I've set up this simple forum as a back-up for the 3G community on

For ease of use you can quickly log in using your Facebook account, up there /\ OR just register an account like usual.

B21a1 Naturally Aspirated Build

General '88 - '91 Prelude Discussion
Post Reply
User avatar
Site Admin
Posts: 113
Joined: 16 Aug 2017, 20:34
Location: Mount Vernon, WA

B21a1 Naturally Aspirated Build


Post by MustardCat » 20 Aug 2017, 04:06

Chris Brinson wrote:

3G Prelude Naturally Aspirated Build Up
By: Chris Brinson

The first engine modification I did on my 1990 Prelude Si 4WS was an intake. I did not get dyno results for the intake but to be totally honest I would be surprised if this modification yielded any significant gains. It mostly just cleaned up the engine bay and the larger diameter intake pipe might have yielded some high end gain.

Next up was an MSD ignition that again, probably did not yield any substantial gains. I did not need to upgrade my ignition at this point but since the price was right (it was free) I did.

It was at this point that I gained access to a dyno or a "chassis dynamometer". For those new to auto performance and tuning this is simply a device used to measure horsepower at the wheels. There are also engine dynamometers that measure horsepower at the flywheel. Unfortunately for us, auto manufacturers measure horsepower at the flywheel. Thus, a 3rd generation Prelude's B21A1 engine's 140 horsepower factory rating is only about 110 wheel horsepower. Not very impressive for a 2.05 Liter engine.

It is important when looking into building up a 3rd gen Prelude engine to know what you are getting into. Frankly, there is nothing impressive about the design of the old B20/B21 series engine. (There is a new Honda B20 engine that has almost nothing in common with the old B20. Thus, I say the "old" B20 engine for the 3rd gen Prelude engine.) The horsepower per liter of the B20/B21 series engine is very low, they do not respond well to naturally aspirated modifications, torque is not great and the engines tend to go through piston rings rather quickly. (The piston ring problem is supposedly due to high piston speeds due to a poor rod/stroke ratio. I am not technically savvy enough to get into a discussion of rod/stroke ratio so I will leave that to the experts.)

Well, given all these negatives is it even worth trying to increase the performance of an old B20/B21 engine? Sure. I just think that it is good to set realistic expectations. If you want to make over 150 HP to the wheels with a naturally aspirated B20/B21 engine and still idle at stop lights, I would suggest reading the remainder of this article and thinking long and hard first.

After the intake and ignition, my Prelude's next modification was a DC Sports header. Since I now had access to a dyno, I first did a baseline run. The baseline run was a depressing 109.5 horsepower at 114.6 ft/lbs. of torque. This number was with a completely stock engine with the exception of the custom intake and MSD ignition. Member "Brady" then helped me install the DC Sports header. The header was used and when it came time to thread the EGR tube into the threads on the header, it would not thread. Not to be discouraged, we just welded a plate over the connection for the EGR tube on the header and capped off the EGR tube itself. Done. We figured the worst thing that would happen is that we would make more power.

The car was driven back over to the dyno and put down a much improved 121.2 HP and 123.8 ft/lbs of torque. Now that is bang for the buck. 11.7 HP and 9.1 ft/lbs. of torque in just an hour or so of work. This turned out to be the most impressive mod we performed on this car. The factory header is just plain garbage. You can see from the dyno sheet that the DC header makes much more power throughout the rpm range. The only draw back is increased noise inside the car.

Next up was a set of Unorthodox Racing cam gears. I can't say enough about the quality of these gears. I had them installed by a Porsche shop in town and the owner raved about the high quality of these CNCed parts. As a side note, these gears are the exact same cam gears used on the B16/B18 engines used in the Civic and Integra. It seems that this is about the only thing that the old B20/B21 engines have in common with B16/B18 series engines.

I was not able to dyno the car immediately before and after installing the cam gears. However, I would not expect merely installing the cam gears to generate any significant power. The purpose of the cam gears is to allow you to advance or retard the timing on the intake or exhaust cam. First we set the cam gears to +0/-0 IN/EX to get a baseline run. Since I am an avid road racer our goal in tuning was to gain power at the upper end of the rpm range. However, no matter how we set the cam gears we could not make power in the upper RPM range. Thus we chose the setting that gave us the greatest gains throughout the power curve. This happened to be +4/-4 I/E on my car. Although it was not our original goal, we made huge gains in low end torque. According to the dyno, we gained over 10 ft/lbs. or 10% under 3500 rpm. (Horsepower shown in graph). The gains were good up through 4700 RPM where the gains totally disappear. The difference in the car on the street was incredible. I found myself accidentally spinning tires pulling away from stoplights until I got used to the increased low end torque. This was yet another excellent modification that had no negative impact on drivability or maintenance.

At this point Brady and I became concerned about our inability to increase power in the higher RPM range. For a road race car this is where you want to be making power. We decided that perhaps the restrictive stock exhaust was to blame. I purchased an inexpensive free flowing muffler and took it to an exhaust shop. They custom bent a 2-1/4" exhaust and welded on the muffler. We dynoed the car with the new exhaust and though it made a few horsepower the high end power that we were searching for still did not appear. We decided to try an adjustable fuel regulator and pressure gauge to see if perhaps the fuel was leaning out at the top end and thus limiting high RPM power. Once again the car was strapped onto the dyno and numerous fuel pressure settings were tried. It was found that there was almost no difference in power no matter what the fuel pressure setting so we just set it back to stock.

Out of bolt on options, I decided to order a set of more aggressive cams and send a spare B21A1 head off for porting and polishing. We did not opt for larger intake or exhaust valves, as the car should remain streetable. We got the head back first and had a local shop swap it on. About $1200 later the head lost 2 HP over our previous dyno session. More than likely, the head really did nothing at all as it is normal for dyno numbers to vary by a couple horsepower between sessions. Now we were frustrated. What if we had gone with larger valves? What if the head was sent to someone with more experience with Honda heads? I personally think that almost no one could do anything to increase power on that stock head but who knows?

Our last attempt to try to make naturally aspirated power on this engine was a set of Crower regrinds. We sent a set of cores off to Crower and they sent us back the regrinds very quickly. I also recall that the price was extremely reasonable. However, when we swapped those cams in and dynoed the car it was more of the same. At best, the cams made 3 horsepower and the top end power didn't budge. The graph shows the slight loss in power from our last dyno with new head in red and the slight gain in power from the cams in green. We did dyno the head and cams on different wheels and tires from the last dyno. In all likelihood, the head did absolutely nothing. Either way, it was a huge waste of funds. That is when I gave up on N/A 3rd gen engines permanently and went to the dark side.

N/A 3rd gen with intake, exhaust, header, cam gears, cams, headwork, FPR, ignition and hours upon hours of tuning time vs. a turbo kit installed the day before with no tuning whatsoever. Look for an article on the turbo install next.

Post Reply