LS swap: rusEfi as a piggyback?

[Noble713]

Ok, when I first joined I was considering rusEfi for my turbocharged 1JZ Chaser. I've decided to dial back my ambitions for that car and instead focus on a more modern chassis.

Without going into too many details, this is a modern, in-production vehicle of Japanese manufacture, with CAN BUS and lots of electronic doo-dads. I want to keep all of the electronics as stock as possible.....while swapping in a Chevy LS3 6.2L naturally aspirated V8. How feasible is it configure a rusEfi to basically intercept the OEM ECU's engine/trans signals, convert to suitable signals/mapping/etc. for the V8 and likewise receive the sensor signals from the V8/T56 trans and pass data back to the OEM ECU?

[AndreyB]

this is a modern, in-production vehicle of Japanese manufacture, with CAN BUS and lots of electronic doo-dads. I want to keep all of the electronics as stock as possible.....while swapping in a Chevy LS3 6.2L naturally aspirated V8. How feasible is it configure a rusEfi to basically intercept the OEM ECU's engine/trans signals, convert to suitable signals/mapping/etc. for the V8 and likewise receive the sensor signals from the V8/T56 trans and pass data back to the OEM ECU?

Need more specifics, as is the question is too vague.

When it gets to communication between the actual engine and ECU it's not much CAN as far as I know.
rusEfi does not have much of piggybacking code, for example there is no code to capture your OEM fuel impulse and control your V8 through rusEfi based on your OEM non-v8 injector pulses. There is no code to convert your v8 trigger signal into your OEM trigger signal to make your OEM ECU happy.

I'd say it's much more realistic to control the engine with rusEfi and produce all the needed CAN messages like instrument cluster - this we have, not extremely configurable but I would be willing to improve this if needs come.

What about trans? No idea what you mean here.

[abecedarian]

I think the intent is to have rusefi actually running the engine, but tricking the stock ECU into thinking everything is okay.

In "lay-person's" language:
The owner tells the manager what it wants. Since the actual work needed is different than what the owner really wants, the manager has the workers cheat, verifies things are working acceptably, then lies to the owner about what was seen and what needed to be done. Everyone is happy... until the Inspector comes along. However, since the Inspector is hopefully lazy and never really looks around, he takes the records the manager gave to the owner, and since there were no problems in the reports, everyone, again, is happy.

Cast of characters:
Owner: Stock ECU
Manager: rusefi firmware and hardware
Workers:
- MAP in, played by the V8's MAP sensor and rusefi ADC circuitry
- MAP out, played by rusefi firmware and unidentified hardware
- TPS in, played by the V8's TPS sensor and rusefi ADC circuitry
- TPS out, played by rusefi firmware and unidentified hardware

Okay, I'll stop there.


The point being that rusefi can MOST LIKELY be adapted to run the engine in question but impersonating it to the stock ECU will not be so easy since many of the sensors on the stock engine are directly interfaced to the ECU. Therefore, rusefi MUST present signals to the ECU which are in accord with the signals the ECU expects to see in its current state of operation, with none out of tolerance. Many ECU's can detect open/short conditions on injectors, ignition coils, sensors, et al, as well as over/under current; the former reflecting gross failures in the device and/or wiring, the latter reflecting degraded connections and other faults. Should that be handled, and if so, how?
How does rusefi detect 'misfire' events, and how does that get sent to the ECU?

Overall:
- possible? Probably.
- possible with current hardware? Not even close.
- lot's more to think about.

[Noble713]

rusEfi does not have much of piggybacking code, for example there is no code to capture your OEM fuel impulse and control your V8 through rusEfi based on your OEM non-v8 injector pulses. There is no code to convert your v8 trigger signal into your OEM trigger signal to make your OEM ECU happy.

Hmmm, sounds like an interesting project. I'm trying to get the university to pay for the service manuals/wiring diagrams for this car (for "research purposes"). They'll be entirely in Japanese but will still help immensely. I assume I'd need to detect the signals of the stock engine/ECU config, or find some resources online about the signals....

I'd say it's much more realistic to control the engine with rusEfi and produce all the needed CAN messages like instrument cluster - this we have, not extremely configurable but I would be willing to improve this if needs come.

Well the car also has stuff like collision avoidance radar. I just figured reverse-engineering functionality for something like that would be a nightmare.

What about trans? No idea what you mean here.

trans = transmission. These cars only come with automatics so if I'm installing a T56 6-speed manual there's probably also some transmission-related signals that I need to reproduce, yes?

I think the intent is to have rusefi actually running the engine, but tricking the stock ECU into thinking everything is okay.
Cast of characters:
Owner: Stock ECU
Manager: rusefi firmware and hardware
Workers:
- MAP in, played by the V8's MAP sensor and rusefi ADC circuitry
- MAP out, played by rusefi firmware and unidentified hardware
- TPS in, played by the V8's TPS sensor and rusefi ADC circuitry
- TPS out, played by rusefi firmware and unidentified hardware

Okay, I'll stop there.


The point being that rusefi can MOST LIKELY be adapted to run the engine in question but impersonating it to the stock ECU will not be so easy since many of the sensors on the stock engine are directly interfaced to the ECU. Therefore, rusefi MUST present signals to the ECU which are in accord with the signals the ECU expects to see in its current state of operation, with none out of tolerance. Many ECU's can detect open/short conditions on injectors, ignition coils, sensors, et al, as well as over/under current; the former reflecting gross failures in the device and/or wiring, the latter reflecting degraded connections and other faults. Should that be handled, and if so, how?
How does rusefi detect 'misfire' events, and how does that get sent to the ECU?

Overall:
- possible? Probably.
- possible with current hardware? Not even close.
- lot's more to think about.

^Yes, something like this, I think. What are the limitations of the current hardware? I think the car has separate Engine and Vehicle ECUs, but until I get my hands on the manuals I'm just guessing on lots of this. So far I've mostly focused on the aspects of the swap (engine bay space, design of engine mounts, steering rack clerance, etc) and the costs.

[kb1gtt]

Trans control is something we have not yet gotten into. I'm a thumbs up from a hardware support side of it. Do you have a schematic of either side of this transmission? Either something that shows the car side and what's connected to it, or even better would be something that shows the internals.

Here's my basic understanding of auto transmissions as it stands now. Some are pure mechanical, so no electronics at all. Some are mechanical with sensors that run to some kind of ECU or electronics to let you know things like shift lights gear selector position, ect. Others have varying levels of control over features like shift points and shift speed (squishy vs hard shift). Then some of the really new stuff has control over how much power goes to the front left tire or the back right, ect. I'd really like to find a 4X4 trans I can control and put under a YJ jeep. I have a 1996 Dodge RAM 1500 that could be a starting point. It's currently RWD and the body is pretty much shot. However V8 and such is still good. How close would an LS and this engine be? Would we potentially have a common transmission? For all I know, this engine could be an LS, I don't really know.

[blundar]

Trans control is something we have not yet gotten into. I'm a thumbs up from a hardware support side of it. Do you have a schematic of either side of this transmission? Either something that shows the car side and what's connected to it, or even better would be something that shows the internals.

Here's my basic understanding of auto transmissions as it stands now. Some are pure mechanical, so no electronics at all. Some are mechanical with sensors that run to some kind of ECU or electronics to let you know things like shift lights gear selector position, ect. Others have varying levels of control over features like shift points and shift speed (squishy vs hard shift). Then some of the really new stuff has control over how much power goes to the front left tire or the back right, ect. I'd really like to find a 4X4 trans I can control and put under a YJ jeep. I have a 1996 Dodge RAM 1500 that could be a starting point. It's currently RWD and the body is pretty much shot. However V8 and such is still good. How close would an LS and this engine be? Would we potentially have a common transmission? For all I know, this engine could be an LS, I don't really know.

Did you even read dude's post? He's putting a LS motor with a MANUAL TRANS into something that sounds a lot like a IS250 or 350.

rusefi isn't ready to do what you want now. You would have to implement a lot of the measures you're trying to do. You might be better off to use something like ProEFI or AEM Infinity (or even the Motec M1?) that are a little bit more mature and expose the CAN module on the standalone for programming if you want to get your hands dirty.