Getting atrocious mileage?

[JohnCh]

@redursidae welcome to the world of se7ens where Alpha-N is the norm.  Even the MBE in new Caterhams employs this older strategy with time-based injector control. It's certainly not as good as a VE-based PID control, but it's certainly possible to achieve good behavior with these less sophisticated ECUs. Over the last 22 years, I've run Emerald and MBE on Duratecs with ITBs and have never had running issues related to the mapping strategy.

 

Your advice is spot on when it's time to replace the Pectel, but I suspect @Vovchandr's issues run far beyond his map or mapping strategy. Fuel milage doesn't plummet by 40% unless there is some kind of failure. He needs to address that first. I suppose it's possible that the Pectel has a driver failure that is intermittently holding open an injector, but other than that, I'd start by looking at all sensors that can trim the injection map, run a compression and leak down, then check injector flow to see if one or more is bad. If the AFR numbers provided above are accurate, I'd be very concerned about bore wash. Given all the issues with this particular engine over the years, I suspect there are multiple things going on. Occam's razor won't apply.

[redursidae]

1 hour ago, JohnCh said:

@redursidae welcome to the world of se7ens where Alpha-N is the norm.  Even the MBE in new Caterhams employs this older strategy with time-based injector control.

 

Even in the single throttle cars?  I would have expected the MBE to be a VE model at least, but nothing wrong with Alpha-N as long as it has the right compensations. I'm very interested in checking out a map if anyone has one to share.

 

1 hour ago, JohnCh said:

Your advice is spot on when it's time to replace the Pectel, but I suspect @Vovchandr's issues run far beyond his map or mapping strategy. Fuel milage doesn't plummet by 40% unless there is some kind of failure. He needs to address that first. I suppose it's possible that the Pectel has a driver failure that is intermittently holding open an injector, but other than that, I'd start by looking at all sensors that can trim the injection map, run a compression and leak down, then check injector flow to see if one or more is bad. If the AFR numbers provided above are accurate, I'd be very concerned about bore wash. Given all the issues with this particular engine over the years, I suspect there are multiple things going on. Occam's razor won't apply.

 

I took a read through the posts and I see what you mean about something beyond the map being the cause. I'm suspicious of the TPS and MAP readings from his screenshot. It's possible he doesn't have MAP if using pure Alpha-N, but the TPS is critical and should read 0% with the engine off. Of course, that's assuming it was set to 0% when the mapping was done, in which case would explain some of the richness. I know the MBE's use load site based axis, mapping volts instead of throttle percentages, but this Pectel seems to be percent based no?

Edited 35 minutes ago by redursidae

[JohnCh]

6 minutes ago, redursidae said:

Even in the single throttle cars?  I would have expected the MBE to be a VE model at least, but nothing wrong with Alpha-N as long as it has the right compensations. I'm very interested in checking out a map if anyone has one to share.

 

Yep! The MAP sensor is really just there for barometric compensation. A TPS controls the tables. I can share my map that was done on a dyno for my 2.4L. I'll shoot you a PM.

[Vovchandr]

17 minutes ago, redursidae said:

 

Even in the single throttle cars?  I would have expected the MBE to be a VE model at least, but nothing wrong with Alpha-N as long as it has the right compensations. I'm very interested in checking out a map if anyone has one to share.

 

 

I took a read through the posts and I see what you mean about something beyond the map being the cause. I'm suspicious of the TPS and MAP readings from his screenshot. It's possible he doesn't have MAP if using pure Alpha-N, but the TPS is critical and should read 0% with the engine off. Of course, that's assuming it was set to 0% when the mapping was done, in which case would explain some of the richness. I know the MBE's use load site based axis, mapping volts instead of throttle percentages, but this Pectel seems to be percent based no?

 

My TPS situation perplexes me. When I bought the car it had an IAC with ITB's with a vacuum rail plugged into all the runners of ITBs going to the IAC. 
 

I couldn't figure out how Pectel controls the IAC so I bypassed it. All vacuum lines are plugged and ITB's idle with air sneaking by the butterflies. 

 

Here is my TPS data

[redursidae]

7 minutes ago, Vovchandr said:

 

My TPS situation perplexes me. When I bought the car it had an IAC with ITB's with a vacuum rail plugged into all the runners of ITBs going to the IAC. 
 

I couldn't figure out how Pectel controls the IAC so I bypassed it. All vacuum lines are plugged and ITB's idle with air sneaking by the butterflies. 

 

Here is my TPS data

Im not saying that this is the root cause of your issue, but the TPS calibration needs redone. The TPS voltage should typically be between 0.5V and 4.5V. Outside of that range it usually means a sensor or wiring fault. You should set the 0 point to the voltage seen with the engine idling and butterflies at the resting position (4.49V for your case). Similarly for the 100 point and linearly interpolate between the two. Changing this will of course change your fueling. I have seen cars where the TPS voltage changes between drives and that’s an indication of a failing sensor or counterfeit sensor. 

 

And regarding an IAC for idle control, I personally find it to be a must for a street car. I run one with my TWM manifold similarly to what you described. The simpler method is using a PWM actuated solenoid, but for this to work the TPS has to read below the threshold set in the map. If you post of pic of the solenoid I can tell you what kind it is.