Basic Engine Tuning 101
08-13-2010, 03:52 PM
#31
Rennlist Member
I think the pumping losses are reduced and you get more net rear wheel hp for the same mass flow. there are probably gains in the areas of scavenging that are much more complicated to explain as well, that would address more mass flow as well. I dont know which would provide the most gains. backpressure reduction or pumping losses, and adding to the list would be the effect of headers , equal length for example which gain in both areas I would guess.
I believe when reducing back pressure, this allows more air mass to flow through the engine so the MAF will see this difference and automatically compensate for it.
As already mentioned, that would not be true if you changed the fuel pressure or injector size.
The LH doesn’t meter the fuel per se, it thinks it knows how much it is giving base on open time and pre programmed amounts. Change the fueling = change the maps.
As already mentioned, that would not be true if you changed the fuel pressure or injector size.
The LH doesn’t meter the fuel per se, it thinks it knows how much it is giving base on open time and pre programmed amounts. Change the fueling = change the maps.
08-13-2010, 03:56 PM
#32
Rennlist Member
I was up to 71psi on the stock injectors to reach 12.5:1 for most of the HP/rpm curve. (at 335rwhp at its best). when I added the stroker, i have a hard time getting above the very rich 11:1 range, with an injector much larger than stock, and at only 45psi. I think the HP level Im at now, (i.e. 372rwhp) might have been better served with the '85 fuel injectors, which are larger than stock S4s. (I cant remember, but stock was 19lb, S3 was 25lb and I have 30lbers now.)
mk
mk
I believe what you're asking though is : If you increased the pressure would the LH and stock chip compensate for it. I believe the maps are tuned to an expected pressure and injector size.
Ken (Porken) found that with his modded chips, the stock injectors were approaching 100% duty cycle at WOT so he increased the pressure (with the '87 FPR) to 55psi and was able to dial back the maps so the injector duty cycle was lower. That implies (to me) that the chips need to be dialed in for the FPR and injector size (or flow) in order to work.
Ken (Porken) found that with his modded chips, the stock injectors were approaching 100% duty cycle at WOT so he increased the pressure (with the '87 FPR) to 55psi and was able to dial back the maps so the injector duty cycle was lower. That implies (to me) that the chips need to be dialed in for the FPR and injector size (or flow) in order to work.
08-15-2010, 09:52 PM
#33
Drifting
Thread Starter
That’s a good segway into what I’d like to explore next. Let’s talk about fueling for a moment starting with the basics and moving up from there. Again, I’ll though out what I think I know… please feel free to poke holes and or correct where necessary.
What are the relevant components? Injectors, pressure regulators, and the LH box, plus everything connected to the LH. Am I missing anything? Okay, there is a tank, some lines and a pump or two and a few other misc parts.
Fueling is basically feeding the engine more or less gas for a given set of inputs by turning on and off injectors. Injectors are rated by flow in pounds of gas per hour, bigger number means more gas (not sure how pressure effects these, but I believe you can only crank it up so much). Is this rating the maximum that will flow through at a certain pressure level if they were on all the time?
Pressure regulators along and dampers keep the pressure in the rails constant unless it a rising rate regulator that increase the pressure or other form of adjustable. Pressure is created by the pump at the tank and is kept constant by a stock a regulator. The regulator does its job by allowing more or less fuel to return to the tank while the dampers smooth out the bumps or pulse created when the injectors click on and off. Higher the pressure the more fuel delivered per given amount of time of injector on time.
Timing for fuel is not as complicated as ignition. Basically all the injectors turn on and all the injectors turn off at the same time (non-sequential). RPMs still come into play, because fuel is metered by amount fed per revolution (or two revolutions) of the crank. At lower RPMs, the injectors fire once for each revolution. At higher RPMs, the injectors fire once for every two revolution so they don’t run out of duty cycle (the ratio of on/versus off where 100% equals on all the time). Ideally, one doesn’t want to push injector past 85% duty cycle or there about.
The injectors are also responsible for atomizing the fuel and there is whole other voodoo science revolving around nozzle and spray pattern design. Ours spray the fuel just outside of the intake valve where is sits and waits until the valve opens. Some fuel sits around longer on certain cylinders.
My understanding is Oxygen sensors (invented by Bosch in the 60’s) sniff the exhaust for unconsumed fuel. Instead of sensing oxygen, it senses demand for oxygen created by unburnt fuel. This makes sense for rich conditions, but I don’t understand how this would work for lean. Once lean, there isn’t any extra fuel left right? Intuition would have it you would need something that senses full for rich condition and oxygen for lean conditions. What is the missing link here?
There are narrow band (used in steady state closed loop driving conditions) that ‘fine tune’ the A/F when used. There are also wide band O2 sensors that are needed for tuning and can sense and measure/report back much richer and leaner conditions than narrow bands. A wideband can be setup to mimic a narrowband, but a narrow band does not have the range necessary for tuning.
What are the relevant components? Injectors, pressure regulators, and the LH box, plus everything connected to the LH. Am I missing anything? Okay, there is a tank, some lines and a pump or two and a few other misc parts.
Fueling is basically feeding the engine more or less gas for a given set of inputs by turning on and off injectors. Injectors are rated by flow in pounds of gas per hour, bigger number means more gas (not sure how pressure effects these, but I believe you can only crank it up so much). Is this rating the maximum that will flow through at a certain pressure level if they were on all the time?
Pressure regulators along and dampers keep the pressure in the rails constant unless it a rising rate regulator that increase the pressure or other form of adjustable. Pressure is created by the pump at the tank and is kept constant by a stock a regulator. The regulator does its job by allowing more or less fuel to return to the tank while the dampers smooth out the bumps or pulse created when the injectors click on and off. Higher the pressure the more fuel delivered per given amount of time of injector on time.
Timing for fuel is not as complicated as ignition. Basically all the injectors turn on and all the injectors turn off at the same time (non-sequential). RPMs still come into play, because fuel is metered by amount fed per revolution (or two revolutions) of the crank. At lower RPMs, the injectors fire once for each revolution. At higher RPMs, the injectors fire once for every two revolution so they don’t run out of duty cycle (the ratio of on/versus off where 100% equals on all the time). Ideally, one doesn’t want to push injector past 85% duty cycle or there about.
The injectors are also responsible for atomizing the fuel and there is whole other voodoo science revolving around nozzle and spray pattern design. Ours spray the fuel just outside of the intake valve where is sits and waits until the valve opens. Some fuel sits around longer on certain cylinders.
My understanding is Oxygen sensors (invented by Bosch in the 60’s) sniff the exhaust for unconsumed fuel. Instead of sensing oxygen, it senses demand for oxygen created by unburnt fuel. This makes sense for rich conditions, but I don’t understand how this would work for lean. Once lean, there isn’t any extra fuel left right? Intuition would have it you would need something that senses full for rich condition and oxygen for lean conditions. What is the missing link here?
There are narrow band (used in steady state closed loop driving conditions) that ‘fine tune’ the A/F when used. There are also wide band O2 sensors that are needed for tuning and can sense and measure/report back much richer and leaner conditions than narrow bands. A wideband can be setup to mimic a narrowband, but a narrow band does not have the range necessary for tuning.
08-15-2010, 11:22 PM
#34
Rennlist Member
A couple of points;
The fuel pressure regulator doesn't hold a constant pressure. The purpose of the vacuum line going to the FPR is to allow the pressure to vary. The intent is to keep the pressure differential between the fuel rail and the intake manifold approximately constant. by holding the pressure differential constant it makes it easier for the LH to control mixture as the flow rate is directly proportional to the open time. basically as the manifold pressure decreases with closing throttle position the pressure assisting the spring in the FPR also decreases dropping the fuel pressure according to MAP pressure, and increases it correspondingly with increasing MAP.
My understanding of the narrow band O2 sensor (wide band function slightly differently as they incorporate a pumping mechanism and additional electrochemical cells) is that it measures oxygen potential on the inside and outside of the exhaust stream, not the amount of fuel present. there is a zirconia membrane with a permeable platinum layer on each side. the oxygen potential of the exhaust gas is balanced against the oxygen potential of the outside air as each gas stream generates a voltage between the platinum and ziconia and these two voltages measured against one another generate the output voltage. as more air is left in the exhaust (lean) the oxygen potential is high in the exhaust steam and therefore when subtracted from the outside O2 potential you end up with a low voltage indicating the lean condition. if the O2 potential is low in the exhaust stream you subtract off a low voltage from the O2 potential on the outside generating a high sensor voltage indicating the rich condition.
The fuel pressure regulator doesn't hold a constant pressure. The purpose of the vacuum line going to the FPR is to allow the pressure to vary. The intent is to keep the pressure differential between the fuel rail and the intake manifold approximately constant. by holding the pressure differential constant it makes it easier for the LH to control mixture as the flow rate is directly proportional to the open time. basically as the manifold pressure decreases with closing throttle position the pressure assisting the spring in the FPR also decreases dropping the fuel pressure according to MAP pressure, and increases it correspondingly with increasing MAP.
My understanding of the narrow band O2 sensor (wide band function slightly differently as they incorporate a pumping mechanism and additional electrochemical cells) is that it measures oxygen potential on the inside and outside of the exhaust stream, not the amount of fuel present. there is a zirconia membrane with a permeable platinum layer on each side. the oxygen potential of the exhaust gas is balanced against the oxygen potential of the outside air as each gas stream generates a voltage between the platinum and ziconia and these two voltages measured against one another generate the output voltage. as more air is left in the exhaust (lean) the oxygen potential is high in the exhaust steam and therefore when subtracted from the outside O2 potential you end up with a low voltage indicating the lean condition. if the O2 potential is low in the exhaust stream you subtract off a low voltage from the O2 potential on the outside generating a high sensor voltage indicating the rich condition.
08-16-2010, 09:32 AM
#35
Drifting
Thread Starter
Well that makes sense and would explain why each damper and the stock regulator have vacuum lines attached. Then on my application (Murf stage one) there is an FMU that increases the pressure even more, but only when it senses boost.
I like your explanation on the O2 sensor, but I still don’t get it. If it measures the amount of oxygen present, then it should only be able to measure lean to stoich and anything rich of stoich would appear that same as stoich… as in no oxygen present.
Following is a decent article on the subject for those with enquiring minds. For the tuner, it’s safe to say the O2 sensor signals the LH via voltage which can be translated back into air to fuel ratios.
http://en.wikipedia.org/wiki/Oxygen_sensor
Have we touched on all the systems critical to tuning? I feel like I have a decent understanding of the principle systems involved, but please feel free to clarify, add, or correct.
I like your explanation on the O2 sensor, but I still don’t get it. If it measures the amount of oxygen present, then it should only be able to measure lean to stoich and anything rich of stoich would appear that same as stoich… as in no oxygen present.
Following is a decent article on the subject for those with enquiring minds. For the tuner, it’s safe to say the O2 sensor signals the LH via voltage which can be translated back into air to fuel ratios.
http://en.wikipedia.org/wiki/Oxygen_sensor
Have we touched on all the systems critical to tuning? I feel like I have a decent understanding of the principle systems involved, but please feel free to clarify, add, or correct.
08-16-2010, 10:28 AM
#36
Nordschleife Master
The narrowband sensor has a catalytic coverter chamber and then a Nerns cell that measures oxygen demand and supply. So, first, the exhaust gas is burned as completely as possible in the catalytic conversion chamber. Next, the Nerns cell measures whether the completely burned exhaust has excess free oxygen or not.
Counterintuitively, in case of a misfire, the oxygen sensor will read excessively lean. This is because the catalytic conversion chamber in the sensor gets overwhelmed and can't burn the entire mixture. So the Nernst cell sees excess free oxygen.
Wideband sensor is a combination of narrowband sensor, an oxygen pump, and a controller. The controller pumps just the right amount of free oxygen to the narrowband sensor to make the narrowband sensor read stoich. The controller then measures how much oxygen it needed to pump there.
I used a lot of semifancy words above, but don't confuse authoritative tone with me knowing what I am talking about. Caveat lector.
Counterintuitively, in case of a misfire, the oxygen sensor will read excessively lean. This is because the catalytic conversion chamber in the sensor gets overwhelmed and can't burn the entire mixture. So the Nernst cell sees excess free oxygen.
Wideband sensor is a combination of narrowband sensor, an oxygen pump, and a controller. The controller pumps just the right amount of free oxygen to the narrowband sensor to make the narrowband sensor read stoich. The controller then measures how much oxygen it needed to pump there.
I used a lot of semifancy words above, but don't confuse authoritative tone with me knowing what I am talking about. Caveat lector.
08-16-2010, 01:58 PM
#37
Rennlist Member
no, the vacuum lines do not allow for variable pressure. they only pull back the pressure by about 5psi, at idle and part throttle when there is intake vacuum. I took some measurements and posted pictures of the Sunx sensor measuring vacuum at idle and part throttle. anyway, this allows the EFI system to have an easier time regulating fuel mixture via the o2 sensor's feedback when the pressure is pulled back.
as a note, i usually dont even run the fuel regulator vaccum lines. I do now, but didnt then.
as a note, i usually dont even run the fuel regulator vaccum lines. I do now, but didnt then.
A couple of points;
The fuel pressure regulator doesn't hold a constant pressure. The purpose of the vacuum line going to the FPR is to allow the pressure to vary. The intent is to keep the pressure differential between the fuel rail and the intake manifold approximately constant. by holding the pressure differential constant it makes it easier for the LH to control mixture as the flow rate is directly proportional to the open time. basically as the manifold pressure decreases with closing throttle position the pressure assisting the spring in the FPR also decreases dropping the fuel pressure according to MAP pressure, and increases it correspondingly with increasing MAP.
My understanding of the narrow band O2 sensor (wide band function slightly differently as they incorporate a pumping mechanism and additional electrochemical cells) is that it measures oxygen potential on the inside and outside of the exhaust stream, not the amount of fuel present. there is a zirconia membrane with a permeable platinum layer on each side. the oxygen potential of the exhaust gas is balanced against the oxygen potential of the outside air as each gas stream generates a voltage between the platinum and ziconia and these two voltages measured against one another generate the output voltage. as more air is left in the exhaust (lean) the oxygen potential is high in the exhaust steam and therefore when subtracted from the outside O2 potential you end up with a low voltage indicating the lean condition. if the O2 potential is low in the exhaust stream you subtract off a low voltage from the O2 potential on the outside generating a high sensor voltage indicating the rich condition.
The fuel pressure regulator doesn't hold a constant pressure. The purpose of the vacuum line going to the FPR is to allow the pressure to vary. The intent is to keep the pressure differential between the fuel rail and the intake manifold approximately constant. by holding the pressure differential constant it makes it easier for the LH to control mixture as the flow rate is directly proportional to the open time. basically as the manifold pressure decreases with closing throttle position the pressure assisting the spring in the FPR also decreases dropping the fuel pressure according to MAP pressure, and increases it correspondingly with increasing MAP.
My understanding of the narrow band O2 sensor (wide band function slightly differently as they incorporate a pumping mechanism and additional electrochemical cells) is that it measures oxygen potential on the inside and outside of the exhaust stream, not the amount of fuel present. there is a zirconia membrane with a permeable platinum layer on each side. the oxygen potential of the exhaust gas is balanced against the oxygen potential of the outside air as each gas stream generates a voltage between the platinum and ziconia and these two voltages measured against one another generate the output voltage. as more air is left in the exhaust (lean) the oxygen potential is high in the exhaust steam and therefore when subtracted from the outside O2 potential you end up with a low voltage indicating the lean condition. if the O2 potential is low in the exhaust stream you subtract off a low voltage from the O2 potential on the outside generating a high sensor voltage indicating the rich condition.
08-16-2010, 08:50 PM
#38
Rennlist Member
no, the vacuum lines do not allow for variable pressure. they only pull back the pressure by about 5psi, at idle and part throttle when there is intake vacuum. I took some measurements and posted pictures of the Sunx sensor measuring vacuum at idle and part throttle. anyway, this allows the EFI system to have an easier time regulating fuel mixture via the o2 sensor's feedback when the pressure is pulled back.
as a note, i usually dont even run the fuel regulator vacuum lines. I do now, but didnt then.
as a note, i usually dont even run the fuel regulator vacuum lines. I do now, but didnt then.
08-16-2010, 09:08 PM
#39
Rennlist Member
I guess if variable is 2,yes its variable. as soon as a little vacuum is removed, like giving it a little gas, the system goes to full pressure. Im sure there is a more subtle variable relationship, but its not like the pressure is changing for different levels of manifold pressure, like a MAP system. 
not exactly sure what your definition of variable is but if the vacuum pulls the pressure back with increased vacuum that seems variable to me. At any rate, I know it can't hold the pressure differential exactly constant but that is the general idea. so you mention it pulls about 5psi at idle, and if we assume an idle vacuum of around 18 in-hg that would be about 8 psi. so not exactly 1 for 1 but the general idea is there. I also realize you can run without the vac attached and the system will compensate fairly well, but that is the design intent of the vac connection on the FPR.
08-19-2010, 09:14 AM
#41
Nordschleife Master
Here's a tuning question to everyone.
Suppose that I am cruising at high gear, low load, and low rpm. Then, if I floor the gas pedal suddenly, the car stumbles and hesitates for a fraction of a second before starting to pull hard. If I don't instantly floor the pedal but instead let it take even just a fraction of a second, there's no hesitation.
It's very subtle, I don't think there's much of a chance to capture this on tape with my data loggers.
Why is this happening and how do I eliminate it?
Guess 1: The port walls act as a temporary fuel buffer or fuel storage. (I think "they" call this "tau".) In steady state at high load, the walls are wet. In steady state at low load, the walls are dry. Both high load and low load states are turned for steady state operation. Now, when there's a sudden transition from low load to high load state, the walls are temporarily dry at the high load state and the high load state is not tuned for this. The car will run lean and hesitate until the walls are wetted.
Guess 2: Some of the many transition parameters in LH or EZK is set nonsensically and doing something unexpected.
Interested in what you guys think.
Suppose that I am cruising at high gear, low load, and low rpm. Then, if I floor the gas pedal suddenly, the car stumbles and hesitates for a fraction of a second before starting to pull hard. If I don't instantly floor the pedal but instead let it take even just a fraction of a second, there's no hesitation.
It's very subtle, I don't think there's much of a chance to capture this on tape with my data loggers.
Why is this happening and how do I eliminate it?
Guess 1: The port walls act as a temporary fuel buffer or fuel storage. (I think "they" call this "tau".) In steady state at high load, the walls are wet. In steady state at low load, the walls are dry. Both high load and low load states are turned for steady state operation. Now, when there's a sudden transition from low load to high load state, the walls are temporarily dry at the high load state and the high load state is not tuned for this. The car will run lean and hesitate until the walls are wetted.
Guess 2: Some of the many transition parameters in LH or EZK is set nonsensically and doing something unexpected.
Interested in what you guys think.
08-19-2010, 10:25 AM
#43
Nordschleife Master
My "warmup enrichment map" has cells labeled ..., 72C, 91C, and 93C. My "acceleration enrichment map" has cells labeled ..., 65C and 72C.
I don't think I need to screw around with the "warmup enrichment map" because with the current values the car drives otherwise fine at temperatures I've tested it and the hesitation effect is present even when the car is fully warmed up after some boosted driving. Is this logic correct?
It would be logical to adjust the "acceleration enrichment map." Before changing those parameters, I'd like to know how exactly does LH use that map. Also, it would be logical to adjust the "acceleration enrichment" settings on the fuel parameters screen, since I based on the manual those interact with the "acceleration enrichment map."
Finally, I have the "WOT enrichment map" set to zeros at and below 3700 rpm, which may also be used to change the fueling when I floor the throttle at low rpm.
The ST2 manual says on page 12:
On the right hand side [of the fuel parameters screen] are adjustments of fuelling when the throttle is opened quickly – transient enrichment. Transient enrichment is necessary to counteract the lean out that occurs in engine when the throttle is opened quickly.
“Base setting” has a default value of 0 and a range of +/-64. Positive values give a larger injector pulse length.
The next four tabs are self explanatory. They scale the enrichment applied to the first four pulses after a transient condition is detected. The default values are the standard Porsche values of 400%, 325% 200% and 100%.
The last tab is “Sustain”. This determines the decay rate of the enhanced pulses to their normal value. The default value is 2, and the adjustment range is 1-10.
Note that these transient enrichment adjustments should be used in conjunction with the “Acceleration enrichment” map. This is explained further in the “Editing acceleration enrichment fuel maps” section.
“Base setting” has a default value of 0 and a range of +/-64. Positive values give a larger injector pulse length.
The next four tabs are self explanatory. They scale the enrichment applied to the first four pulses after a transient condition is detected. The default values are the standard Porsche values of 400%, 325% 200% and 100%.
The last tab is “Sustain”. This determines the decay rate of the enhanced pulses to their normal value. The default value is 2, and the adjustment range is 1-10.
Note that these transient enrichment adjustments should be used in conjunction with the “Acceleration enrichment” map. This is explained further in the “Editing acceleration enrichment fuel maps” section.
EDITING THE ACCELERATION ENRICHMENT FUEL MAP
This map is used in conjunction with the transient enrichment values which are available on the RHS of the Parameters screen. The values on the Parameters screen are set when the engine is at normal operating temperature, and after main map adjustments are complete. The acceleration enrichment fuel map then allows adjustment of the transient enrichment versus water temperature, if required. Autotune is not available on this map.
This map is used in conjunction with the transient enrichment values which are available on the RHS of the Parameters screen. The values on the Parameters screen are set when the engine is at normal operating temperature, and after main map adjustments are complete. The acceleration enrichment fuel map then allows adjustment of the transient enrichment versus water temperature, if required. Autotune is not available on this map.
- What exactly does the "Base setting" on the fuel parameters govern?
- In the maps that have temperature columns, at temperatures greater than the last cell value? Are those tapered to zero or will they asymptote at constant level equal to the last cell.
- Does the "WOT enrichment map" have any curve ***** to it, or does it always give the prescribed enrichemnt at WOT no matter what the other operating conditions?
- None of these settings seem to be influenced in any way by the load signal. Can I thus forget about any differences between regular and super MAF?
John's a busy guy, so anybody in the know please chip in.
08-19-2010, 12:40 PM
#44
Rennlist Member
Sorry, I was incorrect, trying to rely on my memory...
Suggest you edit Accleration enrichment map. If you change the 72degC cell (say) from std value to 100 then there will be an enrichment overshoot. This overshoot is always there when at lower temps (say 40degC) but above 72deg is not with the stock cell value.
I've just tested this modification on my test jig, just to be sure this time :-)
Sorry for the mis-information.
Some of your questions are difficult to answer, for example, the acceleration enrichment is complex. The first time you snap open the throttle, you get the enrichment over shoot (oly at lower Temp 2 with stock maps) , if you repeat quickly on the throttle next time a bit less and then again there's no further overshoot. Wait a little while and you get it again. The code is pretty difficult to work out. We've done what we can, brought out parameters which could be useful to have access to and adjust.
Suggest you edit Accleration enrichment map. If you change the 72degC cell (say) from std value to 100 then there will be an enrichment overshoot. This overshoot is always there when at lower temps (say 40degC) but above 72deg is not with the stock cell value.
I've just tested this modification on my test jig, just to be sure this time :-)
Sorry for the mis-information.
Some of your questions are difficult to answer, for example, the acceleration enrichment is complex. The first time you snap open the throttle, you get the enrichment over shoot (oly at lower Temp 2 with stock maps) , if you repeat quickly on the throttle next time a bit less and then again there's no further overshoot. Wait a little while and you get it again. The code is pretty difficult to work out. We've done what we can, brought out parameters which could be useful to have access to and adjust.
08-19-2010, 12:49 PM
#45
Race Car
When I first started messing with these parameters, I found it useful to set them to something very large or very small, then observe the effects. For example, when fully warmed up, setting the 72deg acceleration enrichment +100, when I would try to start going from a stop, the car died immediately. Basically the LH was dumping a huge amount of fuel when the throttle was opened and killed the engine. I suggest you play around with the settings to see what works best for you configuration, just be careful (i.e. don't be boosting when doing this or something like that).
Dan
'91 928GT S/C
475hp/460lb.ft
Dan
'91 928GT S/C
475hp/460lb.ft