Honda DPFI vs MPFI...
 

OK, I have an '88 CRX DX with a D15B2 and DPFI. From a performance and modification standpoint, everybody seems to recommend upgrading to MPFI, but more than a few people have warned me that it will result in a net loss of fuel economy.

But if I'm not mistaken, the CRX HF was equipped with MPFI. So I'm wondering if the loss in FE that I've been warned about has just been a result of heavy right feet in the drivers in question. Does anybody here have any experience with the DPFI to MPFI swap? I'm especially interested in finding out what happens when driven judiciously and with FE in mind....

The conversion appears to be relatively cheap and easy. So my big issue would be whether it will help or hurt me in the FE dept. Any input will be helpful....

I would venture to believe that its the right foot of the driver... Way back when people used to tell me that cars with 4 bbl carbs were the worst on mileage... In my experience my vehicles with 4bbl carbs got better mileage than the ones with 2bbl carbs. It all has to do with the driver's right foot. Even as a teenager driving my 1980 Bonneville wagon with a big block V-8 and a 4bbl carb I used to average 16 MPG. That was with me driving the backroads @ 60 MPH, and not really caring about mileage because gas was under $1/gallon. It was so right on and regular I used that for my fuel gauge, because the one in the car did not work well. I used to write my mileage down when I got fuel. If I was wondering how much was left I would subtract my current mileage from the one written down, and divide by 16. That gave me my fuel used. I'm sure I could have hypermiled that car at least to the low 20's in mixed driving.

Sounds like we have similar driving styles....

Anyway, I kind of figured it was mainly the driving style of the drivers in question (they were looking for more power after all), but since the swap includes a whole new ECU, I figured that there might be some other stuff going on like more aggressive spark curves or richer mixtures....

Oddly enough, my reason for considering the swap is that I might eventually turbocharge the car just to prove/disprove the theory that turbocharged engines are better on fuel under low load/cruise situations.... The stock DPFI just isn't up to the task from what I've heard.

If ANYTHING causes your mileage to drop by going MPFI, it is going to be the ECU. Remember that the PM6 ECU was mapped for a higher performance AND larger displacement engine. So if nothing else, it might waste fuel by dumping 1.6 liter's worth of fuel into a 1.5 liter engine. You might be able to correct for this with an adjustable fuel pressure regulator. Also, it is possible to rechip OBD0 (and OBD1) ECUs and reprogram them on a dyno based on what you are looking for from the engine.

A word of warning about turbos, though. Honda D-series engines are pretty weak internally. And in stock form, they don't stand up too well to turbocharging. Some people don't care about this, since D15 engines can be gotten either super cheap or free if you blow up a bottom end. But if you expect the engine to last, turbocharging on stock internals is not the best idea.

words of wisedom. although i recommend against turning down fuel pressure.... im sure there are penty of ways to rechip to run a decently tuned mpfi setup.

HEARSAY!!!!!!!!! Nothing replaces proper tuning. Unfortunately too many don't understand this and take the term "Bolt-on Turbocharger" too literally.

Have you ever heard about the Epic-Tuning Hatch? Their goal was to take a Stock D Series to it's power limits. Their goal wasn't to blow it up (I can do that). Their goal revolved around reliability. Countless passes down the track and all that happened was warped rods! When something blows up that means improper tuning. When rods warp how they did that's reaching it's power limits.
375 WHP Stock D-Series

Now that they are done with that goal check out what else they got up their sleeve! INSANE!
AWD D-Series Civic

Hmm... I thought that there were other 1.5L D series engines out there with MPFI. I wouldn't swap the FI from a 1.6 to a 1.5L engine anyway. Different displacement engines would definitely have different needs in terms of fuel and spark and its unlikely that the 1.5L would run efficiently unless the donor engine was nearly identical. I'm still looking for the easy improvements and don't need to deal with transporting the car 150 miles to the nearest dyno shop for tuning at this stage in the game.

Yeah, I'm aware of the toothpick size rods that the D series engines have. Not to worry though. Since I'm not going for performance I'm talking about a very low boost setup (maybe 5 psi max.. if that). And with my driving style, I wouldn't expect any real problems. Owning the car for a month now, I've floored the accelerator precisely 1 time (to confirm that the engine was working properly). Aside from that, I have a very light touch with the throttle. As I understand it, the FE gains that a turbo can produce are all at very low throttle settings anyway, so driving the car like my grandma would tend to keep the internal stresses low even in the face of a little bit of boost. If I drove like Mario Andretti I'd be worried about durability, but I'm pretty confident about this plan.

I'm running the DPFI and have no complaints from a FE perspective. Even looking on the EPA website my DPFI Civic has better FE than the other 4th gen MPFI Civics.

On top of this I'm running the STD Civic which has tamed down cam lobes compared to the other civics so. Therefore keep this in mind when comparing that to your DX.

If I were you I wouldn't bother converting to MPFI until you're going Forced Induction unless you want the practice of working on your car. Otherwise you're going to be wasting money on injectors, and an ECU you'll need to replace when you go FI.

The 1988-1991 HF and 1992-1995 DX have non-VTEC motors with MPFI. The 1988-1991 HF ECU will drop right in. But it will not have proper fuel maps for a DX motor with SI MPFI. Also, you will have to use an EGR valve, since it looks for one. This means you will be stuck using the HF intake manifold, which is not the best idea, either. The 1992-1995 DX ECU would actually be the best choice. BUT, it is an OBDI ECU. Not that converting to OBDI is hard. And OBDI WILL get better MPG than OBD0. However, I know that ALOT of people would rather not do this. Also, I have never tried this. But you might have problems bolting the 1992-1995 distributer (which MUCH be used with an OBDI ECU) to the 1988-1991 head. This problem might be best solved by upgrading to a 1992-1995 DX head (which you can get for a song or free these days).

One of the problems with 'proper tuning' is that you are running with a razor-thin margin of error. ANYTHING that might happen to disrupt this could throw your tuning off enough to cause problems. Also, reliabilty may be better with a lightly turbocharged engine that is properly tuned. BUT, don't expect it to be stock-reliable. The way I see it, a boosted D should NEVER be your sole vehicle.

Wow, the opinions are all over the place on this one! Thanks for all the input guys.

Storminmatt, I'm still a relative newbie to Hondas. Can you maybe give me a quick description of the differences between the 88-91 and 92-95 models? Learning what's compatible and what's totally different will help me out a lot I think.

Of course nothing beats stock reliability. With a Honda there is very little one can do to improve reliability when performance isn't the agenda.

Try to name a few parts you can replace to improve reliability.

Even with performance as the agenda it amazes me how many performance engine builders still put their faith in the reliability of OEM Parts over "aftermarket performance" parts. OEM distributors, headgaskets, bearings, cranks, and transmissions are all parts that few feel you'll need to ever upgrade with sub 400hp performance goals.

oil filter, timing belt, sensors, spark plugs, clutch, bushings, shocks, exhaust. but all this would be with a fresh oem replacement:p

the few non-oem replacements that will improve reliability would be a pcv catchcan, aftermarket head studs, brake pads, and tires

Why not a Fresh OEM d15z1 while your at it :D

The whole PCV issue is because of stricter emissions purposes. This is why 94-97 b18b Integras have the black breather box and 98-2001 b18b Integras don't.

The OEM headbolts are fine on all non performance applications. If the head is off you might as well upgrade to aftermarket headstuds since they cost roughly the same. This is my old car and the current owner is coming up close to 30k boosted miles on the stock headbolts! CLICK HERE

As for brakes do you know that RealTime-Racing's Acura Integra uses OEM NSX calipers on their fronts and OEM Integra calipers on the rears. Once again OEM is more than fine for non-performance applications.

OEM FTW!!!!

because the z1 isnt necissarily more reliable?? the y5 is better anyway. i dont know what your getting at.

i mention the pcv catch can because a decrease in the octane of the air/fuel mixture from an intake manifold that has been tarred up will eventually take the potency away from the motor and over time(many many miles), will hurt reliability.

i mention head studs because eventually head gaskets blow and there is a better alternative for even the non performance application. dont get me wrong, oem headstuds are "fine", but the aftermarket is better. 30k miles is nothing on a honda, boosted or factory.

i didnt say calipers, i said pads. fundamentally different as far as reliability goes.

dude i read your post and even when performance is not on the agenda, at all what i listed does increase reliability on non performance applications. :)

Well, now that you mention it.... I'm wavering back and forth between doing the small things I can so see how far I can go with my existing stock DX engine, or just finding a cheap VX engine to drop in (along with the HF final drive I have). The only thing keeping me on the fence is deciding how much time and work I want to put into the car. I really don't need another big project to deal with in my life.

Obviously it's going to be one path or the other. I just need to figure out which fork to take at this point.

If you want better reliability from a boosted D, then ugraded pistons and rods are a good start. Resleeving the block would help as well if you REALLY plan to take this motor to the top performance-wise, while maintaining reliability. As for folks who believe in OEM parts, these folks are typically racers - people who live their lives a quarter mile at a time, and who don't care whether the car is going to run trouble-free for another 50000 miles.

Way to pigeon hole me Matt.
By the way I'm on my way to Home Depot you want me to pick you up a few more broad brushes?

Like I said, If I decide to try the turbo route, it won't be for performance reasons. I've committed to driving this particular car like my grandma would. In theory anyway, at cruise, turbochargers produce just enough boost to overcome the vacuum produced by the engine (thus increasing the operating efficiency), which is supposed to boost gas mileage by a few percent..... That's all moot for now anyway since I've got other fish to fry first.... For now, I'd be interested in finding out what is involved with converting from OBD0 to OBD1. Is it just a matter of swapping a few wires in the connector, or does it involve pulling a whole new engine harness?

For OBD0-OBD1 you can just buy a conversion harness and its plug and play or you could try and repin but that will be more time consuming and complicated, the OBD0 and OBD1 ECUs use different plugs...you also would need to convert the 02 sensor from 1 to 4 or 5 wire (depending if you go VX) and I believe you would need an OBD1 distributor which you could repin with the OBD0 plug or buy the conversion plug...look on Honda-tech there is some decent info in the FAQ section about the conversion :thumbup:

Cool, thanks for the tips....

I've got another related question now. Since I'm finally homing in on my transmission combo, I want to be sure that any additional changes to my engine won't affect what rpm it works the best at. Currently, I seem to be the most efficient in the 3100 - 3500 rpm range (that's where I got 46mpg while gaining 2000 ft altitude over a 275 mile stretch of I-40, vs, only getting 43 mpg while losing 6000 ft over 150 miles at 2800-3000 rpm).... Would the MPFI conversion change where the engine operates best by very much? I'm going to swap in a DX trans w/ HF final drive so I can cruise at 31-3200 in 4th and reserve 5th for the flat/downhill stretches. It'll put me at 2600 at 75 mph.... If the "butter zone" goes up by even a few hundred rpm, I could end up shooting myself in the foot with the trans combo!

While we're at it, what rpm do the VX engines work the best at?

I didn't read all of the responses to your question cause I don't have the patience. Anyways if you are going to stay dpfi I will recommend one thing from a FE standpoing. Put in the restrictor plate from a std civic. I just did and my fe just jumped, just go look at my gas log. The restrictor plate didn't hurt the drivability. It is slightly sluggish at take off if you floor it but to me this isn't a problem. This tank I am now going to try the std civic ecu (PO9). Stay tuned to my gaslog if you want to see what happens when the tank is through. I figure about a week and a half and I should fill up again.

It's possible, but not likely. The design of many engine components can affect where it's power band lies. While the biggest contributor is the camshaft profile, the intake manifold also plays a role. Longer, thinner runners will move the power band down the RPM range, while short, fat runners will move it up.

The reason I don't think it will make much difference is that the DPFI manifold doesn't seem to be designed with runner length, or even flow dynamics in general, in mind. First of all, the air/fuel mixture has to take a 90 degree turn when entering and exiting the throttle body, resulting in flow restriction.

Then again, those right-angle turns in the airflow are nothing compared to the restriction the tandem valve presents:
https://www.fuelly.com/attachments/fo...130326c2b6.jpg
https://www.fuelly.com/attachments/fo...e2b250d81f.jpg
It's basically a vacuum-controlled choke flap near the top of the throttle body, whose function seems to be to maximize air flow over the main fuel injector nozzle. The ECU has some control over the flap via a solenoid valve, but that's limited to switching vacuum from a venturi on and off. I guess if the flap was presenting enough restriction, manifold vacuum could play a role there too.

Really, I think the most significant issue here is that the manifold's four runners don't seem to be equal in length. The runners for cylinders 2 and 3 have a shorter path to the head than those for 1 and 4. That would mean that the center two cylinders would have a higher RPM power band than the outer two.

Sooo you're saying that the DPFI system is a hogs ear in terms of performance.... But if the MPFI system doesn't have those same shortcomings, couldn't it result in a more revvy engine? (Unless I'm reading things wrong and the tandem valve is actually part of the MPFI system... but I doubt that)

Usually I wouldn't mind that, but if I do install the HF final drive, I want to make sure that I don't do any thing that makes the engine lose low end torque or moves the power range up in the rpm range. If anything, I need to keep it the same or maybe even lower it by a few hundred rpm.

Sorry, I can't give you any really solid assurances one way or the other. I've never put an MPFI swapped D15B2 on a dyno to check its torque curve, so I can only give you an educated guess and how I came to my conclusions. The only way to really know is to try it and find out.

That said, the D15B7 is close to what we're talking about here. It's an OBD-1 D15B2 with MPFI and likely a different cam profile.
The D15B2 makes 92 HP @ 6000 RPM and 89 ft/lbs of torque at 4500 RPM.
The D15B7 makes 102 HP @ 5900 RPM and 95 ft/lbs of torque at 5000 RPM.
Peak HP is nice, but we're after torque here. The additional 6 ft/lbs of torque will help you, but coming 500 rpm later will not. It's really a question of the shape of the two curves... Are they more or less identical, except the 'B7 continues increasing beyond 4500? Does the 'B7 sacrifice low end to get the higher peak value, or is the 'B7 just a more efficient setup, yielding an overall increase in torque?
Anyone have dyno graphs? :/

its likely that the b7 makes 89 ft/lbs at 4500rpm and it just doesnt stop there. you are right, you need to look at the curves of each to really know

I've always had bad luck trying to find torque curves or dyno charts for stock engines (nobody ever gets their car dyno'd until after they mod it...). I was really hoping for a seat of the pants opinion.

Maybe I'll check over on honda-tech.com to see what those guys have to say.

Hate to bump this old thread but I'd like to get some discussion going here again :thumbup:

In the past few months I put in a brand new (rebuilt) D15B2 into my '89 DX, did an MPFI swap, and mated it with a new (rebuilt) Si transmission. I did alot of reading on the swap before hand and it seemed the general consensus was that, overall, FE would be about the same (assuming the Si transmission would potentially offset some of the "gain" from the MPFI swap).

Now that I've got a few hundred miles on the setup I'm beginning to be suspicious that FE is/will be decreased enough to irritate me.

My best tank so far with the new setup was 36 MPG, and I regretfully must admit that this was 90% highway driving. I haven't had to fill up again yet, but judging by the gas gauge (I realize it's not linear) and trip meter I would say I'm not going to get above about 32 MPG on this tank.

I also will point out that I haven't even had to run the A/C yet and judging from my previous experience, heavy A/C usage (unfortunately a must in Dallas) severely taxes the little D15B2 and really hurts MPG.

Before my swap, I could easily get 40 MPG. During the dog days of summer, however, the heavy A/C usage sometimes dipped me as low as 33-34 MPG. If this is the case, I fear many sub-30 MPG from my new setup in the summer, which just won't be worth it (as I can get 27 MPG all day long in my Maxima).

One advantage of the MPFI swap with Si transmission is that highway merging is no longer so nerve-wracking, and I would gladly take this trade-off if I could eek up to about 38 MPG consistently with the swap, but that's not looking too realistic.

How did you break it in?

The Si transmission is hurting your highway milage... It's likely costing you 8-10 MPG compared to the DX gearing, but that doesn't mean there's no room for improvement. Which intake manifold and ECU did you use?

I saw your garage on how you said "Virtually a brand new car!" It's great that your saving a EF since they are getting way more scarce these days. If you invested so much in restoring this car and you are doing 90% highway driving, my suggestion is swap the SI 5th gear to an hf 5th gear. You'll keep your first through fourth for acceleration purposes but have lower rpm's for cruising... best of both worlds!

Yeah, a little. The HF does (as do most '92-01 Civics) have a longer 5th gear ratio (0.694 HF vs. 0.771 DX/Si), but it's super long gearing is mostly a result of its final drive (4.25 Si, 3.888 DX, 2.95 HF fed, 3.25 HF CA/VX). A 5th gear from an HF or newer Civic would reduce your cruising RPMs by about 10%. A DX final drive would reduce the RPMs across the board by about 9%.

This is exactly how it will translate so you can decide for yourself:

Straight SI Tranny in 5th Gear:
1500rpm - 30.86 MPH
1750rpm - 36 MPH
2000rpm - 41.14 MPH
2250rpm - 46.29 MPH
2500rpm - 51.43 MPH
2750rpm - 56.57 MPH
3000rpm - 61.72 MPH
3250rpm - 66.86 MPH
3500rpm - 72 MPH
3750rpm - 77.15 MPH
4000rpm - 82.29 MPH

SI Tranny W/HF 5th
1500rpm - 34.28 MPH
1750rpm - 40 MPH
2000rpm - 45.71 MPH
2250rpm - 51.42 MPH
2500rpm - 57.14 MPH
2750rpm - 62.85 MPH
3000rpm - 68.56 MPH
3250rpm - 74.28 MPH
3500rpm - 79.99 MPH
3750rpm - 85.71 MPH
4000rpm - 91.42 MPH

Wow, I'm getting tooooo used to my new gearing cuz those numbers look high. 0.69:1 and 3.55 rear end gives me 60mph at 1750 rpm. If the TC is locked I'm at around 1175 rpm at 40.

yea displacement is a big factor... when I had a 2000 Civic SI my dad was simply amazed how that 1.6L would cruise @ 80mph @ 4500rpm and still get 30mpg

I assume you're talking about your Durango? Don't forget the last element between the drive shaft and road... Wheel and tire size can play a big role in gearing. '88-91 Civics and CRXs have an overall wheel/tire diameter of about 23", meaning around 875 revolutions per mile.

Yeah, the Durango tires are 31.5"

Now I've got a question about that. If I had the 8.25" ring gear instead of the 9.5" that I have and still had the same tires and 3.55s would it change that final ratio? Like, right now that's a 3.32:1 ratio, would the 8.25 make it a 3.82:1?

I'm pretty sure it does. Just verifying.

The ratio depends on the number of teeth on meshing gears, not their physical dimensions. So, you're going to have to tell me how many teeth are on each gear, assuming the pitch of the teeth, physical location and such allow them to mesh properly.

Wouldn't the tooth size have to be the same, therefore they would increase proportionally with the larger dimension of gear?

That's what I was thinking. 8.25:31.5 vs 9.5:31.5.

The 3.55 ring and pinion ratio stays the same.

That's the opposite of what I was saying...the ring mates to the pinion, and if you change the size of the ring it will have more teeth, so it changes the ring-pinion ratio.

The end result is the same as comparing the ring to the tire as long as the pinion is staying the same.