Port Injection causing engine failure

[LamboLover]

Do the shotguns work then? Not reading about many using them although it sounded full of promise and looks like it has for years. Happy to be guided on the direction the market wants on this.

I have tested MEVD17 sending out CAN packets at 350Hz - ie every time a spark plug fires at 7000 RPM, but it isn't ready yet.

 

[doublespaces]

Look at it another way... You can put a lot of effort in to designing theoretically and practically performing port injection control schemes. I know we've put a lot of effort in to our implementation. But at the end of the day, the port injection itself only exists to mask a fundamental issue with our high pressure system. Specifically the fuel pump unable to keep pressure up with high volume. We don't need a better port injection control scheme (IMHO) what we need is a better high pressure fuel pump.

I may be out of the loop, but has there been any consideration on JB4 control over the shotgun? Does the JB4 rely on a hobbs for the second LPFP as well, or is that dictated by the JB4 itself? I apologize, I've not looked into these topics before.

 

[Bmwfixerguy1]

Do the shotguns work then? Not reading about many using them although it sounded full of promise and looks like it has for years. Happy to be guided on the direction the market wants on this.

I have tested MEVD17 sending out CAN packets at 350Hz - ie every time a spark plug fires at 7000 RPM, but it isn't ready yet.

Yes they work, I'm pretty sure vtt has made over 750whp on DI only.
I know there is a new version releasing soon or just released so not sure if the priors had issues.
I'm gaining more and more interested by the day though

 

[Terry@BMS]

I may be out of the loop, but has there been any consideration on JB4 control over the shotgun? Does the JB4 rely on a hobbs for the second LPFP as well, or is that dictated by the JB4 itself? I apologize, I've not looked into these topics before.

Yes, it can. But on software stuff, Tony and I don't work well together. I'd have to have one on a car in front of me to design the proper control scheme. If it catches on more then we'll come up with something.

 

[doublespaces]

Do the shotguns work then? Not reading about many using them although it sounded full of promise and looks like it has for years. Happy to be guided on the direction the market wants on this.

I have tested MEVD17 sending out CAN packets at 350Hz - ie every time a spark plug fires at 7000 RPM, but it isn't ready yet.

Yes, they appear to work. The missing part seems to be the ability to access additional tables for fueling beyond a certain point(I'm not sure exactly what that means, perhaps related to handling of the higher pressures) as well as integrated control of the second HPFP/LPFP. Currently that control, while two separate issues, resides on a mixture of a hobbs switch and an AIC or similar AFAIK.

Yes, it can. But on software stuff, Tony and I don't work well together. I'd have to have one on a car in front of me to design the proper control scheme. If it catches on more then we'll come up with something.

Fair enough, hopefully we'll see about that.

 

[Terry@BMS]

I have tested MEVD17 sending out CAN packets at 350Hz - ie every time a spark plug fires at 7000 RPM, but it isn't ready yet.

We can't read CANbus that fast continuously at least not on the current JB4 hardware. But, we can read analog signals at 2,000hz and digital at 15,000hz probably. We also have that relay on the coils for 2STEP/NLS we can fire to cut spark. Those take ~100ms to fire though.

It seems 98% of the issue, at least on our JB4 implementation which is much more advanced than say an AIC6 implementation, is more of a theoretical issue rather than a practical issue. I'm not seeing practical issues cropping up with our PI control and we've got hundreds of controllers out there in the field.

 

[Bmwfixerguy1]

Yes, they appear to work. The missing part seems to be the ability to access additional tables for fueling beyond a certain point(I'm not sure exactly what that means, perhaps related to handling of the higher pressures) as well as integrated control of the second HPFP/LPFP. Currently that control, while two separate issues, resides on a mixture of a hobbs switch and an AIC or similar AFAIK.



Fair enough, hopefully we'll see about that.

I wonder how much further along our platform would get if all the vendors were given glow sticks and extacy and locked in a room together for the night..

First n54 to crack 1k would be the following morning around 9am LMAO

 

[LamboLover]

I hear you Terry, but from a control point of view I require my cylinder cuts to be deterministic even if practically a delayed/inferred method mostly works. There is something upsetting folk about the DI and PI solutions beyond a bit of internet hysteria.

 

[Terry@BMS]

Don't get me wrong I'm always up for improving things if it makes sense with minimal drawback.

But in terms of anecdotal evidence for almost 10 years thousands of N54 owners have been filling their plenum with fuel using traditional water/meth kits, with rarely an issue observed, other than the occasional intake manifold explosion if trying to no lift shift without a relay to cut spark.

 

[V8bait]

Water meth is a lot safer then the PI E85 if a DI injector cuts out.

 

[Bmwfixerguy1]

Water meth is a lot safer then the PI E85 if a DI injector cuts out.

What's the science behind that, I'm intrigued. I know you have quite a bit of meth experience.

Also we're you referring to traditional CP injection or fuel?

Also shoot me an email of that diagram we messaged about when you have time

 

[Terry@BMS]

Water meth is a lot safer then the PI E85 if a DI injector cuts out.

I don't see why it would be any different especially when most run straight methanol.

 

[pysical]

I am trying to decide on my fueling upgrade. It seems that I only really have two options. I go with the VTT Double Barrel, or I go with a PI Kit and JB4 for SOME sort of safety. Doing the math, the VTT DB is cheaper and from my low amount of knowledge, it is also safer?

 

[langsbr]

I vote the shotgun. However, there are pros and cons of both types, and PI is much more prevalent. I think with the newer style shotgun, we may start to see more. I know that's what I'm going with.

 

[Jake@MHD]

Do the shotguns work then? Not reading about many using them although it sounded full of promise and looks like it has for years. Happy to be guided on the direction the market wants on this.

DME control of the second pump HPFP in shotgun config might be worth us trying as well.

 

[pysical]

We just need @Chris@VargasTurboTech to get us a group buy on the new kit.

 

[Terry@BMS]

DME control of the second pump HPFP in shotgun config might be worth us trying as well.

Right now as far as I know Tony is using a static table within an AIC6 to map out the solenoid input on boost and engine speed to the flow level out of the second pump he needs to maintain high pressure. But that is going to be hit and miss and require constant adjustment IMHO. What his system needs is it's own PID control loop to set dutycycle to the pump inlet solenoid based on a high pressure set point. If the hardware catches on at some point I'll setup the JB4 for it, should be really easy. But if you can just give him a low speed PWM he can map on boost and engine speed, that would probably save him having to install an AIC6 for it as implemented.

 

[V8bait]

I don't see why it would be any different especially when most run straight methanol.

Straight methanol is significantly higher octane than even e85, and the total % of fueling with the meth is generally much lower than with PI. For instance, I've seen the JB4 PI over 80% duty cycle for no obvious reasons... Boost is modest 24psi, hpfp at target, AFR rich, fuel trims negative. Lose DI at that point and the AFR is going to be much leaner, but still potentially combustible. Lose DI with a sprinkle of meth and it's going to be so lean it may not even combust. Well, unless the person is using like 6 meth injectors. Because that's kinda how PI with 80% duty cycle is.

I think DME or jb4 control of a second hpfp would be nice to have. The current table is 3D mapped based on rpm and boost, and much like PI it still has the ability to hit different hpfp targets since the DME is controlling the factory hpfp. There is room for improvement still just like PI.

 

[Terry@BMS]

The dutycycle on the PI injectors is sort of irrelevant without knowing the boost pressure and low fuel pressure, e.g. the injection pressure. If boost is 25psi and low fuel pressure is 50psi, you're not moving nearly as much fuel as you think you are at 80% dutycycle. I guess we can do the math but it's not uncommon for WMI guys to run dual CM10 nozzles to fuel 600whp+ on E50, also a ton of fuel in there.

My point was anecdotally we are not seeing the problems in practice that some are theorizing. So there likely is some flaw in the theory.

 

[Jake@MHD]

Right now as far as I know Tony is using a static table within an AIC6 to map out the solenoid input on boost and engine speed to the flow level out of the second pump he needs to maintain high pressure. But that is going to be hit and miss and require constant adjustment IMHO. What his system needs is it's own PID control loop to set dutycycle to the pump inlet solenoid based on a high pressure set point. If the hardware catches on at some point I'll setup the JB4 for it, should be really easy. But if you can just give him a low speed PWM he can map on boost and engine speed, that would probably save him having to install an AIC6 for it as implemented.

All of that can be done in the DME.