wstrohm

I have noted a number of posts on this forum that imply the "On-Board Diagnosis Function" of the stock PCME (ECU) of the FD is not well understood by all posters. Did some searching, and found what were apparently direct links to the info that are all now dead. My apologies for this post if I overlooked any links that duplicate the following information.

The factory manual covers a method for reading stored trouble codes from the OEM PCME using a special tool and system monitor. The same system was used on the 1990 - 1995 OBD-I Miatas, and provides a lot of useful information on electrical systems status. The easier technique uses a jumper wire (or paper clip) and the car's Check Engine Light (or "Malfunction Indicator Lamp") as the readout device.

The procedure to read trouble codes is:

1) Disconnect the battery for about 30 seconds to discharge the airbag system's "back-up power supply" capacitor. This will clear old PCME stored faults. Reconnect the battery.
2) Place a jumper wire (a paper clip will do) between the "TEN" and "GND" terminals of the Diagnostic Connector.
3) If the engine is known to run, however badly, start the engine and run it for about a minute. If it will not start, attempt to start it for a few seconds. Next, with the engine off, turn the ignition key to the "ON" position.
4) Observe the CEL. If there are no current stored trouble codes, the CEL will turn on for 2-3 seconds and then go off. If there are stored codes, the CEL will display flashing code(s) identifying the trouble area. (If the CEL stays on, possibly the jumper wire or paper clip is not making a good connection.)

The flashes are read as follows:
1) A break between repeats of the code(s) lasts 4.0 seconds. During this time the CEL will be off.

2) The code(s) can display "tens" digits and "ones" digits. A "tens" digit is 1.2 seconds long. A "ones" digit is 0.4 seconds long. If the code number is greater than "10, the "tens" digit(s) will display first, followed by a 1.6 second break, followed by the "ones" digit(s). Example: Code #34, an Idle Air Control Valve electrical fault, would be displayed as three 1.2 second flashes followed by a 1.6 second break, followed by four 0.4 second flashes, followed by a 4.0 second break, followed by a repeat.

3) If more than one failure is present, the code numbers will be indicated in numerical order.

The FD trouble code chart in the 1994 Workshop Manual lists the following codes:

02 No Crankshaft Position Sensor "Ne" signal
03 No Crankshaft Position Sensor "G" signal
05 Knock Sensor - open or short circuit
06 Vehicle Speed Sensor - no signal
09 Engine Coolant Temperature Sensor - open or short circuit
11 Intake Air Temperature Sensor - open or short circuit
12 Throttle Position Sensor (Full Range) - open or short circuit
13 Manifold Absolute Pressure Sensor - open or short circuit
14 Barometric Absolute Pressure Sensor (in PCME) - open or short circuit
15 Oxygen Sensor - <0.55 volts for 25 secs in closed loop zone
16 EGR Function Sensor - open or short circuit
17 Oxygen Sensor - no voltage change for 120 secs in closed loop zone
18 Throttle Position Sensor (Narrow Range) - open or short circuit
20 Metering Oil Pump Position Sensor - open or short circuit
23 Fuel Thermosensor - open or short circuit
25 Solenoid Valve, pressure regulator control - open or short circuit
26 Metering Oil Pump Stepping Motor - open or short circuit
27 Metering Oil Pump - open or short circuit or sticking of MOP sensor
28 Solenoid Valve (EGR) - open or short circuit
30 Solenoid valve, split air bypass - open or short circuit
31 Secondary Air Bypass Valve - open or short circuit
32 Secondary Air Switching Valve - open or short circuit
33 Solenoid Valve, port air bypass - open or short circuit
34 Solenoid Valve, Idle Air Control - open or short circuit
37 Metering Oil Pump - low battery positive voltage
38 Solenoid Valve, Accelerated Warmup System - open or short circuit
39 Solenoid Valve, Relief 2 - open or short circuit
40 Solenoid Valve, Purge Control - open or short circuit
42 Solenoid Valve, Turbo Precontrol - open or short circuit
43 Solenoid Valve, Wastegate Control - open or short circuit
44 Solenoid Valve, Turbo Control - open or short circuit
45 Solenoid Valve, Charge Control - open or short circuit
46 Solenoid Valve, Charge Relief - open or short circuit
50 Solenoid Valve, Double Throttle Control - open or short circuit
51 Fuel Pump Relay, Speed - open or short circuit
54 Air Pump Relay - open or short circuit
71 Injector, Front Secondary - open circuit
73 Injector, Rear Secondary - open circuit
76 Slip Lock Up Off Signal (PCMT) - open or short circuit
77 Torque Reduced Signal (PCMT) - open or short circuit

Hopefully this data will be of some use to someone.

Sgtblue

iTrader: (16)

I'd also suggest that when obtaining codes (especially if your getting more than one) you repeat the process at least once to be certain you've copied it properly.

In order to clear codes disconnect the negative terminal for 20 seconds and hold the brake pedal down briefly.

danny hahn

Thank you for the write up. got my light to stay on. I have a good gnd as I use it for fuel pump. any other suggestions where to short the TEN besides diagnostic box?

wstrohm

Any chassis ground should do. But if the light stays on for a long time (> 20 secs), either you have a bad TEN -> GND connection, or terminal 2B of the PCME is being held below 2.5 volts for some reason. (The manual says to check for a possible short to ground on terminal 1F of the PCME. That is the FEN terminal in the Diagnostic Module.)

FYI, below is a diagram of the 1994 FD's Diagnostic Connector with additional info.

danny hahn

Dont know if the following data means much, but light stays on when jumped ten to gnd. Voltage between ten to gnd at 2V key on. While running, near battery/alt voltage (.3V loss).

Tried to find short in fen terminal, no continuity between fen and gnd,

At ECU, Voltage terminal 2B to ground 0.78V in key on, can get it to 2.5V while started at 4100rpm.

Checking for possible short at 1F, no continuity between 1F and body gnd.

Hope I followed instructions,please correct me if I didn't. thanks.

Is it normal with key on for the ECU to get warm top side of the box around the number 4 plug?

What I'm trying to do is to use the stock ecu to produce codes to help solve a secondary boost issue.

Changes made to ECU harness: removed aws and egr wires typical of pfc install, and used 1B to power a relay for the pfc a/c fix.

When you mention the manual says, are you referring to FSB section F-166,234,235? thanks

danny hahn

I was looking through the Body Electrical troubleshooting manual trying to look into someone else's instrument cluster when I stumbled upon ECU terminal 2B on page C1-11.

The instructions were to measure voltage at 2b with ECU unplugged while at idle. (flowchart 4, step 2, Tachometer doesn't operate) I would think connectors 1 and 4 would be necessary to start the engine as they address spark and fuel.

I measured 2b with connector 2 unplugged at idle and got minimal .2v instead of the 3-6V. however my tachometer works.

Had to give up on trying to get my old ecu to throw codes for now.

Does the car start without the ECU plugged in?

qle227

anyone know how to remove intake air temperature sensor from 93 mazda rx7. please email me at qle227@gmail.com

MuRCieLaGo

iTrader: (14)

Made a video years ago of what it looks like reading codes with my JDM FD.

https://www.youtube.com/watch?v=RQA_TZKRY8U

wstrohm

MuRCieLaGo,

According to the table above, when you made that video you had a shorted or open MAP sensor. Is that true?

MuRCieLaGo

iTrader: (14)

Yes, exactly (#13).

RE.Amemiya.FD3S

Hey wstrohm, my ECU gives me the "25 Solenoid Valve, pressure regulator control - open or short circuit", i tried searching online to purchase this solenoids but i couldn't get it, does it have another name?

Awaiting for your reply,
Thanks

wstrohm

Sometimes called the "PRC Valve Solenoid," the item is "A" in the first vacuum diagram on this page. It is the first solenoid on the right in the rack, closest to the engine compartment firewall. The Mazda P/N is 1481-13-240A.

RE.Amemiya.FD3S

Wow, You ARE A STAR! Really appreciate it man
and what about the "26 Metering Oil Pump Stepping Motor - open or short circuit" ? is it the Oil pressure gauge arrow meter with the cluster? because mine suddenly stopped working and it doesn't read the pressure anymore!

Thanks a lot

RE.Amemiya.FD3S

Nevertheless my 7 doesn't boost up properly/power loss! "Limp Mode" others told me that it's my vacuum lines, but since the ECU flashes 26 then it's definitely not the vacuum lines, because I replaced the vacuum lines with new silicon ones.

So what should i do exactly? Replace the Metering Oil Pump? which costs $1700 from Atkins Rotary.

The ECU also flashes :

23 Fuel Thermosensor
25 PRC Solenoid (3Way Valve)
44 Turbo Control Solenoid

But No 27 which is the whole OMP.

Please help guys.

Thanks

wstrohm

"Open or short circuit" solenoid could mean the electrical connector to the solenoid has a bad (i.e. corroded or broken) connection, i.e. the solenoid may not be defective, although of course it could be. I would check for poor connections on the:

23 Fuel Thermosensor
25 PRC Solenoid (3Way Valve)
26 Metering Oil Pump Stepping Motor
44 Turbo Control Solenoid

Even though your vacuum lines are perfect, the electrical connections to the solenoids may not be. Also unfortunately, the high temperatures seen by the solenoid rack components can over time degrade the operation of the solenoids themselves, as well as the hoses. There have been tests previously done to verify solenoid operation, and reported on this forum. One example is this thread.

joelanger

so how do you do this on a JDM model with no 'check engine light'?

wstrohm

I would guess that you put together an LED indicator and hook it up as described here. Although the article is for Euro and Aussie MX-5s without a CEL, it should work for the RX-7. Note that I have not tried this, so no first-hand experience.

joelanger

Awesome, I'll give this a go today and let you know if it works or not

silverFDgrl

So, I have a question about this. My car is going into limp mode, and the check engine light comes on. When I check TEN -> GND it blinks once then stays on. My car should always throw code 16 for the EGR, so now it's not even doing that.

We did your test for terminal 2B and it is 1.8v (clearly below 2.5v as you say is needed). So what is the fix? How can I get that voltage back above 2.5?

Thanks

wstrohm

Ouch, my bad, error in my post #4 above, should be terminal 1F of the PCME, which is the "FEN" terminal in the Diagnostic Connector. (Terminal 2B is the IG- terminal of the Diagnostic Connector, which is a pulse-stream controlling ignition timing and the tachometer drive. Not relevant to troubleshooting the constant-on CEL.)

When PCME terminal 1F ("FEN" in the Diag. Conn.) floats or is a high voltage, the CEL should be OFF. A low voltage on 1F turns it ON. When there is a trouble code, 1F ("FEN") pulses low to indicate the code number.

The manual is correct that a constant-on CEL can only happen if PCME 1F ("FEN") is low, i.e. its white/black wire shorted to ground, or possibly a short inside the PCME. With ignition ON and engine off, when the TEN terminal in the Diagnostic Connector is open the CEL should be ON. When the TEN terminal is connected to the GND terminal (following about 3 seconds after turning ignition ON), PCME terminal 1F should be at a logic high voltage (above 2.5 volts) and the CEL should be OFF, unless there are trouble codes.

I apologize for my error in post #4; I have no recollection or rationale for why I incorrectly referred to PCME terminal 2B.

silverFDgrl

Thanks for the clarification. If terminal 1F is below 2.5 volts, do you know any way to fix it and get it back to a logic high voltage?

wstrohm

A few possibilities:
1) There is a poor connection between TEN and GND, which will float TEN and cause the CEL to stay on until engine start.
2) There is a wire-to-wire or wire-to-chassis ground short from the white wire with black stripe (from PCME 1F to the FEN terminal in the Diagnostic Connector). That will pull PCME 1F low and turn on the CEL.
3) There is a defect inside the PCME pulling its 1F terminal voltage low.

What voltage are you measuring on the FEN terminal? Is it at zero volts (a short to ground) or a low voltage?

wstrohm

I need to mention one more possibility for the behavior of the CEL:

4) The FEN signal routs to the Electrical Load Unit, which apparently contains a non-inverting buffer that drives the CEL bulb from its terminal "K." So it is also possible that the ELU's input terminal is a very low impedance due to an internal circuit failure, which is pulling the FEN voltage down. If you disconnect the ELU's connector, does the FEN terminal behave normally, i.e. as follows?

a) With TEN->GND jumpered in the Diagnostic Connector and ignition turned ON, FEN goes low for 2-3 seconds (1.8v on our '94 FD) and then goes high and stays there (10.8v on our car).

b) With no jumper in the Diagnostic Connector, FEN goes low and stays there (CEL would be ON as self-test if ELU were connected) until engine start.

If the FEN behaves as in a) and b) with the ELU disconnected, the problem is in the ELU or its input wiring.

silverFDgrl

The FEN is coming in at 1.9 volts when the key is in the ON position and 2.1 volts when the car is running.

As a side-note we are trying to diagnose why the car is going into limp mode. If the system is reset, the check engine light goes away, the car will boost about 10 times, then something triggers limp mode and it refuses to boost. We have another car here, and even putting that ECU in my car caused the exact same symptoms.

Obviously, getting the check engine light to throw codes would make it MUCH easier to diagnose the limp mode

Thanks for the new info - we will test this tonight after work. Keep you posted

silverFDgrl

UPDATE: so it appears that when testing the above, with the ELU CONNECTED, the FEN went low to 1.8, then went high to 11v for a quick second then dropped to 1.8v and stayed there.

when the ELU is DISCONNECTED the CEL never comes on. we tested the voltage anyway and it starts at 1.6v, jumps to 11 for a sec then goes back down to 1.6v.