When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Hello everyone,
I just rebuilt my motor (first time) and recently got it fired. It started and ran ok, but then started bogging down around 2500 to 3000 rpm. I didn't get any CEL's, so I started researching fuel and spark issues. Then, I thought to just check for codes in case it was stored in the computer. The car displayed 10 codes!!! They are as follows:
5 - Kock Sensor (I bought a new sensor and it is on)
11 - Intake Air Temp Sensor
12 - Throttle Position Sensor (full range)
18 - Throttle Position Sensor (narrow range)
Note: I check and adjusted the TPS and it was within spec after I adjusted it.
23 - Fuel Thermosensor
32 - Secondary Air Switching Valve
44 - Solenoid Valve (turbo control)
45 - Solenoid Valve (charge control)
71 - Injector (Front Secondary)
73 - Injector (Rear Secondary)
(Note; both primary and secondary injectors were sent for cleaning and testing, and have new grommets and seals)
When I put the engine back together and in the car, the wiring harness was less than stellar (crusty), and after pouring thousands of dollars into parts, I didn't get a new harness but told myself that if I had to pull the motor again, I would get a new harness. Does this sound like a harness issue, or could there be one underlying problem affecting everything (like the secondary injectors not firing and leading to tripping the knock sensor and putting the car in safe mode)? I feel like I'm going into a rabbit hole with these problems so any suggestions would be great. Thank you all!
Yes, I am on a stock ECU. That's my first step tomorrow to check continuity of the knock sensor. My idle was hunting as well, so I took off the UIM to a) make sure I have all the vacuum lines and plugs connected and b) bought a new IACV gasket since it looks like it has never been changed or taken off, and I want to clean it out.
When I put the engine back together and in the car, the wiring harness was less than stellar (crusty), and after pouring thousands of dollars into parts, I didn't get a new harness but told myself that if I had to pull the motor again, I would get a new harness. Does this sound like a harness issue, or could there be one underlying problem affecting everything (like the secondary injectors not firing and leading to tripping the knock sensor and putting the car in safe mode)? I feel like I'm going into a rabbit hole with these problems so any suggestions would be great. Thank you all!
Just a quick question - In the process of reinstalling the engine and before starting the car, did you test the fuel system for leaks with the UIM off so you can see all the fuel plumbing connections? If you did this, you most likely had all of those connectors that are reading codes now disconnected, but the key would be ON for the test, and ECU powered up. So it would see all those open circuits and throw codes for them.
If your harness is really crusty, you should get a new one.
Oh, that is a good point! I did do a fuel pressure check with the UIM off and the rats nest disconnected (had to switch out all the old lines to silicone lines).
As a backstory, this car was passed down to me from family, and it was purchased used. The car wasn’t ran for years and before that, it looked like all the gaskets were supplemented with blue gasket sealant. Shortly after I got it, one of the water gaskets failed, and I took the opportunity to rebuild and learn about the car. From Honda and Subaru engines, this has been challenging.
One question is when checking the fuel system, I did have the secondary injectors plugged in. That shouldn’t have thrown a code then, right? Or would the knock sensor code prevent the secondary’s from firing and put the car in limp mode?
One question is when checking the fuel system, I did have the secondary injectors plugged in. That shouldn’t have thrown a code then, right? Or would the knock sensor code prevent the secondary’s from firing and put the car in limp mode?
Not necessarily. Even with the secondary FI's plugged in, it would still throw a code if they were not getting their common +12VDC supply from the EGI Main Relay OR if there's a short anywhere in that circuit's wiring. So if you reference the FSM wiring/schematic diagram, if ANY of these connectors were unplugged, you'll see a code due to an open circuit: B1-01 (@ECU); X-05 (near ECU); B1-02 (EGI Main Relay). If there's a short, there's no easy way to find it other than testing every single connection in the wiring harness.
If the EM harness is crusty, brittle and corroded, I wouldn't be surprised if there are at least a few permanent/intermittent shorts of opens within it. Especially since the harness had to be moved around a bit for the engine removal & install drill. Usually the crusty/brittle insulation will crack & break off, allowing wires that are not supposed to touch one another to short out. This is why it is worth your sanity to just bite the bullet and get a new EM harness.
Another wiring area to watch out for on a used FD RX7 is the Front (F) harness, which runs under the front plastic fender liners. If the car's suspension was ever lowered, or had larger/wider tires installed, it's possible the front tires could have rubbed thru the plastic fender liner and damaged the (F) harness just above it. Take a good look at the plastic fender liners just above the front tires for signs of rubbing/damage, and if they are worn thru, remove the liners and take a close look at the F harness just above it for signs of damage.
I had to be very careful when removing the harness; all sorts of plastic heat sheathing was coming off and had to resolder and rewrap a few wires, and attach a new connector. That should have been my first clue to change out the whole thing. I should've just bought the harness but was thinking I could get by this one time. In hindsight, I knew better.
I changed the brake lines and all the fender liners look good, just covered in dirt (like it was driven in mud or something).
I looked on the forums and it seems like Roy Malloy is the person to go to for Mazda parts. I did send him an email but no reply. Does anyone know if he is still around for a new harness and driver side seat belt? The belt I have is very, very faded.
The codes blown almost look like a scenario where the car has had "reliablility mods". I.e. all the emission and sequential turbo hoo-ha removed.
If RHD there is no CEL. You have to run codes (which is what you have done).
If the original set-up is still existent, then the harness would definitely be suspect.
But you might take a look inside the ecu to see if there is any obvious damage.
It's maybe 32 years old now.
Look for damage to the harnesses where they pass through the firewall. Years of persons fishing stuff through the shared openings may have damaged the harnesses.
If the car was run without fender liners, there could be burn-throughs of the front harness (on either side). Water getting into the front harness can cause major issues.
[As already mentioned above.]
The majority of owners up here (85% plus) have moved away from the original stock set-up.
The components are just too old and you may never know what a senile old ecu is thinking.
Yup, keeping the car stock. What little I drove it before the coolant seal failed, it seemed plenty quick and fun enough. Want to enjoy it for a while as stock. I’ll send Roy an email about the harness.
I’ll be sure to post once I get the harness installed in case others have the same problem.
Yup, thanks to CARB, I have to keep the whole thing stock for the visual test...even if it passed emissions. Such a PITA!
Sorry I missed your earlier post.
Yes, it is LHD, and it looks like the harness was never touched. All the fender liners are intact. The car had 88,000 miles on it with lots of gaskets still in place with what looks like Blue Permatex. I had to replace one iron because one of the bolts holding the oil pan busted through the iron. In hindsight (again), I should've ordered new irons and, rotors, and housings. They measured ok, but barely. I figured what are the odds that I would assemble the engine correctly the first time and it would fire up, but it did. Now, the compression is low for a rebuilt motor, but I'm hoping it improves a little once the seals are worn in. I figure I'll check after 500 - 1000 miles of driving, as I have about 2 miles on the engine now.
The only thing is the ECU pictured below has a Reman for Mazda sticker. It also has the updated fan harness I believe (blue and white connectors). I assume this is the correct ECU?
The 1994 USDM Parts catalogue lists 10 different ecu for USDM cars. But it lists N3A3-18-881A as the ecu for California/ Manual transmission.
The most common USDM ecu is likely the N3A1-18-881.
Probably a New OEM emission harness is the way to go in spite of the expense, but bear in mind it might have been sitting in a box for 30 years as well.
Also the emission harness may be somewhat different than stock for California cars. So save your old one to compare.
Likely most of your 30 year old solenoids will be suspect, but again "new" replacements may have been sitting in a parts bin for many years.
Check that none of the ecu "pins" have been bent, by the plugs being removed and replaced, from time to time.
if you can find a buddy, see if running his N3A3 ecu in your car makes a difference. (Providing he is willing to share.)
Given you ar egetting so many codes, you may have some serious shorting going on, so there is risk of damaging any ecu you try.
If you tested things with the ECU powered but some of the harness disconnected, I would disconnect the battery negative cable for a few minutes to let the ECU reset its codes, then reconnect everything and check for codes again. For any remaining codes, I would disconnect the battery again, unplug the ECU connector and the harness connector (for instance, the intake air temp sensor) and use a multimeter to check for resistance/continuity between the ECU and the sensor connector. Also check that the ground pin on the sensors (like air temp, knock sensor, fuel thermosensor) has low resistance to the battery negative wire, anything more than 2 ohms might cause problems.
Search for 'how to check automotive wiring harness continuity' if you're not familiar with that. Most people are afraid of electrical work, but you can learn it. An extra length of spare wire and something small like a paperclip or a sewing needle can be helpful to get inside the harness pins without damaging the plastic connector housings.
Yes, I agree. I just did that and no codes were on!! I don't know why I didn't think of that sooner.
I took off my fast idle cam because the thermowax pellet wasn't moving, and the paper gasket was just shredded. I ordered a new paper gasket as will as a IACV metal gasket. Once I get those in, I'll fire up the car and hopefully no codes pop up.
I don't know. For my JDM, the ECU would go into limp mode. I would pull to the side of the road, turn the car off, and restart in maybe 30 seconds. Limp mode would be gone for an indiscriminate time (5 minutes, 30 minutes?)
I read later that this was because the ecu would cycle through it's self checks, and throw limp mode again. The reason the times might have been different is that the defaulting sensor may have been still partially working (likely it was the OMP was sending a weak check signal), and so the ecu would pick up the default indeterminantly
OMP signal is one that the ecu takes very seriously.
I can try to find a better explanation.
But I believe it is likely that the ecu would have to be looking at the running engine.
Also, I have searched N3A3-18-881R (on google), and it seems it is some form of replacement ecu for the N3A3-18-881. It is offered alongside N3A3-18-881 on various parts sites. I suspect there may have been a recall. N3A3-18-881R does not show in my parts catalogue, So the change may have occurred after such parts manual was printed. The "California" ecu I have is N3A3-18-881B. It also does not show in my parts catalogue. I suspect it was a follow-up ECU to both the "881" and "881R". I also have a N3E1, which seems to be the one for 1995 California USDM Manuals.
I looked through the 1993 Service Highlights manual I have. Under section F (page F62) it explains the self diagnostic functions of the ECU Probably a similar explanation can be found deep in Section F of the USDM FSM.
It says that after you make a repair of a fault, the ecu will reset if the battery is disconnected for 20 seconds.
There is a table of about 50 possible defaults. As a result of the detected default, the ecu will take various actions, which include, among many other possible actions, making the ecu assume a range of operation. 'Limp mode" may be such an action.
It also says that the ecu will take action on certain faults only after a delay, say maybe 60~120 seconds, without receiving a signal.
These many responses to many different faults, may be why it becomes so hard to really track what is going on.
And this is perhaps why old systems seem to cause the ecu to go haywire, due to, maybe, weak signals, no signals, stop and go signals.
I would add, with a generation of OG rotary mechanics retiring, there are now no shops in our area that will tackle trying to sustain the OEM set up.
Of course, some shops may wish to cross sell new turbo set ups, but to be fair, experience has taught them that is it is near impossible to sustain a pure OEM set up.
So you may be stuck with trying to do the work yourself.
Further Mazda, themselves, or through Mazda Motorsports, has a range of replacement parts, including many of the little formed vacuum hoses for the "Rat's Nest".
Mind you, the pesky ecu might go through its check cycle and throw the codes agin.
there is no cycle, it just checks in real time.
there is a chart somewhere, but most codes are just for open or short circuits, so the check isn't a big one
Hi everyone, sorry for the late reply. I think I had a connecting rod bearing go out on my B18C1 so I started the process of taking that engine out.
Just as an update, I cleared all the codes and the only one that popped up again was the TPS Sensor Voltages. I adjusted those and took it for a test drive, came back, and still no codes!! It looks like I had to just reset the ECU after the the fuel leak check.
@Redbul, I'll still take the California ECU if you have it and I'll still try to contact Ray for a wiring harness. I know there seems to be fewer rotary mechanics arounds and perhaps even less that work on the FD (especially California and our emissions/visual checks), so that's why I'm trying to do and learn as much as I can now. This and Facebook have been a big help so I hope that there are those continue using and checking this. I know I sure do! Thank you everyone!
