Jump to content
SAU Community

Recommended Posts

Has anyone tried to get an aftermarket ECU to talk to an auto trans computer in an r33. I would like to be able to fiddle with the fuel loops particularly in closed loop mode but i still want the engine/trans to behave like they do from the factory.

I have been looking at the wolf v500 and it seems to have the capability to retard ignition or cut fuel based on an eternal signal (say power upshift) assuming thats how the 2 computers communicate . Can anyone tell me the signals between the computers are as simple as a wire for power upshift, downshift, etc or if there is more to it.

cheers

neil

Link to comment
https://www.sau.com.au/forums/topic/145145-aftermarket-ecus-for-an-auto/
Share on other sites

  • Replies 40
  • Created
  • Last Reply

Top Posters In This Topic

there is no such thing as the auto trans computer in the r33

the gearbox logic is done in the factory ecu, so theres nothing to fix.

your ecu needs to be aware its on an auto car and cut IGN timing on gearchange. it needs to know when the auto box is about to change gears

imho no aftermarket ecu will do it

The only reason for the retardation of ignition timing is to lower torque during the change, many cars get away without it. I never noticed any adverse effects.

The auto DOES have its own ECU, it only needs to know throttle position and road speed to correctly time the shift points.

The standard engine ecu sends the TPS signal to the auto ecu, i just bridged the tps signal input to motec to the tps wire going to the auto ecu.

there is no such thing as the auto trans computer in the r33

the gearbox logic is done in the factory ecu, so theres nothing to fix.

your ecu needs to be aware its on an auto car and cut IGN timing on gearchange. it needs to know when the auto box is about to change gears

imho no aftermarket ecu will do it

how do drag dood do it who run Autos?

I've had am aftermarket ECU in my tip FTO for ages with no issues...but it doesn't make as much power as say my line aswe all know PFC is the way to go generally.

there is no such thing as the auto trans computer in the r33

the gearbox logic is done in the factory ecu, so theres nothing to fix.

your ecu needs to be aware its on an auto car and cut IGN timing on gearchange. it needs to know when the auto box is about to change gears

imho no aftermarket ecu will do it

Paul if that was the case as soon as a stand alone ECU was plugged in you would only get 3rd out of the auto it doesn't just guess where to shift.

OK... I have done some searching for myself and found the following

RB25DET wiring diagram

and also

Generic nissan ECU register table

So it looks like there are 2 wires into the main ECU and 2 wires out (not necessarily all to the auto computer) and the registers for the Auto computer include throttle pos sensor,engine RPM and vehicle speed which may come from the main ECU or directly from the sensors.

There is also registers for the throttle closed switch and what i think is Wide open throttle position switch which may or may not come from the main ecu.

P,N,D,2,1 and overdrive are all obviously switches and i think 'shift s/v a' and 'shift s/v b' could be the power/snow switch for the auto because nothing else seems to fit.

Which leaves

kickdown switch - as im guessing the auto comp makes all the shift decisions i would assume this is a signal to the main ecu that the gearbox is going to drop a gear.

Powershift switch - same as above except going up a gear.

ASCD - od cut - ?

ASCD - cruise - ?

hold sw - ?

overrun/c s/v - ?

Can anyone confirm any of this???

no its not the case at all. the drag cars dont use electronic auto boxes, they use mechanical ones. like the one in the vl turbo. it doesnt need electronic crap around it to cut when changing gears. the electronic boxes do. using a manual ecu on an electronic auto like the one in the r33 will not cut the ign timing on gearchange.

its not to do with gear change logic, the auto box knows how to change gears, its do with change gear preparation, that is, stock ecu has logic for when the auto box is about to change gears and cuts the ign timing to protect the auto box and ensure its a smooth change. it wont be stuck in 3rd or anything like that at all.

what you are certainly likely to see

detonation on gearchange

map tracer showing load staying near the middle on gearchange (inducing more possibilities of detonation)

harsh or jerky changes

delays with kick back and change down

kjb is right the back off timing is there to take torque load off the auto box and thats its sole purpose. without, you are placing more stress on the box and what ever other gremlins hide away. by all means go for it.

send us a map tracer of a gearchange, someone ?

no its not the case at all. the drag cars dont use electronic auto boxes, they use mechanical ones. like the one in the vl turbo. it doesnt need electronic crap around it to cut when changing gears. the electronic boxes do. using a manual ecu on an electronic auto like the one in the r33 will not cut the ign timing on gearchange.

its not to do with gear change logic, the auto box knows how to change gears, its do with change gear preparation, that is, stock ecu has logic for when the auto box is about to change gears and cuts the ign timing to protect the auto box and ensure its a smooth change. it wont be stuck in 3rd or anything like that at all.

what you are certainly likely to see

detonation on gearchange

map tracer showing load staying near the middle on gearchange (inducing more possibilities of detonation)

harsh or jerky changes

delays with kick back and change down

kjb is right the back off timing is there to take torque load off the auto box and thats its sole purpose. without, you are placing more stress on the box and what ever other gremlins hide away. by all means go for it.

send us a map tracer of a gearchange, someone ?

You said there was no such thing as an auto trans computer in the R33. This is wrong. If you unplug the trans comp the box goes into limp home mode and only has 3rd. So does it have one or not?

Lets face it, if Apexi called it a feature for auto cars like HKS do there would be nothing to suggest not to use a PFC in an auto car except perhaps the R34s which I believe have the auto computer in the main ecu not as a seperate unit but I could be wrong.

well if it has an auto computer or not, it changes nothing

the auto logic to back off timing is inside the main ecu unit for nissans.

for toyotas its in the seperate computer so this only applies to nissans.

the PFC is a manual ecu only, it is clearly stated in the manual, production documentation and various distributors who sell it.

there is however apexi powerfc auto versions for toyota chaser

as the logic control is outside the main ecu they are able to work with it.

there are a fair few extra functions on the auto powerfc for gaerbox control, speed ramping, change logic and speed based gearsetting

what about getting a piggyback like that microtech? you can set them up to map above the max of the afm, or remove the afm completely as they have a map sensor.

has anyone used this sort of setup on an auto? i have seen it used on manuals, but not auto's (i don't even know anyone with an auto).

yeah the best comprimise that i can think of is either a piggyback system so something like emanage, microtech, or any form of piggyback

or change to a mech auto box. but this is assuming you actually care about the auto box and worry about it. some people dont, so in their case it probably is fine. but i dont accept the "its the same as stock" as it cant possibly be.

would also love to see a map tracer run of a gearchange, still waiting for this. this could yield some info for us and possibly provide a 3rd party hack for cutting timing on gearchange

yeah the best comprimise that i can think of is either a piggyback system so something like emanage, microtech, or any form of piggyback

or change to a mech auto box. but this is assuming you actually care about the auto box and worry about it. some people dont, so in their case it probably is fine. but i dont accept the "its the same as stock" as it cant possibly be.

would also love to see a map tracer run of a gearchange, still waiting for this. this could yield some info for us and possibly provide a 3rd party hack for cutting timing on gearchange

If the timing is the only issue maybe just ask HKS why the F-CON systems disable it :) As you know they are a piggyback system yet actively disable this oh so important feature. I'll find out soon enough when I get my PFC.

What i want to do is to adjust the fueling to improve fuel economy on the highway which is 90% of what i use my car for. A piggy back won't help when the computer is in closed loop mode. I did think about disabling the oxygen sensor to trying to make it stay in open loop all the time but i suspect it wouldn't work and is not my prefered solution. An added benefit with an aftermarket ECU is the possibilty of using a wide band O2 sensor and possibly implementing a lean cruise closed loop mode.

I haven't heard of anyone getting an aftermarket ECU to talk to to the factory auto properly but i suspect that due to limited demand not that many people have really tried.

paul are you saying that main ecu decides when the gearbox is going to change? because i would have thought all of this would happen in the auto computer, with a signal back to the main ECU to say 'hey im about to change gear". Obviously the main ECU decides what to do when it gets this signal. The problem i can see is finding the signal and turning it into something an aftermarket ecu recognises. Looking at the info for the wolf v500 it has a provision to modify the ignition and fuel delivery based on user defined aux inputs, so it would just be a matter of configuring however much ignition retard when that "change gear" signal is present.

Thinking about it further i don't think it would be overly difficult to use a microcontroller like a PIC to capture the signals you want and turn them into a usable output assuming some information is available on how it works..... I haven't been able to find anything so far :)

If the timing is the only issue maybe just ask HKS why the F-CON systems disable it :D As you know they are a piggyback system yet actively disable this oh so important feature. I'll find out soon enough when I get my PFC.

well if its a piggyback then the stock ecu would cut the timing from underneath the fcon so it would simply follow the stock ecu's lead on cutting the timing, piggyback'ing is fine there should be no issue whatsoever with piggyback onto stock auto ecu.

its only when you rip out the stock auto ecu and replace it with a manual ecu, as the manual ecu has no auto logic control. manuals ecu's are any ecu that has no logic control (so any ecu other than stock)

i run my PFC with closed loop feedback off and ive tuned it via a wideband sensor so the economy is pretty good now. its better than what it was with 02 feedback on so its not a big deal. you could simply unplug the 02 sensor and then tune with the piggyback system, i dont see why it would be a problem

i forget how the logic for the auto works exactly but i am pretty sure the stock auto ecu has patterns for gearbox change gear patterns and simply follow these to work out when to cut timing, it didnt know the r33 has a stock ecu & gearbox ecu. can anyone clarify? even if it does, the stock ecu still needs to cut timing so its not different.

has anyone tried it with the wolf3d v400 ?

it's common knowledge on skylines (r32-r34 etc.) that the auto transmission control is done by the ECU. Some cars have a separate ecu for auto trans, skylines don't.

there may be some form of interface between the actual transmission and the car's ECU, but the car's ECU does the important stuff (i don't know for sure because i've never had to work on an auto).

There have been several cases to my recollection of powerfc's blowing up skyline auto transmissions, it's a well documented problem!!!

i run my PFC with closed loop feedback off and ive tuned it via a wideband sensor so the economy is pretty good now. its better than what it was with 02 feedback on so its not a big deal. you could simply unplug the 02 sensor and then tune with the piggyback system, i dont see why it would be a problem

I would have thought that disconnecting a sensor would make the stock computer go into limp home mode...

the standard oxygen sensor is only good if your chasing a afr near 14.7 but with a wideband should be able aim higher with a fair amount of accuracy.... i believe you can go as high as 20:1 but i think you would need to keep a close eye on the exhaust temp around there. You are probably right about not needing to be closed loop assuming it is well tuned.

I have access to SAFC so perhaps it is something i will try first.

thanks

nah if you unplug the stock 02 sensor it shouldnt cause any problems, fairly easy to try. it wont go into limp mode, it just wont run closed loop. limp mode on the stock ecu is simply rev cut at 2500 which is activated by disconnected or faulty airflow meter.

around 15.5s is about the safest light cruise afr you would want to dial in. any more and the exhaust gas temps get really high and risk the engine.

mine are tuned around here and it works really well. i got just on 540ks to a tank on highway driving

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Similar Content

  • Latest Posts

    • I might be way off the mark here.. there must be a T junction under the manifold. That takes manifold pressure through a check valve prior to the T, one side splits out to this bleed off orifice, the other straight to the wastegate actuators. 
    • Carrying out some engine due diligence before my first drive in a 260RS that has been in storage for 10+ years, and found two lines that are susposed to be plumbed into the boost control solenoid. One of the lines (the wastegate boost supply line) to actuate the gates, had a bolt in one end blocking it. The other line (main boost signal line) from the very rear of the intake manifold (RB26DETT) and runs up to the boost controller had this pill in the end of it as seen in photos. Other than the obvious (T28 Ceramic turbos might well be cooked)  Can anyone shed any light on if there is any validity at all with this mod, bleeding a constant fixed rate of manifold boost pressure off through this 3mm hole?  Is it deliberate?  Was this a 90s Jap mod done to RB26's? Or has someone just fitted this without any idea of what happens to ceramic T28s if they are not gate controlled?    
    • This looks great! Awesome to have the 5 speed with an LSD in this. The Recaro's look like SR2's or SR3's. It's a shame they've been retrimmed as they would have looked sweet in their original colours. I think I'd want my agent to be a little more up front and informative too. I've never imported but I reckon it would make me a little nervous, so the more info the better.
    • Thanks for doing that Duncan! Makes you a good person in my books. We don't get kangaroos or wombats here. But we have bats and it's similar. AFAIK it's often the mums with a baby attached that get hit because they drop lower when starting from a tree. If you hit an animal, check on it. https://www.ifaw.org/au/resources/wildlife-rescue-app An app to get the closest wildlife rescue contact.
×
×
  • Create New...