NITROUS OXIDE ( nos / n2o ) advice forum

For the nitrous system I installed a delay box that would keep the fuel side of the system on for a programmed amount of time after the nitrous was shut down. This made up for the accumulating affect of the space volume between the activation solenoid and the nozzle jets. The accumulating affect was only a problem on the high pressure oxidizer side of the system. The amount of delay time that was needed was .7 seconds. Not a perfect solution, but well enough to keep the engine attitude stable.

I was able to test the nitrous system alone, without the aux fuel, by simply limiting max rpm during these particular tests. 6000 rpm was the limit.

Next was the aux fuel system. I had first started by trying to time a cut back of injector pw at the time of aux fuel activation (16 psi). I quickly learned that the cut back was not necessary. I then learned that additional fuel from the electronic injectors was needed as the aux fuel was activated. I quickly stepped up the electronic injectors to max duty cycle right at the point of aux fuel activation.

This is what I learned was happening with the aux fuel system at activation. Dreaded space volume again. The space volume between the aux fuel activation solenoid and the mechanical injector nozzle orifices is filled with air until the aux solenoid is activated. When the aux fuel solenoid is activated, the fuel column, which is at a pressure differential of 45 psi, rams the air from the space volume out of the injector nozzles at 45 psi. This continues to occur until the column of fuel has reached the nozzles. Approximately .5 seconds.

Imagine taking a shop air gun and pointing it down into the throat of the intake port and blasting pressurized air down the intake port. For a split second, this is what is happening when the aux fuel solenoid is activated.

This is a new form of forced induction in itself! Imagine someone using a compressed air bottle to force feed air down an intake port. That's another story though.

The blast of air from the aux fuel system demands a matching shot of extra fuel until the fuel column can reach the orifices of the nozzles. That is why the electronic injectors must be pulsed at max duty cycle right at the point of aux fuel activation. In fact, from the point of aux fuel activation on, the electronic injectors are pulsed at max duty cycle, which means that the actual injector on time must be decreased as engine rpm increases to maintain an 80 to 85 percent duty cycle. At present, I'm using 85 percent as a max duty cycle on the injectors. Fuel delivery appears to maintain stability at that duty cycle. At least, in my situation.

This last problem and solution turned out to be the fix for the entire fuel system arrangement.

Problem: Even with the electronic injectors being stepped up to max duty cycle at 16 psi to augment the shot of air from the mechanical nozzles, I would get a short lean spike at the tail end of this transition period.

This is what was happening. Soon after reaching 16 psi, the electronic injector on time needed to be scaled back, per rpm and map, to maintain the target 85 percent duty cycle. I was simply running out of injector on time to cover the aux fuel transition period.

The Solution: Very simple really. I moved the activation point of the aux fuel system to a lower rpm and map point. This gave me much more electronic injector on time range and time before having to start scaling back the injector on time due to max duty cycle limits. It also gave me the chance to inject much more fuel at the start of aux fuel activation. I am presently maxing out the electronic injector duty cycle at a lower rpm and map, yielding me more fuel per intake cycle. The injectors maintain 85 percent for the remainder of the rpm/map rise.

With all fuel systems activated together, the rpm and map rise is practically straight line with no noticeable upsetting of the engine. There appears to still be now two short lean spikes during the activation or deactivation of each system, but the spikes are very short, reasonable and not enough to upset the engine.

The next simple test will be to program the nitrous fuel side delay box to stay on for the entire pass. This is an easy way to inject about 40 lb/hr additional fuel to see if the engine wants it or not. If it likes it, I'll be ordering larger mechanical injector nozzles. Larger orifices at the mechanical injection nozzles will also mean that the blast of air from the space volume should be with less force, although the same volume, and the time span of transition will be shorter. Possibly more tuning to follow. It never ends, does it.

Here is a video of all the runs made on a Sunday when the tuning first came together. There's still more in it.

http://youtube.com/watch?v=cVn6P_i3c-o

There are videos of better quality at our website.

http://www.drwtransmission.com

Notice how the car leaves the line like a naturally aspirated car. This type of launch should make bracket racing with a turbocharger much easier. Now, instead of the usual drama at the starting line associated with a turbo car, a person can concentrate on a good reaction time.

Very interesting stuff and a very good run.

One thing I find hard to believe though, is that the small amount of air in the fuel delivery system could be enough to influence the overall AF.
Do you have any supporting evidence that rules out other factors being involved?

Regards

I've only come to that conclusion based on all the testing and datalogs I've performed. I have no other evidence.

At first, when I was pulling back the pw because I originally was concerned that the mechanical nozzles were too large, the lean spike was very large and long in duration. It was enough to upset the motor and cause an rpm drop of hundreds of rpm. A very large stumble along with an intake backfire at times. A sure sign of leaness with alcohol. As I stepped the pw back up during this transition, the lean spike lessened and the stumble lessened. It ended up that I couldn't throw enough fuel at it during the transition period. After thinking about this for weeks, the conclusion I came up with was the only one that made sense.

You should see my fuel table. The pw rises smoothly as you would expect and then pow. There is a straight up wall right up to 85% duty cycle at 8 psi boost. The electronic injectors are never scaled back from that point on. 85% duty cycle from 8 psi onward. The resulting data log shows a short duration lean spike only at the tail end of this transition period. The transition period, in total, lasts about .5 to .8 seconds. I recently discovered that the timeline on my datalog is not accurate. The time scaling is about half what it should be. The lean spikes before would start right at activation. Now, the a/f remains constant until the very end of the transition period. After the short lean spike, the a/f recovers to a slightly leaner point than before and during the start of the transition period. 4.5 before and 4.55 - 4.6 after. The large, long lean spike before would max out the reading lean side for the complete transition period. Now the lean spike goes to around 5.46 and is the duration of a point.

Do you have any other explanations for this? I'm open to discussion.

The other point that led me to this conclusion is knowing what a blast of air directed down the opening of a flared tube can do to the total flow characteristics through the tube. The high speed stream of air will draw in additional flow from around the port opening. A perfect reason to port inject nitrous right down into the middle of the port entrance.

Another point to consider. The mechanical injection nozzles that I am using are aerated, drawing in more air from around the nozzle. So the total airflow out the nozzle is more than the fuel line and rail space volume. Much the same way that injecting down into the middle of the port draws in additional air from around the port entrance.

As far as I'm concerned, what I have discussed here is all theory to me. If anyone wants to comment on my conclusions, please do so. If there's a more plausible answer to this lean spike, I surely would like to understand what it is.

I'd suggest that when the aux fuel solenoid opens - it's dropping the fuel pressure to an unacceptable level.

That is an excellent thought. The transition takes place very soon after leaving the line. Actually, at about 4650 rpm. Well before the shift point for second. I'm very busy right at that point, trying to make sure she stays straight. She tends to kick out on the 1-2 shift. I'll need to figure out a way to monitor the fuel pressure gauge without me having to take my eyes off the track. Either video it or track the sender voltage to the gauge (electric gauge).

I'm using a voltage booster for the fuel pump that turns on at 5 psi. There is plenty of room to dial up the booster setting. Maybe that would be an easier and quicker way to find out if the fuel pressure is the culprit. If the datalog changes with an increase of boosting the voltage to the fuel pump, you may be on to something.

Couldn't agree more about the point of nitrous injection and also agree that the arrangement you have would increase the amount of air flow but I'd still be very surprised if it was enough flow and of adequate duration to have a noticable effect on AF. However I doubt I could point you in a more appropriate direction, without some hands on involvement with your project but I think Dave's suggestion would be a more likely bet from what I've picked up so far.

Regards