Thoughts on an easy way to test for crank sensor operation

In chasing my hard starting issue (still having a pressure bleed off issue), I did run a variety of tests to verify that the fuel pump was correctly being activated to prime the fuel rail prior to starting AND in reading the '89 FSM, I found that the ECM SHOULD be activating the fuel pump relay WHENEVER it was receiving a "fuel control signal (reference pulse).

A bit of chasing in the FSM discovered that the "fuel control signal" is a pulsed output of the ICM that is basically a divide by 6 counter with its feed directly from the crank shaft sensor. The crank shaft sensor outputs 18 pulses per revolution, so the fuel control signal starts pulsing after ONLY 1/3 of a revolution. The FSM states that the ECM will activate the fuel pump in this manner while cranking or running.

So you might wonder where I am leading on this...well, the "green" fuel test connector located near the air filter housing is directly connected to the fuel pump. Any time the fuel pump is running, there is 12V present on that connector. In other words, you can use that connector to directly POWER the fuel pump OR monitor it to see if the fuel pump is GETTING power.

In normal startup, the ECM runs the fuel pump for 2 seconds, starting immediately after the ignition key is moved to "ON", then it shuts OFF the fuel pump. As you move the ignition to START, the ECM fires ALL injectors ONCE then does NOT fire them again for 6 pulses of the fuel control signal (ie two revolutions), BUT the ECM has the fuel pump running constantly ONCE the first fuel control signal is presented from the ICM.

By inference, if you place a DVM to monitor the "green" fuel test connector, AND allow the fuel rail prime cycle to complete (ie the voltage to drop to ZERO) before you start to crank, you SHOULD see the voltage pop right back to 12V within a fraction of a second as again within 1/3 of a revolution, the crank sensor will output enough pulses to trigger the ICM to output a pulse to the ECM, which will then activate the fuel pump relay.

If this is the behavior you see, then you can pretty well be assured that the crank sensor is working as are the associate sections of the ICM and the ECM, but given that those module also do other things that can prevent the engine from starting, this is really NOT a good test of all their functionality, BUT it IS an easy test of the crank sensor.

This is a lot easier than manually trying to turn over the crank by hand while monitoring the crank sensor directly.

Yes, I know that the oil pressure sender CAN also fire the fuel pump, but it takes a few seconds for the oil pressure to build up sufficiently to trigger the fuel pump and if you really want to isolate this, JUST UNPLUG the oil sender.

As there are so many things that can cause the 3800 engine to NOT start, you have to start somewhere and the further up the chain you start, the quicker you can eliminate a LOT of things. This test clears several major culprits of no-start/hard-start in one easy test (crank sensor, parts of ICM/ECM, and the fuel pump relay).

So I am interested how the forum feels about this as a recommended 1st step in testing the crank sensor?

David T

David,

I won't speak to the veracity of the information but assuming it is correct, that is a very well thought out technical article in it's own right. I love the logic.

My question is: What year models 3800 engines does this apply to and is it only for Reattss? Might it also apply to Other Buicks and Oldsmobiles?

TJ55

David,

My question is: What year models 3800 engines does this apply to and is it only for Reattss? Might it also apply to Other Buicks and Oldsmobiles?

TJ55

I could not venture to guess if all 3800 series engines have the same fuel pump test connector as our Reatta's but given that the engine was widely used and that the connector gave the factory a way to power up the fuel pump and fill the entire fuel line with gas before attempting to start the car at the end of the production line, I would say that there is a good potential that this process could apply to all 3800s, at least those contemporary to the production runs of the Reatta.

I CAN say with confidence that the '89 Riviera is identical to the Reatta as the '89 FSM covers both vehicles and in this specific the documentation lists NO differences.

David T.

So are you saying that if 12V power is applied to this green wire, it will bypass the 18x ring process and should start?

Today I had a disaster with my Reatta dying on me in the middle of traffic.

No, applying +12v to the green wire will ONLY ensure power to the fuel pump. My test was a way to easily see if the 18X/3X ring, CPS, the associated ICM & ECM circuits and the Fuel Pump Relay are functional as the green wire, assuming that the oil pressure sensor is unplugged, will ONLY switch to +12 when cranking if those elements are all working. It is NOT an exhaustive test of the ICM & ECM as those modules have circuitry outside of that needed to activate the fuel pump relay (green wire is simply a pigtail off the Fuel Pump Relay output). If the green wire switches (only takes 1/3 or a rotation of the crank), that pretty much clears the CPS (18X/3X ring & sensor) as suspect, HOWEVER, the CPS, as well as the ICM & ECM COULD BE INTERMITTENT, so the test is really only airtight when the car cranks, but won't start, YET the green wire switches to +12v during cranking.

If the green wire shows to be switching to +12 during cranking (again with the oil pressure sensor unplugged as that sensor has a backup mechanical switched +12V feed to the green wire/fuel pump activated by sufficient oil pressure), then check for sufficient fuel rail pressure (15PSI NOT ADEQUATE, spec is 35-45PSI), also check for spark at one of the plug wires.

The only time in my 25+ years of ownership that I ever had my Reatta die was a CPS failure. The car cranked perfectly that morning, ran perfectly and about 1/2 way to work, it just totally died and refused to start afterwards (cranked just fine).

this really ticks me off if the CPS died again because I replaced it 2 years ago, less than 8k miles, from Autozone. It was a pain to remove the balancer and install this. Im tempted to go replace the coils and ICM box first, because I don't know the history behind them. I was at 1/4 tank of gas yesterday when it just all of a sudden died and I coasted to a side street thinking the gas gauge was overstating how much was left. So I called Hagerty for emergency gas. They gave me three gallons of gas, and it started right up. I figured it was just low on gas. So I whipped a U turn following the tow truck. As soon as I hit the main road it cut out again and died!!! People were very angry that I wasnt moving even though I had hazards on. I coasted halfway off the road into a bus stop.

Multiple cars were not paying attention and almost sideswiped my car! Called the tow tuck driver back and he refused to come back and help even though he was still within my sight! I called Hagerty and complained he abandoned me. They said they would call another tow service and have a cop escort traffic away from my car. Another two hours rolls by, and no police and no tow truck, also no automated text from Hagerty.

By this point, I was on 3% cell battery and my car battery was dead from the flashers. Hagerty's dispatched the SAME damn tow driver who charged me $15 because he claimed I was over my coverage limit. He towed my car home where it now sits in my driveway.

Now I'm estimating either ICM/CPS cooled off enough and started which allowed me to take off just long enough to die again. All with a car with 51k miles.

also check your fuel pressure, regulator, etc. - when i first purchased my Reatta, i replaced my crank balancer, sensor (the rubber on the balancer was cracked bad). i thought i had it solved. a couple days later, the same thing happened (start & die at some point, or no start). turns out the fuel pump was bad.

if ok after that, i'd go with the ICM "goo" check and connection problems there.

that's what happened to me...

also check your fuel pressure, regulator, etc. - when i first purchased my Reatta, i replaced my crank balancer, sensor (the rubber on the balancer was cracked bad). i thought i had it solved. a couple days later, the same thing happened (start & die at some point, or no start). turns out the fuel pump was bad.

if ok after that, i'd go with the ICM "goo" check and connection problems there.

that's what happened to me...

Good suggestions. Where can I buy a fuel pressure gauge to tap into the schraeder valve on top? I'd almost like a permanent one if that's possible.

Harbor Freight?

You can borrow a fuel tester from a FLAPS. One at HF no longer worth what they're asking for it.

If you have access to a logic probe or O Scope, you can definitively test the ICM and CPS. Ronnie's site tells how to make a makeshift logic probe and do some testing.

Sean, I bought mine from Harbor Freight. You're welcome to borrow it if that is easier.

Greg

HF has the one i have on sale here for $19.99 and online (at least for my zip code): http://www.harborfreight.com/catalogsearch/result?q=fuel+pressure+test

and your can usually borrow one from a FLAPS, but it may be something you want to have on hand for more than a day or two. Ronnie has a nice one at TRS, plus any parts you may need: http://thereattastore.com/fuel-pumps/

and if you can borrow ol'yeller's - so much the better!

I've had two vehicles die on me due to fuel pump failures. 1st one was a 80'ish Ford Tempo.

Just cruising along a highway and then.... nothing. Engine just died like someone had just turned off the ignition. Had AAA tow it. to a shop who diagnosed a fuel pump failure.

2nd one was about a year and a half ago. I was in my '97 GMC Sierra. Same thing happened. Cruising along at highway speeds and then... nothing. Attempts at restarting got me nowhere. The tow driver even got under the truck and banged on the tank thinking maybe that would dislodge any debris or get the pump motor going again. No dice. Had the driver tow me 144 miles to my house where I replaced the pump, filter etc. There's a post here somewhere on that topic.

Even though my vehicles never started again after the pump failures, it doesn't mean that you aren't having intermittent issues with your pump. Maybe even have some debris at the pump filter that falls away when the pump stops then gets sucked back on once the pump starts to suck fuel again.

It's possible that the fuel pump relay is causing you problems.

By all means, check the fuel pressure. You can also swap the relays on the firewall and see if the problem moves with the relay.

John F.

Back in the day I used to watch the tach while cranking, if the ignition was good I could see the needle move. Another way is to pour about a 1/2 oz of fuel into the octopus and crank. if it fires briefly it is a fuel problem.

Just had a thought that if you had a starter switch on about a 10 foot cord you could crank with the key in RUN so the dash would be live and could watch the tach.

All that said the only thing I have seen heal/fail like that and take the whole engine out suddenly (like a switch) is the ICM or the ICM connector (IMNSHO the pins are too short).