So, other than being limited to the factory ignition coils, why do people say that the factory ecu can't or shouldn't be used when doing a swap?
Is it because of the auto trans or ???

 

Well, where can I go to find out?

 

Hmm it may be possible but you're going to need to trick the ECU into having everything that it wants to see to run properly. You're going to need to run all the factory ignition,fuel injectors, turbos, maf sensor, etc. I did this with a 13brew swap into an FB. You will need to look over the pin-out of the factory 20b ECU. I don't know where to find it.

An example of what I mean; I wired up the factory 13b rew ecu into the FB chassis. I had an external fuel pump wired to a 12v switch standalone. The ECU will throw a limp mode code and the car ran like crap because the fuel pump relay trigger inside the ECU did not detect a relay. Wiring up a dummy relay fixed the issue.

If you are swapping a 20b and spending all the time and money on doing a swap the right way, why would you want to run a factory ecu. Who knows how will work out with a manual transmission. You're much better off with a standalone. If you can figure it out, all the power to you. A friend of mine makes over 650hp on a factory rb25 ecu with an SAFC swapped into a 240sx.

 

It's more of a curiosity. It'll be a couple years before I'm ready to get into the swap.
Everywhere you look people say the factory ecu won't work, but noone explains why. It seemed to be tribal knowledge with no facts.
I just like knowing reasons behind ideas so I can make my own decision. People said turboing a 6port na was a bad idea. But I beat on my setup with only safc for control for a couple years until I went full standalone.

 

Most things about a 20B seemed to be tribal knowledge untill I jumped in and started de-mystifying the process. Stuff like cooling capacity needed, what water pump to use, front covers and so on.

I know there's a few JC Cosmos that have been converted to manual transmissions. I'm pretty sure that most of us 20B guys would love to see what kind of electrical trickery can be done to make a NF01 ECU work properly with a manual (I certainly would!), it's just that time and resources can be finite for side projects, especially when a full standalone can do the job with no fussing about.

My current understanding is that the NF01 ECU goes into Limp Mode without the ECAT wired in as well. As for how it specifically behaves, I am very much in the dark as I don't have one to play around with, but I'd fathom a guess that it'd be similar to a S5+ with a dying OMP. When I get the time/means to translate the Cosmo's FSM (another side project...lol), we'll learn a bit more about it then.

 

The frustrating amount of hrs spent figuring out how to make the ecu work could be spent earning the money to pay for a standalone lol. This is why they say it won't work. Anything is possible, its a matter of opportunity cost.

 

I'm starting a "20b fund" beginning next year for my fc. And while it's accruing the monies I'll need, gonna be doing bodywork and research. So, I got time.
There's a guy local to me that makes regular trips to fill containers of Japanese goodies. He can get me a full clip when the time comes.

 

Thanks for the info by the way.

 

as others have said, it isn't that it "won't work"
It is that a stock 20B makes 300whp. Swapping it into a RX7 takes a lot of time and/or money. It just doesn't make sense to do all that with an un-tuneable ECU that is a pain to figure out and limits you to about that same power level.
You can buy a used LTx12 for <$1k pretty easy. Installation can then be done in an afternoon, and you have the flexibility for up to about 700whp. Why would you hassle with a stock ECU when you have that option?
Even if you have all the free time in the world, you're still better off with a standalone, and use the time saved to learn bongo drums or find a girlfriend or something.

 

Too add..anyone doing any kind of engine swap will benefit from learning how to wire and setup their own ecu. It can and will get fustrating but in the end, the knowledge gained is invaluable.

 

To add to the above, you would be stuck with the stock 550/550 injectors, must run the N390 AFM and boost sensor, and be limited to 5 psi of boost.....Basically the Cosmo is a 3 rotor (or 2 rotor) using all the same technology as the S5 TII with a set of sequential twins and automatic trans.

 

^Hmmm didn't realize that a stock 20b only ran 5psi to make 300hp.

 

that is a good first question. the Cosmo was never sold outside of Japan, and thus all of the factory service info was in Japanese, and its still rare.

this is one reason that the stock ecu is hard, the wiring diagram wasn't easy to find and its in a foreign language, and there is a lot of it.

we never tried it either, it will set code(s?), and limp home of some sort is likely too.

it is easy just to leave the trans wired in.

it'll hit ~8ish on the first turbo, but by redline 5psi is optimistic. ReTed pulled the turbos off of his 20B, and made 237rwhp, my friend has a 20B running the stock ecu and turbos in his truck, and it made 247rwhp....

the stock Cosmo wiring is huge, and if you don't get a front clip, making the stock ECU run would be an immense job.

even with a front clip its still a huge job, as half of the engine control wiring goes through the dash and body, so you need get it running with no codes as a clip, take it apart, remove the chassis wiring, and install in your car.

there are a few people who have done stock ecu over the years, and its a huge, frustrating, long tedious job if you LIKE wiring.

the stock ECU doesn't give you much power, and its got lots of limits (think boost cut), but it does run nicer than any standalone 20B i have seen to date, although i haven't seen one in a while.

 

IIRC, the dyno charts for the 20B-REW in the JC Cosmo FSM show 0.7 Bar, which is just shy of 10psi in a JC Cosmo. Otherwise, Banzai Racing is correct in needing the airflow meter and Boost Sensor from a JC Cosmo. You could substitute a S5T2 sensor as they're scaled identically tho.

As far as the Q4A-EL Transmission is concerned, I have a theory that it's just a beefed up and re-geared version of the R4A-EL found in the 13B Cosmo and FD and operates identically. Considering that Automatic FDs get funky with Limp Mode with a Power FC, perhaps the same logic applies with the 13B Cosmo and by extension, the 20B Cosmo as well. Banzai, your thoughts on this?

For the amount of wiring work it would need, I personally would build a harness from scratch. Nothing annoys me more than poorly executed wiring jobs, especially when adapting and 'blending' two harnesses together in manner intended here.

I've taken a look at the diagrams and pinouts for the NF01 ECU and it doesn't seem TOO complicated. Once I get back on my main linux compy, I'll take a crack at translating them.

 

Having a 20B and no access to the ECU is like carrying a set time bomb around in your pocket with out knowing the exact time it will go off! My water temp sensor for the ecu malfunctioned and engine started running very strange. This would have beed a pain to figure out without access to the ECU.

 

this is why u font want stock ecu.....and this aint all of it! this is just the crap *** BS that is directly associated with the engine... (solenoids and sensors and emissions)then u have the transmission and so on

 

Most of the above is very true but a 20b running in its original sequential turbo set up is a fun car to drive, torque from 1800 rpm through to the sec turbo power pull. Nice.

 

I know this thread is old but if anyone is still reading it,
I picked up a 20B front clip last year and have been going through the ECU pin outs and have most of them identified. I could post a pic if anyone is interested and also wants to chime in on helping me finish it.
I can start and run my front clip with a fuel pump connect to the battery directly.

 

Over the last few months, I've translated most of the Cosmo's Engine Control diagrams and made them searchable in both English and Japanese. The only thing I haven't put in are wire colors for everything due to the original materials being a bit blurry/smeared. Perhaps we can collaborate on this so we're not duplicating the same work?

 

I'd be more than happy to share notes on the topic as I currently have another 20B engine in an FD right now that needs some more work before it will run especially on the wiring end of the project. I have all the pieces of the puzzle except some diagrams and literature. PM me and we can exchange phone #s if you like.