Probing injectors to PCM harness

Tiny
HENRY JOHNSON2
  • MEMBER
  • 1994 MAZDA PROTEGE
  • 1.8L
  • 4 CYL
  • 2WD
  • AUTOMATIC
  • 240,000 MILES
1. When I check the injector harness, with the key to on, I get 12 volts on both pins. Is that supposed to be? Much of the internet says only one should have 12 volts.

2. When I check continuity from the injectors to the PCM harness, I get:

A. Continuity on 9 out of the ten sockets for the white/red wire.

B. Continuity on 9 out of the ten sockets for the yellow wire.

C. Continuity on only 1 out of the ten sockets for the yellow/black wire.

Two of the injectors have the yellow wire and two have the yellow/black.

The problem I am trying to solve is that the car starts and runs perfectly but will not start after I shut it off. After it sits awhile- hard to say how long- it will start again and run perfectly, like a new car. I have changed everything from one bumper to the other so now I have to dig into the wiring and PCM.

Thank you,
Gerry Connery
Sunday, March 27th, 2022 AT 12:01 PM

5 Replies

Tiny
CARADIODOC
  • MECHANIC
  • 33,871 POSTS
The confusion comes from when you expect to see certain voltages. These three diagrams cover the Engine Computer and controls. Unlike most other brands, you will find 12 volts at the injectors with simply turning the ignition switch to "run". The blue arrows show where 12 volts comes from the ignition switch and turns on the main relay. On other brands, that doesn't occur until the engine is rotating, (cranking or running). That's done to insure the injectors are turned off if a fuel line is ruptured in a crash so raw gas doesn't get dumped on the ground.

Once the main relay is turned on, it switches 12 volts onto the circuit with my red arrows, ending at bullet # 11.

Moving to the second diagram, # 11 continues to the red arrows and includes appearing on the wires that are the same color at all the injectors. That appears to be a white / red tracer, (stripe). That's the wire you should see 12 volts on all the time. Where the confusion comes in is the four individual yellow wires. Those are the ground, or in this case, the control wires. They're only using two switching circuits. Each circuit fires two injectors together at the same time. The Engine Computer grounds those wires when it wants to fire those injectors. With 12 volts on the white / red wire, and 0 volts on the yellow wire, the difference is 12 volts and the injector pops open its valve so fuel can spray out. At the precise time, the computer removes the ground on the yellow wire to turn the injectors off. This is when you'll find 12 volts on both wires. It's coming in on the white / red wire, passes through the coil of wire inside the injector, and will be seen on the yellow wire, as long as the connector is plugged in. If you were to unplug the injector, you would no longer see the 12 volts on the yellow wire.

The important point is the difference in voltages on those two wires at each injector. If that were to be 12 volts all the time, like you read online, the injectors would be held open continuously and it would bleed fuel pressure off to 0 psi. That's one way we remove fuel pressure before working in the system. That also leads to a very long crank time to get the engine started next time. That fuel pressure is supposed to hold for days and even weeks without dropping significantly.

This whole story is irrelevant though. The engine runs fine so we know the injectors and its wiring harness are okay. Even if the computer had an intermittent problem related to the injectors, that would only affect the two injectors on that one yellow wire. The symptom would be different. We need to look at what everything has in common when the engine doesn't start. Most notably that's the crankshaft position sensor, (green arrow). I could be wrong, but as I recall, when the crank sensor fails, you won't have injector pulses or spark. There's also a pick-up assembly inside the distributor. On most models from the '90s when they use both sensors, they both have to provide timing signals to get spark and injector pulses. If either sensor fails, both systems are dead. Around the early 2000s we started seeing engines that would still start and run in a back-up mode when one sensor fails. Regardless, the next time the problem occurs, check for spark first before the problem goes away.

For the benefit of others researching this topic, this article might be of help:

https://www.2carpros.com/articles/how-to-test-an-ignition-system

The best clue is in your last paragraph. Crankshaft position sensors and camshaft position sensors, (distributor pick-up assembly), are notorious on all car brands for failing by becoming heat-sensitive, then they work again after they cool down for about an hour. The misery can come from trying to figure out which one is failing. You can start by looking at this page:

https://www.2carpros.com/articles/diagnostic-trouble-codes-for-1995-and-earlier-mazda-cars-obd1

to see if you can read the diagnostic fault codes. If you are lucky, you may find code 02 or 03 to direct you to the failing sensor. If there is no code related to either of them, that is not conclusive. Often on some models, (Chryslers, for example), they need some time to detect the missing signal, as in when a stalled engine is coasting to a stop. Simply cranking the engine is often not long enough for that to get detected. In that case the missing signal is real easy to determine if you have a scanner, but not just by reading fault codes. It's also important to read and record the fault codes before you disconnect the battery or run it dead, otherwise the codes can self-erase, then that valuable information will be lost.

There's two symptom descriptions we see over and over related to crank and cam sensors. One is the engine starts and runs fine for perhaps as much as 15 minutes, then stalls, leaving you sitting on the side of the road. It will restart after about an hour on a cool day. The more common description is the engine runs fine as long as you're driving, due to natural air flow keeping the sensors cool, then, after stopping for a few minutes, such as when getting fuel, the engine won't restart. Engine heat has time to migrate up to the sensors leading to the failure of one of them. Again, the engine will restart after waiting up to an hour.

Let me know if this gets you looking in a different direction, or if I can clarify anything.
Was this
answer
helpful?
Yes
No
Sunday, March 27th, 2022 AT 8:08 PM
Tiny
HENRY JOHNSON2
  • MEMBER
  • 4 POSTS
Randy B,

Thanks for your great reply. The 1994 Mazda Protege does not have a crankshaft sensor according to Mazda. It's funny though, one is listed in my Mitchell "Service and Repair Manual". However, the sensor inside the T2T53171B distributor (Mitsubishi J815) is sometimes listed as a cam or crank sensor. I just ordered one from China which will take a couple of weeks to arrive. I have already bought two distributors from WAI Global, #38403, and it seems odd that two new distributors would be the problem, but who knows.

I did not get the three diagrams you mentioned, but you didn't think that was the problem. I would like to be clear on the injector readings. When the key is turned to run (car not running) I have battery voltage on both terminals of the unplugged harness and, with the car running (harness still unplugged), I get full voltage to both terminals. Of course, the car starts and runs great so I guess it doesn't solve anything. I did not probe the harness when the car was running, and the harness plugged in.

The spark was very strong when the car wouldn't start, and it would not start with the fuel pump jumped. This is an OBD1 model and there haven't been any codes tripped. I have been checking and probing all the wires I can, and I think I'll pull the fuse box under the hood and see if anything looks suspect.

A couple of observations:
1. When the car won't start, not even a little, I can disconnect the fuel pump wiring harness under the back seat and the car will try to start for a second or two. The same thing happens if you disconnect a couple of injectors. I guess the fuel firing sync is not right.
2. Sometimes the car starts with a little burp and smoke, but it clears up pretty quick.
3. The car has never stalled after a start, and I have actually taken it on a 150-mile round trip.

I'll get back after I put the new sensor, J815, into the old distributor, T2T53171B. WAI global was supposed to get back to me, but hasn't, I'll try again. This has been five months, four mechanics, including a Mazda dealership, and lots of parts, but the hunt continues.

Thanks again.
Was this
answer
helpful?
Yes
No
Monday, April 11th, 2022 AT 6:28 PM
Tiny
CARADIODOC
  • MECHANIC
  • 33,871 POSTS
Rats. I'm sorry that I didn't catch the missing diagrams, and after I sweated for hours over my keyboard to format them for posting. Well, okay; entire minutes.

I just noticed the connector I think you unplugged is X-14. I expanded the pertinent part of the second diagram in the fourth one to try to make it easier to read, and I added the orange arrow pointing to X-14. The double dashed line indicates all the other terminals are in the same connector. That's the four injector wires and two wires for solenoids. Based on where those connectors are located in the circuit, you'll still have 12 volts on the feed wire, (red arrows), and on the second terminals for each injector, (purple arrow), as you described. It comes through the coils of wire inside each injector, then shows up on the second terminal. That 12 volts on the second wire will also be there when the plug is connected because the computer doesn't ground them momentarily to fire them until the engine is cranking or running.

The camshaft position sensor is the one in the distributor. I circled the crankshaft position sensor in the fifth drawing. It's shown in the same place for the single-cam engine and the double-cam engine.
Was this
answer
helpful?
Yes
No
Tuesday, April 12th, 2022 AT 2:50 AM
Tiny
HENRY JOHNSON2
  • MEMBER
  • 4 POSTS
Randy B,

You will not believe what the problem was! The distributor parts listing for this car is incorrect on many of the parts look-ups, Advance, Autozone, Reilly, WAI, Spectra and many Ebay sites. The 1994 Mazda Protege, DX, SOHC, 1.8L with the T2T53171B distributor has a Hall Effect sensor and many of the parts listings, WAI# DST38403, Spectra MZ12 claim to fit my car, but they all have optical sensors and will not work on my car. Advance Auto's lookup shows the 38403 as an exact match to the T2T53171B and they will not make good on their mistake. I think the mix-up may have come from Mazda switching from the T2T53171 to the T2T53171B in midstream.

The cam sensor in my car is the Mitsubishi J815-Hall Effect- and the cam sensor in the wrongly listed units is the J927, an optical sensor. You cannot mix and match these units. I ordered a J815 from China and when I put it in the old T2T53171B distributor, the car runs great and starts up after you turn it off. Apparently, with some old cars, they will start even with the wrong distributor and that fact makes you go off in many different directions because, in your mind, you have just put a brand new distributor on and that can't be the problem- not. Of course, I am stuck with two worthless distributors because nobody will stand up and admit their error.

Thanks for all your information but, as usual, it sometimes turns out to be the simplest thing.

Gerry Connery
Was this
answer
helpful?
Yes
No
Tuesday, May 3rd, 2022 AT 12:00 PM
Tiny
CARADIODOC
  • MECHANIC
  • 33,871 POSTS
Wonderful, dandy news. This will have to be stored in my memory banks. I'm very happy you figured this out.

Please come back to see us with your next problem.
Was this
answer
helpful?
Yes
No
Tuesday, May 3rd, 2022 AT 12:12 PM

Please login or register to post a reply.

Sponsored links