CRX HF lean burn
 

I should probably look to posting this question on a Honda board, but since they're mostly ricers I figured I could at least get away with asking it here.

Well, I was following this thread for a bit: https://crxcommunity.com/viewtopic.ph...ffe26e1d1c7750

..and I thought about how the Civic VX engine (D15Z1), although much more powerful than the CRX HF engine, still uses so little fuel. I then wondered, without doing the HF to VX engine & ECU swap, could a compromise be reached with the HF engine? That is, what would it take to get a CRX HF to be lean-burn? Obviously the car would still lack the V-TEC and horsepower of the VX, but it would certainly save more fuel if it had the lean-burn advantage. Any thoughts? Could a VX ECU be modded to work?

lol, or I suppos I could do just a little bit of research first, to prevent me from looking like an entire ***.

I guess I should omit the "lean-burn" phrase, since that is apparently reserved for an engine that keeps 4 intake valves at bay, while using just one intake valve per cylinder during times when only minimal power is required. Instead what I would like to propose is the feature of the VX that allows the engine to shut fuel to the cylinders off when the car is just coasting, since some of us don't like shutting the car off every time we coast. I think every car on the planet should have a way of shutting fuel off when coasting for significant periods of time.

decel fuel cut? like leave it in gear and use less gas, or neutral coast fuel cut? the latter, just use the key. the former is a little more complex.

It would be wiring, ECU and a few sensors like the $170-$300 O2 sensor... but this is a good question! maybe throttle position sensor? I think the 2nd generation crx hf had very simaler fuel injection set up as the VX does, 1st generation crx hf have a carburator of course and are already basicly a lean burn with carburator.

A couple of things. First of all, lean burn is not restricted to 16 valve engines. There is no reason why it couldn't be used on an 8 valve engine. This is especially true of the HF, where the intake valves are off to one side of the cylinder. This tends to do a good job of promoting swirl, which is precisely what the D15Z1 does when not using both intake valves.

As for lean burn, this is going to be something difficult to apply to the HF. The problem is that the ECU has no provisions for it. As long as it is running in closed-loop mode, it is going to try to maintain a stoichiometric mixture. And any attempt to lean things out beyond the ability of the ECU to correct is going to make things lean all the time - not good.

As for shutting off fuel to the cylinders while coasting, this is actually a typical feature of pretty much ALL EFI systems. Your HF probably already does it.

I know that my 1985 crx hf cuts fuel while coasting in gear as long as the engine is still above a set RPM, my 1983 honda civic also does this, you can clearly see this in the wiring diagram for that car, I can dubble check but I'm pretty sure that my 1973-78 civic shop manaul also shows the fuel cut solinoides, 3-4 of them on the carburator allowing it to compleatly cut fuel, and a vacuum switch on the intake manifold that trips them when the throttle is closed and the engine is reved.

My HF cuts the injectors at above 1200 rpm with the throttle closed when it's warm, 1500 when is cold. You can feel it happen when the rpms drop below you can fell it surge forward slightly when the injectors open

I think you could simulate a lean burn system on a car other than the VX pretty easily, but I am not sure how well it would work, or how hard on the engine it would be.

If I was going to try it, I would buy an Innovate Motorsports Wideband O2 sensor and controller. These controllers usually come with two sets of analog outputs, that can be programmed to simulate a narrowband O2 sensor, such as the one used on the HF. However, it would be very easy to set the controller to output one signal that had the narrowband crossover point at 14.7/1 AFR, and the second one at something higher, like 15.5-16.5/1 AFR. Then it would just be a matter of switching which one of these two signals fed into the O2 input on the hf ecu. When it was set to the higher AFR input, it would use the closed loop mode to attempt to hit this AFR instead of 14.7, which would result in a "lean burn mode".

It is possible this would just confuse the ECU, and cause a CEL, though.

For in gear fuel cut, I checked the shop manuals and the 1973-77 1.2L and 1.3L civics had fuel cut over 12mph with a warm engine, good oil presure and high manafold vacuum, those engines do not appear to be cvcc engines, the 1977 cvcc engine might be the first cvcc (lean main mix with a rich mix next to the spark plug giving an overall leaner mix) that also had fuel cut while going over 12mph with high manifold vacuum, and it appears to be 3 solinoids that were triggered that cut the fuel.
does anyone know if a vx ecu would work in a 2nd generation crx? are they simaler enough that it would plug in?

There is no plug-in capability. Not only do the wires not terminate in those locations on the ECU, but the plugs are different pin numbers and shapes.

As for the VX, shutting off the fuel to the engine: That only happens in deceleration, not coast. If you are in neutral and the fuel is shut off to the engine the engine would no longer be turning. (And that would be true for all engines.)

I was thinking the same thing as I was reading through the thread. An ideal setup would be a controller you could adjust on the fly. That way you could adjust the mixture based on current conditions. You should be able to run a leaner mixture in the cool months than the heat of summer. Moisture in the atmosphere can also play a role. An easily adjustable setup would let you tweak things as you go rather than being forced to dial the mixture back to a single safe-for-all-conditions level.

Nah.
The O2 sensor error code gets set when the throttle is closed and the engine RPMs stay up but the O2 signal doesn't swing lean. What significance does that have? Those conditions occur when you're coasting down in gear - foot off the gas, but the car's momentum keeps the engine RPMs up. In that situation, the ECU cuts off fuel flow (as was discussed) by not opening the injectors. In that case, there should be no combustion occurring in the engine and only pure air being pumped through. The O2 sensor would normally swing over to a full lean reading... If it doesn't, the ECU knows something is wrong, sets the error code and triggers the CEL.
So, as long as the wideband controller flips the simulated output over to full lean when the sensor reads air, it should work fine.
Further, the HF ECU doesn't use any adaptive fuel maps like newer cars. Its fuel tweaking is limited to the flip-flop action when in closed loop operation.

[QUOTE=bobski;92293]
Nah.
The O2 sensor error code gets set when the throttle is closed and the engine RPMs stay up but the O2 signal doesn't swing lean. QUOTE]

Yeah, I haven't looked into how the ECU determines that the O2 sensor is bad. Generally, it seems like closed loop systems only allow a certain percentage change from their stored fuel curves. I was theorizing that the ECU might trigger the O2 sensor code if it was reading rich, and the closed loop was unable to adjust enough to make it lean (which is what could possibly happen in the setup I described above). However, if the O2 sensor code is only triggered like you describe, this should not be an issue.

I am kind of interested in trying this setup out, since I already have the wideband sensor, but since I also have a VX with the stock lean burn, I am not all that motivated. Oh well, maybe it can be a future project on the Fit...

Good point... I don't know how far it will alter the injector duty cycle to satisfy the closed loop code. I imagine it would have a fair amount of wiggle-room to account for variations in manufacture and the effects of component wear. The ECU doesn't have any means of measuring the exact fuel pressure or accounting for partially clogged injectors... It only has the O2 sensor reading.

The fact that the plugs are different is the least of your concerns. After all, LOTS of people modify their 1988-1991 Hondas to use 1992-1995 ECUs. Rather, the BIG problem is the fact that you are going to get check engine lights due to the lack of VTEC on the HF motor. And even if you can somehow rig up a way to bypass this, the fuel maps will be wrong. And reprogramming a P07 is apparently different from reprogramming other OBDI ECUs. Not that it can't be done. But it's not like people generally know how to do it.

Yeah... For one thing the board architecture is completely different. The P07 uses two processors where the other ECUs use one.
I don't think there was enough demand for the P07's additional hardware features to make it worth decompiling and reverse engineering the code.

I just had a thought as far as the O2 sensor tweaking goes.
If closed loop operation turns out to be too limited, you could install a MAP signal scaling device like an SAFC and set it to lean out the mixture. Normally, closed-loop operation would override any tweaking by the AFC, but together with the wide band controller it would work perfectly. Adjusting the AFC would move the closed loop code's window of operation up and down the A/F ratio scale, and the wide band controller would set the actual target point within that window.