Marine Power 6L Problems

[IDBoating]

I don't have any documentation for it so I am not sure on all of the details

If you provide the engine's serial number to MP, can they provide info on that specific engine? Most manufacturers of serialized complex equipment keep a record of each serialized unit. Just a thought....

[chpthril]

That makes a lot of sense. This is probably a stupid question but does the throttle body remain completely closed until the IAC reaches 100% open?

I did notice that the IAC gets to 100% approximately at 2000 rpm where the issues is happening so it would make a lot of sense that the throttle body just hangs up a little as it tries to open initially if that is how it works.

Seeing the IAC @ 100% (full open) with the throttle open, seems normal. Its in anticipation of the throttle closing. The prevents a rapid closing and high engine vacuum condition, which increases emissions. Once the throttle closes, the open IAC can bring the idle down slowly. Once the engine is off idle the IAC is really out of the loop.

[lee]

Have you asked Marine Power if they would send you a new flash for your ECM?

[ericinmich]

I worked on some cam phasor controls at GM, not the duel equal (small block) but the true variable DOHC ones. I can't tell what the sample/data rate is on the snapshots but everything looks good except for a step change in the cam angle.

The TPS desire vs actual being off by a few percent is normal. The changes in MAP look normal. The change in limiter value is likely normal, it looks like the controls is set up to change from 'idle mode' to 'non-idle mode' at 2000 rpm. This is why the cam is commanded to move and also limiter value changes, etc.

I would like to see a trace on a working engine. If the update rate on the snapshots is long then it could be that the cam phasor change shown is normal, then I would lean towards a fuel issue.

[IDBoating]

The TPS desire vs actual being off by a few percent is normal.

Agreed, but his claim that the actual "sticks" while the desired continues to change is what caught my attention.

[ericinmich]

Agreed, but his claim that the actual "sticks" while the desired continues to change is what caught my attention.

Yea, its tough with seeing the real data and also understanding the update rate. I would also see if the throttle follows desired if the engine is not running (ign on). If that seems to track then I would rule out the ETC. Sure wish it was a closed loop system so I could see some O2 values during the problem. My hunch is that its going lean.

[cpropes2005]

Logged some more data yesterday. Like eric says it is really difficult to see the fine detail of what is causing what by just looking at the snapshots. I really wish I could plot all of the outputs vs time, overlay them and then zoom way in on the issue but the software doesn't support that. It does give me a ton more info than I had before though. When watching the live stream the boxes get highlighted when they are updated so you can kind of get an idea for sample rate. It isn't super fast but looks like the sensor readings update sequentially at around 5 to 10hz. Some of the other generic info like the rev limiter data only updates maybe once every 30 seconds or so. A higher sample rate would definitely be helpful.

My latest working theory is that like eric says the ECM is programmed to switch from "idle mode" to "non-idle" mode at 2000 rpm and there is a bug in how it makes that transition. I wish the rev limiter data would update at the same rate as the other fields so I could confirm but it looks like in idle mode the cam angle doesn't change more than a degree or two and in the other mode it has the freedom to operate across its range so you can kind of tell which mode it is in even though the rev limit data doesnt update in real time.

I think it is actually making it too rich when it happens. The injector pulse width goes way up plus I can sometimes smell gas. If you go way back to my original post I also have a much more infrequent issue where the boat will all of a sudden start idling very rough and die, followed by strong smell of gas. Waiting a minute or two and restarting in neutral with a little extra RPM "fixes it" every time and the boat runs perfect after that. I haven't been able to make it do this with the scanner attached but it seems like that issue is just a prolonged version of the momentary hesitation so maybe somehow it gets stuck in the wrong mode and cant reset itself until I restart and rev it.

To add one more piece of info when I turned the key on yesterday I got a continual beep alarm. After a minute or two it changed to two beeps every seconds. I went ahead and started it and it continued with the two beep alarm for about another minute and then went away. I ran it for awhile after that while I was taking the data with no more alarms. This is the first time I have ever gotten a beeping alarm since I have had the boat and the only thing I had done since the last time I ran it was connect the scanner and poke gently poke around near the ecm looking for the engine identification plate. I know this means I should have some stored codes but I am having trouble getting the scanner to pull them.

It still seems a lot like a programming issue so I think I am going to go forward with trying to get the ecm updated. I have tried to contact Marine Power once before with no luck but I have more info now so maybe I will give them a call before I send it off to Bob. At a minimum it would be nice to have the password for the tune so that I for sure have the option of going back to stock if I need to.

[cpropes2005]

I ran the throttle up test with the engine off and it does look like there might be a slight issues with the throttle response right around where 2000 rpm would be. I pushed the throttle up as slow and smoothly as I could because I wanted the desired and actual throttle positions to be as close as possible. Up to about 31% the 2 values map really close to each other with them showing the same value probably 90% of the time or an actual value that is just .2 to .4% higher. At 31.6% it seems like the actual value hangs up just a little bit like I had seen on some of the other logs with the engine running and the desired value gets about 1.5% out of line with the actual before they come back together as the throttle gets up to 36%. I know a couple of % difference is normal but it just seems interesting that the only time there is more than a .4% difference on this slow throttle up test is right when I am having the problem. Obviously if I throttle up faster the delta is larger.

I already removed the ecm to send it out but I think I will put it back in and see if cleaning the throttle body helps at all since that is an easy test. I will report back what I find. Thanks for all of the help so far guys.

[ericinmich]

I am still leaning towards going out lean (punny). The inj pulse width is going up due to the MAP going up. The MAP is going up because the cam and throttle ate moving. You can get a gas smell from lean as the engine can be partialing (few cylinders not burning) and the unburnt fuel gets passed out the exhaust. One test would be to put a heat gun on each exhaust tube while keeping it in the misfiring situation. I guess over fueled would also show a heat embalance. Dumb question, are you sure all spark plugs are fresh and gapped correctly? Lean mixtures and larger gaps take higher voltages to fire..