imzjustplayin

First off, I am asking for someone to clarify some questions I have about octane. If you can, please verify the points I made in the following paragraph.

From what I know, one uses a higher octane fuel or even octane booster in turbo and supercharged conditions in order to prevent detonation.

Detonation is caused by the fuel/engine getting too hot which would be the case when you're forcing more fuel and air into the combustion chamber.

Running a higher octane makes the ignition of the fuel more difficult which reduces the likelihood of detonation.

Detonation can also be caused under high loads even with out super/turbo charging.

Modern engines can retard and advance the timing of the engine, if it detects knocking, it will retard the timing; from what I understand, by reducing the "available performance" through retarding the timing (less optimal) it reduces the temperature of the engine and or fuel which in turn should stop the detonation.

The ECU in the Honda Civic VX ('92-'95) and the HX ('96-'00) can control timing on the fly, if it detects knocking, it will retard the timing.

Lean-burn will engage at highway speeds on the VX and HX Civics and run from 15:1 all the way to 25:1 for 49 states and to 20:1 for cars sold in california.

The recommended fuel for the VX and HX is regular


Question:
Can improving the cooling system remove the need for a higher octane fuel? Running 87 fuel, your vehicle currently knocks when running up a hill, if you improve the cooling system through a variety of methods, can and will this remove the need for higher octane fuel or will it continue to knock under the same conditions?

In lean-burn mode, does the ECU dynamic tune the spark timing in order to get optimal fuel economy but at the same time will retard it upon detection of detonation?

Supposedly when the car detects detonation, it will retard spark timing, after this has occurred, when will it advance the spark timing again? After the car shuts off? After a set temperature? Every friday? When the ECU is reset by draining it of its power by say unplugging the battery?

Going by what I think I know, that high temperatures causes detonation and the ECU will tune accordingly to prevent said event, when the car is in lean-burn mode, would the usage of higher octane fuel permit the longer duration of lean-burn mode and or increase the likelihood that the car will maintain its maximum permitted air-fuel ratio? What about improving the efficacy of the cooling system?

Relic1

I get a little long winded about this cause I've done a lot of thinking about it. (plus I know a little about the VX)

The VX has no knock sensor, I'm not sure about the HX... but I don't believe it has one either. So no knock detection there.

an easy to to think about octane, a higher rating burns slower. The lower the rating the more rapid the flame front, which means the gases may explode before they are supposed to or may burn faster than the ECU expects them to burn.

cali-spec VXs, use a standard 4-wire O2 (93-95 versions, 92 versions use a 5-wire but lean burn is disabled) they run the standard 14.7 that every other car runs.
I'm not sure about HXs.

improved cooling system does VERY little for the octane requirement, since the cooling system has little to do with the rate at which the fuel burns. But it may prevent pre-det, which is caused by hot spots in the cylinder. But only slightly. (this is why the VX and HX heads can not be bored out, the cooling passages are very close to the combustion chamber and exhaust passages to help cool faster)

during lean burn the ECU dynamically reads the A/F and adjusts the timing to match built in tables. Unfortunately in 90-91 (when the VX was engineered) the technology simply wasn't there to produce (in typical honda inexpensiveness) a way to dynamically monitor for knock and EGT temps for the ECU to be able to adjust the ignition settings. (see the newer hondas for this technology)

while running through the RPMs and load based tables, the ECU switches between the cells a lot, typically the switching of cells readjusts the settings. The knock sensor circuit feeds it's processed signal to the processor to determine what retard is needed. But as the engine changes situations so does the signal, so it's on going, so there is really no set time limit to the settings.

Lean burn mode is only entered when the ECU sees low load and medium RPMs for an extended period (a few seconds, to the ECU this is a long time, it operates in the microsecond range) so anything outside of those conditions and the ECU drops out of lean burn. Octane has little to do with when it kicks in.

If anything higher octane may screw up the lean burn mode... Higher octane burns slower, so if the ECU is expecting a rapid burn, the fuel may still be burning on the way out of the cylinder, which will look rich to the O2 (since some of the fuel will not be completely burnt) so the ECU may pull more fuel to get to lean burn, which will burn even slower and slower... till it drops out of lean burn thinking that there is a problem, then re-enter it seeing the conditions, and back and forth...

nowtype

1) One cannot assume that higher octane means faster or slower burn time. That is simply a fact verified by many fuel comparison burn time sheets.

2) Anyone have any actual DATA that shows the VX burning at 25:1 AFR? Based on my experience not with the VX but tinkering with cars, I'd suggest that this is a grossly over exaggerated internet folktale.

3) The second you throw in higher octane fuel, you've absolutely killed the point of being an economy car especially one so focused on fuel economy. The difference in price of the fuel is the difference between that 2-5mpg.

Answers:
1) If you can get decrease heat soak, you will see a very small difference in knock threshold. Enough to worry about if you are seriously heatsoaked. Average unmodified Honda Civic, not really worth wasting your time on. Don't forget, a drop in your intake temps causes higher density charge, thus your cylinder pressure will increase, thus making it more prone for detonation.

2) No knock sensor, no retarding of timing. The ECU is never optimizing spark timing really. ECUs can't think. They just work off a set of parameters. Map Voltage, rpm, IAT, ECT, battery voltage, throttle, and a few other minor ones and based on those conditions it tells the injectors/spark plugs what to do.

3) It will retard as long as the knock voltage is down. Once it goes up, it retards again. Most cars are built to tolerate small amounts of knock. Knock sensors are really just a safeguard if something fails or if minor detonation is occuring. If you have major detonation, a knock sensor is just going to keep your engine alive a little longer. End result will the engine will fail.

4) I highly doubt higher octane fuel is going to see any significant increase in fuel economy for you. It is something that could only be verified by testing in my opinion. I think if it made a difference, Honda would have advertised to people stating that if they ran premium, they'd get an extra 5mpg. I'm guessing in the end, regular vs premium the results will be the same. These cars don't live on the brink of detonation. They don't want them to. It's not good for them. So a switch will be pretty pointless in the end.

imzjustplayin

Well assuming that 25:1 AFR was true, wouldn't that be pretty close to the brink of detonation? If I were to tune it to the brink of detonation, then improving the cooling and using a higher octane should alleviate the risk of detonation and or even let me tune farther than I'd be able to with regular, correct? I use premium fuel anyhow so I wouldn't care about the cost difference between regular and premium, premium is also suppose to have more detergents than regular. The price difference between regular and premium was much more noticeable when gas was cheap, now that it's so expensive, the small price increase is negligible.

Also is isn't so much about saving money as it's about not having to fill up so often, saving a few MPG even when you're already getting 50 would mean that much longer between fillups. Basically, I'm not sweating the price of premium so that is why I'm interested in what the gains would be when tuning with premium and otherwise.

Relic1

If you're tuning the car, using several different techniques and sensors, then premium may help. But not on the stock ECU.

imzjustplayin

That's kind of sad, I know this isn't a lexus forum but I was wondering if an LS400 had a knock sensor or not. If it didn't, that too is sad but would definately justify the civic not having one. Though to think that the electronics in cars has been so UNsophisticated for such a long time, it's kinda sad. These technologies should have been around 20 years ago, integrated into the car..

Relic1

The more expensive honda (Interga) typically had knock sensors to monitor knock.

All honda's post 01 have a much more sophisticated system to monitor A/F and knock.

A side note: From what I've heard and read, Honda has plans for even more cutting edge technology in the very near future.

baller status

Relic1 beat me to it, but I still have more to add.

The OBD2 heads have knock sensors. Honda is very conservative pulling a ton of timing when the ECU detects knock.

You don't need a knock sensor for VTEC-E. VTEC-E engines are no more prone to knock than any other. Knock sensors are used when you want to run a less conservative timing map.

All of you are making the assumption that the VTEC-E engines are conventionally designed engines requiring conventional afr ratios. These engines are specifically designed to be able to run lean. The VTEC-E swirls the air in the combustion chamber so that the fuel spreads out evenly. If the mixture burns evenly then there is no knock. Knock happens when parts of the mixture spontaneously (not really, actually it's heat and pressure) combust apart from the spark ignited combustion.

Each time I hear someone talking about how lean the VX engine is, the number gets bigger. First I heard 18, then 20, then 22, and now 25. But my question is, what wideband O2 sensor can accurately read leaner than 18? The 5 wire VX O2 sensor can't! So to run leaner you'd be relying solely on perfectly tuned ignition maps in open loop.

Running lean or overly rich slows combustion requiring more timing. Running higher octane slows combustion requiring more timing. So you can see that running higher octane does not help. Anyway why would you ever consider running a different octane than what was recommended by the manufacturer?

If you want a modification that will allow you to get the absolute most out of your fuel octane, I suggest filtering the oil vapor from the crankcase gases before they are reintroduced to the intake and intake manifold. Use a baffled catch can with an oil-air separator media. If you are going to do this on the tube going to your intake, make sure it can match or outflow the stock system, otherwise you'll increase pressure in the crankcase reducing power in the high RPM slightly but most importantly you'll be forcing oil past the seals.

baller status

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Relic1 &raquo;</TD></TR><TR><TD CLASS="quote">The more expensive honda (Interga) typically had knock sensors to monitor knock.

All honda's post 01 have a much more sophisticated system to monitor A/F and knock.
</TD></TR></TABLE>

The B series had knock sensors as early as 95 and maybe earlier. To my knowledge all 96 and up D series have knock sensors. My engine certainly does.

imzjustplayin

The EJ7 (Civic HX) has more stringent emissions requirements and so doesn't do lean burn as aggressively. So I was wondering what would happen if I put in a 5 wire (wideband) O2 sensor with an ECU that only needs a 4 wire O2 sensor? Is there a way to make the ECU accept the wideband O2 sensor?

I just want to know that if I were to try and tune the ECU, would I be able to tune the ECU in such a way so that I could get just as lean mixtures as I would be getting with the VX? Also since it has a knock sensor, and I have the fuel tables set an aggressive lean mixture, will it automatically adjust so that it doesn't knock? Does it have to knock if the mixtures are really lean?

Relic1

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by imzjustplayin &raquo;</TD></TR><TR><TD CLASS="quote">The EJ7 (Civic HX) has more stringent emissions requirements and so doesn't do lean burn as aggressively. So I was wondering what would happen if I put in a 5 wire (wideband) O2 sensor with an ECU that only needs a 4 wire O2 sensor? Is there a way to make the ECU accept the wideband O2 sensor? </TD></TR></TABLE>
no, the ECU would set a fault code for the O2, and have no idea how to control the wideband.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">I just want to know that if I were to try and tune the ECU, would I be able to tune the ECU in such a way so that I could get just as lean mixtures as I would be getting with the VX?</TD></TR></TABLE>
not without redesigning the head slightly and moving the cooling jackets closer to the combustion chamber to help pull away some of the additional heat created with that lean of a mixture.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">Also since it has a knock sensor, and I have the fuel tables set an aggressive lean mixture, will it automatically adjust so that it doesn't knock? Does it have to knock if the mixtures are really lean?</TD></TR></TABLE>
OBD1 ECUs do not have a long term trim, so any time the ECU enters those load/RPM cells it will knock until the ECU pulls timing. Very not good for the engine in the long run.
Several of the more popular tuning programs have the option of running a wideband input into the standard O2 line and running closed loop off the lamda tables. This could be used in this way, but going crazy lean like the VX/HX will result in melting aluminum...

imzjustplayin

"OBD1 ECUs do not have a long term trim, so any time the ECU enters those load/RPM cells it will knock until the ECU pulls timing. Very not good for the engine in the long run.
Several of the more popular tuning programs have the option of running a wideband input into the standard O2 line and running closed loop off the lamda tables. This could be used in this way, but going crazy lean like the VX/HX will result in melting aluminum... "

What do you mean like the...? I thought we were talking about the VX/HX? There was a graph somewhere around here, I don't have it saved on my computer unfortunately, but it clearly showed an increase in temperature and NOx at 15:1 and 16:1 A/F ratio but then as the ratio went up, they all gradually decreased. It makes sense though, as there is less fuel in the combustion chamber, only so much heat can be made, I don't see how running a 40:1 air fuel ratio could produce more heat than 15:1. I'm not quite sure why honda wasn't able to keep the lean burn so aggressive as they used to but I think the reason is because not only emissions standards, but because the catalytic converter doesn't function so well with only one type of emissions but needs a combination of everything in order to eliminate everything.


This thread further discusses the phenomenon I was talking about, at least in terms of emissions.
http://www.gassavers.org/showthread.php?t=6624


hah, I found the image, it was in that thread I linked to.




Modified by imzjustplayin at 11:14 AM 6/6/2008

Relic1

baller status: the L1H1 and L2H2 are rated for normal operation up to 20.6:1
lean burn = 16-16.5:1 when under slight power or zero load... when slightly decreasing speed it will jump to around 19-20:1, decreasing speed with zero throttle will shut off the injectors like it does in any honda.

imzjustplayin: (great image, I've never seen that before)
reason for me stating that; Most people's thinking...
if you have a VX or HX, why not use the stock ECU and be done with it. (great mpg from the factory)
If you don't have one, and you want to tune for mpg, using a tuning program, as you suggested, things could go bad fast. (hence my warning)

If you have a VX or HX and are looking to tune it for even better mpg figures, that's a very different story. (I'm guessing that's where you're going with this)
All though figuring out exactly when you could drop the A/Fs and how far you can go with it would take some serious thought, some controlled experimenting and mapping out the stock ECU via wideband in a stock car...

(I'm not seeing temps in that pic)
leaner doesn't have to mean hotter, but for the purposes of automotive discussion, to produce enough power to accelerate slightly or maintain speed in a car you need to be in the 11-17 range.
I've played around with logging my EGT and wideband while at idle. I know it's a different circumstance, but I don't have a test driver for when I'm playing like this.
What I saw was that the EGTs tended to plateau around 16.5-17:1, then decrease slowly starting around 17.5. The car would start to buck and miss at around 18.5:1
I was also playing with a higher compression engine, so the values may be shifted slightly. (which may line up with what you said, I'm not sure how CRs and head design change the EGT readings at the same A/Fs)

The reason the cali-spec VX didn't have the lean burn (and 93-95 ran a 4-wire O2) was that lean burn created too many emissions related problems for cali, so lean burn was disabled.
I'm sure it was the same for the HX, only since emissions is a growing concern, I'm sure honda slowly made the criteria for lean burn more and more strict.

All though based on that pic you posted, I'd thnk they'd want to aim for ~17:1. Where HC and NOx are the lowest and mpg is the highest.
Then again, heat, cat converter efficiency... and all the other emission gases are factored in as well I'm sure.

imzjustplayin

NOx is produced from higher combustion temperatures, unless something very very different is happening, when the temp goes down, so does the NOx.

wHOAlly

i dont have an answer to the OP's question, but some things ive noticed while tuning cars.

i usually get lean misfires ~16+AFRs, some cars can run at 16 smoothly, but even slightly leaner and it will start to misfire and get rough, ill usually target 15-15.5.

imzjustplayin

So it's really all about whether or not it will misfire. If a car doesn't misfire at at a given AFR, if you put a load on it, will it then misfire?

imzjustplayin

bump

Relic1

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by imzjustplayin &raquo;</TD></TR><TR><TD CLASS="quote">So it's really all about whether or not it will misfire. If a car doesn't misfire at at a given AFR, if you put a load on it, will it then misfire?</TD></TR></TABLE>
more load increases the chances of missfire or not being able to keep up. (missfire, stall or both)

imzjustplayin

We have this "Dr Power wagon" crap 2 cycle engine. When it's shifted into a higher gear, load is high and going up a hill, I will hear what I'd describe as what would be knocking/pinging on a vehicle experiencing high load. It's much higher pitched because I think it's a smaller motor, I guess I could describe it sounding like a chain hitting a metallic object.

Anyhow I was wondering, is this the same thing? Isn't a misfire different from knocking/pinging? I thought misfires were caused by weak sparks, poor coils, and high resistance sparkplug wires while pinging is preignition of the fuel mixture due too high of combustion temperatures and the way to solve it is to retard the timing? I think this is where my knowledge of misfire/preignition stops.

baller status

Lean misfires occur when there is so little fuel that it doesn't spread out evenly and you get abnormal combustion.

The VX/HX motors are an exception to that chart and the conventional wisdom concerning afr radios. They were purposely designed to be able to run lean mixtures. Closing one intake valve forces the air to enter at one "corner" of the combustion chamber and swirl around the outside until it reaches the closed intake valve. This stirs up the fuel mixture allowing you to run very lean w/out lean misfire.



The HX, like the other OBD2 D-series heads, has increased quenched area. When the piston comes up, air gets squished at the outsides of the combustion chamber and squirts inward. As a result, the HX requires less ignition timing advance to achieve MBT.

imzjustplayin

What MBT? Also while it has increased the quench area, did it actually improve efficiency in the sense of whether it uses more fuel than the D15Z1? Also why is it advantageous to not need to advance the timing as much? Is it because it gives more room for timing advance in the event you want to run more lean? If there is more quench area, you can run leaner, advance the timing more, and not get a lean misfire opposed to what you'd normally have with the older OBD-I D15Z1 head?

rorik

Really interesting thread.
MBT is maximum brake torque, it is the highest amount of torque (power) that can be achieved as a result of ignition timing.
In other words, too little timing or too much timing for whatever load cell on your map will drop the power, mbt is the perfect amount.

I don't really know how you find the perfect balance between timing and fuel for the best efficiency. It seems like a lot of people have gotten good results with just leaning out almost until they get misfires while keeping the timing at mbt. (now more advanced with the leaner mixture)

Now that I think about it, I've noticed that on the non lean burn vtec motors, they have one lobe on the cam smaller than the other, I never knew this was to facilitate swirl to increase efficiency/lean burn capability..So how much more does the lean burn motor really swirl the mixture? What else is different about it vs the gsr/b16, etc?

Another thing I've been wondering about for a while, is it possible to have the injector size, timing and duty cycle optimized to achieve a stoich mixture in the top part of the cylinder while the rest of the cylinder is just air? I see that being possible by triggering the injector for only the last part of the intake stroke, so the top of the mixure is at stoich while under it's leaner until it's just air at the bottom. As the power requirement increases, the injection timing would advance and the duty cycle would increase to fill a higher % of the cylinder from the top down with the stoich mixture. This would not be achievable at all load/rpm points, but it should be during cruise..??

imzjustplayin

Well I was thinking about how on higher performance vehicles, people go with bigger CC injectors so that they don't have to keep them open as long, but then have to upgrade the fuel pump as well because the PSI in the system would be subject to dropping from what it's used to. I read on diesels they use very very high pressures in their direct injection in order to precisely choose when the combustion occurs. So I'm thinking that the only way to achieve your idea is to 1. Have Direct Injection or 2. (though 1. seems better) Is to have I think higher CC injectors w/bigger pump and have the duty cycle of the injectors be shorter and or keep the stock injectors but increase the PSI in the fuel system so that you don't have to keep the injectors open as long. So with your idea, the longer you have to keep the injectors open for a given amount of fuel, the less atomized the fuel will be and the earlier you'll have to inject the fuel into the intake just so that there is enough fuel to burn, but then that'll make the idea of having a pocket of air on top of the cylinder and fuel near the spark that much more difficult. So in short, I guess the answer would be Direct injection and or a much higher PSI fuel system where you could squirt the fuel at the last second before the intake valve closes. So the injector and or fuel system would need to be capable of injecting fuel in a small fraction of the time compared to the conventional system.

SikofFalln

This maybe a little off base, but isn't the only difference between the Y5 head and the Y8 head the valve train? The Y5 has roller rockers and what not while the Y8 has conventional rocker arms? I know the cams are not interchangeable because the rockers have different ratio's but couldn't you just bolt the cam caps from a Y8 with the cam to the Y5 head and get rid of V-Tec-E all together? The heads seem to be of the same casting, and I found on Majestic honda that they share the same part #.