Cloud9...68

I'm trying to fire my engine up nearly two years after I started my project to convert it to a mostly-track car, but it's won't fire. I first found that the DME relay was dead (as was my spare), because the fuel pump doesn't turn on when I turn the key to the start position, so I jumpered the DME connectors in the fuse box.

But it still wouldn't fire, so I checked for a spark at one of the plugs, and nothing. So I proceeded to do a blink test, and while the CEL illuminates when I turn the key to the start position, the CEL never starts blinking when I depress the throttle to the floor, no matter how long I hold it there. So I checked the voltage going into the DME by disconnecting the DME's harness, and checking the voltage between pins 18 and 24 on the connector with the key in the start position. I get +12V (positive lead of my VOM on pin 18, negative lead on pin 24). So, I'm getting power to the DME, but nothing seems to be coming out.

I know that a bad or improperly installed reference sensor is a common cause of a no-spark issue, but the fact that the blink test doesn't even work has me fearing it's something worse, like a dead DME. I never touched the DME while I had it disconnected, nor did I do any of the things the manual warns about (cranking the engine with the engine ground disconnected, reversing the polarity of the battery cables, etc.), so I don't understand why just sitting for nearly two years would cause it to die, so if anyone has any other ideas of things to check, of of any way to test the DME, I'd really appreciate it. Thanks.

kwikt

Cloud, have you checked or looked into your theft alarm system maybe. Just an idea.

Cloud9...68

Not yet, but it's on my list of things to check. Do you have any direct experience with a problem similar to mine? From what I've read, there are relatively few things that can cause a no-spark condition in these cars:

1. A dead DME
2. A dead coil
3. Broken rotor
4. A problem with the speed/reference sensor
5. A problem with the anti-theft system.

There's a nice description in Clark's garage showing how to bypass the anti-theft system while fooling the car into thinking it's still active, which I'll try after checking to make sure I've installed the speed/reference sensor correctly. But to be honest, of the items in the list (and please anybody who knows of others, add them), the only one I touched that might explain both the no-spark and the lack of a blinking CEL is the speed/reference sensor. So, I'm going to start with that, while waiting for the spare DME somebody has mailed me.

MB968

There should be a resistence check you can make on the speed sensor. There should be something in one of the manuals about this test. Sometimes the Hayes is easier, just because it has a lot less in it. This type sensor, if memory serves, is just a coil of wire around a magnet. If the coil of wire malfunctions (shorts, breaks, etc.), the resistence will be way off. So it is easy to tell if they are OK.

Cloud9...68

MB,

I'll have to check on that. But it's very straightforward to test whether the speed/reference sender is delivering a signal to the DME: I just have to see if I'm getting any voltage between pins 48 and 49 on the DME connector while cranking the engine. I'm planning on checking this tomorrow.

kwikt

I thought reference and Hal sensors can only be checked with an oscilloscope. Did you happen to disconnect and remove your DME unit and reconnect at the passenger foot board? Don't forget the couple relay and fuses on the driver side leg area. Not sure what they are though. Check all grounds also. Sounds like you are missing a ground some where. I think there is a diagram were all these points are. If you were closer Cloud, I have a spare coil, hal sensor, DME relays, reference sensor with the shim glued to it, throttle body sensor etc. you can try. Adjusted mine a few times already when swapping flywheels.

Cloud9...68

Kwikt,

On the more general 924/944/968 forum, someone pointed me to a thread where a method to check the speed/reference sensor using a diode of the appropriate voltage rating (I think it would be 3V) was described. But I thought a simple multimeter could be used to just check if it is giving a signal, although you can't get an accurate voltage reading this way - it just tells you if it's alive.

I removed the engine in April of 2010 to rebuild it, so yes, I had to unplug the DME from its connector. I didn't know about the relay and fuses on the drivers side, though. I'll take a look. As far as grounds, I kind of doubt that's the problem, because I only removed the ground cable that goes to the back of the engine, and was very careful to re-attach it correctly. Plus, I replaced my power cables with the Higher Connections kit, including the auxiliary ground cable, so I'm pretty sure I'm well grounded.

I also don't think it's stuff like the coil, hall sensor, etc., based on the fact that the CEL isn't blinking when I try to do the blink test, which tells me I'm not getting any output from my DME. Also, I don't understand why those items would have gone bad from just sitting for nearly two years (although I guess the same thinking goes for the DME as well). Somebody is sending me his spare DME, which I should get on Monday. While I'm waiting for it to arrive, I plan to use the weekend to check whether the speed/reference sensor is sending a signal, I'll bypass the alarm, and I might as well confirm whether I'm getting power to the coil (need to read up on how to do that...), and whether the rotor is in correctly.

Also, I've discovered that my DME relay isn't bad after all. The reason my fuel pump isn't turning on when I turn the key to the start position is that I'm not getting any voltage across pins 85b and 87 on the fuse box where the DME relay plugs in. These are the terminals that energize the coil in the relay to close the switch for the fuel pump's circuit when the key is in the start position. So I'm going to have to trace back why I'm not getting any voltage across these pins, although I don't see how this has any connection to my no-spark problem. So, I'm getting neither fuel, nor spark. Other than that, everything's great (other than that, Mrs. Lincoln, how was the play...).

Yes, I wish you were closer, too. The bottom line is that this has been a huge undertaking, and I'm going to have to retrace a lot of my steps to track down these issues. I'm hoping to find something that explains the multiple issues I'm having, although I realize it's also possible that they each have separate causes.

kwikt

I can't remember but isn't there a starter relay also in the fuse compartment? I do not have my 968 with me at the moment so I can't help you by looking at my car. . My son has it at FSU in Tallahassee. This is rotor at position at TDC.

Cloud9...68

The problem isn't with the starter - it cranks just fine. There just isn't any spark (or any fuel). It isn't inconceivable that I installed the rotor incorrectly, though, so that's on my list of things to check.

MB968

Cloud, you may already have ruled out the crank position sensor. But, just in case I ran the same kind of check described in my Mazda diagnosics manual on a used 968 crank positon sensor (my 968 engine ran with this used one until I took it off due to a cracked connector). If you check the resistence between the center pin on the sensor (yellow wire on my sensor) and the two outside pins, one should give on the order of 980 ohms (a black wire on my sensor). My Mazda diagnostics manual allows about a range of 10% on the resistence reading (the sesnor is different, so for the Mazda the acceptable range of resistence is 520-580 ohms. If you check resistence to the other outside pin, it should show no continuity (ie extremely high or infinite resistence, no reading). This pin is the sensor ground.

And, I like the idea of connecting LEDs to check the sensor at the ECU. My TecGT has this feature built in, and at a glance lets you know when you crank the engine, that the ECU is getting the signal it needs to start the engine. If you get this working for the DME, you'll have to let us know how it goes. It may be a feature that I'd like to install for later diagnostics.

User 41221

PSA... be careful when using Clark's alarm bypass info. If its not specifically for the 968, you run the risk of burning up your wiring harness. The earlier versions of the 944/951 use different wiring than the later versions, and I can't recall if the 968's changed after that.

Cloud9...68

Yes, I discovered that. But it seems to me that the fact that I'm getting +12V at socket 86 of the DME relay connector in the fuse box, along with +12V at socket 18 of the DME connector whenever the battery is connected, and +12V at socket 27 of the DME connector when the key is in the start position, seems to argue against the alarm module shutting off power to the DME. In other words, since socket 86 sends current to the coil in the DME relay which closes the circuit that sends juice to the DME, I would think that the alarm immobilizer would shut off current flow to socket 86.

I seem to be getting power into the DME - the problem is that nothing is coming out, as evidenced by the fact that the CEL doesn't flash on and off when I try to do a blink test, and I'm not getting any spark.

The other weird thing I'm seeing is that my tach needle doesn't bounce at all when I crank the engine, which seems to point to the speed/reference sensor, but I was very careful about gaping it when I had the engine out, so I'm assuming it's OK. And even if it were bad, I don't see how this could cause the blink test to not work. That's what the blink test is for, after all - to identify problems with sensors and components, like the speed/reference sensor.

Any ideas on things to check for that might explain why my blink test isn't working would be very welcomed. Thanks.