1100GS pops, bangs and backfires under load or overrun

TL;dr version, the bike runs fairly smoothly when cold, but as soon as it warms up there's a really bad misfire/backfire when the engine is put under any sort of load. What does this sound like?

Hello, I'm pretty new here and I'm just bringing a '97 1100GS back into operation after what appears to be a fairly significant layup. The fuel in the (original) tank was disgusting, there was plenty of rust, the fuel pump filter sock was choked with fine rust and bits, so I've replaced the pump and filter. Most of the fuel pipes were tie wrapped rather than clamped, and many of them were perished almost to the point of splitting. Throttle cables were all sticky, the fast idle cable wouldn't return, etc etc, so I've replaced them all too.

Anyway, I got a replacement tank, re-plumbed everything, filled with fresh fuel and took the bike out. It started easily enough, and settled down to a ridiculously high idle (1800 or so). Smoothish at idle, but under any sort of load it just stuttered and surged and if put under a lot of load the revs build veeeerry slowly in the lower gears, but with lots of popping in the airbox and a backfire or two if held at sustained high throttle in a higher gear.

Today I set the valve clearances and rocker end float, nothing was too far out apart from the exhaust valves a 'tad' on the tight side on one cylinder but broadly correct. Balanced the throttle bodies, again they weren't far out, but the air screws were way too open leading to the high idle.

I had some very weird behaviour once I dropped the idle down at first: the engine would idle very smoothly then every second or so it would appear to do a massive misfire and then recover, but it was doing it far too regularly for it to be a simple misfire, it was as though the engine management was trying to pull the idle down by cutting the fuel or something. As I went around the engine to troubleshoot anything misplaced, it settled right down into a smooth idle.

So I've adjusted the throttle cables, the air screws, and the engine was running beautifully and smoothly, and revving cleanly off load. I couldn't take it out for a test there and then as I had other things to do. This evening, I took the bike out, it ran pretty smoothly when cold (I could accelerate through the gears without any misfires/stuttering) but as soon as the first dash appeared on the RID/temp gauge it started misbehaving under load again.

[edit] I also forgot to say I notice that it was missing/surging when engine braking from a reasonable speed in a lower gear (so putting it under deceleration load, if that makes sense).

I'm tempted to say it's ignition, but what do people think? It's my first BMW boxer, so whilst I've got tons of experience with Japanese bikes of all ages, I'm not used to the foibles of these yet.

try pulling the front cover off , where the belt is , and look at wiring for hall sensors.

theoneandonly said:

try pulling the front cover off , where the belt is , and look at wiring for hall sensors.

Click to expand...

I shall do. The wiring all over the bike is pretty crusty, to be honest. Lots of cracking sleeving around the loom in places etc. Does hall sensor failure or wiring issues manifest itself like this? Also, forgot to say, the tacho doesn't jump whilst all this misfiring is going on..

Got the cover off, the wiring that I can see looks ok:



What does concern me slightly (but I can’t verify without making up a Hall sensor tester) is that the plate looks to be hard up against the adjustment on ‘fully advanced’:



I’d say the hall sensor is not original looking at the cleanliness of the plate and wiring, so someone has been in there before..

Would an incorrectly timed Hall sensor cause the running issues?

Co2 pot probably needs adjusting…………

Ian J Hartley said:

Co2 pot probably needs adjusting…………

Click to expand...

Speaking from experience?

thecarfixer said:

Speaking from experience?

Click to expand...

Sorry, I meant Co potentiometer. Nowhere in your right up did you mention adjusting it.

Early generation of Motronic used a Co potentiometer, later models (1150 series), the motronic evolved to using a lambda probe in the Exhaust to tell the system when to cut back fuel when the bike goes into closed loop.

Ian

No, you’re right, I haven’t adjusted it as I’d assumed it was set up correctly.

I don’t have a CO meter, so setting it might involve some guesswork! I know that the code plug determines whether the ECU is set to use a lambda probe in the exhaust or refer to the CO pot, and I know where the lambda socket is in the wiring loom.

Tempted to weld a bung in the exhaust to fit a lambda sensor and code it to use the lambda map!

I’m taking it for granted that as you have replaced much of the fuel system because of poor fuel and hoses, you also checked the injectors for a nice spray pattern……..

I haven’t as of yet. One would assume if the injectors are supplying fuel for cold start enrichment, they will be ok, but they could be problematic at the shorter pulsewidth of hot running. But then I’d expect the idle to be poor, which it is not.

that wiring looks good to me . mine was cracking , and shorting out , sounded like the injectors were cooking bacon!as the ignition switched on and off .( pulsing the injectors)

theoneandonly said:

that wiring looks good to me . mine was cracking , and shorting out , sounded like the injectors were cooking bacon!as the ignition switched on and off .( pulsing the injectors)

Click to expand...

Yeah, makes sense. The tacho is rock steady too. I suspect that it's been replaced as it looks fairly shiny in comparison to the retaining bolts.

https://www.ukgser.com/forums/showthread.php/405964-R1100GS-Misfire-Help-Needed-Please

The symptoms this poster reported sound quite similar to mine, so I'll pull the injectors and take a look. Probably makes sense to put some Bosch EV14 injectors in at any rate, given how everyone seems to rave about the improvement they make. @Ian J Hartley deserves credit for mentioning this point, I'll update if an improvement is found.

Check the resistance across the injector terminals, they should be 16. Any higher or lower or uneven and it’ll pop and fart and run like a pig.

Both 15.5 ohms. BUT, I notice evidence of water/corrosion on the right hand side injector terminals, and the connector pins also.

So I'll clean those with a little diamond file I have, and re-test.

Cleaned the injector terminals. Not much better, if at all.

Pulled the same injector and cranked the engine, there is (obviously) fuel being pulsed out of the injector, but I wouldn't say the spray pattern is much of a pattern (a tiny jet).

I think I'll get the injectors out to be cleaned and retest.

Injectors came back from Injectortune (https://www.injectortune.co.uk), I've used them in the past and they do quick service. The raw figures don't tell the story, the spray pattern was 'ok' and they flowed 312cc - cleaning them has bumped them up to 318cc @ 3bar.

Fitted them to the bike today, and could immediately tell it's better. It happily revs out, has a good torquey midrange and felt nice. But the idle wasn't great. (see https://thumbsnap.com/i/Lh6dLKeE.mp4 to get a good idea).

I went over the throttle bodies again, slack in the throttle cables, then balance with my carbtune with the brass air screws, after which I tried to set up the throttle balance by getting the vacuum to be equal as possible at 1500 then 2000 rpm. Try as I might, I couldn't really get it to idle smoothly - as per the video it was showing a fairly regular miss. I kept playing with the cable adjustment, and it all suddenly smoothed out again and stopped the regular misfire. Back to the same game as before!

Took it out for a run, and in general it runs alright, although I stalled it when gently feathering the throttle and slipping the clutch and it didn't really want to start without a good amount of throttle. Settled down and ran 'ok' for the rest of the test. What I have noticed is that if you tickle the throttle just above idle - literally taking up the slack in the cables - it goes really stuttery and misses lots. I'm pondering whether the TPS is knackered right at the start of its range..

In between this, I've also done some test adjustments of the CO pot, and can see the behaviour if I go too lean, and too rich, so set it roughly half way between those points where it seems to idle nicely.

So afterwards I've done some more reading and read about setting up the TPS, so checked the voltage on pin 1, engine off but ignition on. 700mV! I've adjusted it down to 340mV tonight. Quite clear that someone has been fiddling all over with this bike - to no benefit at all.

I took it out for a test from a cold start this evening, it started easily and settled to a decent high idle on 'choke' and ran smoothly. Idles pretty well, but has a noticeable stutter/surge at very low throttle openings (i.e. cracked just above idle again). The bike isn't unrideable anymore, but it's still fairly unpleasant on 'tip in'.

I think my next plans are:

* Measure the behaviour of the TPS to see if it's had it - at over £200 for a new one it's not something I want to 'trial and error' but as a primary sensor for load to determine fuel and ignition timing, I think it's important
* Build a little LED test tool to set up the Hall Effect Sensor as I don't think the timing is right (given how wrong everything else has been set up, this won't be right)
* Rebuild the throttle bodies - the rattle like hell, so I can't imagine very low speed running is ever going to be right until they are tight as they should be

Rebuilding the throttle bodies is very worth while. I did this on my R1100GS and found that it made throttle balancing much easier and accurate. I used the 'Dan Cata ' kits as sold by motobins. There are a couple of YouTube videos showing what to do!

Also worth checking that none of the throttle cables are damaged. A fraying or sticky cable can affect things.

GORILA said:

Rebuilding the throttle bodies is very worth while. I did this on my R1100GS and found that it made throttle balancing much easier and accurate. I used the 'Dan Cata ' kits as sold by motobins. There are a couple of YouTube videos showing what to do!

Click to expand...

Good to hear - the bike clatters a fair bit from the Tb's at idle, so that's an important job.

GORILA said:

Also worth checking that none of the throttle cables are damaged. A fraying or sticky cable can affect things.

Click to expand...

They're all new, I replaced them as the throttle operation was heavy and sticky..

0.7 on the TPS probably was VERY quick and drank fuel

It really should be set to 0.38 ish Volt

Slightly above rather than slightly below

This is the baseline for your injection it basically sets the value for the injector duration and everything is up from there

If the ECU believes the throttle is open wider than it is? It will inject the amount of fuel according to the throttle angle it is being told

Less than the 0.38volt and it reduces the fuel amount accordingly leaving a weak initial set up