scalbert

Well, the resistor trick did not work 100%. It did provide significant help but on a long trip last week to Raleigh, from Atlanta and back, I did hear some knock a few times when getting on it. It mostly occurred during cooler starts or while cruising on the highway for some time. Regardless it is not welcome.

I was curious how or why it occurred only when the VTEC solenoid engaged; about 5000 - 5500 revs, so I decided to investigate this. I did so by monitoring my O2 voltage at WOT. It drops right after the VTEC change-over occurs and climbs back up by 6k revs. I also enlisted the help of my fellow east coast 6-Speed SC owner and his dyno with A/F ratio measurements confirmed this.

Also talking with Comptech they have seen the same thing and the reason why they are working on a piggy back computer to help this situation. This also occurs with the S2000…

But before completely resorting back to the lower boost pulley and I needed one more try. Obviously backing off timing a few degrees was not enough to overcome the slight lean condition during the VTEC change-over. Nor could I bump up fuel pressure anymore as it has, and would, set a code. So I needed a new route; more fuel during VTEC...

Then I remembered an old trick; use a cold start valve to add fuel when needed. So I picked up a Bosch cold start valve for an '85 BMW 535i for $40 and began installing it tonight:



A cold start valve is basically a fuel injector which is meant to stay open for ten to twenty seconds at a time. Versus a regular injector which is meant to be pulsed.

I picked up a simple relay which will be tied into the VTEC control solenoid signal. This will close the relay and open the cold start valve delivery the needed fuel. Based on some simple number crunching this will provide more than enough fuel to quench the lean condition and starve off knock. It might actually be too much but I can always restrict the flow to this valve.

More pics of the parts, more to follow tomorrow night along with results if ready. Obviously this is targeted at very few people but it might benefit some.

Manifold Inlet:


I will probably weld in a bushing to offset the injector in order to decrease the amount it intrudes into the passage. Also I will be polishing the elbow tomorrow evening to match the manifold inlet.

allmotor_2000

Interested to know how this turns out... scalbert's custom AIC!

scalbert

Crude but hopefully effective...

allmotor_2000

Steve... simple suggetion to help regulate flow out of this injector. Use a NOS fogger for fuel. This way you can vary the size of the jet... hence control the amount of fuel. Use a relay that is activated by BOTH the VTEC and WOT (from the TPS). When both conditions are active (an AND situation)... whoa... fuel

scalbert

Actually a simple needle valve works wonders for regulating and the use of a cold start valve is less than a NOS solenoid/nozzle set up.

Plus it is not like these valves are not meant for this type of duty...

types1967

well good luck scalbert hope everything works out ok.i just saw the pics of the piping of the supercharger elbow i wondered if you cleaned it up by polishing it it would give you more power since the inlet tube from the blower looks like it needs a little cleaning up.

Jens H.

Interesting Interesting ,I hope that the fuel is distributed somewhat evenly .Jens

scalbert

I agree with ya Jens, that is a concern but not more so than the older TBI system. So it should work.

One other concern I had was pooling in the center chamber. Some fuel could get drawn into this region and "fall out"...

Jens H.

well with those old school CSI etc. systems this additional injector thing was just to get the beast to run .I spoke to Types1967 about the idea of going with water injection which would have additional benefits above and beyond just controlling detonation issues.My supercharged Miata at 6.5 psi boost has some detonation and the plan is to get water injection on this thing to get rid of it.An intercooler is another option but the additional plumbing and high cost have turned me off .High quality water injection (could have a little kick with alcohol) has the benefit of cooling the intake charge and cleaning up the combustion chamber of deposits which further reduces the chance of detonation etc.My other concern is that the ECU may "see" the rich mixture and try to dial down the main injectors to compensate .I am not sure if this would be an issue but it is worth a thought .In effect this may kill the potential benefits of the add on injector .But it is worth a shot as I am not sure how fast out ECU's really react to changes in the exhaust mixture .Jens

scalbert

I'm very familiar with H20 injection; I had the ERL Aquamist system on my old GTP. But just to test this concept I am going low budget at this point. I'll have only about $60 into it in the end.

As for the ECU seeing the condition, it really should not. Since a simple switching O2 sensor is used (non-wide band) the readings are not reliable for the ECU at WOT. Plus, at this rev range it should certainly be in open loop ignoring any feedback signals. At least this is in theory...

Jens H.

Believe it or not i have seen these things throw check engine lites for some pretty minor problems.It should be interesting to see how this trick works out.Jens

allmotor_2000

The reason I suggested the NOS fogger is that you can even pulse the solenoid... fuel regulator.

Anyways... under boost there should be NO problem with fuel going where it should

The ECU might throw a 'running rich' code... remember I had that problem Steve while running the Vortech FMU? Worst-case... just regulate the fuel until you have like a 12.0:1 A/F and no DTC.

Jens H.

That's what i was getting at.
Jens

scalbert

I've set the rich code by bumping pressure up too much. But since this will only add extra fuel when at heavy load the ECU would not see this as it doesn't monitor the O2 signals; open loop.

My idea is to not run much richer; I just want it at the same level it runs at 4500 revs...

Jens H.

Sounds cool to me i like it.jens

scalbert

Just a few new pictures. I have the elbow polished for the most part and the JBWeld is drying on the bushing. All should be done it in the next few nights.



New cold start injector installed:

juice

I was thinking the same thing. A needle valve would probably be the way to go.

haiduc

Greddy makes something called the Rebic IV.

Basically its a small stand-alone for additional injectors. You can set the duty cycle and connect up to 8 additional injectors for a FI car.

Just a suggestion, albeit a more pricey one than the aforementioned.

types1967

hey steve,allmotor,jens ive found the easy way just run some sunoco 100 like i did and no detonation whatsoever.problem is 4.29 a gallon so figure extra 22-25 a week for piece of mind.or just run the 100 when you want to have fun at wot.i use it and love it car really flies but again their are cheaper alternatives like what steve is talking about.i think what im going to do is buy a stand alone engine management system from aem and have it dyno tuned.yes its expensive but to really get the full benefit from the blower our cars need an ecu we can program.plus i wouldnt have to worry about detanation issues.

scalbert

There are a few additional injector controllers out there that work well. This or a piggy back computer or even a prgrammable DFI system will be the way to go. This is more of a proof of concept study to be sure that extra fuel during VTEC engagement will cure the knock issue when running 5+ psi boost.

Plus, since this is only an issue for a short period between 5k - 6k revs a full programable system may be an overkill.

types1967

HEY STEVE I SENT YOU A PM

haiduc

Good point, it's not like you're going to be getting another pulley so you can run 8lbs.

$700 is on the pricey side.

Doc.Booty

After reading this thread, i feel REALLY stupid.

allmotor_2000

I would not run 100-octane fuel on a daily basis... depending on how much you drive that can add up to a LOT of money. It's more rewarding to tune your car to run on 91-octane and then have the availability of extracting mroe power on 100-octane. When I ran at Bakersfield (6.5psi), I ran VP Racing Motorsport 103. I could have run 10psi and had no detonation... but that still wouldn't fix a lean condition

ferizzo

ABreece, thats cuz we are!

haiduc

AEM EMS?

randomwalk101

hey scarbert, why don't you cut off that valve a little? It kind of pertruding into the intake too much??

scalbert

There is no way I would cut that... Who knows what required items might be affected not to mention the spray pattern.

scalbert

I just put it back together and wired up the relay control but I still need to make up the new fuel line before I can test it. But I did discover on flaw that I can address easily enough.

I hooked up an LED to give me a visual indicator that the circuit to the cold start valve has closed. I basically have a relay that is closed with the VTEC solenoid control signal. This all works great and the LED comes on when the VTEC solenoid is engage right at 4800 revs. And it continues to be lit until the revs drop just below 4800 revs. That is the problem.

During a shift the throttle would be closed for a moment with little airflow. But in the current set up the cold start valve would remain open dumping fuel into the manifold. The most drastic case would be if I got on it, took it up to about 6500 revs and coasted down. This could leave the valve open longer than desired running rich for a moment.

To fix this I'll use my Hobbs pressure switch I was using to spoof ECT. I can run the control signal through it and then onto the cold start valve. This way fuel will only flow when boost is above about 3.5 PSI and when VTEC is engaged.

scalbert

I got everything finished up last night and took it for a ride and so far so good. But I will reserve complete satisfaction until I get about 1000 knock free miles on it with varying climate conditions. But at the first hint of any pinging the lower boost pulley goes back on, most likely permanently.

But right now it seems good and strong, I even lowered the fuel pressure a little to help the bottom end. I do have the control signal to the cold start valve wired through the Hobbs switch so when I let off when above 4800 revs the extra fuel flow shuts off.

I have about $90 into it which if this pans out might be a bargain to those who might want to run a little more boost from their blower but not get into a full computer.

What also turned out nice is that I had some new -6 AN fittings in my office drawer that have been there for about five years. This way I did not need to cut any lines. I just removed the line provided by Comptech and installed the one I made up. Turned out nice and clean and I even found a new injector harness for the cold start valve. I tried to make it as clean looking as possible and with the engine cover back on it should not look any different.

Overview shot of the installation area:



Cold Start Valve installation with wiring and fuel line:



New fuel line fittings at dampener:



New fuel line fittings at FPR:



Hobbs pressure switch set at 3.5 PSI



I didn't take pictures of the boost gauge and engagement light since they are strapped to the steering column and not the nicest looking installation. But I have a gauge mount coming in, might be here today, which will make a permanent home for the gauge and light. I went with a steering column mount instead of the A-pillar since I only plan on this one extra gauge and the Sport-Comp Autometer gauges match the factory silver pretty well so it should look pretty clean. Once installed I will take a snap shot of the gauge mount.

A little later I'll do up a quick schematic of the install and post the image.

Jens H.

Nice work Scalbert it get's an A in my book .Jens

scalbert

Thanks Jens, that valve flows pretty well. I opened it at idle (to purge it) and it imediately stalled the car Hopefully this will be a long term solution.

scalbert

Schematic:

fender4

from one engineer to another

ModAddict