ECU mods and Engine torque control

[svxboy]

I have the wiring diagram to the SVX's ECU and was wondering what the Engine torque control wire was for and what it went to? Could this be the switch that kills engine power at lower RPM's? It says in the diagram it should read 5 volts off and 0 volts on. I wonder if installing a switch or a 5 volt power supply to this lead would eliminate the "engine torque control". Anybody out there with good electrical knowlegde and is familiar with the SVX's ECU? I am just looking for comments and suggestions on the subject. I am in search of trying to eliminate some of the low rpm lag that this car has and to better its QT's. Thanks, Greg

p.s. I know eddycat and cdigerlando would have some possible knowledge on this subject.

[Chiketkd]

...but I wanted to see if anyone knew how to splice in this switch. thinking about splicing in a switch before going back to the dragstrip in late October.

Longass also mentioned something about the 'torque control' switch in a previous thread:

Originally posted by longassname
I don't know what smallcar offers. I certainly don't need to go to them to find a spst switch (single pole single throw). It's not for changing the shift patterns. That can't be done propperly without reprogramming the tcu and there's no reason to do it anyway. There is however what is called the "torque control" signal sent from the tcu to the ecu during shifting. When the ecu sees this signal (it's actually a pull down = circuit is connected to ground) it pulls out a bunch of ignition timing to reduce torque during the shift. This makes for a soft, silky, luxurious shift which is comfortable for everyday driving and kinder on the transmission. Decreasing torque at any time; however, is not good for drag racing. In fact it makes the sifts seem painfully long when your mind is operating in that hyperfast compressed time drag mode. By putting a switch on the wire for the torque control signal you can dissable it while drag racing. I don't recomend it as the next "free mod" for everyone to run out and do to their svx which they realistically never drag race. If you actually go to the track however it seems like a no brainer. People with 5 speeds of course no longer have the torque control signal as they no longer have tcu's.

Any ideas? Where is this wire that runs from the TCU to ECU?

-Chike

[NikFu S.]

How the heck did you find this?

[mbtoloczko]

The torque control pin is shown on the ECU pinout diagram. There are images in my locker of the ecu pinout. After I switched over to a 5MT, I experimented with disabling torque control. As LAN said, its a pull-down type signal, so it holds high (5 V) automatically, and when torque control needs to be enabled, the pin on the ECU is grounded (normally the TCU). Enabling torque control has no effect until the motor gets above a certain throttle position and/or engine RPM (I've kinda forgotten). When it does enable, the motor falls completely flat on its face. Its so drastic that the first time it happened, I thought the engine had just died. I doubt its the cause of low rpm lag you mentioned. If you want to experiment with enabling the torque control (so you know what it feels like), all that you need to do is splice into the wire going into the torque control pin on the ECU, and then ground that wire.

[Chiketkd]

Mychailo,

Those ECU pinout diagrams are perfect and just what I was looking for. So after I splice into the engine torque control wire, I should ground one of the wires? Would the firewall make a suitable grounding point?

Also, you talked about the motor falling flat when torque control is enabled - what happens when it's disabled? Are the shifts super firm and crisp?

This is the reason why I'm interested in making the switch...

-Chike

[svxboy]

WOW Chiketkd, I barely even remember writing that initial post back in '01! How fast time flies by. It sounds like from the other posts that the TCU is responsible for grounding out the signal of the torque control pin whenever the transmission is shifting therefore causing the retarded ignition timing. The part that I still don't understand is to bypass this, could it be just simply disconnected or a switch put in-line or does a 5 volt power supply need to be connected to trick the ECU. Its already known the wire cannot be grounded, otherwise the torque control would be enabled all the time.

Greg

[Chiketkd]

Greg,

Glad to see you're back from the great beyond.

Yeah, when thinking it over I realized that disconnecting the wire may not be the solution as it'll cause the pulse to drop from +5 volts to 0.

Doesn't seem like a there's a simple solution around this...

-Chike

[b3lha]

Read mbtoloczko's post again. It's a pull down signal which floats at 5V.

The ECU grounds the wire to enable Torque Control while shifting. Just disconnect the wire, or put a switch in line with it if you don't want this to happen.

It might improve your drag times, provided the gearbox doesn't explode the first time it tries to shift at full torque.

[mbtoloczko]

Originally posted by b3lha
...
The ECU grounds the wire to enable Torque Control while shifting. Just disconnect the wire, or put a switch in line with it if you don't want this to happen....

Yep, that's how to do it. Chike, I only tried this after I switched to the 5MT, so no data on how it works with a 4eat. I would be curious to see what happens with a 4eat. My thought is that having it manually disengaged is going to be pretty hard on the clutch packs.

[Chiketkd]

Originally posted by mbtoloczko
Yep, that's how to do it. Chike, I only tried this after I switched to the 5MT, so no data on how it works with a 4eat. I would be curious to see what happens with a 4eat. My thought is that having it manually disengaged is going to be pretty hard on the clutch packs.
I'll definitely consider doing this. I'm sure the 228 ft-lbs of torque our engines make at peak would not be enough to cause damage to our clutch packs. But only way to know for sure is to cut that wire...

-Chike

[svxboy]

OK, lets get this straight. If I want to bypass the torque control system with an auto tranny, I would have to disconnect or install a switch between the TCU and terminal #20 on connector B59 to disable the grounding action performed by the TCU. Then I would have to install the splice between terminal #20 on connector B59 and #7 from connector B60, at the ECU, so it reads 5 volts all the time, getting its source from the PS switch supply. With a manual tranny, there is no need to install the switch because the TCU is already removed and no grounding occurs. The grounding occurs at the TCU not the ECU.


Does this sound right? This has been a mystery to me for years.

mbtoloczko, what of power increases did you notice when you did this mod?

Thanks, Greg

[Phast SVX]

Originally posted by svxboy
OK, lets get this straight. If I want to bypass the torque control system with an auto tranny, I would have to disconnect or install a switch between the TCU and terminal #20 on connector B59 to disable the grounding action performed by the TCU. Then I would have to install the splice between terminal #20 on connector B59 and #7 from connector B60, at the ECU, so it reads 5 volts all the time, getting its source from the PS switch supply. With a manual tranny, there is no need to install the switch because the TCU is already removed and no grounding occurs. The grounding occurs at the TCU not the ECU.


Does this sound right? This has been a mystery to me for years.

mbtoloczko, what of power increases did you notice when you did this mod?

Thanks, Greg


Greg!!! whats up man, didnt know if you still had the SVX or not. We should get together sometime, im at school in whitewater. Give me a call
608-712-1906

phil

[svxboy]

whats up man! My youngest brother goes to school their too. I call ya sometime soon.

Greg

[oab_au]

Not a good thing to do on it's own.

The torque is reduced during changes, to allow the box to do a "soft shoe shuffle" during the change. The TCU does other things to make the change unnoticable. During the 1st to 2nd change the TCU drops the line pressure, and pulses the band on and off a few times to slowly brake it down to speed, instead of just banging the band on.

If the torque is not reduced during the change the band will slip due to the low line pressure and the pulsing of the band pressure.

Removing the dropping resistor, does not keep the line pressure high during the shift as the TCU acts on the A solenoid to modify the line pressure, during the shift.

The way around the problem, as I see it, is to leave the resistor connected, so that the line pressure is varied in line with throttle pressure. This allows the TCU to preform the other duties that it does based on throttle pressure, such as engine braking, ect.

The Small Cars vacuum switch is then used to turn the TCUs control wire to the A solenoid off when the throttle is past 20% or so. With this line disabled during the shift, the line pressure will not be modified down by the TCU. It will just be set by the throttle position. As the throttle is down during hard acceleration, the line pressure will remain high during the change, to prevent the slipage.

This TCU line to the A solenoid should not be just disconnected. It should only be switched off under hard throttle. At other times, we need to allow it connected to preform other necessary duties, such as, varing the pressure with cold ATF, dropping the pressure when selecting D or R from cold, ect.

This change on its own will allow the box to survive hard full throttle changes. Then and only then, could the torque control be modified for a full torque change.

I would recommend that the same Small Cars kit be used to switch the Green/Yellow wire, from the TCU plug C wire no.8. off the A solenoid, and through a resistor to ground, when the vacuum switch activates. Otherwise the TCU will see the open circuit wire and post a code.

Worth a try?
Harvey. ;)

[oab_au]

No body think this is worth a try, or didn't I explane it clearly:confused: I think it will allow faster, firmer changes and allow the torque control to be switched out for hard acceleration.

Harvey.;)

[svxboy]

Makes total sense. Why have more power going to the tranny that isn't going to be put to the ground properly, through the slipping bands in the tranny. Could I just install a multiple throw switch that simaltaneously disables solenoid A for full line pressures shifts and at the same time eliminates the torque control function, for use when dragging/racing?

The other question was as to when using a manual tranny when the TCU is removed, is the torque control always active since the ECU now recieves a 0 volt signal at that terminal? In my friends 5spd SVX, power seems to fall off fast after 3000rpms.

Also, is there a way to manually control the Air Resonance valve in the intake manifold? Is this unit at all disabled with the replacement of a manual tranny? It too could be the cause for the loss in power on the higher end if it doesn't open properly.

Greg

[oab_au]

[QUOTE]Originally posted by svxboy
[B]Makes total sense. Why have more power going to the tranny that isn't going to be put to the ground properly, through the slipping bands in the tranny. Could I just install a multiple throw switch that simaltaneously disables solenoid A for full line pressures shifts and at the same time eliminates the torque control function, for use when dragging/racing?

___________________________________________
You could Greg, be a bit crude, It would be better to use two Small cars vacuum switchs, one on the TCU line to the A solenoid, and one on the torque control line. That way the TCU unit could be adjusted to work at say 20% throttle and the other on the Torque control line, could be adjusted to switch at full throttle. At light throttle openings, it would be normal, at higher throttle openings it would firm up the shift, and at full throttle, firm, full torque shifts.
__________________________________________


The other question was as to when using a manual tranny when the TCU is removed, is the torque control always active since the ECU now recieves a 0 volt signal at that terminal? In my friends 5spd SVX, power seems to fall off fast after 3000rpms.

___________________________________________
With the TCU removed the Torque control line is pulled up to 5 volts by the ECU, it's the TCU that pulls it down to 0 volts to reduce torque. If it is removed it should stay up at 5 volte all the time. Does your friends SVX have the exhaust resonator still fitted, won't make top end power without it.
__________________________________________

Also, is there a way to manually control the Air Resonance valve in the intake manifold? Is this unit at all disabled with the replacement of a manual tranny? It too could be the cause for the loss in power on the higher end if it doesn't open properly.
____________________________________________
Anything is possibable, but the way it is set up is to open at the right change over point, changing it to some other rpm won't help. No the manual replacement won't affect the operation. If the valve does not open it will stay in the Inerta mode, fall flat over 4000rpm. You could check if it is working by useing a light to see the arm under the manifold, that works the valve, raise the rpm to about 4500/5000, the arm should move.


Harvey.;)

[Myxalplyx]

Bizzump!

May as well bump this up from the grave since it is all the talk and rave right now. ;)

[NikFu S.]

How the heck did you find this?






Lol. :tard:

[TomsSVX]

don't bother... i have tried only to find the car completely falls on its face when jumping the wires (giving it a 5v source). More or less it operates without any torque control with 0v signal and and activates it with a 5v signal...

Tom

[Trevor]

Tom,

Not surprised it did not work, considering the source of the info. :rolleyes:

To pick your brain/knowledge --- In respect of the SVX, is torque induced via the interruption of fuel or ignition? :confused:

I have always understood that some, but not all cylinders, are deprived of ignition for the short periods involved. This would be plausible as control would be practically instantaneous, as opposed to a probable slight inherent delay.

Please, can someone confirm the exact mode of operation. Please, no opinions, only exact evidence, e.g. pointers towards verifiable published statements, from an absolutely reliable source exhibiting authority.

[TomsSVX]

cannot verify but my suspicion is that it drops all timing advance

Tom

[Myxalplyx]

cannot verify but my suspicion is that it drops all timing advance

Tom

I can verify that fuel is reduced or eliminated between shifts based on my dyno runs I did from 1st to 3rd gear. The air/fuel ratio would go off the chart between 1st - 2nd gear. This is not the case with any of my other cars, Subaru or not. Of course this could be an isolated event so I can not generalize (speaking about the air/fuel ratio going way lean).

[Dessertrunner]

Does any ones car stall when the speed is above 20mph. Eg if you quickly clutch in and lift your foot off the trottle does the engine die. If you hit the breaks does that effect if the engine shuts down. When the pins on the ECU are left floating for example not connected to either +5v or ground they could drift between the two. Mine stalls some time for a couple of days then never for two weeks or more. Its a pain and we need to sort it out. If we can all run trials we should be able to figure the event or sequance that causes the engine to shut down.
Tony

[TomsSVX]

thats the infamous 5mt stall. It is a fuel cut. responsible when the car is moving but the transmission is not engine braking. Thus the engine dies... I am working in finding a 100% solution.

Tom

[Dessertrunner]

I think they are related, for example the ECU looks to see if the engine is in P & N, I would assume (might be wrong) that if it sees this condiation it sets up a starting engine map condiation. Also you would think that ECU would not allow full power under this condiation.
We all need to find out what infomation the ECU is getting that may effect is management under this condiation and then I might be able to put a scope on the imput wires to see if anything goes grazy just before the engine dies. Is your fix manchinal or electronic Tom.
Tony

[TomsSVX]

no, it will require rewriting the code in the ECU if it is even possible. A fuel cut cannot be stopped by means of a mechnical modification. I have tried numerous ways of wiring the circuit and none seem to work better than the other. I am at a loss until I can tap into the ECU myself and do some testing

Tom

[Trevor]

Does any ones car stall when the speed is above 20mph. Eg if you quickly clutch in and lift your foot off the trottle does the engine die. If you hit the breaks does that effect if the engine shuts down. When the pins on the ECU are left floating for example not connected to either +5v or ground they could drift between the two. Mine stalls some time for a couple of days then never for two weeks or more. Its a pain and we need to sort it out. If we can all run trials we should be able to figure the event or sequance that causes the engine to shut down.
Tony

Leaving an input floating is bad news, in that if the input has a relatively high impedance, the open circuit could induce extraneous noise. This would tie in with the sporadic nature of the problem.

As it would appear that the pin can not be grounded, a small capacitor of say 0.1 mfd connected pin to ground may provide a solution. It is possible that a capacitor as it charges at turn on, could close the input circuit for an instant, but the value suggested should not cause this problem.

As an alternative, a resistor of a value largely exceeding that which would trigger an input, could be tried.

[TomsSVX]

Leaving an input floating is bad news, in that if the input has a relatively high impedance, the open circuit could induce extraneous noise. This would tie in with the sporadic nature of the problem.

As it would appear that the pin can not be grounded, a small capacitor of say 0.1 mfd connected pin to ground may provide a solution. It is possible that a capacitor as it charges at turn on, could close the input circuit for an instant, but the value suggested should not cause this problem.

As an alternative, a resistor of a value largely exceeding that which would trigger an input, could be tried.
Edit: Nevermind I re-read it and found it... I don't think that is the issue here. As this should be left alone (torque controll) and we should be looking at a way to keep the engine management from cutting the fuel when the car is rolling but no accelerator is applied in a neutral state. I have noticed ont eh wideband that my idle will bounce back from dead lean once the car comes to a complete stop. This is an indicator to me that the ECU is trying to engine brake. Jumping the neutral pin should take this issue away for anyone else but right now it is not working for me

Tom

[Trevor]

Edit: Nevermind I re-read it and found it... I don't think that is the issue here. As this should be left alone (torque controll) and we should be looking at a way to keep the engine management from cutting the fuel when the car is rolling but no accelerator is applied in a neutral state. I have noticed ont eh wideband that my idle will bounce back from dead lean once the car comes to a complete stop. This is an indicator to me that the ECU is trying to engine brake. Jumping the neutral pin should take this issue away for anyone else but right now it is not working for me

Tom

This thread is a convoluted mess. :(

My suggestion directly relates the post (28) originated by Tony and which I quoted within . Also post 24 is directly applicable.

Whatever, no input should be left open as this could result in all sorts of problems. ;)

[TomsSVX]

its not left open. It is merely "OFF" rather than being switched on when needed. People think it works the other way and needs to be switched "off" by grounding the pin. Thats my experience with trying this.

Tom

[Dessertrunner]

Guys right now I can't get mine to stall or miss behave, my question is if I get the Subaru anyliser and plug it in when I am next having problems it should show me if the fuel is cut because it shows the duty time on the injectors. Has any one tried this or do you think it is a waste of time.
Tony

[Trevor]

its not left open. It is merely "OFF" rather than being switched on when needed. People think it works the other way and needs to be switched "off" by grounding the pin. Thats my experience with trying this.

Tom

Tom,

I am aware of what you are saying. What I am pointing out is the danger inherent if an input remains with no connection, when not used and idle, i.e. is left open. This represents an anomaly which you can not afford to have existing, in view of your obscure problems.

When talking things electrical, the words "off" and "on" are best not used, because they can be ambiguous/confusing. Energised not energised in respect of a circuit or device; closed open in respect of a circuit, switch or contacts; present or not present in respect of inductance, capacity, voltage, current etc.

If an input connection/pin is not used, the associated input circuit is open, and can receive an induced electrical signal without there being actual contact. When connected to a sensor the input is not open, as there will always be some load present, which in affect closes the circuit. This will constitute both resistance and capacity.

Note that shielded input leads are used as a means of shielding any loop which could be open to an induced input. In effect the input is not open, even though the sensing device is not conducting significant current.

I am saying that when the input circuit is idle, it should not be left open. It should be closed, by means of a relatively high resistance, and better still in addition, a small of amount of capacitance. Otherwise there could be extraneous affects relative to any operative problem.

Be sure I am not presenting this as definitive, but rather as something which in the circumstances must be taken into account.

End of my story, Trevor. ;)

[Trevor]

Guys right now I can't get mine to stall or miss behave, my question is if I get the Subaru anyliser and plug it in when I am next having problems it should show me if the fuel is cut because it shows the duty time on the injectors. Has any one tried this or do you think it is a waste of time.
Tony

Tony,

This would be very worthwhile, as it would positively confirm the exact method used in order to control torque. Without this certainty, surely all is darkness. :)

Cheers, Trevor.

[Dessertrunner]

I have a question to help me work through the problem. Does the stall occur when NOT BRAKING. As far as I am concerned I don't think my engine has died when not under brakes or recently applied the brakes. May be the key is the cars brakes. The reason I ask is that in the cold today I had a miss and the engine was down to 500 rpm but did not stall.
The logic follows that some action we take tells the computer to cut fuel. What is it? Does the ECU see the Brakes when applied?

[Speedklix]

I also do not jump it. I had wired it per whatever I had read at the time, I think it was huck's directions. The car didn't like to drive and threw a CEL, until I cut the wire.

Is it being implied that I could be harming something by doing this... perhaps it could give me some light to my current issue. Which looks like a failed maf. Side note.. I also found no need to trick the gear to release the key from the ignition... or I was too tired to remember splicing it :confused:

[Nomake Wan]

Interesting... though I think the whole 5MT Stall should have its own topic. This topic began with addressing ATs, and believe it or not some of us still have ATs in our cars. :rolleyes::p So the torque control thing would be useful for me to figure out as something I could do to squeeze a little bit more power out of the drivetrain.

Also, I can't talk about the ECU much... but I did read a lot about the TCU. According to what I read, the TCU does look for ABS signals when sending information to the ECU. But since you guys are talking about MT cars, I don't see how brakes could be affecting what's going on. From what I know so far of the ECU, it does not receive signals from the ABS. Only the TCU cares about what the brakes are doing.

Back onto the real topic... from what I think I'm reading in Trevor's posts about the original topic (as it is all professional electronic speak which I am not fluent with), if we wish to tinker with the torque control, we should disconnect the torque control wire between the ECU and TCU, then add a resistor and capacitor to the ECU side of the wire, then ground that new circuit? Is this correct? From what I gathered, you warn that leaving the wire off and causing the voltage to float by itself is not good.

Hopefully I am interpreting correctly. :o

[TomsSVX]

braking involves 2 things and could be of either.

Using the brake pedal to slow down

Using the engine to slow down.

The auto uses the engine to slow down. When you are slowing down with either the clutch depressed or the trans in neutral the engine is not braking as the ECU is expecting. It still cuts timing and it still cuts fuel like it would if there was still a connection to the driveline. SO, until we can change the maps to adjust to this difference, it will stall

Tom

[huck369]

braking involves 2 things and could be of either.

Using the brake pedal to slow down

Using the engine to slow down.

The auto uses the engine to slow down. When you are slowing down with either the clutch depressed or the trans in neutral the engine is not braking as the ECU is expecting. It still cuts timing and it still cuts fuel like it would if there was still a connection to the driveline. SO, until we can change the maps to adjust to this difference, it will stall

Tom

Not if you wire up a clutch position switch to read "Nuetral" when the clutch is depressed....as then the ECU thinks it's in nuetral.....never had stalling on either of my 5 speeds wired this way.

[crazyhorse]

Slick solution Huck
This also tells the ECU it's "safe" to start the car.

[Dessertrunner]

I the car would start easier as well as the ECu is set for a start.

[Trevor]

Back onto the real topic... from what I think I'm reading in Trevor's posts about the original topic (as it is all professional electronic speak which I am not fluent with), if we wish to tinker with the torque control, we should disconnect the torque control wire between the ECU and TCU, then add a resistor and capacitor to the ECU side of the wire, then ground that new circuit? Is this correct? From what I gathered, you warn that leaving the wire off and causing the voltage to float by itself is not good.

Hopefully I am interpreting correctly. :o

Nomake you have blown a breath of fresh into a thread which is now a spiders web.:)

Yes Nomake, you are basically correct. :eek:

However the added components would be best, added at the ECU connector, rather than at the end of any wire remaining disconnected from the TCU. Also to be exact, it is the input which would be floating if left disconnected, and therefore open to receive any "floating" inductive voltage.

Not to worry, you have explained everything as near as damn it. :)

Special thanks, Trevor. ;)

[TomsSVX]

trevor, had it beaten into my head by someone MUCH better with electronics than I. I understand what is going on now. Yet still do not have a good solution

Tom

[Nomake Wan]

Also to be exact, it is the input which would be floating if left disconnected, and therefore open to receive any "floating" inductive voltage.
Just have to ask... you say to add it "at the connector," but to do so would require a wire be placed into the particular pin on the connector anyway... so wouldn't that mean that the circuit to be added would in fact rely on a wire going into the ECU connector at that specific pin? ;)

I had to do that with the "Power Mode" mod, splicing a wire into pin 4 of the TCU and grounding it. I would assume that this modification we are speaking of would be damn near similar, with the addition of a small circuit of resistor and capacitor.

On that note, Trevor, do you have a suggestion as to circuit specifications? That is, exact specifications for a resistor and capacitor to be added. With my new living conditions I now have an easily-accessible electronics store, so I would be willing to take some risks in the name of progress. :D