Cold start problem with Mercruiser 5.0L MPI

[iceman]

I've posted about my engine problem earlier this year with my Mercuiser 5L MPI. That turned out to be a bad fuel pressure regulator. On a test cruise after that was fixed, I ran into another problem with the engine stalling when I tried to start it after three days docked. It stalled while the rpm fluctuated up and down, on a down cycle. I ended up replacing the IAC (idle air controller) and that may have helped but I'm still hearing rpm fluctuation during idle. I also got my hands on an engine scanner, and recorded some some data while I started the engine. I'd like to get some opionions from you guys if it looks "normal"

On the attached plots, I'm showing battery voltage and engine speed. The engine rpm is oscillating up and down like it did when it stalled on me, but in this case (in my driveway) it didn't stall. Is this kind of rpm behavior normal for an engine at idle? Or should it be more constant at 600 rpm?

Also you'll see the battery voltage drop to near 8 volts during startup. That could be normal since the current draw goes up so high during cranking, but could it be an indicator of a weak battery?

Finally, you can see the spark advance in the second plot fluctuates alot during idle, from 2 to 15 degrees. Is this normal?

I welcome any opinions and insight, as I'm trying to learn so I can diagnose this engine myself as much as possible. Marine mechanics are not very accessible where I'm located.

Tom

[ss3964spd]

Fun with engines, right Tom?

In short, non of the fluctuations you are seeing are normal except, perhaps, the voltage drop when you crank the engine for starting. Engine stater motors draw a LOT of power from the battery. What I don't know off the top of my head is how much of a drop is normal.

The graphs are interesting. I can tell you that my 5.7 MPI settles at around 600, cold or hot. IF the RPM's on mine fluctuate they only do so by, maybe, 10 or 20 RPM. Yours should be dead steady as well.

Voltage at idle fluctuations - got me, I've never checked but I suspect it, too, should be fairly steady. The only way I've ever checked is with a multimeter across the + and - battery terminals at idle. You should see somewhere between 13.5 ish to 14.1 ish if I recall - but that assumes a steady idle because output will rise and fall a little with RPM. Of course, Alt voltage output will be lower at lower idle RPM's.

It could possibly be the battery showing a weird load to the alt/regulator (and engine computer) and thereby screwing with both their little brains. Should be simple enough to prove out: either swap in a known "good" battery or take yours and have it load tested.

Spark advance is a little curious. With the engine at idle it really should not be all over the place like that. If you increase the advance at idle the idle RPM will, absolutely, go higher, and a 13* swing is huge. Since the spark advance is completely controlled by the ECM, and the ECM needs relatively "clean" power from the battery, there is a possibility that flakey battery power is screwing with the ECM, casuing the advance to move around, thus causing the idle fluctuations.

What I would do first. Using a multimeter and with the engine idling, place + probe on + bat terminal and touch - probe to different points on the engine - specifically the distributor base if you can reach it. You want to see readings in excess of, say, 13V. If you have less you'll have to determine why; could be a tired battery, bad grounds, tired alt/reg. While yer at it ensure all engine electrical connections/harness connections are tight and clean - ya might even unplug them and plug them back together.

Dan

[iceman]

Hey Dan,
Yes engines are fun, but I'd much rather be running it out on the water than squeezing inside of the engine compartment trying to figure out what might be going wrong! Although I've got to admit I'm fascinated with the data that you can extract from these MPI engines.

Your feedback is kind of what I was thinking, that the rpm, voltage and spark advance fluctuations shouldn't be happening. I have two batteries, and the main one is continually charged with a solar panel, but to charge the other one I have to hook up AC. I can't remember now if I was running on one or both when I took that data sample. Maybe I'll charge them both up and try that test again. I can also do the voltage checks you suggested.

The weather has been terrible for going out on the water anyway lately. The rain is gone but the winds have been high. Right now (10:30AM) it's gusting above 20 knots at my usual launch ramp, and forecasted to get worse!

Tom

[iceman]

Still trying to sort this out. My batteries are Nautilus NC-27s, with a rating of 730 MCA each. My engine spec says I need 950 MCA, so I guess I should always start the engine with both batteries selected.

So I did another test: I topped off the batteries with distilled water, and then charged them last weekend. Today I ran the engine with both batteries and took some more data. The plots are attached. The voltage dropped down to 9V while the engine was cranking. I still see RPM and voltage fluctuation, but then I ran the throttle up and back down to idle. The RPM and voltage seemed much more well-behaved in the latter two cases. And that was after about 2 or 3 minutes of runtime.

Could the batteries be weak enough that they're a drain on the alternator so when the engine is cold there's not enough power to go around? Then when the engine warms up, or maybe the batteries have had a chance to charge a little, things smooth out?

I'm trying to decide whether to try and replace them.....