Miata Cam Angle Sensor: Fuel but no spark?

And now, the next installment of the ongoing saga of "How can this Miata berkeley me over?" When we last left our intrepid hero, he had gotten oil pressure back, he now had  the car running and idling and the wideband sensor reading properly.

Fast forward a week and I try to start it to go shake it down and start street tuning it. Except now it won't start. Cranks properly, and I can smell fuel, but it doesn't pop or attempt to start.

My diagnostics so far:

-Figured at first that maybe I didn't have enough timing in it for cold starts but adding some timing had no effect.

-Installed an inline spark tester (actually 2 different ones, to verify one was good) and have no spark. Sparkplugs are wet with fuel, leading me to believe I have injector pulse.

-CAS connector had some cracking in the insulation where the wires enter the weather seal, so grafted on a new CAS connector from a junkyard car, thinking it was maybe shorting/grounding out/high resistance. No effect. (Car was slammed in front end and engine is jammed against firewall, so cannot grab CAS from it)

-Ohmed out the coil pack, all readings nominal.

-Battery power present at coil pack connector with key on

-Battery power and ground present at CAS connector.

-Backprobing the Cam Position Signal (which controls spark) on the CAS results in no voltage while cranking with it plugged in.

So, am I correct in believing that my CAS has gone and died on me? From what I read, this controls both fuel (Crank Position Signal) and spark (Cam Position Signal). Is it possible for one signal to die from this thing without losing the other? I know that they are prone to calling it quits without any warning, and as far as I can tell, this is the original 28 year old, 140k mile sensor from the car but I just don't want to shotgun a $350 part at the car without being extremely certain its the issue.

Note: CAS is installed on 2000 engine in a 1990 chassis with MegaSquirt. 

Thank you all.

No.  If your crank angle sensor died, the ECU would not know the engine is turning, so it would not fire the injectors. You need to find something between the ECU and the coil driver.

Hmmm, the plot thickens. Checked my crank position signal voltage on the CAS and had 5V while cranking. Double-checked the cam position signal at the CAS while cranking and now had 5V, when previously I had none. Hooked everything back up and hit the key and the car tried to start, although being loaded up with fuel resulted in a nice backfire.

So, got spark back and it just refused to run, not popping or sputtering or anything. Which really baffles me, because the car ran and started great last weekend, I touched nothing on the car and now it refuses to run. Put it back to base map, thinking maybe I screwed something up, and start turning the CAS with a timing light on it and got it running again, figuring I would have to reset the idle and all that.

Now, TunerStudio keeps going offline and it keeps losing sync. I'm utterly lost and confused, because it ran fine a week ago with no issue and now it's acting all sorts of berkeleyed up. Getting ready just to trailer it to a shop and have them sort it out

Sure sounds like an ECU issue... wonder if you fried the coil drivers some how?  If you still have the stock ECU, plug that back in and see if it will run the car. 

In reply to ShinnyGroove :

Not really an option, because there is no way a stock ECU will run this thing with 700cc injectors, and I deleted the AFM

In reply to NickD :

Where did you get the MS?  If DIYAutoTune, call them up. Their tech support is pretty good.   I’m sure there’s a diagnostic check for coil drivers, but I wouldnt know how to do it. 

Are you running all the 90 sensors on the 2000 motor?  Still might be worth puttinf stock injectors and AFM back on to try to run stock ECU. 

Also, good luck finding a shop that can sort out your Megasquirt issues. 

So, I continue to be baffled. Wanted to check and see if maybe it was a bad cable, as DIYAutoTune lists that as a possible cause for the issue. Went out and tried to start the car and it cranked and tried to start once. Then went back to having no spark again. I then noticed that the power LEDs on my MegaSquirt are lit when the key is in the Off and Accessory position, then when you turn the key to run or crank, the power LEDs go out and you lose communication via TunerStudio. What is going on here?

I was thinking that that could have something to do with the confusion you had when tapping into the ignition wiring, but that was for the gauges right? So that *shouldn't* affect anything. 

Have you been able to check power and ground feeds yet?

One other thing to check if this is an MS2 or MS3 based unit - this has an output test mode where you can command the coils to fire with the key on and engine off. See if you can get spark in output test mode.

In reply to NickD :

Does the MS get unswitched power (and thus the power LED stays on even with the ignition turned off)?

How good is your battery? Could it be that cranking the engine drops the voltage enough for the MS to go "thanks but no thanks"?

BoxheadTim said:

In reply to NickD :

Does the MS get unswitched power (and thus the power LED stays on even with the ignition turned off)?

How good is your battery? Could it be that cranking the engine drops the voltage enough for the MS to go "thanks but no thanks"?

Battery check is a good call. I can't tell you how much I kicked myself for the hours of work I did on my old motorcycle only to find out the battery didn't have enough juice to keep it running. 

One thing to know - if you don't have a good signal from the CAS, the ECU will still fire off a prime squirt. This can wet the plugs even if the injectors aren't firing as part of trying to start. So "plugs are wet" isn't a definitive test of CAS health.

If I were in this situation, I'd be talking to the MS. If you're not sure if the CAS signal is getting through, then ask the MS.

I also found that once plugs got wet, the engine started getting harder to start. I've been known to dry/heat up the plugs with a propane torch to work around that...

Okay, so saying that the MS is powered when they key is off and off when the key on was a bit of an oversimplification, as I learned last night. While trying to do more diagnostics (thinking it was maybe a bad ignition switch) I was able to see what kind of weirdness it's doing. Let me try and right this in a manner that makes sense.

Install the EFI relay or hook the battery up or plug in the MegaSquirt and the MS is powered down with the key off and in Accessory. Rotate the key to Run and the LEDs light up in the MS, although a little on the dim side, but when you turn the key to Run the fuel pump does not prime. Turn it to crank and it will crank and sputter once. Then  when you let off the key, the LED is now very dim on the MS in Run and Crank. Turn the key back to Accessory and the LEDs go bright and the fuel pump will prime briefly. Turn the key to Off, LEDs are still bright. Unplugging the ignition switch does not turn the MS off, making me think it isn't a berkeleyed switch. Cycling the key results in a still-bright LED when you turn to Accessory but no fuel pump prime. Hit Run and Crank, no fuel pump, dim LEDs. If you unhook the battery, remove the EFI relay or unplug the MegaSquirt, it will act normal and then go back to this weirdness as soon as you crank it.

What baffles me is threefold: Why does the fuel pump only prime when I go from Run to Accesory? Why does everything else on the car function normally (Radio and gauges still turn on when they are supposed to) on the key cycle? And what changed (as stated, everything was fine a week ago, it sits untouched for a week and everything goes to E36 M3.) to make it so wonky?

Battery voltage is not an issue, because we have a brand-new, much larger battery hooked to the cables for right now, and it still does it. Also did it when jump starting the car from a running Subaru. Don't think it's related to my original gauge installation, in which I accidentally tapped into a power wire for the door buzzer, resulting in the gauges being on whenever the doors were closed and off whenever the door was open (Unhooked it from there, reconnected the wires properly, and attached the gauges to the power supply for the cigarette lighter). I thought maybe it was an ignition switch, but like I said, everything else acts normal, and unhooking it does not power down the MS. Also, my father suggested maybe my used alternator had a shorted voltage regulator back-feeding, as he's seen that before, but unplugging the alternator had no effect.

Keith Tanner said:

One thing to know - if you don't have a good signal from the CAS, the ECU will still fire off a prime squirt. This can wet the plugs even if the injectors aren't firing as part of trying to start. So "plugs are wet" isn't a definitive test of CAS health.

If I were in this situation, I'd be talking to the MS. If you're not sure if the CAS signal is getting through, then ask the MS.

 

I'm pretty sure that the CAS is out of the picture now. I don't think it was ever the issue, I just used it as a scapegoat, honestly. It has a good connector on it now, it is producing CKP and CKP signal and has power and ground. 

Wow, this is one upset little car.

Two rules for dealing with Massive Electrical Weirdness.

1. What did you just do?

2. Check your grounds.

One thing that can go bad while sitting ("car ran fine, then it wouldn't start") is the main relay. This doesn't sound like it, though. Dim lights on the MS mean it's not getting full voltage, which really looks like a grounding problem.

NickD said:

Keith Tanner said:

One thing to know - if you don't have a good signal from the CAS, the ECU will still fire off a prime squirt. This can wet the plugs even if the injectors aren't firing as part of trying to start. So "plugs are wet" isn't a definitive test of CAS health.

If I were in this situation, I'd be talking to the MS. If you're not sure if the CAS signal is getting through, then ask the MS.

 

I'm pretty sure that the CAS is out of the picture now. I don't think it was ever the issue, I just used it ias a scapegoat, honestly. It has a good connector on it now, it is producing CKP and CKP signal and has power and ground.

Maybe, but someday someone's going to come across this thread while googling their own problem and maybe it'll help them.

Keith Tanner said:

Wow, this is one upset little car.

Two rules for dealing with Massive Electrical Weirdness.

1. What did you just do?

2. Check your grounds.

One thing that can go bad while sitting ("car ran fine, then it wouldn't start") is the main relay. This doesn't sound like it, though. Dim lights on the MS mean it's not getting full voltage, which really looks like a grounding problem.

It's also one upset big owner.

What did I do is easy: Nothing. Well, I put a new fan thermoswitch in it (broke the old one off by accident when pulling the engine out after oil pressure debacle) but I highly doubt that's the cause. Just to be sure, I even unplugged it. I put the new CAS connector on as a reflex when it wouldn't start, because I knew the insulation was cracked down by the base of the connector and thought maybe something was open/grounded/shorted.

Where are the grounds on a 1990? I know there is the one on the PPF. One that goes from the bellhousing shim to the body by the master cylinder. Two small ones that I have grounded to the intake manifold from the injector harness.

So you were working on the car last week. Did you lean on something? Disconnect something and forget about it?

There are grounds all over the place. PPF, brake master, a bunch under the dash. There's a diagram in the factory manual.

Waaaaaaaaaaaaaaaaiiiiiiiiiiiit a second, I just had a lightbulb moment!

I'm running MiataRoadster's phenolic intake manifold spacer, which puts a 1/2" phenolic spacer between the manifold and the head and phenolic washers under the intake manifold nuts. So my grounds that I have on the intake are likely not grounding, correct? Last week when it was running, we had the fan thermoswitch ground to the cylinder head via a wire due to the broken sensor. Is it possible that that jumper was thus serving as a ground for other things and now that I've hooked the thermoswitch up normally, they aren't grounding properly? I think I either need to jump that thermoswitch connector back to the head or relocate those grounds off the intake manifold.

In reply to NickD :

Just ground the intake manifold using a jumper cable and see if that improves matters? I don't think there is such a thing as too many ground connections, but then again the intake should be at least somewhat grounded via the mounting studs and the brace if you still have the brace on it.

BoxheadTim said:

In reply to NickD :

Just ground the intake manifold using a jumper cable and see if that improves matters? I don't think there is such a thing as too many ground connections, but then again the intake should be at least somewhat grounded via the mounting studs and the brace if you still have the brace on it.

No brace, I have way too much going on in the engine bay already, so I 86'd that johnnie for more room. And the mounting nuts don't contact the manifold, because there is a phenolic washer under each of them as well. 

In that case I would definitely put in a temporary ground, see if it helps and if it does, fit a less temporary ground connection.

In reply to BoxheadTim :

That'll be the first thing I try when I go home tonight. I just hooked it to the manifold because my BP-5A manifold had some unused bolt holes various things that were no longer relevant. Didn't even think of the phenolic spacer. I'll just move them over to the cylinder head.