My (Longwinded) Theory

Well...  A couple of years ago I bought a 6th gen VFR.  I'd been looking at them for years and really figured that it was the bike for me, ie decently sporty, mostly comfortable and had hard bags.  I have a tendency to research the hell out of things, so I did as much as I could before I picked it up.  The only thing I really came across before buying was that a lot of folks weren't happy with the throttle response.

At the time, I was tuning an RS125 two stroke for my buddy, and had a decent grip on how different jettings did different things, so I wasn't too worried about that, as with a power commander, I could rejet without taking anything apart.  Nothing to it!

So after a couple of years, I stepped up and got an 09 black and white.  It's pretty sharp, if I do say so myself.  But I might be biased.   :wink:

So I rode it for a bit, and had to agree with the general concensus on the throttle.  It was pretty horrible.  Synching the throttle bodies helped a bit, but it still wasn't anywhere near where I felt it should be.  Hopping on my thirty year old carburetted bike really highlighted the difference and I decided to put the PCV on the short list of things to get.

I picked one up this spring and fired in Cozye's map for the 06+.  It helped a lot, smoothed out the low end, the downshifts and got rid of that annoying bouncing thing it would do under constant throttle around 5k.  LIfe was good.

And then it started getting 23 mpg and diluting the oil, which I thought was a bad thing.

First thought was that there was something wrong with the fuel system.  I rigged up a little bench test to check the injectors for leaking, and they all came up okay.  Fuel pressure regulator checked out okay, and the pressure on the rail was inside the limits.  Fuel system was all good, which pointed to more of a design problem.

Here's where we get to the meat of my theory.  

The 5th gen bike is fuel injected, but was originally designed as an open loop system, which means that the injectors spray a predetermined amount of fuel based on the throttle opening.  Works just like a carburettor, in that the same throttle opening will always result in the same amount of fuel being delivered.  (factors like temperature and barometric pressure will change if, of course, but bear with me)

In an open loop system, all of the components need to be carefully calibrated so that a given input will yield a consistent output.  The system can't compensate for any changes that happen in the process, so tolerances during manufacture need to be quite tight, especially in volume production, if you want to build a number of systems that perform consistently.

In a closed loop system, like the 6th gen, the output of the system (exhaust gasses) are measured and the amount of fuel delivered at the intake is varied to produce the desired air fuel ratio.  This means that production tolerances can be opened up and parts that perform slightly differently can be used, as the system can compensate for any differences as it goes along.

The bonus for this on the engineering side is that the Honda team was no longer restricted to using a single model of fuel injector for the bikes, thereby reducing costs as parts availability opened up.  

However, the only problem is that vehicles that use a closed loop system still switch back to open loop during certain conditions, ie below operating temperature of the O2 sensor, medium-hard acceleration and throttle overrun (engine braking).  This means that they go to the preprogrammed base map in the ECU and operate just like the open loop system, with the amount of fuel delivered depending mainly on throttle position instead of the results of the combustion.

In a car, the computer monitors the fuel delivery it makes during closed loop and compares it to the open loop delivery (base map).  Every time the car is run, any differences between the two are stored as short term fuel trims.  If the short term fuel trims show a consistent trend (ie higher than the base map for three consecutive trips) then the short term trims are moved into what is called a long term fuel trim.  The long term trim is added to the base map to bring it closer to the closed loop value, and is used to keep the open and closed loop fuel delivery as close as possible.  An OBD2 scanner with live data will show these values, which can be handy in diagnosing car problems.

From Kaldek's experiment with the Healtech diagnostic unit, it doesn't look like the VFR has a long term fuel trim, which means that whenever it goes open loop, it reverts to the factory fuel setting, plus whatever short term trim has developed since you started up the engine.

This wouldn't be a problem if it weren't for the larger tolerances on the fuel injectors.  When I was testing mine, I noticed that it was a completely different model than the 12 hole injector Kaldek had installed on his bike.  This means that there is a very real possibility that it has a different flow rate.  I need some confirmation from Denso on this to make sure, which is why this is still a theory. :wink:

Given that there is a real possibility that there is a variance in flow rate between the different injectors, it means that Honda would have had to set up the base map to ensure that the engine gets enough fuel from the lower flowing injectors.  Any injector at the top end of the rate would then flow more fuel at that point in the map, resulting in a richer mix.  As manufacturers tend to go with a conservative (read rich) map anyway, this means that the higher flowing injectors would result in a really rich mix.

Once the bike hits closed loop operation, the ECU can check the exhaust and reduce the duty cycle  to the injectors, reducing the fuel output and bringing the combustion back to where it should be.  The two bikes would then run identically at this point.

The problem comes when the bike goes open loop while you're riding, under the three conditions above.  Once it's warmed up, it's really only the acceleration and engine braking that will affect things.  For instance, I downshift when I go into a slow turn, which causes the ECU to switch to open loop mode.  It reverts to the base map and, because my injectors flow more fuel than the ECU is expecting, means that the engine starts running rich.  One symptom of rich combustion is a really snatchy throttle at initial roll on.  So when I get back on the throttle, I have to fan the clutch or chance ending up in the oncoming lane.  Sound familiar?

The other case is when you're accelerating.  I have an Ultra Gauge in my car, which shows when the ECU is in open or closed loop.  If I'm going up a hill and accelerate enough for the slushbox to drop one gear, it goes open loop.  You really don't have to hammer it to get it to run open.  On the bike, accelerating up an on ramp would be about the same.  Then when you get up to cruising speed and settle down, it would switch back to closed loop, realise that it's running rich and drop the amount of fuel, resulting in that little droop that many of us are familiar with.

The ECU does have a short term fuel trim, which will modify the base map and smooth things out.  However, this is lost when the bike is shut off and you're back to square one.

This is further compounded by the PAIR valve, which can also cause the engine to run rich, for instance when you're up to temperature but idling.  On my bike, the PAIR is open whenever the bike is below temperature, and when the throttle is closed above temperature.  Say I come up to a stop in traffic and close the throttle.  The PAIR opens and dumps fresh air in the exhaust, right in front of the O2 sensors.  They report a lean condition and the ECU richens things up, resulting in a snatchy throttle when you start off.  Gotta be one of the dumbest systems I've come across in terms of rideability.   I can see where it would reduce emissions and help bring the cat up to temp, but I'd rather have a bike that works as it should.  Or as well as my old, low tech one...

Anyway, to sum up, I think that Honda may have used injectors on our bikes that have different flow rates and calibrated the ECU (mine is from Keihin.  What happened to MotoGP trickle-down?) for the bottom end of the range, resulting in the ones that flow more running too rich when the bike is in open loop mode.  This accounts for the wide variance in rideability of the 6th gen VFR and the huge variance in mileage.

Thoughts?


PS
If you're interested in researching fuel injection, http://www.autoshop101.com has some good technical articles on it.

Interesting, is there any way to remedy this or are we just stuck with the situation?

It's pretty easy to fix with a power commander, but the commander would have to be set up for that particular bike, either on a dyno or with an autotune.  Using a map that was developed for another bike would work if the two bikes have similar flow rates, but if they're different there may not be much of a benefit.  This could be why we see folks reporting mixed results from the posted maps.  Just my opinion, though.  I could be out in left field with the whole thing.  :o)

I thought cozy's map was made for 02 -05, while people tried to use it on 06, they are different fuel mapping.  The 6th gen with good fuel(not available in the U.S anymore, will yeild anywhere from mid 30's to mid 40's mpg wise (Stock).  If running a map thats only getting 23mpg , Id get it out of there. But in just correcting the stock lean issues in early throttle, normally will lose 2 or 3 mpg, cause your enrichening the fuel in that area. If enrichening other areas, though differnt pipes you could lose even more MPg.

Cant comment on the flow rates of the Injectors, but Id say they speced pretty much the same on 6th gens. They all are 12 hole injectors, a quick parts check when tell ya if the injectors are the same with differnt model years cause they wont cross over over supecede.

Just a quick check, 02-05 is differnt from 06-08 as far as injector part numbers, course the exact differnce unknown

It may be longwinded, but it makes sense to me and I don't think you're far off the mark.

So in your opinion, what do you think of the PAIR block-off mod and the deactivating the O2 sensors mod?? Both together and/or separately?? Also with or without a Power Commander??

@spud786

You're right, cozye did develop his map for the 02-05, so It wouldn't work as well on the 06+ models.  I looked through it before I installed it, and there weren't any drastic modifications that would have accounted for the MPG drop on my bike.  That's why I figured it was more a problem with the base map being too rich to start with.  

From what I saw with the RS, a bike that was too lean would spin right up without making a whole lot of power but it would be smooth at the initial openings.  It was when the needle was too rich that it demonstrated the snatchy throttle that I see on my VFR, which is why I'm thinking that it's too rich to start with.

As for the injectors, my guess on that came from working in electronics manufacturing.  My company would have a part number for  a specific value of component, like a memory chip.  However, there were multiple suppliers who made memory devices that fit the specs, and we would buy from all of them, which increased availability and reduced cost.  Then the boards would be built and sent out.  One could have a chip from Samsung, one from Hyundai, one from Kingston, they would all perform slightly (very, very slightly in most cases) differently, but they would all be considered to have the same part number in that location, even though the parts themselves were different.

When Kaldek was going through his problems with his bike, he got a set of injectors from an 06 and posted up some pictures.  They were twelve hole injectors, with a grey body and labelled as Denso 4480.  According to the parts fiche, my bike has the same part number, but the actual injector in there is a twelve hole with a red-orange body and labelled as Denso 8FIJG.  My guess is that Honda sent their requirements to Denso for an injector with certain physical dimensions, electrical characteristics and a flow rate between two limits.  Denso makes a number of injectors that fall within those categories, and whichever ones they have on hand when Honda places an order is what gets installed in the bikes that day.  When Honda switched from the all open loop on the 5th gen to the closed loop system on the 6th gen they would have opened up the flow rate tolerances on the injectors,  thinking that the ECU would compensate for the differences, which is why we see more of a variance in throttle response on the 6th gen bikes.  It's a perfectly logical move to cut production costs, and would have worked properly if the ECU had the long term fuel trim to adjust the base map over time.  

Seems to me that the 02-05 had injectors that had one large hole and the 06+ part has twelve holes.  Helped increase atomisation, makes for a smoother burn and, ironically for me, better fuel efficiency.

@ Auspañol

Thanks!  I spent a while going through it before posting to try to get my throughts to come out right.

I think the PAIR block off would definitely smooth the throttle.  When the PAIR valve opens after the engine is up to temperature, it causes the motor to run rich and results in the snatchy throttle off the start.  However, it probably helps to keep the catalytic converter at temperature.  Blocking off the PAIR may degrade the efficiency of the cat over time resulting in a performance drop.  

Deactivating the O2 sensors will cause the ECU to stay in open loop all the time, which definitely smooths operation on my bike.  Most of the noticable jerkiness/surging that I see is caused by the ECU switching between open and closed loop mode.  Say it's running open loop from the base map, which is overly rich for the injectors I have.  I get up to speed, it goes closed loop, takes a look at the O2 sensor readings and goes "Hey!  That's way too much gas!" and cuts way back on the fuel.  This makes it feel like I either backed off the throttle a bunch or ran out of gas for a second before it stabilises.  The only problem, and this is where I got into problems, is if it runs open loop all the time, and you have a rich injector in there, your mileage is going to drop significantly, to the point where the excess fuel dilutes the oil.  That's something that should really be avoided.

This can be fixed with a Power Commander by leaning out the entire range of the bike.  This is not something that I would recommend unless you have some type of sensor at the output of the engine to ensure that you're not running too lean.  Too lean melts pistons and just isn't a good idea...

My long term plans include taking the PAIR off again (i've run both with and without, sometimes on the same day) and getting cat-less headers to eliminate the cat degradation problem.  Also either an autotune or some dyno time to set up the power commander properly.  Dynos are pretty scarce around here, so I may be saving up for the autotune.

Cheers!

Your theory and logic were explained extremely well so that even a mechanical midget like me can follow it.  You made one remark that in my inexperienced opinion is the one obvious reason that dictates this entire problem.  EPA mandates are controlling Honda's (and all other bikes sold in the USA) manufacturing processes, decisions, and performance results relating to fueling, exhaust, and tuning.  If concessions are going to be made (nothing is perfect), it's definitely not going to be made by the EPA, it's going to be a hit in performance...

[image: View Post]dawson, on 04 November 2012 - 10:31 AM, said:

@spud786

You're right, cozye did develop his map for the 02-05, so It wouldn't work as well on the 06+ models.  I looked through it before I installed it, and there weren't any drastic modifications that would have accounted for the MPG drop on my bike.  That's why I figured it was more a problem with the base map being too rich to start with.  

From what I saw with the RS, a bike that was too lean would spin right up without making a whole lot of power but it would be smooth at the initial openings.  It was when the needle was too rich that it demonstrated the snatchy throttle that I see on my VFR, which is why I'm thinking that it's too rich to start with.

As for the injectors, my guess on that came from working in electronics manufacturing.  My company would have a part number for  a specific value of component, like a memory chip.  However, there were multiple suppliers who made memory devices that fit the specs, and we would buy from all of them, which increased availability and reduced cost.  Then the boards would be built and sent out.  One could have a chip from Samsung, one from Hyundai, one from Kingston, they would all perform slightly (very, very slightly in most cases) differently, but they would all be considered to have the same part number in that location, even though the parts themselves were different.

When Kaldek was going through his problems with his bike, he got a set of injectors from an 06 and posted up some pictures.  They were twelve hole injectors, with a grey body and labelled as Denso 4480.  According to the parts fiche, my bike has the same part number, but the actual injector in there is a twelve hole with a red-orange body and labelled as Denso 8FIJG.  My guess is that Honda sent their requirements to Denso for an injector with certain physical dimensions, electrical characteristics and a flow rate between two limits.  Denso makes a number of injectors that fall within those categories, and whichever ones they have on hand when Honda places an order is what gets installed in the bikes that day.  When Honda switched from the all open loop on the 5th gen to the closed loop system on the 6th gen they would have opened up the flow rate tolerances on the injectors,  thinking that the ECU would compensate for the differences, which is why we see more of a variance in throttle response on the 6th gen bikes.  It's a perfectly logical move to cut production costs, and would have worked properly if the ECU had the long term fuel trim to adjust the base map over time.  

Seems to me that the 02-05 had injectors that had one large hole and the 06+ part has twelve holes.  Helped increase atomisation, makes for a smoother burn and, ironically for me, better fuel efficiency.

Thanks for expounding on your thoughts... which by the way you express in a very clear and concise way...

I can corroborate your experience with variability in certain aspects of parts in large production-line style manufacturing... as I have experienced it first hand on the 6th gen as well... in my case, I grafted my '03 frame onto an '07 I got with a unsalvageable title (crack in steering head from a fron-on collission) in order to update to the VTEC II version with it's VTEC engagement/disengagement hysteresis (much better idea) and avoid having to pour loads of cash into fixing several major problems that had developed on my '03 (clutch basket, timing chain, warped heads, milky oil, snuffed suspension, wire harness surely suffering from overheating... etc etc)... selling of the parts left over would pay for the purchase of the '07... and of course when I discovered the CPU didn't come with the wreck (with the wreck already paid for and in my possession, the P.O.'s mechanic had accidentally thrown it away with the broken front cowling), I realized I couldn't use mine from the '03... then there was the HISS issue (I'm in Spain)... I found a cheap 2nd hand CPU in the US (no HISS issues) and had it sent over.... low and behold the pins in the CPUs connectors were not the same... Kaldek kindly sent some over from a spare harness he had, but they weren't the same either... an electrical engineer friend of mine here mentioned that it's perfectly feasible that providers of certain subcomponents can be different from one continent to another... we managed to find a solution and the bike fired up first pop (one also has to splice in an extra 12V supply as there are two 12V cables in the US CPU and only one on the Euro and Aussie versions...

That's why your theory makes sense when we see so many people claiming other peoples' Power COmmander maps don't work for them.

Go figure...

doesn't anyone make a remaper for the hondas stock part??

on a HD.. you get a big bore kit.. new intake. new pipes.. and you will need MASSIVE changes in the map.. so you pug the bike into a computer with a map adapter / storage
..and presto you bike has a new new map..
no powercomander needed.