75 914 Won't Start

[WRX]

I have a perplexing problem with my 75 2.0.

From the beginning:

Last year I bought the car and it ran GREAT! In the fall, I started to have problems with the starter which I tracked to a bad ground. I was not able to fix the strap before I stored the car for winter.

Fast forward>>>>>>> It is now winter and I want to turn the engine over. I jump the ground to the engine block and car cranks, no start. In addition, my el cheapo battery charger/booster smokes hard under the load. I attribute this to the cold temps.

Fast Forward>>>>>> It is now spring and I want to drive!!! I change ground strap, put in new plugs and fuel filter, and crank. No start (grrrrr). I now shift into troubleshooting mode. (I use the Haynes and the Bosch FI book by Probst and a cool d-jet site that I found through reading the Pelican site).

The mode of non start is as follows:
Car cranks and will fire intially, but not start. If I keep cranking, it never really catches. If I pump on gas furiously, I can get it to sputter and backfire thru the carb at about 3000-4500 rpm, but it will not idle and will die if I drop off the pumping of the pedal. Car will start and purr like a kitten for 10-15 seconds with a shot of starting fluid.

I have done the following:

-Replaced fuel filter
-Replaced plastic fuel lines with steel and new rubber (The old lines looked kinked)
-Replaced all vacuum lines (Holy crap is that stuff pricey!)
-Replaced Spark plugs and wires
-Replaced Dizzy cap and rotor
-Cleaned and inspected Trigger points. Continuity/non-continuity check OK. (Don't know if they bounce or not)
-Checked that with key on, I can hear injectors click when the throttle moves (Both sets)
-Fuel pressure checks with car cranking (30 psi). System pressure decays in about 30 seconds, but comes right back when I turn key on.
-Fuel pump operates
-All connectors are connected
-Verified all sensors through wiring harness and they checked out OK. CHT reads 2500 Ohms at my room temp(~60F) Also, I have seen the CHT resistance go down when the engine is warm from my gas pumping starting regime. All other voltages and resistances within spec.
-Checked MPS, good primary and secondary coil resistances, and holds vacuum.
-Checked grounds at back of engine.
-Double checked timing (Although that probably didn't change itself)

I have purchased a new CHT sensor and will install just in case.

I hesitate to do the FI jar test for safety reasons.

I am wondering if it is possible to partially blow up an ECU (As opposed to it just dying).

I am pretty much at a loss at this point, and I have a sneaking suspicion that this is either a pricey fix (New ECU) or I am missing something idiotically simple!

PLEASE HELP!!

Thanks,
WRX
(I didn't know my head could smash THROUGH a 914 hood!)

[Dave at Pelican Parts]

It seems obvious that you are not getting enough fuel. I would try to verify your fuel pressure gauge--I had one that read higher than actual pressure, and it drove me nuts trying to figure out what was going on!! (If the needle bounces more than is reasonable, it may be that the actual pressure is quite low. An experienced friend [Hi, B!] pointed that out to me, and my mechanic determined that the pressure was indeed low.)

Have you made certain that the MPS electrical plug is plugged in and making contact? Ditto the trigger points.

Symptoms like this seem to be caused frequently by something not being plugged in...

And it's always a good idea to verify the valve clearances, the point gap (if you have points), and at least the static timing (somewhere near TDC).

Not sure where to go from there.

--DD

[WRX]

Thanks for the quick reply Dave!

I will re-check the fuel pressure with a calibrated gage, although the pressure readings were rock solid and repeatable at 30psi. You never know.

I thought the same thing about the MPS, but I checked it at the unit and through the harness and got identical readings. Ditto for the trigger contacts and the CHT.

The only thing I did not test was if there is any trigger point overlap due to wear. I will probably replace them just to be sure.

I run the XR 700 system and have checked static timing as well as the optical trigger adjustment.

It keeps gnawing at me that my POS battery booster may have zapped the ECU in some way. Maybe some EE geek can tell this ME geek if it is possible to partially kill the ECU. (I always thought they just died, but I am getting the enrichment circuit to work so who knows.)

I plan on checking voltage at the injector connector while cranking to see if they are getting any juice.

The fuel pump seems to be working, it it possible that something ewlso on the relay board may be kaput?

I guess there is no way to avoid doing the FI jar test at this point. I will also check the valve clearances if for no other reason than to give myself an annoying job to keep my mind off this problem.

Thanks,

WRX

[Dave at Pelican Parts]

I think a relay board problem would show up as either no-gas-at-all, or an intermittent problem.

The ECU can be partly bad, one of my spares has a number of bad circuits in it. One of the more common failure modes is that the FI converts itself to a CIS system--the injectors turn on but never turn off again. There may be failure modes that kill it on the lean side.

You can rent a D-jet tester from Pelican. It mostly tests the ECU--it is not a test for full functionality of the system but is a good ECU tester. You'll have to call for details, though. Or perhaps Bleyseng or one of the other PacNW folks has one you can borrow?

When you do the injector-in-a-jar test, unplug the XR700. That will keep the sparks from happening. And make sure you've got a non-dry-chem extinguisher handy! (See previous thread about the small engine fire.)

--DD

[3D914]

WRX,

Something in your notes above is bugging me:

-Fuel pressure checks with car cranking (30 psi). System pressure decays in about 30 seconds, but comes right back when I turn key on.

Fuel pressure should not drop off that fast. You don't mention what it decays to, so if its just a coupe psi or so - no problemo.
However, if its dropping 10-15 psi then I would think something is not up to par.

Dave - any thoughts on this?

[WRX]

The fuel pressure goes to zero when the key is turned off. In researching this, it points to a bad fuel pump check valve.

My understanding is that this could present a hard satrt problem if the lines to the fuel rail go dry, leading to extended cranking, but I guess I discounted it because I can see that the pressure comes right up and I do have flow. (At least at the cold start valve.) Also no bouncing or "sponginess" in the gage that would indicate trapped air.

I would like to fix this however. I don't think you can buy the check alone, does this mean I need a new pump?

Another thought is unless the pressure regulator closes completely, the pressure would decay by going back to the tank through the return line. So bad regulator? (I would say that it does hold steady pressure)

Lastly, the pump in the 75 is a 2-port design (Supply and pressure only) there is no D port on this pump. Does that make a difference?


Thanks,

WRX

[Dave at Pelican Parts]

There is no R port--the pressure port is the D port.

I don't recall there being a specific check-valve in the pump; I thought the rotors served as the check-valve? If that's the case, you may have a pump problem of some kind, where the pump can deliver pressure but not volume? Hrr... Naah, I think that's a red herring.

OK, get to your FLAPS and buy a "noid light". Plug it into the injector connectors and see if there is a signal going to them at all when the starter is cranking. There definitely should be! The "injectors in a jar" test will also tell you if there is signal going to the injectors, but only if there is fuel pressure and if the injectors are also working. You can use it as a substitute for a noid light... But those are <$10, I think. Or you can pay $35 on evil-bay for a 'fuel injection tester' that is just a noid light.

--DD

[pbanders]

Simple way to test fuel system pressure retention (from Bosch literature). Quote:

Retention of residual fuel system pressure is important in any EFI system. Loss of residual fuel pressure will result in fuel vapourisation. If vapourisation occurs in the fuel supply lines, problems with hot start and idle quality will result. Once the fuel pump has shut down the system pressure will settle. After 20 mins, system pressure should be no lower than 1.0 bar (~15 psig). Loss of residual pressure can be caused by many faults including,

- Leaking Fuel Pump Check Valve
- Leaking Fuel Pressure Regulator
- Leaking Fuel Injectors
- Leaking Cold Start Injector
- External Leakage [hose connections, injector bodies etc]

To isolate the area of leakage, run the fuel pump and then shut it down. Clamp the fuel return line, if the pressure holds, replace the fuel pressure regulator. If the pressure does not hold, clamp the fuel supply line. If the pressure holds, the fuel pump check valve is leaking. If the pressure still does not hold with both the supply and return lines clamped, then the pressure loss is through the injectors and/or cold start injector. As a general observation, rapid loss of residual fuel pressure is generally attributed to leakage through either the fuel pump check valve or
fuel pressure regulator. Fuel injector leakage will normally cause slow loss of residual fuel pressure. Leaking fuel injectors however normally cause the most noticeable hot starting problems as not only does the fuel system need to be purged due to the vapourisation, but also the engine is essentially flooded due to the leakage.

[pbanders]

Oh, as for the non-start condition, do what DD says about getting a "noid light". Does it flash? Then you're getting injection pulses from the ECU. But are you getting fuel spray? Many times, the injectors are so clogged that nothing comes out. Pull the injectors and do the injector-in-a-jar test. No spray? That's your problem.

[WRX]

I think the Pressure regulator is not closing completely when the pump shuts off. I can hear and see gas coming into the fuel tank through the return line.

While this is certainly a problem, It will probably result in hard hot starting (I only wish I could get that far!)

Sorry Dave, port D must be "druck" which is german for pressure?

[Dave at Pelican Parts]

Yup, that's the D port on the early pump. Some literature lists it as "damper", but there ain't no damper in a stock 914 fuel system...

--DD

[3D914]

WRX,

On the "noid light" at the local flaps - I may be the only one experiencing this, but most of those noid lights are set up as 12V lights, and since the injectors are only getting around 3.5V - it can be difficult to see it light. I found I had to close the garage door, etc, and eliminate very bright light just to see the noid light come on.

Just my $0.02

[pbanders]

Gerard, did you get a "Bosch FI systems" noid light? The one I have flashes decently, got it from AutoZone.

[WRX]

OK, the noid light shows..............Flashing!

I will rip it apart tomorrow to do injector testing.

Some quick questions.

- I can't believe that all 4 injectors would be so clogged all at once as to stop engine from running.(Remember, the car ran fine last fall) However, I did not use stabil in the system (The only vehicle I didn't prepare properly!) before winter. Could all the injectors clog fom just sitting?
- If I find no spray, or little spray, how do I clean them? In my search fro a noid light I came across a Tool called an EFI pulser which appeared to be a voltage signal generator with an adjustable pulse duration (4ms, 10ms, continuous). Would this tool allow me to clean the injectors outside the car in a safer place?
-Even though I have injector pulses, they could still be of the wrong pulse duration if the CHT or other sensor is off, Right?

Anyway, I am encouraged that the injectors are getting a signal because it is looking a lot less like a bad ECU.

Thanks,

WRX

[Dave at Pelican Parts]

It's unlikely but possible for the injectors to clog from just sitting. Fuel will eventually have its more volatile fractions evaporate, turning it into sticky nasty varnish. But it's unusual for this to happen in just a few months.

You can take them to an injector cleaning service and pay some $$ to have them cleaned. Probably the safest way if you're worried about gasoline. If not, you can hook them up to ~3.5V and run carb cleaner to the fuel hose under some pressure. That will hopefully work its way through. Alternately, you could hook them up "backwards" with carb cleaner under pressure at the nozzle end. This can help "back-flush" the crud out of the inlet if there is any there.

Yes, the duration of the pulses could be wrong enough to cause a non-start. But at least you now know the injection is trying to open the injectors. To me, the next step would be the injector-in-a-jar test. All four injectors, two large jars (or four smaller ones), a disconnected optical ignition gizmo, a fire extinguisher, and a friend to work the starter while you look at the injectors for spray pattern.

--DD

[WRX]

I will do the test.

I always get that Final Destination vibe when surrounded by gasoline fumes.

When looking at the rubber fuel hose last week, the old crap sprang a leak and I was standing in about 2 gallons of fuel before I could get the leak stopped! (All kinds of cool end of life visions on that one!!)

Should know something by tomorrow.

WRX

[silver2.0]

I don't mean to kick a dead horse, but did you check your valves as Dave suggested? Just a thought, but when I dropped a couple seats in my heads, I lost compression on 2 cylinders and it barely ran. I could kind of start it, but it ran terrible. Just a thought. You could just as easily check this with a compression test, and that would at least tell you if they were too tight, and you were loosing compression. Sounds like you are very with it, and you will figure it out for sure. Good luck!

[WRX]

OK,

I took off the injectors, doused them with carb cleaner, ran compressed air thrpugh, was able to actually spray cleaner through the injectors.

Hooked up to car, injectors in jar on driver side. Floor gas and cranked for 5-10 seconds. Got about 5 ml in the container (Both injectors) Visual was a very light spray with good pattern.

Is this what I should be seeing??

BTW, the injectors didn't have any visible gunk in them.

I am afraid to put this back together until I can verify that there still isn't a problem.

Thanks,

WRX

[Dave at Pelican Parts]

That's pretty much what you should be seeing. You don't have to floor the throttle, though--that will make all (half?) the injectors squirt out "acceleration" pulses which pretty much are extra injections of fuel that happen when the throttle opens.

Anyway, sounds like you've got fuel going in.

--DD

[pbanders]

OK, you've got fuel. Put it back together, and if you have compression and spark (at the right time) it will start and run.