Nitrous system lean spike

[oldbones]

So, I need some input. I'm running a plate on my 5.0 and I'm getting a lean spike either from a bad fuel pump, voltage drop or line pressure drop. I have attached data logs showing when this is occurring. One sheet I only sprayed fuel and the other one is with the both of the solenoids disabled.

Is this just a tired pump? A quick pressure drop from the fuel solenoid opening? or a Voltage drop issue?

Any feed back would be greatly appreciated.

 

[oldbones]

And one with the System Spraying Fuel and N20 (with the N20 Delayed about 1 Second from the fuel.)

 

[dennisn]

Where are you pulling fuel from the rail? What other mod's. Running e85? According to your first pic. your getting a lean spike without even spraying any n20???? That right there should tell you something. Your delay's not going to matter cus this spike is not caused due to nitrous getting there first. How big a shot you trying to spray?

Almost looks like a fuel pressure drop at the rail when the selonoid opens.

Oh and if you do wind up using the delay once you work everything out you'll probably only need like a 200 to 300 ms delay. (.2-.3 sec)


There should be no real voltage drop from those selonoids opening. Their not pulling that much amperage.

 

[mastwolf]

Overfueling on the hit can cause what looks like a lean spike too. Similar to a cylinder misfire when the fuel does not get burned at all.

 

[oldbones]

Where are you pulling fuel from the rail? What other mod's. Running e85? According to your first pic. your getting a lean spike without even spraying any n20???? That right there should tell you something. Your delay's not going to matter cus this spike is not caused due to nitrous getting there first. How big a shot you trying to spray?

Almost looks like a fuel pressure drop at the rail when the selonoid opens.

Oh and if you do wind up using the delay once you work everything out you'll probably only need like a 200 to 300 ms delay. (.2-.3 sec)


There should be no real voltage drop from those selonoids opening. Their not pulling that much amperage.

It's pulling from the rail and it is on E85 with a 125 shot. It has JMS bap set to 18v at at about 65%. I set the delay out that far for testing purposes. That's good know about the solenoid draw I'll focus on fueling then.

 

[oldbones]

Overfueling on the hit can cause what looks like a lean spike too. Similar to a cylinder misfire when the fuel does not get burned at all.

Interesting, so I need to have the fuel and N20 hit the cylinders at close to the same time to see if the lean spike disappears? It has a .057 N20 jet and a .043 fuel jet. It doesn't have a stumble that I can feel during acceleration.

 

[oldbones]

I'm going to play with the bap and see if adding fuel pressure will change the lean spike. I'll report back with more data logs.

 

[basspro302]

Coil up some extra nitrous line

 

[oldbones]

Coil up some extra nitrous line

It won't help, it already has a electric delay for the n20 solenoid. Which is set about one second from the fuel solenoid engagement. It's set that far out for diagnostic purposes.

 

[oldbones]

For testing purposes I turned up the BAP to 22 volts at 100% and here is the resulting data log and lean spike. The N20 is not turned on.

I'm going to try and move the fuel solenoid as close as possible to the plate and data log again. It'll probably be a few days before I get it done.

 

[basspro302]

Overfueling on the hit can cause what looks like a lean spike too. Similar to a cylinder misfire when the fuel does not get burned at all.

This is true. The o2 won't see raw fuel as rich.
What did you start your delay at?

 

[oldbones]

This is true. The o2 won't see raw fuel as rich.
What did you start your delay at?

Post 2 has the system turned on and the car spraying n20 and fuel.

Fuel starts at 29.6 seconds and than N20 Starts at 30.8 seconds.

 

[basspro302]

Did you try and start from 0 seconds delay and work your way back? Or did you just dial in a delay for the n20?

 

[oldbones]

Did you try and start from 0 seconds delay and work your way back? Or did you just dial in a delay for the n20?

I started from the opposite spectrum and threw delay at it and then data logged.

I'm going to move the solenoid and mount it directly (via a 90 deg female to female 3an fitting) onto the plate and dial down the delay. Then I'll data log it again, I should have it done by the weekend. I'll post my results up. Half of the problem is that I run the Innovative PSN-1 which is a wideband nitrous controller. I have the safety perimeters on the controller for the A/F set tight, so the controller can react quickly to any lean conditions and that spike wont even allow the system to engage.

 

[dennisn]

Hey here's something to maybe try. Test it as you've been without the nitrous and don't let the fuel go into the plate. Let it dump into a gas can or something and see if your spike goes away. If it's still there you have a pressure drop. If it goes away maybe it's rich but o2's aren't picking up unburnt fuel like others have mentioned. No need to move your selonoid that's what your tdr is for. Figure out what's actually happening then go from there to try and correct the issue.

 

[oldbones]

Hey here's something to maybe try. Test it as you've been without the nitrous and don't let the fuel go into the plate. Let it dump into a gas can or something and see if your spike goes away. If it's still there you have a pressure drop. If it goes away maybe it's rich but o2's aren't picking up unburnt fuel. No need to move your selonoid that's what your tdr is for. Figure out what's actually happening then go from there to try and correct the issue.

That's a great idea, I'll give that a try.

 

[oldbones]

Ok, here is an update. I turned off the nitrous and setup the fuel solenoid to dump fuel into a container and not inject it into the engine. Also, the BAP is set at 18v at 100%. Here are the resulting data logs.

I'm thinking about measuring the acoustical properties of the system, which are the pressure spikes the system is subject too. Not that it will solve my problem, but because I'm curious. Here is a link with some basic explanations: http://turbolab.tamu.edu/proc/pumpproc/p25/P25-Tut1.pdf I'm not a ME, but this stuff interests me.

and

 

[dennisn]

your last logs aren't really lean is it? Stft are under 1.0 which is slightly rich not lean right? I think I know what your looking at as its not as rich at the hit. I'm not a tuner and don't pretend to be one. Maybe someone with a little more knowledge can chime in.

 

[04compgt]

A dry kit with a properly sized injector and tune will not do this , I hate wet kits for this reason.

 

[oldbones]

your last logs aren't really lean is it? Stft are under 1.0 which is slightly rich not lean right? I think I know what your looking at as its not as rich at the hit. I'm not a tuner and don't pretend to be one. Maybe someone with a little more knowledge can chime in.

The pictures I post are measured AFR in lambda. The PCM targets .82 lambda on my tune. So, at .82 measured lamda the STFT would be 1.0. If the measure lambda is higher then the .82 targeted value then the STFT would read <1 and if it is too rich then the STFT are 1<.

The dip at the hit only goes to a measured .80 and then goes lean to a measured .898--at the lean spike.Then over the pull the fuel pressure stabilizes slowly and the PCM stops chasing its tail. First gear has the most swing, because the pressure change events happen so quickly.

I might put a side branch helmholtz resonator right at the fuel rail adapter and try to smooth the pressure swings, so the PCM get the STFT correction directly after the initial pressure change, without such a drastic change in pressure spikes. Here is a link to an explanation: http://turbolab.tamu.edu/proc/pumpproc/p25/P25-Tut1.pdf