454 mAG efi TWINS-ecm OR efi PROBLEM??
 

A real world brain drainer here. Twins, '97 454 Mag EFI's, .030 over, rebuilt, no hours. One engine is great but the other is like a alien from Mars!! idle speed is ok for both, but the port engine wanders, doesn't respond to throttle movement just off idle. Runs approx 400 rpm slower for any given throttle position. When throttle is advanced 1` click- no change in rpm... 2nd click the engine may accelerate to 1600 rpm & then slow down almost to idle again. EGT is higher on the port engine & reaches dangerous levels @ 3000 rpm even under no load. (see chart).
All readings taken with a Mercruiser scan tool & duplicated on both engines. No codes are ever recorded other than the "checksum" number which means nothing to me......
Computers have been returned to Arizona Speed for checking as there seem to be timing issues on the port engine also. Even though the timing says it is at 28* advanced per a timing light on the crank, the scan tool says only 18-20.1* advanced..... (starboard says 28*) but 3000 rpm timing is a matched 40.1*.
If the distributor is turned ever so slightly at 2000 rpm the timing goes straight to 40* or more!!

HELP!! EFI Wizard needed here ( many of these same indications were there before the rebuild also). Distributor gears are new & shafts tight. ?????????

we try to attach again...

have you checked the fuel pressure? Sounds like fuel starvation.

On the timing; did you put the engine into service mode when setting the initial timing to 8 degrees? The scantool and the ECM assumes 8 degrees of initial timing when displaying the ignition advance. If the initial advance was something else the ECM and scantool will show a false value.

fuel pressure is 45# at idle, increasing to 55# at full boost. Injectors are clean, two fuel filters (before & after Aeromotive A2000 pumps).
Yes, base timing was set in "service mode". The base timing is holding as is the 28* if read on the crank, but the scan tool shows 18-10.1* & varying at 2000 rpm which it should not do. The second engine is steady at 28* indicated on the scan tool.
????????

Tried swapping the ECMs between the engines?

I sent the ecm's back to Arizona Speed to check them out. No I did not swap them.... yet!! That's next..after the IAC's are tested and reset.

I don´t know what the ignition curve is supposed to look like on a blown engine but 40 degrees advance at 3000 rpm and above is definitely too much on the standard MPI engines. They usually are around 30 degrees in the 3000-5000 range. Also ,in 2000 rpm area, if the ECM thinks the advance is something different than it really is (18-10 instead of 28) it may do all kinds of funny things in trying to find the optimum advance. It may explain the hunting .

Another thought; If the ECM thinks the advance is something different than it really is in one part of the rpm range but you have a match in another part , the SW must be screwed up.

all good ideas to do first, you might also hook your engine up to a vacuum gauge, compare to two engines, this will let you know if you have any cam- are valve related problems, should be smooth and steady, if the needle is bouncing something is up, also spray carb cleaner are use propane to check for vac. leaks, around the intake, lower and upper, injectors, and throttle body, might also take a look at the spark plugs, I know you dont have any hours on it but its worth a look, let us know, also check your knock sensor's for damage
Mark

:eek:
I guess I should have added that this engine did exactly the same things before the rebuild, I just did not have a scan tool at the time AND I was dumb enough to think it was normal as my other engine had a dead cylinder in it.... I was trying to "limp" thru the summer/fall last year with a "semi-working" boat.....
then the cold came, did the rebuilds & didn't start them until I got the cast off my leg in the middle of May.... it's hard to launch a boat with a cast on!!
Since the problem was there before, it seems unlikely to be a vacuum leak... but that is an idea to look into.

I will first reset the venturi plates to identical openings and reset the IAC sensor to the same initial voltage as the good engine.... therefore starting a "new" baseline of reference after the computers are verified by Arizona Speed (they got them this afternoon). I am learning, but it is hard to find "marine" mechanics that REALLY know what they are doing around here.....
so I am looking for the OSO expert help desk!!:D
Good ideas, but keep it coming.... especially about the timing variation @2000 rpm & the EGT trouble..... see the "bad" engine consumption is higher, so it is not lean & should be actually cooler EGT not hotter!!

Advancing the ignition too much should overheat the engine. 40 degrees sounds way out to me.What does the starboard engine have in the 3000-5000 range? The ECM SW is a prime suspect. unfortunately I can´t read your scan tool data but I assume they show reasonable and similar values between the engines.

I continue to think out loud here. IF the ECM checks out OK then the next step would be to check all the sensors. On the scan data you can see if the TP, MAP, IAT, IAC and ECT deliver reasonable values(and similar in between the engines).Further, the knock sensor is especially important since it has a key role in limiting the advance from getting too high. In order to check that one you have to check the input reference from the ECM to the sensor with a DVOM and, if that checks out, swap the sensors between the engines and see if the advance curve becomes normal on the port engine and/or the problem moves over to the starboard engine.

On marine mechanics, I guess it is almost as tough in Oklahoma as it is here in Stockholm, Sweden. Fortunately there are good books .......

I guess I need to find a Merc manual for all the test/calibration proceedures....???? where??

Here for example; www.motocom.com/marine/6
I think you can buy them at West Marine also.

BTW, Amazon.com is cheaper. They have the Clymer Merc 95-97 manual

Thanks Mat.....

I really thought there would be more "input" from others also....
so I guess EFI really is a strange bird to most everyone!

Disconnect the lead to the knock sensor, the ECM will go to a preset value.

Timing is way too high. I'm running 28 - 30 degrees.

The following are links to what I went through and tried, perhaps some or hopefully one idea will help you, even though we had disimilar problems. Most of this is ECM troubleshooting:

http://www.offshoreonly.com/forums/s...threadid=48133


http://offshoreonly.com/forums/showt...threadid=47022

I'll check back often and see what i can do to help!!

Good Luck!

Kirk

I agree the EFI is tricky to troubleshoot. However, having two identical engines and swapping things between them in a structured way and observe the result is basically all you need to figure out what is wrong.

What's happened to dennis moore??? I fully expected him to weigh in on this problem????????
Dennis?????

Blown- By the time you learn about this boat & all the things you learned from the last one you are going to be one hell of a mechanic.

More people end up in "the house" so to speak with Jack Nickleson (remember the movie???) than become good mechanics!! Unfortunately, I am now having to go back to work full time & 100 miles away from the boat!! Only on weekends to fight this thing & I still want to play some (the wife insists on it!).

Pantera- My timing IS set to 28* on both engines, just one of them indicates on the scan tool that the spark advance is low at 19-21* when the crank is showing 28*, Then a 3000 rpm, both engines are now at 40*+ !! Az Spd has my ECM's & I have been trying to reach Shoffner all week with no luck..... I gotta get a life!!;)

Blown Formula, when you say that " the timing is set to 28 "degrees and "the base timing is holding 28 degrees" what do you actually mean? Please explain how you set the timing and where the 28 degrees came from. Do you first set the timing in service mode to some 8 degrees and then read 28 degrees at the crank at 2000 rpm being out of service mode?

The base (service mode) timing is set to 8*. The 28* is @2000rpm in normal mode read at the crank with a adjustable timing light. Then I check "spark advance" display in the scan tool & see one engine @ 28* @2000 rpm & one showing 19-21* same rpm. Check the crank & it is still at 28*???? Since the tool is reading the ECM, is the ECM doing something screwy??? Both read 40.1* @3000 rpm spark advance on the scan tool. There is no distributor gear slop as both gears are new. Seems like way to much advance at 40* to me.... still waiting for my call back from Az Speed.... I have a Poker Run next week & I need to solve this!!

Pantera- did you ever solve your problem??? What was it?

The scan tool and ECM have no idea where your base timing is. They can only reference the timing provided by the ECM, so moving the distributor should not change the value displayed but it sure as heck will change the actual timing as shown on the balancer. I don't trust adjustable timing lights, you really need to degree the balancer from 0 up to 40 or so. In a pinch you can put a reference mark at 2.5" from 0, this represents 36 degrees on an 8" balancer (actually it should be 2.513 but you cant measure that close) of course 1.25 will be 18 degrees. This is a good reality check.

As far as manuals, I promise the factory manual is better. Any Merc dealer can get them for you, we do normally keep them in stock.

I know this might sound crazy but two brain burners that I finally figured out.I had some prochargers on a 312 and had the exact same problem. Put a vacum guage on both motors and compare, one of my motors had a vacum leak, and it caused the ecm to search all over for a known idle quality,it was moving the timing air motor and it never performed as good as the other,and never left a code until I found the leak a upper intake gasket was pushed out ,after I fixed that ,they both ran perfect.Also for your guy's info I ran into a friends boat that was blowing huge amounts of oil out the breathers dip stick tube ect we thought it was detonating, it turned out the intake was supercharging the inside of the block by way of the egr valve and valve covers

MBAM/ Marc- I will get some degree tape Monday.... and check that. I agree the ECM doesn't know where it starts at but with base values the same (same service mode base timing) then they should both advance to close to the same place according to the scan tool. One advances to 28* as it should- the other goes to 19-20.1* per the scan tool.... so does that mean the ECM is screwing up at 2000RPM? The ECM also does not read vacuum so that should not effect that but may other things (run up/down, wild, etc...).... just need ONE firm answer!! :D

At least this problem is not a 1* taper in a valve guide problem like Pantera24 was.... what a nightmare!! Glad he finally found it, but it had to break before they located the trouble.....

thanks guys.... and Iwill put a vacuum gauge on it too.... although my Boost/vacuum gauges do agree with each other.

The engine does read vacuum/boost through the MAP sensor. can be checked with the scantool if it delivers reasonable and consistent values between the engines.
Another idea; crossfire or some screw up with the sparkplug wires?

I understand the MEFI 1 does not read boost or vacuum but the MEFI #3 does...... wiring was checked.....
onward and upward!!

On the standard MPIs the ECM reads manifold pressure also with the MEFI1 SW. If the ECM does not read manifold pressure it is probably more sensitive to leaks. I think you really need Arizona speed & marine to help out in the trouble shooting.

Blown Formula. I can now read your attached chart with data. A few comments;
TPS; all values look normal
MAP; all values look normal. Do I understand you correctly that the MAP input is not used by the ECM in this set-up?
IAC POS, etc; Obviously different between the engines. Is it trying to compensate for something?
Base timing @crank; 28 already at idle?! not normal on a standard MPI engine. usually fluctuating betwenn -10 and 10. However, Az S&M may have done something. Stabilizing around 28 @ 2000 rpm is OK. Did you ever check timing at crank at@ 3000?
Spark adv per scantool; 18-20 on bad engine at idle. That is high.I dont understand how it can be different from what you measure at the crank!?
I read in the manual that if the ECM fails , the ignition module (in the distributor) will take control of the timing and advance it to 27 degrees. Is that what is happening????
I think you need Az S&M to comment on all this.

I think you misunderstood- Scan Tool spark advance
Base timing is 8-10*
2000 rpm #1=19-20.1* #2= 28*
3000 rpm #1= 40.1* #2= 40.1*
Used a timing light on #1 @2000 rpm & see 28* on crank???? not 20.1*
The MAP is used and read but cannot adjust fuel based upon BOOST. The ECM doesn't allow it, only the MEFI-3 does.

Yes, I think the IAC is maladjusted & will adjust it asap.

Per Jim Shoffner-Az speed. The merc ecm's MEFI-1 is pretty sloopy in regard to timing. If there is an rpm difference at around 2000 rpm, the ecm has "slots" to set timing and may indeed allow that variation. He also wanted me to take some MAP voltage readings at speed and report to him. He also believes the IAC & TPS as well as the throttle blades need to be readjusted. He is sending me the values with my ECM's. He did make a couple of adjustments on them to help out, but with the new MAP readings, we will have to redo them again.

Just to ensure that we are on the same page here;
The scantool based advance values (19-20.1 and 28.5(I think ,very small text ))in your table pertains to 2000 rpm not idle as it shows in the table. Further you have 40+ degrees on the scantool at 3000 rpm. Do you have any scantool values for idle?

Then going to the values measured at the crank, they show 28 for both engines, both at idle and at 2000 rpm. No values at 3000 rpm. do you have any values at 3000 rpm?

Ok now....
@ idle the spark advance reading oscillates so rapidly both + & - that you cannot get a value there on the scan tool in normal mode. In service mode @ idle it is 8-10*... this is normal per Az Speed.
The crank timing is 28* @2000 rpm. The crank balancer is not degreed that far. I purchased timing tape to install this week on both engines. I can tell it advance quite alot though. Also, Az Speed sez that is not to much advance in an "unloaded" test. Loaded it will be much lower. I will find out!!