How NOT to Design A Fuel Pump Connection ALL R12s

This thread describes how BMWs inept (charitably put) design of the fuel pump controller connection on most R12 bikes can leave you with a dead bike, especially if you ride in the rain. I am by no means the first person to comment on this (as a web search will show you) but most have talked only about how to fix it. Much of this thread is about how the BMW design is a failure waiting to happen, with recommendations on pre-emptive maintenance.

What happened to me? My 08 RT simply crapped out today in nice weather while going on an errand. It died and could be restarted, multiple times. Sometimes it quit again almost immediately or in a couple hundred yards and sometimes I could ride it 5 miles with no problems. My GS-911 showed a 10168 code, electric fuel pump failure, so I pulled the left side cover to examine the fuel pump controller and connections, a well known problem.

Let's start with a broad view photo to set the scenario. The silver circle in the photo is the well that contains the fuel pump controller. In the photo, the controller has been removed. Notice that the fuel pump controller/well is on a slope,at the front of the fuel tank and that the removed controller occupies the low spot in that well. Therefore, fuel pump controller connections are at a low point in the well where any water that penetrates will collect. A smart engineer would make a well with an electrical connection self draining so that any water which penetrated would also flow back out and not build up to short the connection. Really smart engineers wouldn't put the connection in a well, let alone a well at the front of the tank where heavy rain can reach it while riding. BMW's engineers aren't smart enough to have done either.

Now the Controller

Here's a photo of the original (silver metal housing) fuel pump controller from my 08 RT. The blue connector clips to the fuel pump connection at the bottom of the well. That connection point will be shown in more detail in the next post.

Note the gasket on the SIDE of the controller that is intended to keep water and grit that can collect in the well from penetrating to the connection to the fuel pump. This is a major design flaw that causes many of the so called fuel pump controller failures because it is has only SOME of the properties of a compression gasket like it should be. It lacks any vertical locator pin to keep the gasket in place and prevent it it from spreading to the surrounding ridge. It also is clamped by only the two controller screws and two clamping points do not define a compression plane. The one of my bike was partly adhered to the side of the well and had apparently spread when "clamped". BMW doesn't even bother to grease this during assembly despite the fact that silicon grease is widely used in many industries and applications to help ensure rubber gaskets seal properly on installation.

The somewhat hard-to-see black electrical connector connects to the bike wiring that provides power to the fuel pump controller. I see no issues with the design of this connector.

Closeup of Fuel Pump Controller Well

Here's a closer photo of the fuel pump well. Note the ridge that surrounds where the fuel pump controller goes. The SIDE gasket of the controller is intended to seal on the inside of that ridge.

Also note that the ridge does not go all the way around the controller, Instead, the wall of the well itself provides part of the sealing area. This is the reason that a proper compression gasket could not be fitted (it would have to be on the BOTTOM. not the side and would require a different well design). It's a first class design f**kup, IMO.

Note also the two metal pins surrounded by a ridge in the fuel pump controller well. That ridge around those pins is the only half way intelligent part of the design. Its purpose is to prevent water that gets into the well from getting over that shallow ridge and into the controller connection to the fuel pump pins. However, it is very inadequate. If enough water gets in, it simply flows over the ridge and penetrates the connector, causing corrosion. That corrosion may have my bike to quit (but later examination also showed the board inside he controller had a little corrosion and some of these controllers also apparently fail from heat overload.) I didn't measure the exact currrent drawn by the fuel pump but it is reported to be 7A, and that's plenty to help accelerate corrosion and too much to successfully put through a waterlogged, corroded part.

And because the well is on the low side, putting the bike on the sidestand increases the chance of water overflowing that ridge and shorting out the connection. A smart design would instead put a critical connection at a high point (in addition to keeping water out, allowing for drainage, etc).

You can see two screw holes outside the well for the screws that hold the controller in place. Those screws were also corroded when removed.

(I have since learned that a BMW replacement controller comes with 2 new screws and a new gasket)

What Happened to Mine and Why

In my case, enough water and road grit penetrated the controller well to cause corrosion at the connection at the pins to the fuel pump. (Later examination showed some corrosion on the controller board andd some controller are also reported to fail from heat overload)

That water and dirt penetration of the inner well and to the controller circuit board is prima facia proof that the design doesn't work. If the design was any good, the controller would work with the outer well full of water because it wouldn't penetrate to the pump connections

Most R12 bikes have a similar design AFAIK but I'm also no ultimate expert on all years and models. This can happen to yours if you ride in the rain (what probably killed mine) or are foolish enough to spray a pressure washer into the front of your bike. My bike gets washed about 3X/yr with a hose and bucket. Of course, if you live in the desert or never ride in the rain or wash your bike, you're not likely to have this problem. But I ride in the rain while touring unless it gets so heavy I simply can't see - I've got about 1000 rain miles this year- less than many of you, no doubt.

AFAIK, the only BMW response to this problem has been to make powder coated controllers and a couple service campaigns which would not have changed what caused my bike to fail on the road. The whole well needs a complete rework on all models, IMO. Certainly periodic failures have been reported on many models. (There is a website showing UK stats for the GS and well over 100 failures were reported in 2009 and that many already this year.

How Does One Prevent Water-Caused Controller Failure

I believe it is possible to reduce chances of water-caused (but not heat caused) failure by routine inspection and here is what I intend to do going forward.

The left side cover comes off every 6K to do valves, etc so that is a good time to inspect. Inspection consists of:
1) Pull the 2, T-20 screws that hold the controller in place
2) Lift the controller and examine beneath it to see if water or dirt have gotten into the controller well.
3) If water or dirt are present, remove controller, clean well and connections.
4) Reassemble using silicon grease on the controller side gasket.

Based on what I've seen, I recommend taking a hard look at this before any long ride that may involve rain or takes you away from your friendly dealer.

Also, carry a jumper cable or spare controller - see below in thread

Thanks, that was very interesting! (And I don't even have an RT )

How Does One Fix Controller Failure on the Road?

Well, your first problem is figuring out whether or not its the controller or something else entirely. If you're a decent mechanic, you will probably have a fair idea of what's wrong with your bike when it dies for some electrical reason, but if you're mechanically challenged, you're basically SOL. Doing routine service on your bike will get you familiar with how its put together and give you confidence to deal with problems on the road, if you carry a decent tool kit.

I happen to be a decent mechanic, who, though not a pro (though I have been at times), has wrenched on a lot of stuff for nearly 50 yrs. I also have my GS-911 and a netbook with the GS-911 software and the RepROM disc loaded to its drive on the bike when I travel. The GS-911 gave me the diagnostic code that took me right to the place to look for the failure. I knew I was going to find 1 of 2 possible problems before I ever pulled the left side cover to access the controller.

In my case, the controller was intermittent and was reinstalled after removing corrosion from its connections; it started the bike a few times but couldn't keep it running. It eventually failed entirely with no voltage at the output so I cut off the connector and made a jumper to get the bike running. That works ifne but must be unplugged when not in use or it will quickly kill a battery. (DeOxit from any Radio Shack, Home Depot, etc works well for removing corrosion, BTW) Note that the RepROM says to replace a corroded controller, not repair it, and in many cases it may be damaged internally (like mine) even if the exterior looks OK.

You can do 1 of 3 things on the road
1) Clean the connections and reinstall the controller and see if the bike runs OK- only a partial chance so carry at least a bypass cable
2) Bypass the controller entirely so the fuel pump is "always on". There are 2 ways to do this. One involves a bypass cable that connects to battery and the other involves cutting the blue connector with its wires off the controller and patching it directly into the power supply to the controller.
3) Install a new controller if you have one.

I've decided to replace the controller, to do 6K inspections on the controller well, and to carry a spare controller on the bike. Looks like a little can of DeOxit would also make a useful addition to the toolkit.

Article on making a bypass

http://www.advrider.com/forums/showpost.php?p=2768376&postcount=21

Above is an article on doing a bypass on one model. They're all pretty similar though color codes for wires may differ.

Commercial Source of a Bypass Cable

Here's a link where you can get a pre-made bypass cable. These are cheaper than a controller and work fine. (The controller is about $150 and the cable about $25) Just remember to unplug it when you stop riding or your fuel pump will RAPIDLY kill your battery due to the current it draws!!

http://forums.bmwmoa.org/showthread.php?t=38757&highlight=fuel+pump+bypass


You can also make one of these by splicing the blue connector cut from a dead controller onto any wire that you can connect to battery or an outlet rated at 10A or more (like the front accessory outlet on my 08 RT but note that many GS bikes and earlier RTs only supply 5A at the front outlet so will not run the 7A pump reliably from that connection point).

Great Article on Fuel Pump Controller Problems

http://www.motorcycleinfo.co.uk/index.cfm?fa=contentGeneric.nvxszdkebklxuaer&pageId=233223

The above link is to a great article that reaches the same conclusions I do about prevention but doesn't detail the design faults as clearly

First, thanks for the taking the time to report this problem, post a solution, and take photos to illustrate it. The effort is appreciated.

My question is, what's directly below/behind the "well" that the controller sits in?

Could a few small drain holes be drilled in the bottom of the well to prevent water from accumulated in there? Adding drain holes has several times solved similar problems for me over the years. I remember I bought a new Volvo in the mid-80's and every time it rained hard the taillight housing would fill would water and blow out the taillight/stoplight bulbs. A few strategically drilled drain holes solved the problem.

It looks like drilling a drain would simply pentrate into the interior of the fuel tank so drilling is a recipe for one of several worse problems.

If someone knows otherwise, please say so because it would be an easy design improvment that would negate the need for 6K inspection of the controller well.

Knowledge is power. I don't believe I will ever have a need for this information. That said, in the event I do - this information will be invaluable.

Thank you for taking the time to run it down and post it with good pictures to support the data.

Good thread.

Excellent article. Thanks for taking the trouble to share your experience and knowledge.

FYI, I have a 2009 R1200R. I just checked the RepRom DVD Service Manual for both your model and mine. The fuel pump, and its controller are located under the tank on the Roadster (R1200R), and the unit is oriented upside down compared to the RT. To get at this assembly on the R1200R, the fuel tank must be removed completely. On the R1200R, both the location and orientation of the fuel pump/controller assembly would, I think, prevent the problem you had. I have had the tank off on my bike, and the reason I checked the manual is because I was sure I remembered seeing the fuel pump assembly located on the underside of the tank.

Also, the reassembly instructions for both models call for lubricating the sealing side gasket with silicone, as you suggest doing.

I didn't check the GS instructions, but, being more along the lines of a naked bike, I'm guessing the fuel pump is located more like on the Roadster.

EXCELLENT POST!

I just pulled the controller on my '07 RT for the 1st time and, I'll be darned, there was a small puddle of water down in the "well" (in addition to all the crud packed along the side of the controller). The visible "high water mark" under the controller was just up to the base of the plug, no where near enough to overflow the connector but, none the less, water was getting past the seal.

Thanks for the heads up!

Great info on the R1200R- with an inverted well the design would work a lot better because its not fighting gravity to keep water off the connections.

What about other models? Anyone else got a controller at the bottom of the fuel tank?

The failure is very well known on GS models which you will see if you do a web search.

racer7 said:

What about other models? Anyone else got a controller at the bottom of the fuel tank?

The failure is very well known on GS models which you will see if you do a web search.

Click to expand...

My hunch was wrong. Rather than load the DVD up again, I took advantage of Max BMW's online parts catalog (if you haven't checked that out, it's a great tool). The GS is like the RT. Maybe the Roadster is the only one with the pump under the tank. Actually, even on the Roadster it's not really upside down, more like a bit off vertical, but the tilt still favors water draining rather than sitting in the well.

As I reported in another post, I carry a spare board on my bike. This is due to the failures I heard about and the fact that it is an easy field repair. I replaced mine with the new one and I used RTV to seal it in place and I don't think I'll ever have a problem now. But, I do carry the old one along with an antenna ring for the EWS failure. I suspect I won't need either, but maybe I'll be able to help a fellow rider.

Great pics.

racer7 said:

The failure is very well known on GS models which you will see if you do a web search.

Click to expand...

Yes and no. The GS failures primarily fell into two areas, both covered by service campaigns from BMW. The first was a bad O-ring or a bad installation of an O-ring that effected very early '05 and earlier GS. The second was part of the all model recall where the failure mode was as likely to be heat instead of water. I don't know what changes BMW made to the controller, because the service campaign didn't effect earlier models.

I've an '05 GS as my primary transportation that sees lots of wet riding and has never had an issue. I am aware of the possibility and know how to bypass the controller if necessary.

racer7 said:

I believe it is possible to reduce chances of this failure by routine inspection and here is what I intend to do going forward.

The left side cover comes off every 6K to do valves, etc so that is a good time to inspect. Inspection consists of:
1) Pull the 2, T-20 screws that hold the controller in place
2) Lift the controller and examine beneath it to see if water or dirt have gotten into the controller well.
3) If water or dirt are present, remove controller, clean well and connections.
4) Reassemble using silicon grease on the controller side gasket.

Based on what I've seen, I recommend taking a hard look at this before any long ride that may involve rain or takes you away from your friendly dealer.

Click to expand...

Thanks, I'm going to check mine this week. Great information.