Please respond if you've NOT had a 1200GS/GSA fuel pump controller failure

[Stig]

Had 3 now - No probs

John T
- 1200GS (05)
- 7500K
- AWR
- NO.. But it was a dog (replaced with one below)

John T
- 1200GS (05)
- 24000K
- AWR
- NO.. But oil seal issues etc

John T
- 1200GS (07)
- 7500K
- AWR
- NO

However.. I've always kept the area well doused in WD40 or 3in1

 

[Tim Cullis]

Failure of two in the first 600 miles? Possibly nothing more sinister than the the law of large numbers; two fail in every 5,000 say = 0.04% or zero point four in every one thousand? Your guess is as good as mine....I guess.

No, 2 out of the 105 known failures, not 2 out of an imaginary 5000 bikes.

That's ~2%.

I always did enjoy permutations and combinations

it doesn't initially burn out it fails and cuts out, usually it will then work again if the bike is allowed to cool down.

Actually, sometimes it works again for a short time, sometimes it doesn't. With me it didn't.

 

[babu]

babu
2005 R1200GS
16000 km
All weather
Stored under roof, indoors november-march.

Fuel pump ok so far.

 

[davedyer]

2007. 1200GS.
6500 miles.
Used all days, all weather, all year.
Garaged.
No probs at all.

 

[Wapping]

No, 2 out of the 105 known failures, not 2 out of an imaginary 5000 bikes.

That's ~2%.

I always did enjoy permutations and combinations



Actually, sometimes it works again for a short time, sometimes it doesn't. With me it didn't.

Tim, you are cheating again.

You asked Dahoum for an explanation of the two 'early' failures. I gave you one (possible) reason.

You simply stating that 2 out of 105 is about 2% (which is correct) is not an explanation. It is simply a percentage. If you get more failures reported but none (or more) of them 'early' your 2% figure will drop (or rise or stay the same) will it not?

For instance, let's say you get 210 bikes but no more early failures the ratio will fall to 1%. Either way, you are no closer to detailing your reason for early failure than you were when you posed the question. In short, you do not know.

To compound your possible error. You may have found the only two 'early' failure bikes in the whole of the UK, but mixed them into only a data base of 105 known failed bikes owned by UKGSers. This creates you enormous problems when drawing any conclusions and, in passing, gets you no closer to answering your question to Dahoum. Or does it?

As you seem to like extrapolating numbers forward, based on your current study, predicting how many UKGSer bikes will fail. I guess you would agree that the early failure ratio will (by your own reckoning on the current numbers) not exceed 2% or is that conclusion somehow wrong? Whether or not 2% is an acceptable 'early' failure rate is, of course, an entirely different matter.

 

[aceranic]

1200gs
model 2007
5000km
all weather
road only

 

[Barrmill Biker]

Barrmill Biker
'04 R1200GS
30000 miles
All weathers
Used most days of the week
and as much as possible at weekend

 

[Wapping]

Message to Tim:

You might like to flick through this:

It may help with the sums.

It's more complicated than it seems (now there's a surprise) but pleasingly full of Greek*

http://en.wikipedia.org/wiki/Failure_rate

However, the possibly interesting (if inconvenient) truth is shown in Wiki's example at the bottom of the article. The maths is easy to follow. The example highlights the importance of mixing in the data from items that do not fail. The example also mentions the relevance of statistical errors (the simple example makes no subsequent adjustment) - The adjustment process would, I assume, bring in (or I assume more likely exclude) the two early failures you seem so happy with....as little more than freaks or indeed 'normal' chance failures in large numbers.

-----

*or maybe it's Double Dutch

 

[Gizmo*]

Not wishing to spoil everyones fun but depending upon how Tim intends to proceed the statistical accuracy of the data might not be important if it can be shown that the relay failure is a safety issue one.

the VOSA code of practice does not state that it has to be a certain percentage or indeed number of failed units it simply says "common to a number of units" and clarifies this with "It will be a matter of VOSA/supplier judgement to decide whether the number of units affected is sufficient to justify invoking the Code. A fixed numerical limit cannot be specified as the decision will take account of the degree of seriousness of any possible hazard involved." so the number of bikes unaffected or percentages could be totally irrelevant !!

They also have power to get info from BMW

“VOSA may wish to seek information from a supplier about safety defects that have been brought to their attention. In these cases, the supplier will pass all relevant information available, and cooperate with VOSA in establishing whether a defect is present."

Under this clause BM would have to give details of number of replaced controllers which would be accurate and not a forum guesstimate


This only applies to going to the VOSA though and you would need to show that a fuel pump relay failing is a safety issue so its worth having info on cases where this has happened and left a rider in a dangerous position rather than counting how many units haven't failed.

Please note that in that Code of Practice the onus is also on the manufacturer to invoke the procedure under the same criteria and if the data is presented to BM they should do so if they feel it is a genuine safety issue, if its later discovered that they had the data and did nothing VOSA would ask questions as to why. Most manufacturers prefer to go down the disclosure route !.

HTH......

 

[Allan Davies]

Allan Davies
GS 1200/54
25K
All weathers + a bit of fording/offroad
whenever possible

 

[Jockser]

05 1200GS
30000 km
All weathers
Garaged
No probs

06 1200GSA
35000 km (& counting)
All weathers
Garaged
No probs

 

[Tim Cullis]

I guess you would agree that the early failure ratio will (by your own reckoning on the current numbers) not exceed 2% or is that conclusion somehow wrong? Whether or not 2% is an acceptable 'early' failure rate is, of course, an entirely different matter.

Just found another.

Three failures at less than 1000 miles out of around 110 recorded so far.

 

[Wapping]

Just found another.

Three failures at less than 1000 miles out of around 110 recorded so far.

So now 2.72% of the bikes (owned by UKGSers) that suffered failed fuel controllers failed 'early'...So what? Is that good or bad or irrelevant? Whichever, it still gets you no closer to answering your question to Dahoum...

Gizmo's reply starts to get closer to the nub, perhaps, as we still don't know how you intend to present your findings / neatly collated lists.

The 'safety' attack is, of course a possible route. However, given that many of the failures (not all) resulted in a bike not starting it might be tricky to say / prove that the failure presents a significant risk?

Have you been doing anything with the non-failure list and / or asked 'your list' members whether they have owned an earlier (or subsequent) BuMW 1200 that hasn't failed?

I only ask as I suspect 110 bikes (an arguably large number and undoubtedly annoying / inconvenient number when failure occures) might to some (Vospa / BuMW for instance) look awfully small against the total numbers of GS's sold in the four and a half years since the mark went on sale.


====


PS I had an acknowledgement, from the King of BuMW, to my insnorance letter....no such courtesy from the miserable Jock at RBS.....

 

[Tim Cullis]

it still gets you no closer to answering your question to Dahoum...

Hey, it's not my responsibility to try to answer the question I posed to Dahoum

For the sake of those (one or two) trying to follow the thread, Dahoum reckoned that removing corrosion off the top of the unit would prevent failure. Nice theory, but I then asked how he explained the failure of two bikes with less than 600 miles on the clock. He couldn't.

Other schemes proposed in the past include slapping on toad slime, sticking it in a couldron and chanting "hubble bubble, toil and trouble".

Have you been doing anything with the non-failure list

No. I'm not really interested in how many bikes haven't yet failed.

I know BMW is aware of the failure thread. This thread serves no purpose other than possibly being something that BMW could take comfort in.

 

[araspitfire]

No. I'm not really interested in how many bikes haven't yet failed.

I know BMW is aware of the failure thread. This thread serves no purpose other than possibly being something that BMW could take comfort in.

So... this is really all about an agenda???

No one is interested in figuring out whether or not there really is a problem?

Or if we are all victims of the "most will cry, few will proclaim happiness" reason for distrusting our gut instincts on matters such as this...

I would say that if you get a hundred guys that say I've got a few miles on my bike, and not had a failure... That number should be used multiplied times 10 (say)... and the number of people reporting failures should only be multiplied by 2 (say). A much higher percentage of people will read these two threads who have had failures, than of those who haven't...

Only BMW will know how many controllers are moving off the shelf, unfortunately... and we don't trust them obviously so their actions on the 'problem' are disregarded. (yet they have the true stats)

Like many, I do feel that there probably is a problem (though I don't carry a spare), but I also know that how I feel is directly related to the numbers of reported failures... seeing lots of reports of highish mileage non-failures brings things back into perspective...

This is important IMHO... in my years, I've seen so many development resources wasted dealing with "problems" that were really not problems, but were blown out of proportion by the community at large, due to the very loud voices of a few unhappy people, and the very quiet multitudes of those that don't care....

It's all well to try to convince BMW to fix the controller... but if the true failure rate is closer to the 'norm' than our perception... wouldn't it be better to allow them to continue development on things that will make the bike better.

Most people don't do adventure travel because they feel the world is a very dangerous place... based on seeing only stories of danger and destruction on the news every day... but people (like you) who have traveled around a bit must know that our perceptions from 'the news' and reality rarely match up...

Al...

 

[Alte Wolf]

My 06 12 GSA no fuel pump failure 18000 miles. Bike now in possession of my youngest bro 3

My 08 12GSA OK so far, early days I suppose.

Bro 2 has 06 GSA also and as far as I know no fuel pump controller failure either.

 

[Tim Cullis]

No one is interested in figuring out whether or not there really is a problem?

You can't be serious!
- 118 failures so far
- some owners have had two failures
- two owners have had three failures.

It's not about an agenda. I decided off my own back to compile a list after experiencing three failures first hand in Morocco (two on the same day, one with my 1200GSA, another on a 1200GS), and then reading about some poor sod stuck in Tajikistan for two weeks.

If you read the various threads about fuel pump failures (as I've had to do to compile the list) you will find that many people have complained in the past and asked why someone hasn't done something about it. Until now, nobody did.

I'm still adding to the list. I haven't decided yet how I am about to proceed, though I've been offered some suggestions.

I'm a reasonable guy and recognise that BMW may not be able to produce a quick fix out of the hat, but there are a few things that could be achieved without delay:
- free parts/labour on FPC failure even if bike is out of warranty
- repayment of bills where owners have had to fork out of their own pocket
- dramatic lowering of the price of the part where owners purchase them for peace of mind.

Tim

 

[Wapping]

For the sake of those (one or two) trying to follow the thread, Dahoum reckoned that removing corrosion off the top of the unit would prevent failure. Nice theory, but I then asked how he explained the failure of two bikes with less than 600 miles on the clock. He couldn't.

The truth is you cannot explain them either. There again, you set the rules.

You observation about the two early failures might only be true if the two (now three, I think?) bikes' control units that failed had no corrosion....Do you know that they didn't?

Simple low mileage is perhaps no indication of anything. For all you know said bods had forgotten to tell you they lived in a salt mine, the bikes being alive with corrosion. Even then it might not be true. The two (three?) might simply have failed for a reason totally removed from the cause (s) the other 107 failures. Your lists, sadly, simply remain as lists at the moment.

The truth is, I suspect that the early failures are little more than statistical blips, balanced by failures (or no failures) in some extraordinarily high mileage bikes. No different perhaps than the bod whose engine let go of its life oil at an early date; sh1t sometimes happens. I think that is why some statistical modelling excludes the top and bottom ranges.

Be that as it may, at some time, you will I guess have to do something with your list. What's coming up on the menu?

 

[Wapping]

You can't be serious! 118 failures so far with more to add in. Some owners have had two failures. Two owners have had three failures.Tim

Forgeting of course:

(a) That some owners have had no failures at all.

(b) That some 'failure' owners might have previously (or now) owned (or own) 1200GS's that hadn't (haven't) failed at all.

It's still just a list Tim.

I share your concerns that 118 people are annoyed, upset, inconvenienced.

That they broke down in the desert sands or on the Croydon by-pass makes little real difference....

......or is it a point scoring, I can piss higher up the wall than you can, thing? I broke down in Rabid-el-Flea-bag, that's 50 points.... you only broke down in Nether Wallop, that's 10?

 

[Tim Cullis]

The truth is you cannot explain them either. There again, you set the rules.

Err.. they failed because they failed. And knowing why a unit might have failed is no comfort.

What is clear is that the problem has been going on for many years and hasn't been resolved as the many failures with 08 models demonstates.

If you really don't care, butt out.