• Welcome, Guest! We hope you enjoy the excellent technical knowledge, event information and discussions that the BMW MOA forum provides. Some forum content will be hidden from you if you remain logged out. If you want to view all content, please click the 'Log in' button above and enter your BMW MOA username and password.

    If you are not an MOA member, why not take the time to join the club, so you can enjoy posting on the forum, the BMW Owners News magazine, and all of the discounts and benefits the BMW MOA offers?

1997 BMW R850R Sudden No Start - Not Just the Hall Sensor!

Do a compression test, both plugs out, throttle held open.
Report back

I am very happy to report the results of this compression test -- 170PSI, both cylinders identical.

I'm amazed, considering the bike's condition when I got it. Guess these engines really do live up to their near-bulletproof reputation.
 
I finally figured it out.

The throttle bodies were pretty dirty inside. Idle air passages were mostly clogged up and the whole interior of each body was coated in carbon and oil filth. I thought for sure that was it.
Cleaned out both throttle bodies, installed a new set of Tills injectors, changed the intake boots and all the throttle body O-rings...and still no start.

So that either meant the Motronic got fried when the Hall Sensor went, or perhaps it wasn't getting enough fuel pressure. On the latter front, I had a suspect. I had purchased a Holley in-tank nylon fuel line to replace the original (and quite hardened) U-hose on the end of the fuel filter. Identical to what Beemer Boneyard sells but about 4 inches long, not 10. It seemed to fit perfectly when I installed it.

I removed the fuel cap and stuck my dad's snake camera down into the fuel tank. The fuel level was juuuust low enough for me to see that, when I turned the ignition on, fuel was squirting out of the connection between the fuel filter and that little nylon fuel line.

That explains the sudden failure! The line must have ruptured somewhere, which is a shame. A thread on ADV Rider talked about the suitability of these cheap Holley lines, which is why I bought one. One user even pressure tested one to 50 PSI and found it to be suitable. Guess that isn't always true. Thankfully it was cheap, less than 6 bucks, so I'm not too broken up about it.

Last night I ordered the stainless steel U-pipe kit from Nushings. Cheaper, and frankly a better solution than OEM, as it is stronger and allows all the in-tank fuel lines to be replaced regularly with less than 1 foot of standard submersible line.

Once that arrives I'll drain and open up the tank, and post another update, but I'm pretty dang sure that's been the problem this whole time.
 
Alright, the new U-hose kit didn't fix it. I'm running out of ideas again.

I also tested for fuel injector spray while cranking the bike -- both brand new Tills injectors are spraying fuel with what appears to be a conical pattern. But the spray doesn't look like a fine mist; it looks like larger droplets and there was not much of it. I've never checked an injector's spray pattern before but a Youtube video I saw on oilhead fuel injector cleaning sure looked like they are supposed to spray a lot more.

So I have great compression, great spark, great air, and I do have fuel, but it might not be enough fuel. The bike doesn't sound like it's firing at all when it cranks on a fully charged battery and will not start at any choke position or throttle position.

What do I check next? A failing fuel pump? A failed fuel pressure regulator? A bad or choked up fuel filter? Did my Motronic get fried when my Hall sensor broke?

Both the fuel pump and fuel filter are almost brand new, replaced just a couple months ago with less than 50 miles on them since. The pump still primes exactly as it always did when the bike is turned on, and when cranking.
 
Check your fuel pressure at the lower hose (output)
Insert a tee and a pressure gauge.
When the pump primes you should have 43 psi

The injectors should spray, not dribble
 
Check your fuel pressure at the lower hose (output)
Insert a tee and a pressure gauge.
When the pump primes you should have 43 psi

The injectors should spray, not dribble

Rented a fuel pump test tool from O'Reilly and did this today and it came with a brass T-fitting, which I spliced into the output line. Both when priming, and with the bike cranking, the pressure gauge read a whopping ZERO psi. Tested the tool with an air gun to make sure it would take pressure (it did). Tried again. No pressure buildup that it could measure in that output line. I made extra sure by poking the Schrader valve in the T-fitting and I got maybe a drop of fuel to come out, so there was definitely no pressure or very close to no pressure.

Also double-checked and the bike will fire up just fine on starting fluid. Won't stay running after it burns off though. So it must be a fuel issue.

I then did some further testing by draining and removing the fuel tank and pulling the fuel pump assembly with sending unit plate out. My dad and I cut the top off of a smallish HDPE container, dunked the inlet end of the fuel pump in it, and filled it about halfway with fuel to immerse the inlet end of the pump. He held it on the right cylinder head of the bike, with the fuel pump harness plugged into its normal place.

First I tested this way by having the fuel pump output line redirect fuel back into the container, and after a few on/off priming cycles to fill up the pump and filter the fuel stream coming out of the output line was as thick as the internal diameter of the line itself and had no bubbles or anything. So the flow from the pump and filter assembly visually looked very strong.

We then plugged the output line back into the bike, going to the fuel distributor and pressure regulator. With the pressure gauge still plugged into the output line I then cycled the fuel pump on and off with the killswitch multiple times. We did see fuel come back through the return line, but still got ZERO indicated pressure on the inline gauge.


This bike sat for three years with a tankful of E10 gas in it and was (supposedly) run only a few times during those years. Can rust and tank junk kill a fuel pressure regulator? Make it stick open?

I ordered a new fuel pressure regulator already, since it really can't hurt to replace another part that was soaked in nasty gas for years. But are there any other suspects?

Edit: One thing I should also mention is that when priming the pump by flicking the killswitch off and on, it doesn't always pump for the same amount of time. Sometimes it's very short, maybe a quarter of a second, and the longest it pumps is maybe a full second. Is that normal? It was doing this even when the output line was redirecting fuel straight into a container and not into the distributor.
 
Last edited:
If the regulator is stuck open, pinching the return fuel line (upper) would allow for pressure on the fuel rails.
Try this test for a brief time to see if the bike will start.
 
If the regulator is stuck open, pinching the return fuel line (upper) would allow for pressure on the fuel rails.
Try this test for a brief time to see if the bike will start.

Once I get my set of metal fuel quick disconnects in the mail in a few days, I'll try this out.
 
Well I forgot to try pinching the return line, GSAddict -- and the bike still won't run even with a brand new fuel pressure regulator. It will still run on starting fluid. On the positive side I was able to confirm and document the "cheaty way" to replace a fuel pressure regulator by removing the battery box.

The old regulator was pretty dirty, and unlike the new one it has no filtration screen -- and when I got the bike the original fuel intake screen on the fuel pump was ripped and the BMW-branded fuel filter was dated 2006. It's entirely possible some crud or the decaying fuel tank liner or a decaying fuel filter got into it and fouled it up before I ever got my hands on this machine. The pressure regulator diaphragm being partially stuck open could explain why it never ran well when it did run.

Every part of this fuel system has now been replaced with new parts except the distributor and the fuel tank. At this point the only option I can think of is that my brand-new fuel pump is bad.

I have three options:
1. Get a new pump under warranty, as it's a Kemso pump and they provide lifetime warranties. But I have to prove that the pump is bad before they'll send me a new one, which means taking it out of the tank anyways and then photographing it in a test rig.

2. Replace the bad Kemso pump with a used OEM (VDO) pump I got with my used fuel sending unit. It's in great shape and was pulled from an R1100R with only 16k miles on it and still has a good intake screen as well.

3. Buy another new (more expensive) pump, like a new Walbro or Quantum. Requires spending even more money which I would like to avoid.

Any advice? I'm leaning towards just installing the VDO pump and seeing what happens. The Kemso was a $37 fuel pump and unfortunately I suspect that may just be too cheap to be trustworthy.

Edit: I did a bit of research and the Walbro fuel pump that came in this bike is a universal part. There's no manufacturer guarantee that it'll work with my motorcycle. I had thought about reinstalling it, but considering how long it sat in nasty ethanol fuel I'm not going to risk it, especially since I'd have to buy a new intake screen for it and change the fuel pump terminals.
 
Last edited:
I think you're doing something wrong in the testing. You're not going to have a spray pattern if you have zero fuel pressure. You established a good flow from the pump. If you have fuel returning past the regulator to the tank through the upper hose, and a spray pattern, it's practically impossible to not have pressure.

Lower/forward fuel pump plate outlet to lower distributor pipe. Upper to upper. Lower is the one with pressure. Take the hose off the lower distributor pipe and run it into a jar. See if you have strong pump-fed flow. Then try to block it off.
 
I agree with Anton.

Remove the lower line on the tank and deadhead with a pressure gauge that has a 150psi range.
Cycle the pump for the 2 sec start cycle.
If all is good tank side you will read up to 120psi (the pump has it's own built in internal over pressure regulator)
Report back
 
I agree with Anton.

Remove the lower line on the tank and deadhead with a pressure gauge that has a 150psi range.
Cycle the pump for the 2 sec start cycle.
If all is good tank side you will read up to 120psi (the pump has it's own built in internal over pressure regulator)
Report back

When I first used the testing kit, I was using a T-fitting plugged into the lower line, not a dead-end fitting. I also blew compressed air into the gauge to make sure it would actually read pressure after I got none at the line.

I'll rent the tool again and test with a dead-end fitting. It makes sense that I wasn't testing pressure directly from the pump with a T-fitting; what I suspect is that this cheapo pump just isn't providing enough fuel pressure, or enough flow, to run the bike now, and never provided enough to run the bike correctly anyways.

Anton -- I never tried to block the "send" line but I did get what looked like plenty of flow from it into a container when testing, so I'll try to block it off with a pressure gauge as GSAddict detailed.
 
I have very good news -- the bike runs again, and seems to be running properly this time!

Here was the process: I dead-ended the fuel "send" line with the rented pressure gauge again as you folks suggested. Surprise, surprise -- 0 PSI from the cheapo fuel pump. I even tried draining the gas out of the tank via the fuel pump and only got a very slight trickle of fuel out of the send line.
Long story short, I swapped in the used OEM pump I had and it fired right up.

All these parts swaps and work, because of a bad fuel pump with less than 60 miles on it! My guess is the pump gradually started failing the moment it went in the bike.

I then did a relatively quick throttle body synchronization and idle adjustment with my dad's Carbtune. The bike still seems to vibrate quite a lot whenever the engine is running, which increases directly with engine RPM. I still have yet to drain the somewhat overfilled oil and do a valve adjustment so both are on the list of things to do very soon. After those I plan to do a very careful TB sync. Any other ideas as to what would cause it to have what feels like rotary or lateral vibration?

Also the low fuel light seems to be half-lit all the time, which is odd. Not sure if I blew the relay again or if maybe I got the float stuck when I replaced the pump. I'm going to look inside the tank again with a snake cam and figure it out.

Thanks for all your help, GSAddict and Anton. At the very least I can now use this machine to get places while I slowly work out all the other less important issues.
 
No prior mention of this, so - Have you pulled the big brass air screws and cleaned off their tips, and cleaned out the passageways in there? Also look for a crapitation buildup in the bores of the throttle bodies.

Pauls1150, how does one look for crapitation buildup in the bores of the throttle bodies, short of pulling them?
 
Pull the plastic tubes (by loosening the hose clamps at each end) that go from the airbox to the throttle body.
Note that there is an O-ring on the throttlebody side, don't lose or damage that.
 
Back
Top