficial for the OE location, but the difference compared to front mounting would be small; I’ve proved that in my own testing. He’s looked at a front mount alternative also; due to packaging limitations there are many trade-offs that make it less suitable for the average user. I’ve measured improvement with a GP intercooler and thermal dispersant coatings, but due to the dynamic nature of the driving environment, ultimately it doesn’t make enough difference in the overall scheme of things. It boils down to how you value the deltas. What are the degrees of freedom? What value do you determine is statically significant? Would I rather spend X dollars on a 2% increase in performance output over the stock IC or spend that amount on a 20% improvement that doesn’t involve the IC?

    A water to air intercooler has few application advantages, that is at the drag strip and short duration drifting where the reservoir can be cooled with ice. Bell Intercoolers - Technology and FAQs One could argue the W/A’s radiator location could offset the A/A's inherent advantage, but in the long run heat saturation would interfere. Especially in a location where ambient temps exceed 93F the majority of the year.
     
  11. Dr Obnxs

    Dr Obnxs New Member

    Jun 11, 2009
    1,158
    3
    0
    A Man of Wit and Charm! (Just ask my wife!)
    Woodside, CA, up in the hills and trees.
    Ratings:
    +3 / 0 / -0
    For those that care to read it....

    Check this out. It's some testing I did long, long ago on some top mount ICs.

    What it comes down to:
    • The stock IC is pretty darned good. It has very low thermal mass, and very good air flow charecteristics.
    • Lots of the aftermarket units have too high thermal mass, and have very poor airflow charecteristics.
    • There are lots of ways to get better power per dollar than what aftermarket ICs offer.

    Now, the W2A in an interesting beast for the Mini for sure. There are two heat exchange events, and that's bad. But you can place one of the radiators front mount very easily (Water flows to front mounts much eaiser than intake air). The heat extraction potential of a front mount is really, really good. So good in fact that I'd guess with good implementation, it would out perform a crappy location (like on top of the Tritec engine) for an A2A, no matter how good the core design. There's just not enough cooling air to flow through there. As far as eventual heat soak from a front mount, it seems to me that it's not inevitable, but rather something that should be managed via proper design.

    You can also look at some of the packaging issues with W2A. Look at Defenders of Speeds Laminova core design. There you get very good heat coupling in the engine side water to air in a fairly small package. Look at the Gintani design that didn't use Laminova cores. There you decrease intake volume (better throttle response). While a Gintani geometry unit with Laminova cores wouldn't be cheap, I bet it sure would work really, really well.

    Matt
     
  12. TGS91

    TGS91 New Member

    May 8, 2009
    1,593
    18
    0
    Sales Dude
    St. Louis, MO
    Ratings:
    +18 / 0 / -0
    #12 TGS91, Sep 28, 2009
    Last edited: Sep 28, 2009
    I was referring to Defenders of Speed who have a production R53 Water/Air unit. DoS Gen III W2A Intercooler for R53 MINIs (7 core)

    My understanding is these have been thorughly tested on the MINI and several other makes (Sub WRX's, VW's, etc)

    They have a R56 version in development
     
  13. TGS91

    TGS91 New Member

    May 8, 2009
    1,593
    18
    0
    Sales Dude
    St. Louis, MO
    Ratings:
    +18 / 0 / -0
    Looks like you sent this while I was composing.....What he said! :lol:
     
  14. k-huevo

    k-huevo Club Coordinator

    May 6, 2009
    683
    159
    0
    Pipe Creek, Texas
    Ratings:
    +159 / 0 / -0
    What an expert has to say “…the water intercooler largely stores the heat in the water until off throttle allows a reverse exchange. Some heat is expelled from a front water cooler, but the temperature difference between the water and ambient air is not large enough to drive out much heat. Another way to view the situation is that ultimately the heat removed from the air charge must go into the atmosphere regardless of whether it's from an air intercooler or a water based intercooler. The problem with the water intercooler is that the heat has more barriers to cross to reach the atmosphere than the air intercooler. Like it or not, each barrier represents a resistance to the transfer of heat. The net result; more barriers, less heat transfer.” This is a physical law that must be dealt with no matter the core design of the W/A. We would like to believe in a new design that could defy former held truths, but it hasn’t been created yet, “…virtually all cores will produce essentially the same efficiency results. Perhaps a core with slightly less flow area per linear inch, or one with longer tubes, will need perhaps 5% more tubes to equal the best of intercooler’s with regard to flow loss and efficiency. Not a very important difference.”

    We all place values on products that have nothing to do with dollar/worth ratios; you are welcome to your “added value” choices. When I look at the cost of the intercooler in the link above, I see money that could be spent on modifications that will produce much greater results than marginal increases in inlet air-cooling efficiency could come close to providing.
     
  15. Dr Obnxs

    Dr Obnxs New Member

    Jun 11, 2009
    1,158
    3
    0
    A Man of Wit and Charm! (Just ask my wife!)
    Woodside, CA, up in the hills and trees.
    Ratings:
    +3 / 0 / -0
    I'm a physicist

    and know something about physical laws. It's just math here. It's true that if you had a A2A and a W2A with the ambinet exchange at the same place for both the fact that you'd have two exchanges would screw the pooch for the W2A. But with the ambient exhange in a high pressure high flow area and the A2A on the Mini in a place that basically sucks, it's not the same comparison. The fact that the heat extraction from a larger water radiator in a high flow area can move a lot more heat out of the water is what will win in this situation.

    To be generous, the stock Mini location gets what 10-20 square inches of air flow into the scoop? A W2A radiator in the front of the car gets what about 100 square inches? And that air flow is normal to the radiator, while the air flow past the scoop isn't normal to the scoop. That makes the effective ratio of airflow even worse for the stock location.

    This back of the envelope calculation means that the W2A radiator just has to remove 1/5th the heat per unit area to be as effective as our A2A. That's not much of a challenge.

    There's another thing with W2A that most don't think about. With A2A, it's really all about on boost cooling, as the thermal mass (if the IC is good) is pretty low. That's a nice thing about the stock IC, it recovers from heat soak pretty well. Now because the W2A has a heat resevoir (the water) it can still shed heat from the system when the car isn't on boost, as long as it's moving. The math gets really messy here, but this means that depending on boost duty cycle W2A systems have the potential to burn off the stored heat in ways that the A2A never could. Yes, it also means that there are use cases (high duty cycle, low speeds) that can store a lot more heat than A2As will, and take longer to recover. But the only think I can think of where this would really be the case is in a poorly vented dyno room, or some really, really turny course that had very long lap times or very short breaks between runs.

    Bell makes really good cores, but in this case, with the geometry of the top mount on the Mini, I don't think it's correct to say that A2A is the better way to go. It's a cheaper way to go, and does OK for most, but I think the best way to go for our cars, money no object, is a well designed W2A. It's just too bad I don't have the "no object" money!

    Matt
     
  16. BlimeyCabrio

    BlimeyCabrio Oscar Goldman of MINIs
    Lifetime Supporter

    May 4, 2009
    3,532
    2,896
    113
    Professional Facilitator and Alignment Consultant
    Holly Springs, NC
    Ratings:
    +2,896 / 0 / -0
    MINI Intercooler discussions are so addictive... I just can't get enough, no matter how many times we've had the same discussion...

    I will say... 90 degree approaches are why there's such an opportunity for water/meth injection, when done right...
     
  17. TGS91

    TGS91 New Member

    May 8, 2009
    1,593
    18
    0
    Sales Dude
    St. Louis, MO
    Ratings:
    +18 / 0 / -0
    Right now I have a stock IC on a R56 and debating adding a A2A or W2A

    A2A is going to run me $700-$800 and a W2A is going to run $3K

    My question is bang for the buck. I have never really got clear on actual HP gains. I *think* a A2A is going to get me 5-10WHP and I have heard claims of 20-30WHP on the W2A. Is that fairly accuarate?
     
  18. drPhilGandini

    drPhilGandini MINI of the Month May 2009

    May 4, 2009
    113
    0
    0
    Ratings:
    +0 / 0 / -0
    There might be a slight disconnect here (or I'm just showing my ignorance of turbo powered cars.) Although an IC works regardless of the source of heat (SC or Turbo), the people posting here all have experience with R53s, including the original post by DrMike. You might expect to get different results with an R56 for many reasons. The OEM IC on an R56 is already front mounted in a relatively high flow location--the R53 IC is mounted on top of the engine and lies flat. Hello...can we say "bad?"
    That said, I'm with Paul (BlimeyCabrio), these discussions are always enjoyable, especially when the big guys start swinging the bats. DrObnxs has done things with thermal coating on the IC, and they seem (relatively) cost effective, although most people would know by now that I share Keith's (KHeuvo) POV that the IC is perhaps not the best place to spend money expecting a great return.
     
  19. TGS91

    TGS91 New Member

    May 8, 2009
    1,593
    18
    0
    Sales Dude
    St. Louis, MO
    Ratings:
    +18 / 0 / -0
    I'm comin to that same conclusion Dr. Phil. I'm hoping more R56 owners come over to MA to have these types of discussions
     
  20. minispeedrcr

    minispeedrcr New Member

    May 20, 2009
    54
    0
    0
    Ratings:
    +0 / 0 / -0
    I'm running the Gintani on my race car, not sure if its the same configuration as over the counter. Mine has a really large front mount radiator and a large resivoir tank in the rear of the car that holds more than a gallon of water. The only test I have done is with a scan guage that read IAT. This was done at Cal Speedway in june with 95 degree outside temps. During the race I was getting consistant temps of 124deg. Race is 30 minutes long.
    vince
     

Share This Page