Getting the Most From 91 Octane Pump Gas with a Honda K Motor, Part Four

As we mentioned in previous articles, the Hondata K-Pro is a user-programmable stand alone ECU that fits in the factory ECU case.  It is a modified OEM ECU.  The core ECU must be gotten from a junkyard.  It is from a 2004 RSX Type S. 2002-2004 MT K20A2 is the one to look for.

Here Shawn gets on his keyboard and works his magic on our K-Pro. Shawn is a careful meticulous tuner with a lot of K engine experience.


So the results!  On 91 octane pee water, our engine pumped out 303 whp and 213 lb/ft of torque, pretty impressive with a car with full exhaust and a long intake.  The interesting thing is with the addition of the high flow cat, the power dropped to 260 whp and 181 lb/ft!  This shows how much back pressure can kill a well built K Engine!

Another interesting thing is we noted that the engine was pulling at 10 in/hg vacuum in the intake manifold at wide-open throttle. This indicates that our engine was seeing a significant intake restriction at high rpm.  This engine has a longer intake and exhaust than the engine we built for an Ariel Atom a bit back and made 9 less hp although that engines dry-sump might have been part of the reason for the added horsepower.

We feel that the inlet restriction was costing us quite a bit of power and in future builds, we will be using a larger 80mm throttle body, porting the inlet of the intake manifold larger and going up to a 3.25″ diameter intake pipe.  We feel that this may yield at least 10 more hp. Another thing we would like to try is flex-fuel and running E85. This would add probably 15-20 more hp as well as enable us to up the compression ratio more.

We are still really pleased with our engine, it is really amazing fast, fuss-free with a wide powerband.  Vtec allows it to idle smoothly and have good bottom end with a powerband that extends to 8500 rpm. We really can’t wait to apply the lessons we have learned with this build to another K Series engine! Stay tuned, now that we have an outstanding engine in our RSX project car, we will next work on the suspension and brakes to make this a well-balanced street sleeper!

For more on our Project DC5 RSX check out here!


CSF Radiators

Church Automotive Testing

Eimer Engineering


XForce USA

Canton Racing Products


Drag Cartel

Supertech Performance



Advanced Street Performance

PRL Motorsports

Earls Performance Plumbing

JE Pistons

K1 Technologies

King Bearings

4-Piston Racing


WPC Treatment




    1. Well, a mass airflow meter car should always have a recirculating system. Speed density you can do what you want. A track driven car might need a vent to atmosphere system because it is run at high loads and RPM a lot and thus have to deal with more blowby.

      1. if I have a track car with a recirculating system (mass airflow), what would indicate that switching to a vented system would be a good idea? Lots of oil in my catch cans? Anything else?

          1. Can you give us any more details on the cat? IE: diameter, CPI?
            I’m really surprised to see that much of a difference, but obviously it messes with the reflected wave too (I remember that being mentioned in the Porsche build too)

          2. Magnaflow P/N 59959 inlet and outlet 3″ body diameter 4″. The cell count is not specified.

  1. Happy to see this car finally hit the dyno. I went and bought my first Honda, a 2010 Si coupe, partly because of the first articles of this project back in november ( and because my winter car died ). On paper they are very similar cars but the RSX is way prettier, I do like the close ratio gears on the Si though.

    Does the DC5 share any of the suspension and steering problems of the 7th gen civic?

      1. Ooof, well my 8th gen isn’t much better, the drivetrain is 90% of the reason I bought it, I’m not swapping my miata as a summer car.

        I might do sway bars and coilovers to help with the handling. The steering feel is tragic but I’m not sure there is anything I can do about it.

    1. @Nicolas, I’d recommend doing the Hondata Flashpro ASAP. Buddy of mine has a 2008 Civic SI and the throttle hang when shifting is horrendous. Of course, the Flashpro does a great job improving throttle response and bumping up power a bit. Half the reason I got the Hondata Kpro for Project S2000 was to get rid of the VTEC torque spike. So annoying.

  2. Mike, does anybody sell a black anodized aluminum radiator? I would think that a black radiator would be far superior at absorbing heat than a reflective one.

    Actually, I’m really surprised that they don’t have black anodized radiators in motorsports like F1.

    If anybody can do a ‘back-of-the-envelope’ calc, I would love to see it. I guess you would just need to know the surface area of a typical radiator. Too lazy to do it myself, but if I was doing an ultimate ‘motorsport’ build and they sold black radiators, I would absolutely study the price difference and see if it was worth it.

    From my Heat Xfer course in college:

    “Black objects, on the other hand, generally absorb and emit radiation very well, and have emissivities close to 1. An object with ε = 1 is called a perfect blackbody. The best absorbers are also the best emitters. Black objects heat up faster than shiny ones, but they cool down faster too.”

    1. Black body radiation is also about radiation – it doesn’t have any significant effect on convection which is the dominant heat transfer mechanism going on here.

    2. the way I understand it, black only helps heat transfer when the only acting factor on it is a temperature difference. as soon as you introduce other factors, such as air moving through the radiator, the color emissivity is so small it becomes irrelevant.

      Another thing to keep in mind, is color emissivity goes both ways, so if you have stuff thats hotter than the radiator (turbo, headers, etc), it’ll absorb heat faster too.

      1. Just FYI, but I was thinking this would probably be more of a benefit in motorsports. In F1, cars sit on the grid with no movement for a significant amount of time. I would imagine that any kind of improvement would be worth while. You could also just paint the radiator black, but I’ve never seen even that in F1.

  3. I ran some general numbers and came up with a thermal resistance increase from anodizing of the radiator of about 2% at max power for the car in this article. This translates to a coolant temperature increase of about 1 degC over a new unanodized radiator and is probably comparable to the increase in thermal resistance after the radiator surface has oxidized for a year or so. The reason for this result is that the thickness of the anodizing layer is small (~0.1mm) so, even with poor thermal conductivity, the impact is small due to the thin layer and the large effective area of radiator cooling fins that it is spread over.

    1. Of course, it’s probably not worth it for the typical enthusiast. But, in motorsports like F1, 1-2% is actually significant. It only prove the point that a motorsport radiator should be either painted black, or anodized black. I would probably get some serious kudos if I was working for an F1 team, because 2% can be reduced to a significant lap-time improvent.

      In racing, strangely enough, it’s the littlest detaiils that have the most benefit.

  4. No, I just used the thermal conductivity of alumina. As one of the previous commenters said the radiative effects of black are small compared to the conduction; there may even be some extra heat absorbed from the engine because some engine parts are hotter than the radiator though the fan and shroud will block most of it. If the anodizing is colored with a dye, the thermal conductivity of the anodizing layer will be largely unaffected; if the black has a carbon content the thermal conductivity of the anodizing layer will be increased by an amount that depends on a number of factors and the offset from no anodizing will be reduced. Uncolored/ no carbon loading is the worst case.

  5. To settle the point I ran the numbers for black radiative cooling and came up with 0.01% of the dissipated power in the radiator. assuming that 50% of the radiator intake surface was emitting.

      1. Motorpsorts are like anything else in life. A small initial change can result an a huge difference when it is multiplied in time. Honestly, there are aeronatical enginners chasing fractions of a second on a lap. Nothing is wasted in this endeavor. Every improvment is important over the course of a 2hr race, those improvements are amplified non-linearly.

      2. Adding a layer of anodizing reduces thermal transfer from convection (i.e. heat removal to the air flowing through the radiator) by about 2% at max output power for the car in your article. There’s a reduction in the reduction of about 0.01% from radiation from the front of the radiator, leaving about 1.99% degradation. The net result is still a small reduction in radiator performance compared to what it will do when it’s new and shiny; it will probably perform about the same as it will after the car has been in use for some time without anodizing and the radiator is no longer shiny but grayer and oxidized. Anodizing also has a smaller coefficient of thermal expansion than aluminum, which means it cracks when cycled over the temperature ranges seen in car radiators. So you’ll end up with a coating that has cracks in it and atmospheric oxidation growing underneath the anodizing in patches, further reducing radiator effectiveness. Will it make any practical difference? Probably not for a street car unless you’re pushing the cooling system to it’s limits, but spending money on something that degrades performance and will look bad after normal use doesn’t seem very useful (except, maybe, for a show car that is hardly run).
        As the last commenter said, motorsports are a different matter. All the pictures that I could find of F1 car cooling systems use bare aluminum radiators.

  6. With some things, observation provides the answer without having to do any math. If F1 cars and LMP cars don’t have black radiators, it’s because it doesn’t help. Why does it not help? Possibly the process of creating the black surface hurts convection heat transfer more than gaining from radiation. And/or, the surrounding environment isn’t cold enough. Or worse, the surrounding environment is hotter than the radiators as could be the case with tightly packaged engine bays with 1000degC exhaust manifolds and making the heat exchangers black actually hurts performance.

    1. Oh yeah, MotoGP bikes also do not have black radiators. Again, the absolute pinnacle of racing with extremely high budgets.

  7. The part number XF0-KST-UNV-350 appears to be for the K Tuned thermostat housing. Any chance you could double check the part number for the Varex muffler?

  8. 303whp NA from a 2.4L street engine on pump gas is just bonkers. Imagine telling people you would be able to do this 20 years ago. They would not believe it could be street-able in any way.

    1. The amazing thing is this is all easily available off the shelf parts! No engineering, experimental specs and months of waiting for special stuff!

  9. It still seems bonkers!
    At least it appears to be a hub dyno so we don’t have the tyre/roller losses that are normally more significant than transmission losses.
    Does the dyno has a means of calibration that’s done regularly?

    If not, has the build or similar been on a calibrated engine dyno?

    It’s a result that you should be proud of with any pump style fuel.

Leave a Reply

Your email address will not be published. Required fields are marked *