• 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?

Sudden, brief loss of power - 1987 K75C

JamesInCA

Member
Symptoms: When moving at speed (happened anywhere between 15-50 mph), sudden (VERY) loss of power, with no response to throttle. Bike slows, of course, but engine somewhat curiously appears to remain "on." No odd behavior of dash lights. After a few seconds, power and throttle response return. Occurred about 6-7 times in ~60 miles of riding, but "clumped" into three instances, where it happened once or twice a few seconds apart. (27-mile each way commute, with a little midday driving in between.)

My first guesses were: Fuel delivery problem. Suddenness of onset and resumption of normal response suggests electrical cause. Or maybe water in gas? That pernicious fuel-pump connector under the tank? Of course also suspect fuel filter/hoses/pump. Absence of funkiness with dash lights (and a much less catastrophic-feeling failure) suggests it's not a recurrence of my previously jiggly kill switch.

Testing today found a faulty throttle position switch, but not in a way that could obviously cause the observed behavior. The switch tests in the "open" condition. That is, on a normal switch (or so says the one on my parts bike), the outboard pin and the center pin should be closed when the throttle is closed, and open when the throttle is open. (The center and inboard connector should be closed at WOT.) The presumed-faulty switch tests open regardless of throttle position.

So, this shouldn't cause what I'm seeing ... or could it? We know some people remove this switch, which would leave the bike in the position I found in testing: circuit permanently open. If, however, the switch's internals are loose, such that the contacting bits could jiggle into a closed position, it seems to me it could cause what I experienced. Can anyone confirm an occurrence of this? Is anyone familiar with the internals of this switch? I'll pull it apart tomorrow; just ran out of daylight today.

Today's diagnostics also included:
- Check battery terminal connections: fine
- Check ground to frame: fine
- Check spark plug & fuel injector connections: snug
- Computer connector: snug, contacts clean
- Pull & inspect spark plugs: fine
- FPR vacuum hose: lookin crusty, replace next week
- Crankcase vent hose: also crusty; order & replace
- Check & reseat tank connector: Looks fine
- Check for spark: Fine on all three zylinders
- Check fuel pressure: 36-37psi (stationary, of course)
- Drain tank; no evidence of water or crud in gas (vibration damper newish; < 10k miles)

In progress, to be finished on return of daylight:
- Replace fuel filter
- Replace in-tank fuel hoses with appropriate new OEM hoses
- Replace fuel supply hose (tank to fuel rail)
- Check that tank vent pipe is clear

On receipt of parts order:
- Replace crankcase vent hose
- Replace FPR vacuum hose
- Replace fuel return hoses

(Edited to add other diagnostics performed.)
 
Your description of the bike's stopping as "more catastrophic" when the kill switch failed suggests to me an electrical connection problem that causes the fuel pump to stop pumping. If the kill switch is turned, the lack of ignition causes the motor to stop pretty abruptly, while I think the motor's running-to-not transition because of the fuel pump stopping would be a bit smoother as pressure in the fuel rail declined.

I would reexamine the tank connector and the wiring, as well as any other connectors in that part of the wiring harness.
 
David, thanks. Fuel pump power and that very connector were my first thoughts as well. It's a terrible design. I've reseated it as tightly as is possible, and replacing it with something that connects more positively is in order as well. In the meantime, I've also marked the wire positions so I'll be able to see if any work themselves loose from their current, tight position. The in-tank connections were very tight; no problem there.

If the problem does recur, I will rig a way to continuously monitor fuel pump power, either via the fuse or something around the tank connector. What's difficult here is that I can't reproduce the issue on demand, and when it does occur it's so brief (and hazardous) that there's no way to stop for diagnostics. Or maybe I ought to rig that up now, though I don't want to introduce anything potentially problematic, particularly in the area of that connector.
 
James,

I have a vague memory of reading about a K-bike tank connector failure within the last few months, but I can't remember if it was here or on the IBMWR mailing list. Edit: It was Scott's experience, mentioned below in Post #6.

How about tapping the fuel pump and ground wires to also run a 12v relay with a pair of NC contacts; the contacts would be held open when the fuel pump circuit has power, and would close if the power fails. Wire the relay contacts to a warning light up on the dash. If the power fails (at least as far up-circuit as you can make the connections), the warning light will illuminate. You might see it even in the middle of your where'd- the-power-go? full attention mode.
 
Last edited:
Classic symptoms of the under tank plug problem

So I had an 85, but unless the Germans dramatically redesigned that 4 pin connector they are problematic and reseating them hard accomplishes nothing permanent. This problem has been well covered here by people smarter than me, but the core design of the male tank plugs and the female 'tube' connectors in the wiring harness is at the root of your problems IMHO.
 
It is not just the connector that goes bad, but the pass through for the wires. It is part of the float assembly. My K75 would just shut off after 15 minutes and would not restart for 10. There is a solder connection inside the float assembly block with wires soldered on both sides, when it got hot, the fatigued solder joint would open up. Had to replace the float assembly. This may or may not be your problem.

Once I was sure it was a fuel problem, I ran a pair of wires from my fuel pump out the filler cap vent to a volt meter. The rode around until it quit and verified the voltage had gone away. Should you try this, make sure you a have good connection (as in lugs) bolted to the pump. You do not want any sparking in the gas tank! And don't ride with the fuel cap open.
 
Yeah, I'm hoping it's not that pass-through. I had another go bad, but fortunately I had the parts bike on hand to scavenge from. Otherwise, they're pricey as it's the whole dang float unit.

Anyway, fuel pump power is back in the #1 spot for causing this problem. The throttle position switch tests fine off the bike; looks like the open condition on the bike is likely due to the adjustment of the throttle shaft. It's not likely to be failing in the way it would have to in order to be the cause here.

I had thought of running wires out the gas cap to get a direct read on fuel pump power. It's just kind of ... terrifying? Still, even with voltage metered at the FP lugs, you'd have to compare it to voltage outside the tank (but after the tank connector) to verify that it's the float unit that's bad. Or check continuity from the connector to the internal fuel pump wire terminals, which was how I verified my earlier one was bad. But it was permanently dead, not intermittently.

In this case I'm not seeing a correlation with temperature or running time, but more of a jittery behavior. It comes, happens a few times, and then goes away until it comes back some random time later. I'm hoping, fingers crossed, that it's more likely in the connector. I have a four-wire Posi-Lock connector on order. Not likely to do anything else til it arrives. I finished reassembling the tank internals and fuel supply hose today, but I'm not going to ride it around until I've at least made a plausible stab at a fix. So maybe next weekend. :cry
 
Quick update, as I hate finding unresolved mechanical threads years later.

I've now had the motorcycle back together and running for a couple of weeks, after some work interruptions, weather, etc. On the day the problem first appeared, it occurred in 3-4 clusters over a total driving distance of about 60 miles. I parked it and didn't ride it until completing a few plausible fixes, as the problem was really quite hazardous in live traffic. Since reassembly, I've ridden about 450 miles under varied conditions without recurrence, so I'm cautiously optimistic it's resolved.

Work completed is:
- Replaced fuel filter (just because)
- Replaced internal fuel tank hoses (they were old; it was time)
- Replaced external fuel suply hose (ditto)
- Removed throttle position switch
- Replaced gas tank electrical connector

I believe it was one of the last two items that fixed the problem, and most likely the connector. Unfortunately, I won't ever really know (provided it doesn't recur). The problem felt like an intermittent electrical contact, and the under-tank connector is a known weak point. I snipped it off, as well as its mate on the harness, and replaced them with a Delphi Weather Pack four-wire plug.

I notice a bit more "burping" -- not really backfiring -- as a result of removing the throttle switch, but it's not bothersome. I may or may not replace that.

The crankcase vent hose, vacuum hose + protective spring, and fuel return hoses will all be replaced next time I have occasion to be behind the air box.

On a side note, I like the Weather Pack connectors, though I wish they had a real strain relief for the wires emerging from the back of the connectors. They are relatively economical if bought in bulk (5-10 connector batches), and a reasonably good crimp tool is available for $30. The contact pins and seals are available in three sizes spanning the range of 10-20 AWG, and are rated for 20 amps. I do suspect the Deutsch line might be slightly more robust and I may switch to them as my sunk cost in the Delphi is low, but the Deutsch-compatible tooling appears pricier and more application-specific. I may also check out the Delphi Metri-Pack series, as they have a line of smaller ones, have strain relief, and work with the same inexpensive crimper. (The Weather Pack connectors are not, shall we say, "compact." But, 20 amps.)

Edit: BTW, the fuel pump mounting bracket, vibration damper, and screen are about a year old, and in good shape. No crusty rubber in my tank.
 
Sudden brief loss of power - 1987 K75C

During these episodes did the tach continue to show engine RPM or did it drop to zero ( even though engine is still being cranked over by the rear wheel ) ?
If the tach went to zero your hall sensor may be going intermittent . That happened sporadically to my K75 after I'd run it enough to get it hot . As soon as it cooled off for a few seconds as I coasted off the road , it would run fine again for a while ...

Sounds like you may have already fixed it , but if not, perhaps something to investigate ...


Todd
 
The crankcase vent hose, vacuum hose + protective spring, and fuel return hoses will all be replaced next time I have occasion to be behind the air box.

The crankcase vent hose can be easily changed without removing anything else, but sometimes removing the three screws and the coil cover gives lots of extra room for the task.



:dance:dance:dance
 
I was having power loss on my 1991 K75. It turned out to be the thermostat failing and not sending information. My mechanic installed a new thermostat and it was fine.
 
During these episodes did the tach continue to show engine RPM or did it drop to zero ( even though engine is still being cranked over by the rear wheel ) ?
If the tach went to zero your hall sensor may be going intermittent . That happened sporadically to my K75 after I'd run it enough to get it hot . As soon as it cooled off for a few seconds as I coasted off the road , it would run fine again for a while ...

My tach stopped working the last time it saw rain, a few months ago. I haven't bothered to pull it apart and attempt cleaning, as the speedo does work and I don't routinely need the tach. But it's somewhere on the list to do.

The issue didn't appear to be related to ambient temperature, engine temperature, or running time. The day it happened, it happened first after about 40 minutes of riding; some hours later, when I went out for lunch, it occurred within a block of leaving the parking lot. On the way home that evening, about 30 minutes in.

When yours restarted after coasting, how did you restart it? Did you have to stop and use the starter, or did it just come back as you were moving? Mine did the latter, but I'm not clear -- because things moved very quickly -- whether it only restarted when I engaged the clutch a bit (i.e. a sort of rolling bump start), or if it spontaneously came back to life while I had the clutch fully pulled in.

The crankcase vent hose can be easily changed without removing anything else, but sometimes removing the three screws and the coil cover gives lots of extra room for the task.

Thanks! I'll take a look at that this weekend. I'd figured to just do it all when I have a reason to pull off the air box.

I was having power loss on my 1991 K75. It turned out to be the thermostat failing and not sending information. My mechanic installed a new thermostat and it was fine.

What did that power loss "feel" like? Sudden and jerky, or progressive? How long did it last? Was he able to demonstrate failure of the thermostat, or just that replacing it fixed the problem?

I did a quick check of resistance on pin 10 and it was in range for the ambient temperature. I didn't heat the bike up and check again, but it's worth doing.
 
one more idea

If the brake telltale light comes on after the stall, it means a complete electrical interupt. That is likly a ignition switch interupt and a cleaning is in order. It has happened to me several times on my 85 k100s. Usually when I am away from home.
 
If the brake telltale light comes on after the stall, it means a complete electrical interupt. That is likly a ignition switch interupt and a cleaning is in order. It has happened to me several times on my 85 k100s. Usually when I am away from home.

Yes, I've seen that, and it wasn't that. No interruption of dash lights during the power loss, and no extra lights on when power came back. When that happened to me it was a loose run/stop switch, which I cleaned and tightened up, and has been fine since. That experience was rather more terrifying and abrupt, which is why I guessed this one was more likely a fuel delivery issue with a root electrical cause localized to the fuel supply.
 
Sudden brief loss of power K75

When my engine cutout , I instinctively ( ingrained spanish 2 stroke dirt bike reflex ) pulled in the clutch and coasted off the road . It immediately restarted at the side of the road and continued on ( for a while ) before it happened again . Ambient temp was 70s . My first failure was probably 25 miles from home, then failure 15 miles after that, then 10 miles after that , then 5 miles . The last time it didn't want to start immediately , though after sitting for a minute fired right back up as if nothing happened ! This was to be a 300 miles ride day and at that point I was spooked about it cutting out in traffic.

Since it was dry day and I suspected that it was heat related ignition failure since failures were becoming more frequent .... and I wanted to continue , I removed the T shaped cover over the Hall sensor to cool it off then kept on riding . It did fail again , but this time after 150 miles or in traffic and then restarted immediately. To me that pointed to a heat relationship to hall sensor. I replaced it with used unit and it hasn't hiccuped in the past 25,000 miles . Fixed !
 
Quick ignition switch test..

Turn ignition on (don't start the bike.)
Wiggle key. Don't turn it, just wiggle it in the on position.
If the dash/instrument lights flicker, your ignition switch needs disassembly and cleaning (you can't clean it without disassembly.)
 
When my engine cutout , I instinctively ( ingrained spanish 2 stroke dirt bike reflex ) pulled in the clutch and coasted off the road . It immediately restarted at the side of the road and continued on ( for a while ) before it happened again . Ambient temp was 70s . My first failure was probably 25 miles from home, then failure 15 miles after that, then 10 miles after that , then 5 miles . The last time it didn't want to start immediately , though after sitting for a minute fired right back up as if nothing happened ! This was to be a 300 miles ride day and at that point I was spooked about it cutting out in traffic.

Since it was dry day and I suspected that it was heat related ignition failure since failures were becoming more frequent .... and I wanted to continue , I removed the T shaped cover over the Hall sensor to cool it off then kept on riding . It did fail again , but this time after 150 miles or in traffic and then restarted immediately. To me that pointed to a heat relationship to hall sensor. I replaced it with used unit and it hasn't hiccuped in the past 25,000 miles . Fixed !

Thanks, that's a very helpful description for what it could look like in a running context. I think I'll do a pre-emptive Hall sensor heat test, as it's still chilly here in SoCal, so even if the sensor is failing it might currently be getting enough ambient cooling to conceal that.
 
Sudden Loss of Pwr: K Bike: Try "contact enhancer" on big computer connector

I've solved this problem with NAPA "contact enhancer" CE-1, $70, aka Stabilant 22. The eye dropper bottle comes with "Q tips" to touch the posts and or the pinching finger tips of the computer connectors and others. The fluid will make a hissing sound as it cleans off the slow buildup of 15+ years of corrosion that also servers as a growing insulator. This stuff makes fuel injectors behave better when you touch their connector parts. It has helped numerous older fuel injected cars and bikes I've owned. I carry the bottle and Q tips to rallies in hopes of helping people like you. I've helped a few.

Sometimes just opening and closing a connector causes enough "rubbing" to clean off corrosion. Connectors are best designed if they wipe as plugged in and out. New parts installed also perform the wiping of the other connector, so the new part appears to have solved the problem when the old part with cleaner connectors would have been fine too. I also think many K bikes go for sale because the owner has trouble like yours and can't "trust" them anymore and then try some expensive troubleshooting . . . when clean connections could really make a big difference.

My 85 K100RS with 146,000 miles hasn't needed many parts, mostly just connector cleaning when trouble pops up. I have done several of the K bike tech page switch cleanings etc. as well.

Waiting a week or so for the bottle and spending $70 years ago felt bad, but since then using it has felt so good every time, many old Volvos, an old BMW 525i, several old K bikes, some old Chevy Suburbans. The fuel injection, computer and signal connectors are real low voltages so connector resistance can wreak havoc when beyond a decade old.
 
Quick ignition switch test..

Turn ignition on (don't start the bike.)
Wiggle key. Don't turn it, just wiggle it in the on position.
If the dash/instrument lights flicker, your ignition switch needs disassembly and cleaning (you can't clean it without disassembly.)

It passes the wiggle test, thanks.
 
I've solved this problem with NAPA "contact enhancer" CE-1, $70, aka Stabilant 22. The eye dropper bottle comes with "Q tips" to touch the posts and or the pinching finger tips of the computer connectors and others. The fluid will make a hissing sound as it cleans off the slow buildup of 15+ years of corrosion that also servers as a growing insulator. This stuff makes fuel injectors behave better when you touch their connector parts. It has helped numerous older fuel injected cars and bikes I've owned. I carry the bottle and Q tips to rallies in hopes of helping people like you. I've helped a few.

Sometimes just opening and closing a connector causes enough "rubbing" to clean off corrosion. Connectors are best designed if they wipe as plugged in and out. New parts installed also perform the wiping of the other connector, so the new part appears to have solved the problem when the old part with cleaner connectors would have been fine too. I also think many K bikes go for sale because the owner has trouble like yours and can't "trust" them anymore and then try some expensive troubleshooting . . . when clean connections could really make a big difference.

My 85 K100RS with 146,000 miles hasn't needed many parts, mostly just connector cleaning when trouble pops up. I have done several of the K bike tech page switch cleanings etc. as well.

Waiting a week or so for the bottle and spending $70 years ago felt bad, but since then using it has felt so good every time, many old Volvos, an old BMW 525i, several old K bikes, some old Chevy Suburbans. The fuel injection, computer and signal connectors are real low voltages so connector resistance can wreak havoc when beyond a decade old.

It's possible that cleaning the old connector might have done the trick, but I noted quite some time back that it's a mechanically unstable design in the first place. The male pins are held in place mainly by 25-year-old squishy rubber, and I've observed the action of closing the connector pushing them out of place. Replacing it with something more physically stable was already on the "when I get to it" maintenance list.

Just to reiterate, there's no evidence at this point that the problem remains unsolved. When it occurred, it happened several times in 60 miles. Since I've resumed riding after changing a few things, it hasn't happened in 600+ miles. I'm going with "solved" until otherwise indicated.

Oh, and I picked up a teensy bottle of Deoxit Gold at the local Fry's Electronics. It's marketed as a cleaner & contact enhancer like Stabilant 22 and has been used for this purpose for years by musicians. I intend to put it to the test on my dodgy tachometer.
 
Back
Top