All good points - thanks.

I'm not 100% sure but I believe these are applicable:

https://www.newtis.info/tisv2/a/en/...plug-in-comb-type-solder-connectors/connectors-from-x6/x6038-no-adapter/vrDXJ2s

https://www.newtis.info/tisv2/a/en/...nfo/power-train/engine-electronics/interfaces/bit-serial-data-interface/oYPEvM2

Definitely not my strong suit, so if you have any input it'd be greatly appreciated!

 

Not sure why but a couple of posts didn't show up for me until now - my last post was a response to HabbyGuy post #25.

@ScottAlexander - I don't have a good answer, but to clarify, the water pump has always been running when the ignition is turned on, so the alternator being disconnected at both connections hasn't changed that. Your previous suggestion about plugging in the new water pump when I got it, to see if it acted differently than the old one was a good idea, but I got restless waiting for parts and ended up pulling the old water pump and thermostat out before the new parts arrived, so it wasn't until I put the new parts in that I found out it worked the same as the old ones.

Anyway, I appreciate you guys sharing your ideas - it's a big help. I'll report back after the new alternator is installed.

 

I looked at the Newtis diagrams.

It looks like the only component you didn't replace yet on that bus is the oil sensor.

Does the oil sensor function perfectly in the iDrive?

 

Another thought:

Did you go into the eBox yet under the cabin filter and check the DME and area for wetness, corrosion, the 5 fuses in that black little box?

 

Yes, the oil sensor was showing max oil level in the iDrive and seemed to be working correctly. I did also check the fuses in the eBox and they were fine. There was no sign of wetness or corrosion around the DME, but I didn't dismantle anything for a thorough inspection - I was a little hesitant to start pulling things apart for fear of damaging something or loosening a wire connection somewhere.

 

The diagrams take kind of a "black box approach", showing some inputs and outputs, but no real way to tell which are interconnected.

Ideally, we'd know what the environmental conditions are that would turn on the water pump, and check those "inputs" (assuming they're discreet inputs to the DME). But thinking back to the odd symptoms you describe (disconnecting one unrelated device changes the DME's output to another device), I'd have to say it's very likely that the DME is just cooked. I wouldn't be at all surprised if you DID find out that your DME was water-damaged, since having seemingly unrelated "systems" messing with each other is the kind of thing that can happen when the water (and the impurities that it causes) start creating new and exciting circuit paths on those low-current modules' circuit boards.

 

Thanks. Not what I was hoping to hear, but that does make sense. I'll probably install the new alternator as my hail mary, and if that doesn't work, take a closer look at the dme to see if there's any evidence of water intrusion. Odd that it was working fine pre-repairs and now it's not working - it's been inside the garage the whole time, so there's definitely been no new water intrusion. Appreciate all the expert advice!

 

Brainstorming:

You just put on a new valve cover gasket.

Is it possible when you took off the valve cover you stressed one of the 3 wiring harnesses that goes into the eBox?

For example, when I take off my valve cover, I take off the plastic top of my eBox. This allows me to lift up and lift out the 3 black wiring harnesses going into the eBox. This process gives me more room to lift up the valve cover and get it out.

Is it possible during your valve cover lifting you damaged a wire connections going into the eBox?

 

That's entirely possible. Plus I may have pulled on them when replacing the starter (I unbolted the intake manifold and pulled it back to get access without undoing all the connections). I didn't see any signs of anything disconnected though. And I did check continuity from the alternator back to the DME, the oil sensor back to the DME, and the water pump back to the DME... that about maxes out my electrical testing capabilities.

 

Alrighty, so I got the new alternator/voltage regulator today, installed it, turned on the ignition, and the water pump starts running - not a good sign. Sure enough, I cranked it over and it wouldn't start. So I unplug the water pump and it starts right up and runs smoothly. As I'm letting it warm up (monitoring voltage and temperatures), I decide to test something that scottalexander mentioned in a previous post, about the voltage rising with the IBS unplugged. I unplugged the IBS to test that but the voltage stayed at the same 13.7... still thinking about this.

So, once the car was warmed up a bit I plugged in the water pump and ran back to the drivers seat expecting to have to use the throttle to keep it running, but this time it keeps running smoothly. Seems odd, but then I realize the difference might be that I'd forgotten to plug the IBS back in. I decide to leave it running for 10 minutes or so... stays smooth. Then I plug the IBS back in... still stays running smooth. This could be that it's so warmed up that it's running smooth like it has a couple of times in the past.

And now for the cherry on top. I turn off the car and I get a big red gear warning light and the iDrive screen tells me I have a transmission malfunction. It's been in park for weeks! I do a scan and there are about 20 different fault codes. Full disclosure - a couple of years ago I had a similar transmission error while driving, and a buddy without a BMW specific scan tool cleared the codes and it's been fine since. Hadn't given it much thought until now. Anyway, I cleared all the codes, and I'm back to my all familiar 2E7C - BSD data bus communication fault. Going to focus back on this.

BTW, the car starts and runs fine right now but it's fully warmed up. Code still there, water pump still runs when the ignition is on without the car running. This is driving me nuts...

Will take a closer look at the DME tomorrow.

 

You say, This could be that it's so warmed up that it's running smooth like it has a couple of times in the past.

Or could it mean that the car is confused between closed loop verses open loop?

What sensor is malfunctioning turning the water pump on? Would broken thermostat wiring or broken coolant temp sensor keep the car from actually starting?

Next:

You say, a big red gear warning light appeared and the iDrive screen tells me I have a transmission malfunction and there are about 20 different fault codes.

That’s characteristic of an electrical malfunction (e.g. one of the other forum member got this recently when his mechatronic harness was not plugged in all the way).

Is it fair to say this problem would have never occurred if you had not replaced the starter?

There are only two wires on the starter.

There is only one harness on the throttle body (is it plugged in all the way)?

What happened when you replaced the starter that gives us a hint?

 

Exercise:

I watched a youtube video recently where a BMW 5 series would not start. No shop could diagnose the problem.

The problem ended up being a faulty MAF sensor.

Once unplugged the car started up.

 

Just thinking out loud...

Why does the pump ever run? It runs when the car says the water is too hot. It doesn't run when the water isn't too hot.

Perhaps the coolant temperature sensor is wonky. Perhaps some other sensor input that makes up water pump runs is doing something bad.

Like maybe it isn't completely broken, but it is doing strange things. It could be telling the the DME the coolant is real hot at startup and we better run the pump ASAP and maybe that wonkiness goes away as the car really warms up.

I'm assuming the car is designed to fail-safe if the coolant temperature sensor is bad. You could surf for what is that operation and disconnect the sensor and see how it behaves.

 

@scottalexander: Your ideas make sense. I thought about the thermostat a while back, and it was one of the components I unplugged at some point but unplugging it didn't make a difference (this was pre water pump/thermostat change, but I imagine the results would be the same). I would think that broken thermostat wiring or a broken temp sensor could start the water pump prematurely, but I doubt it would stop the car from starting - just a guess. Also, on my scan tool I can monitor the temps from the two coolant temp sensors, and the oil temp sensor, and they all appear to be working correctly - I'm not sure that it could tell me correct readings through my scan tool and give the DME faulty inputs at the same time. And while all this makes sense logically, would something like a faulty temp sensor cause a BSD fault if it's not on the BSD bus?

With the transmission fault plus the others, yes, that makes sense. I don't know which is the "mechatrinic harness" but I can look closer and see if there are any loose connections. The faults haven't come back, so hoping that's a separate issue, but wanted to mention the faults in case it prompted any ah ha moments.

You say "Is it fair to say this problem would have never occurred if you had not replaced the starter?" - Again, seems logical since the problem manifested itself after installing the new starter. As you point out though, there are only two wires on the starter. One is the power, and one I assume is to engage or disengage the starter with the flywheel. Of course BMW could have some other function/sensor in there. Maybe if the DME thought the starter was engaged with the flywheel it would shut down the engine so as not to grind gears?

You ask "There is only one harness on the throttle body (is it plugged in all the way)?" - appears to be if we're talking about the same thing. There are a bunch of wires that connect underneath the intake manifold, sort of like a junction box, and then come out as one big harness that goes to the DME (I'm not sure exactly what point is considered the throttle body vs. the intake manifold). If something was a little loose it might be hard to tell, particularly because it's hard to see. I want to also mention, in case it could end up being a clue, the only thing that was unplugged from there during my repairs was the cranckcase breather hose which has an electrical connection to the harness (it broke when I was removing the intake manifold).

You ask "What happened when you replaced the starter that gives us a hint?" - very good question... trying to come up with a very good answer, but it's not materializing

Do you think the MAF sensor could be a problem? Seems unrelated, but interesting. I'll try unplugging it.

@twh - Your ideas make sense. As mentioned in my reply to scottalexander, the sensors appear to be working, but maybe there's something weird going on. I'm not sure what other sensors might control the water pump other than the temp sensors... I'll have to search on that one.

The questions I keep coming back to are:
What is causing the water pump to run?
What is causing the BSD fault code?
Why does the car run fine when warmed up?

BTW, it's not like the car doesn't start, it's more like it starts for a split second and shuts down. Even when it's semi warm and not running well, it's like someone is flipping a light switch off and back on again that shuts off the spark to the engine, which is why if it's throttled up it can stay running through that because the engine is still in motion when the switch providing the spark is turned back on. I'll do more trouble shooting this afternoon. Thanks guys - your input is really helpful!

 

You say, Why does the car run fine when warmed up?

It appears the car is not running well in open loop. When the car is warmed up it is running ok in closed loop (and then reads the info generated by sensors).

Brainstorm:

Is it possible the car is never in open loop and actually starts out in closed loop?

 

Idea:

When you replaced your starter, you said you didn't fully remove the intake?

Is it possible you moved the intake and damaged the two O2 sensor wires or O2 harnesses that sit near the top of the intake manifold?

 

I think I'm following the logic, but applying it to this scenario is a bit outside my wheelhouse. Are you saying that when it's in open loop it doesn't consider the inputs and when it's in closed loop it does consider the inputs? I'm not sure why it would start up in open loop, or why if it starts out in closed loop it wouldn't run fine (assuming the inputs were correct). I may need a tutorial...

 

Regarding the O2 sensors - that's entirely possible. Would it be easy to troubleshoot?

 

New experiment:

Usually for your car to start you have to unplug the water pump.

In the past we've been talking about a few sensors that could be malfunctioning.

When was the last time your fan worked?

If you put your AC on, can you get your fan to operate?

If you unplug your fan, can you get the car to start?

 

Do you mean the fan that blows heat/A/C in the cabin or the electric cooling fan for the radiator? Guessing you mean the cabin fan since you talked about the A/C, but "unplugging" makes me second guess that. The cabin fan works fine with heat or A/C on and a couple of days ago when the car was running a long time the radiator cooling fan came on briefly. Both seem to be operating normally, although I haven't unplugged anything.

I haven't had as much time to troubleshoot as I'd like, but I do have a few additional inputs. The car has all the same symptoms but it's definitely running better since I put in the new alternator. It will still stall when it's cold but it's much easier to keep it running. Today I was even able to start it cold with the water pump plugged in. Still ran rough and died eventually, but an improvement. I pulled out the DME and inspected all the wires, harnesses, plugs, etc - everything is squeaky clean - no sign of any water intrusion or corrosion. I also unplugged the MAF and after a few seconds of adjustment it ran fine with that unplugged too. I was also able to monitor the MAF on my scanner too to see it was working. I also decide to try an additional scan test function - operating the water pump from the scanner (I should be able to turn it on and off at two different speeds), but the water pump continued on as if I wasn't making any changes. The more I think about it, the more I think the water pump is running in fail safe mode because it isn't communicating with the DME.

Anyway, I also looked at everything I could think of to try to find loose connections and everything seemed OK, BUT I did find one thing that made me go down another rabbit hole - one of the ground wires on an ignition coil connector had been pushed back so it may not have been making a good ground connection. I found an interesting thread on coil grounding issues on an e90post.com:
https://www.e90post.com/forums/showthread.php?t=1428829

The issues are different but involve badly grounded coils, alternator issues, and BSD codes - here's the cut and paste of that outcome.
Root cause:
Aftermarket Delphi coils sending voltage spikes back through the "ignition" circuit due to insufficient insulation.
Side effects:
BSD codes, alternator damaged by the coil voltage spikes, causing the alternator to send its own voltage spikes through the charging circuit
Diagnostic path and fix:
1) Replaced alternator after observing it spiking voltage on charging circuit; new alternator does not spike voltage, but BSD codes still present (and related drop to battery voltage when present)
2) So tested other components on BSD line - all good; then tested ignition circuit and observed a separate voltage spike on that circuit coming from the coils (not a misfire in combustion chamber, so no misfire codes). Replaced coils with genuine OEM coils and these voltage spikes disappeared.
Tech said that the aftermarket Delphi coils apparently aren't insulated as well as OEM as they have seen problems with them. Side by side of aftermarket Delphi coil vs OEM Delphi coil show a difference in the insulating material on top and who knows on the inside. OEM is the top one in the pic.
Before the part replacements, tech was also able to replicate the problem at will. Picked it up today and so far so good. Will update if anything changes.

Has me wondering if there's something similar going on with my vehicle. I don't quite have enough of a grasp on how the systems work together to figure it out. Also I was able to push the pin on the ignition coil back into place and it still had the same issues. Maybe it could be a coil related issue though - I did pull all the coils and replace one, so at least the theory matches up to something that could have happened during my initial maintenance/repair job that started all this. Maybe it's a wild goose chase though. Will do more testing tomorrow.

Thanks for sticking it out with me on this hellish trouble shooting session!

 

Earlier I was referring to the radiator cooling fan.

If you have to unplug the water pump for the car to start, it would be another experiment to leave the water pump connected and see if you can unplug the radiator cooling fan and see if the car starts.

Why? Because there appears to be a sensor malfunction somewhere.

When you unplugged the MAF you say the car started. Was the water pump plugged in at that time and the car started with the water pump plugged in?

If the water pump was plugged it and it started, that points to the MAF sensor being the culprit.

You say, The more I think about it, the more I think the water pump is running in fail safe mode because it isn't communicating with the DME.

What happens when you try to turn on the water pump using the manual method? For example, you need to bleed the cooling system so you put the key in, turn on the ignition, (but don’t put your foot on the brake) and you press the gas pedal for 10 seconds. Does the water pump turn on for 12 minutes like it should?

If you had ISTA software we might be further along the progress curve.

 

Thanks - I'll try unplugging the radiator fan and see if that makes a difference. I'll also redo the MAF test... I don't remember the exact conditions after running so many different tests.

The water pump does not turn on using the manual method (or more specifically, nothing changes, since the water pump is always running).

I looked into ISTA software a while back and remember it being either expensive, or hacked - I may have to revisit. I believe there's a lot of threads on it.

 

The ISTA software is free.

Go to BimmerGeeks and download it.

You have to download the Standard Tool set first (e.g. INPA) and install them.

You'll see it out on BimmerGeeks

 

Thanks - I will check that out.

Latest testing update: I tried to start the car cold with everything plugged in and it actually ran for 5 seconds before it died. Then I unplugged the MAF and it ran for 20 seconds or so and died. Plugged that back in. Unplugged the fan, ran for almost a minute then died. Plugged that back in, ran for about a minute then died. I suspect the variability is due to the engine heating up, or it could be random, but in no combination did it run properly. I'm intrigued that the symptoms (at least the engine running performance) seem to be getting less severe, with the noteworthy improvement coming after the alternator replacement.

Since I'm pretty convinced there's a DME - water pump communication issue, I decided to go back and retest continuity from the DME to all the components on the BSD bus (water pump, IBS, oil sensor, alternator). I hadn't actually checked the IBS wire previously because it was hard for me to hook up my tester from the trunk and also access the wire at the DME, but now that the DME is pulled out it's easier. Anyway, all the connections checked out OK. I used electrical component cleaner on the water pump connector again to make sure it was clean (I'd done it originally because a lot of oil had dripped on and around the water pump). Anyway, all the BSD components are still disconnected. I hook the DME back up, hook the battery back up, and I decide to just connect the water pump to see if it can talk to the DME without anything else connected (I'd tried this all the way back in the beginning before changing any components out). With just the water pump connected, I turned on the ignition, and to my surprise the water pump didn't go on! Started the car and it ran fine. Monitored temps and voltage and when the engine coolant temp got to around 200F the water pump turned on - more evidence that he DME and water pump are now communicating. I need a break anyway, so I'm planning on letting the car cool down, then seeing if when I turn on the ignition the water pump stays off (hopefully), then I'll add BSD components one by one and see if any of them lead to the activation of the water pump. Sort of mid diagnostics at this point, but an interesting development.

 

Yes, that is a new development.

What caught my eye was when you said, " I used electrical component cleaner on the water pump connector again to make sure it was clean (I'd done it originally because a lot of oil had dripped on and around the water pump).

We just learned there is a lot of oil on the engine and electrical components.

The front of that engine and the all the electronics have to be spotless. Remove the serpentine belt and clean it.

When you cleaned the water pump connector from oil, I'm wondering if that is a large portion of the problem. Maybe the other portion is oil all over:

- intake and exhaust cam sensor connectors
- intake and exhaust vanos solenoid connectors
- coolant sensor connector (on the block next to the mickey mouse hose)

How about cleaning those.