Today I got the GS-911. This is a USB to BMW computer adapter and the associated software.

I ordered it last week for a few reasons:

• Double-check my diagnosis about the fuel pump
• Reset the “service engine soon” light after I replace the oil
• General maintenance

So, here’s what this has to say:

Yep. I guess I was right.

All the rest of the data is wicked cool too!

Of course I can’t really play a whole lot since the bike isn’t running… but I’m looking forward to using this in the future!

Every once in a great while the bike stalls on downshift. It happens once every 200-300 miles… I’m wondering what the fault code is for that? I wonder if it was the fuel pump starting the fail?

You know I’ve been trying to repair my motorcycle over the past couple of weeks. I’ve narrowed the problem down to what I’m pretty sure is the fuel pump being shorted out.

The problem arrises when I look at the parts fiche:

I went ahead and ordered part #1 — the entire fuel pump assembly. I didn’t want to dick around with fitting a new pump to the thing since that involved hoses and clamps that require more tools to buy. It winds up being maybe $40 more than the pump plus extra parts but it eliminates a lof of the uncertainly.

That’s not the problem.

Part #10 is the the part that worried me. It’s the fuel level sensor. It’s somehow attached to the pump.

Looking at this I was wondering how this all fits together with the pump and how it comes out, and more importantly, how it goes back together.

I was all ready to just go in blind and worried when I found the shop manual for my bike. (I had one for my earlier iteration of the bike — that one had a float instead of a new-fangled, and as it turns out fault-prone, electric sensor)

It seems that the business end of the sensor (#1 below) attaches to the fuel filler (#2) and the only interaction with the pump is the plug you see on the upper-right of the picture up there.

The only thing I have to do since I’m not screwing with the sensor strip is unplug #3 from the fuel pump assembly, swap in a new one, then plug the existing fuel sensor back in.

Easy-peasy.

All I have to do is wait for the wrench (#1 above) to come in from the US warehouse. Booo… I ordered it on Thursday night and it’ll take 2-7 days (business days I’m presuming) to come in and then get shipped to me.

C’est la vie.

But knowing what I’m up against — or as the case me be not up against — is a nice thing. Now I’m just hoping my diagnosis is correct!

Tonight I tore into the bike again trying to root out the cause of the bike woes.

Online I heard a suggestion that the throttle position sensor might be screwed up. That was an easy check. I pulled the connector and threw the multimeter on it. It’s a simple potentiometer and simple to check with the ohmmeter. 4.3k rail to rail with the center tap going from 1.2k to 4.1k. Right around in spec.

Next up was the fuel pump again. Further tearing it up I got to the fuel pump.

1.1 ohm across the leads seems low, but the final test is current draw. I rigged up a janky setup and tested it.

For a brief instant I saw 15A on the Fluke. Thankfully it is specced to handle 20A for 30s so this wasn’t a problem. The 7.5A fuse inline fuse blew on cue.

The pump is, essentially, a dead short.

That would explain no gas getting pumped.

I guess I’m in the market for a new one now…

While I was down in California the fuel pump controller bypass cable arrived in the mail.

To recap, what this does is hook the fuel pump directly into power — more or less as direct as you can make things.

To verify that power is in fact there I double checked with the power on and the cable plugged into the aux power jack on the bike:

Sure enough, power is flowing. Looking on the bike seat you can see the controller I pulled off the bike. This one is a new-style one that’s fully potted.

It hooked up no problem…

The plug closest is the fuel level sensor plug. Hidden behind the bodywork is the fuel line itself.

Crank.

No fire.

Even with the fuel pump connected directly I didn’t get anything. I got brave and pulled the fuel line. If things were working I would get a fountain of raw gas shooting out the now vacant hole.

Nope.

This pretty much rules out everything except the pump itself.

At this point I have to decide if I want to replace the pump myself. It’s a $150 part along with $100 of specialized tools. It’s still cheaper than the BMW-branded pump plus all the dealer markup.

Oh… and if I got a tow to get to the dealer…

Yesterday I wrote about my motorcycle problem.

Today I dove deeper into the fray.

This morning before work I pulled the fuel injectors. It’s a simple job that takes one screw per side. When I tried to start the bike I got nothing out of either one. I should have figured it wasn’t the injectors because the chance of both going bad is quite low.

I was really getting ready for a “don’t try this at home” situation when I would try to clean the injectors by soaking them in gas in my ultrasonic cleaner. Wouldn’t be a good thing to do, but I figure if I’m outside I’d likely have walked away from it.   Really though… just because I would try that don’t do it yourself. Don’t say I didn’t warn you.

With nothing coming out the injectors I can safely say I have a fuel issue. Air and spark have both been verified. That, and it’s somewhere upstream from the injector ports.

After work I pulled the fuel supply from the injectors. If I get gas pouring out the hose when I power up the bike then I can safely say it’s the injector. If I don’t then it’s further up from there.

I didn’t get anything. Just a dribble when I pulled off the hose — but that’s just what was there from before.

The big empty hole is where the injector goes. The fuel hose it the is poised above it with the flange for the screw to hold the whole assembly down.

At this point I’m running out of options. Between the tank and the fuel hose is just a few things: fuel pump, fuel pump controller, distribution tubes. There’s also the carbon canister, but that should not interrupt the flow of fuel, just perhaps make things run rough.

The tubes and hoses are not likely to be the problem. They might constrict the flow, but shouldn’t cause a dead stop. The pump is an option, but I’m hoping that’s not the case.

The fuel pump controller, on the other hand, is a known source of problems.

The torx bit is pointing to the fuel pump controller. It has the smaller of the connectors on top of it.

What this guy does is modulate the speed of the fuel pump. There’s no distinct problem to running the pump at 100% all the time other than it’s wasteful and loud. However, if something leaks under it there have been many reports of it spontaneously frying itself. Someone took it upon themselves to make a cable to bypass it. I ordered one up from BurnsMoto to try this option. It’s another $40, but it can save my bacon if I’m in the boonies and it dies (even if it’s not the problem now).

So, I have two options I see:

1. Fuel pump controller is broken. I’ll know as soon as I get the bypass cable and try it out. If that’s the case I can order a new controller for $150 and get back on the road.
2. Otherwise, it’s the fuel pump. This is a bigger task since I need special tools to get into the tank to replace it. I’ve not heard any systemic problems with it so I’m crossing my fingers that it’s not the problem. If it is the pump I can likely pull the tank and drive it to a BMW shop to save the tow. Cost for this option is likely $330 + markup + labor. It’s still cheaper than with a tow.

Wish me luck!