fuel fuel injection cut off when engine brakes the car

[Tapani]

This may lead you to the real answer:

http://www.subaru-svx.net/forum/showthread.php?t=49188

In forced power mode at least the OWC 3-4 is not engaged - thus no engine braking.

Kind regards,

Tapani

[92snowmachine]

Originally Posted by THE NEW GUY

But measuring duty cycle tells you what percentage of the period is already high. If Duty cycle goes to cero, then for sure "on time " is cero.
In my opinion, duty cycle is the most efficient way to know if any injector is receiving injection signal. But you need a reliable DMM. An oscilloscope is too expensive.
Other good way to know if there´s fuel injection is reading O2 signal. If O2 signal goes to cero, then there´s no fuel injection.
It´s too cold outside and winter is comming hard here in Montreal... I think I´ll leave this meassurements for next summer.

well i guess i will wait till next summer to tell you what's wrong with this. maybe then you will tell us what you are trying to accomplish.

[THE NEW GUY]

Originally Posted by 92snowmachine

well i guess i will wait till next summer to tell you what's wrong with this. maybe then you will tell us what you are trying to accomplish.

oh! I forget to answer that question... I was thinking that, if the injection cut off does not work in forced power mode, then I could add a relay to disconnect the injectors during this condition and improve the fuel saving while driving in forced power mode. During this activation of this relay, all the signals going to the injectors are redirected to a power resistor to avoid a CEL. A micro PLC controls all of this.

[immortal_suby]

I must be missing something. Don't you need Torque converter lockup to be able to cut off fuel injection without the engine stalling out? If you force FI cutoff without the TC locked (as is the case in forced power mode) I believe the engine will just stall since nothing is forcing it to turn.

[THE NEW GUY]

Originally Posted by immortal_suby

I must be missing something. Don't you need Torque converter lockup to be able to cut off fuel injection without the engine stalling out? If you force FI cutoff without the TC locked (as is the case in forced power mode) I believe the engine will just stall since nothing is forcing it to turn.

aja. Nice apreciation.... I'll try it tomorrow: forced power mode, release the throttle and turn off the engine for 3 or 10 secs, then turn key back to ignition position (right before starter position)
EG33 shoud remain reving!
if not, then you are right!
It just anwer my question!
I'll post tomorrow about it. How did'nt I think about this before?

[icingdeath88]

You know that will also turn off the TCU right? The TCU which is what controls whether the torque converter is locked up or not.

What exactly are you trying to do here?

[Huskymaniac]

I have often wondered about this myself and, based on observations, here is what I think happens. When doing normal engine braking (TPS=0 and RPM>Idle), the FI duty cycle drops to some predetermined minimum. However, when the cruise control is on, I think the ECU allows the duty cycle to drop to zero for at least some of the cylnders to keep the car from continuing to increase speed when the engine braking is insufficient as in the case of a steep hill.

I have noticed this effect on multiple cars. I drive on highways through some steep hills and I typically use cruise control to get good gas mileage. When the cruise control is on, the car always manages to keep from running away when going down a steep slope. However, if it is off and I take my foot off the pedal, the car will continue to accelerate down the very same hill at the very same speed.

[Tapani]

There's engine braking availble regardless of the TC lock. The one way clutches are operated by the TCU. These are disengaged in forced POWER mode, at least according to my short testing.

Dig out the manual and the training materials available online - it's all there.

My wide band goes to full lean (16) when the fuel is cut off - this happens regardless of TC locking, but of course the torque must be transferred via the OWCs for the transmission to rotate the engine.

If you cut off the ignition you loose control of the TCU as icingdeath indicated.

Get a TCU with modified maps if you want to change the shifting - you will not loose engine braking.

Tapani

[THE NEW GUY]

First: I just did the test I mentioned before: go down a hill @ 100kmph with forced power mode on and gas pedal completly released. Turn engine off, disaceleration increased, turn key back to engine position and engine was still turning @1500rpm. EG33 started again without the need to energize the starter. disaceleration decreased.
Second: Torque converter is independent of the TCU. Only TC lock depends on the TCU. Even you are on tc lock condition, down a hill, if you turn off the engine the tc lock will disengage because there's no electric power. (tc lock is controlled by a solenoid valve powered by the TCU) But, if speed is enought, hydraulic pressure in the tc will keep the engine reving; then the gearbox is forcing the engine to continue turning, like in engine brake condition, until the rpm falls below a centain value (typically 1000rpm) At this point, centrifugal hydraulic pressure on the tc is too low to keep transmiting movement to the engine. Engine is free to stall. Engine brake disappears because ther'e no more connection between the gearbox and the engine. (car still moving slowly in D position untill stops)
third: I agree with Huskymaniac about the FI cutoff when cruise control force it. This makes me think that with the original ECU, we do not have FI cutoff not even in normal transmition control. The only way to be certain is to connect the instrument and check both: O2 signal and duty cycle on the injectors.
The goal of all of this is to confirm that our mpg can be improved by adding FI cutoff during engine break conditions.

[92snowmachine]

the only way to be sure there is fuel cut off is to check for ms. duty cycle will show zero when on time is low. oxygen sensors going lean is also not a good test for presence of fuel only amount. they measure a very narrow window and anything outside of this will falsely lead you to believe there is no fuel when in fact it is just lean.

[dbarnblatt]

Originally Posted by THE NEW GUY

First: I just did the test I mentioned before: go down a hill @ 100kmph with forced power mode on and gas pedal completly released. Turn engine off, disaceleration increased, turn key back to engine position and engine was still turning @1500rpm. EG33 started again without the need to energize the starter. disaceleration decreased.
Second: Torque converter is independent of the TCU. Only TC lock depends on the TCU. Even you are on tc lock condition, down a hill, if you turn off the engine the tc lock will disengage because there's no electric power. (tc lock is controlled by a solenoid valve powered by the TCU) But, if speed is enought, hydraulic pressure in the tc will keep the engine reving; then the gearbox is forcing the engine to continue turning, like in engine brake condition, until the rpm falls below a centain value (typically 1000rpm) At this point, centrifugal hydraulic pressure on the tc is too low to keep transmiting movement to the engine. Engine is free to stall. Engine brake disappears because ther'e no more connection between the gearbox and the engine. (car still moving slowly in D position untill stops)
third: I agree with Huskymaniac about the FI cutoff when cruise control force it. This makes me think that with the original ECU, we do not have FI cutoff not even in normal transmition control. The only way to be certain is to connect the instrument and check both: O2 signal and duty cycle on the injectors.
The goal of all of this is to confirm that our mpg can be improved by adding FI cutoff during engine break conditions.

What about cars modded with manual transmissions.... where engine braking is the norm... or do you guys want to see results with automatics?

[92snowmachine]

Originally Posted by dbarnblatt

What about cars modded with manual transmissions.... where engine braking is the norm... or do you guys want to see results with automatics?

it would be a better test to do the shutdown with a manual car. the difference between engine braking and engine off would definitely be noticeable.

[LetItSnow]

Originally Posted by THE NEW GUY

The goal of all of this is to confirm that our mpg can be improved by adding FI cutoff during engine break conditions.

I'd be surprised if the engine didn't see fuel cut on engine braking.

Something to consider: If there isn't a complete fuel cut on engine braking, in the event of a manual fuel cut, what do you think the ill effects will be?

I've got the feeling that the trouble you'll find will outweigh any benefits you might achieve, and that this will be a placebo mod at best.

[THE NEW GUY]

92snowmachine: I'll use the oscilloscope then... you may be right.
dbarnblatt: this experience will benefit manual and automatic transmissions, since we all have the same mapping in the ECU. Except the ones with ECOTUNES chips.
92snowmachine: abslotutelly right! Would somebody with MT do the test of turning off the engine while going down a hill and back on after 5 secs? do you feel any difference on the engine brake disaceleration? I suspect you will.
LetItSnow: I think the ecu does a complete FI cutoff only in cruise control; just like Huskymaniac said. About the installation of an additional FI cutof, I did this before in a '94 Peugeot 605 2T; it worked smoothly since ecu did not realize injectors was disconnected. there's no kick when injection restarts, but you will notice the increased power of the engine break (which is good for me). The FI cutoff works 2.5 secs after conditions are true and last until you press the gas pedal again or rpm crosses certain preset point (in this case 1200rpm, right after tc lock disengages). coolant temp is considered, as it is vehicule speed. All this analog data goes to a PLC (programmable logic controller). I use Logo 12/24RC, from Siemens who have two fast counters (for rpm) and four analog input (0-10vdc) built in. You can even program a new power signal to the tcu to make it switch faster... etc.

But the price for the PLC is $100.-... Anyways I'll prepare my connection schematics and PLC program just in case... for next summer.

[oab_au]

I am sure it does the fuel cut off all the time, even in N.
It is to prevent the fuel on the walls of the manifold from vaporising under the vacuum conditions, and causing it to go rich as soon as the throttle is opened.

Harvey.