Getting a bit fed up with this bike. Seems like every time i go to use it (Which, granted, is not very often), I have to spend a fortnight fixing it before i can take it out.

1983 K100 RT 8V. Bike refuses to start. Bad battery, replaced with new one, and engine just floods and refuses to fire at all.

Have had the plugs out, all VERY wet. Dried off, cleaned, and burned off the fuel with a torch. At this point, if i disconnect the fuel pump and the injector electrical plugs, and try to turn it over, it will catch and fire a couple of times.

Try again, with above reconnected and it floods again. Fuel runs out of the exhaust where it joins the back box.


Today, I threw a few hours at it, and discovered the following:

If i plug everything in, quickly tap the starter, and let the fuel pump prime, THEN, unplug the fuel pump, choke off, throttle about half way open, and press starter again it actually will start. It runs rough as hell, will rev, but smokes like anything. Eventually the remainder of the pressurized fuel will run out, and it will die. It wont idle. Suspect an amount of fuel still in the down-pipes.

Have tested the fuel pressure with my inline gauge, and am getting a consistent 36psi. So not the regulator.

The bike lives under a cover in front of my house, so conceivably has got a bit damp over the last few months.


------

Now, my assessment of this is that it MUST be a sensor issue, causing far too much fuel to be delivered. It all worked fine when i last used the Bike (November), and i'd suspected temperature sensor issues (It wasn't - It was bad valve adjustment), so i'm pretty sure thats ok.

Before i start ripping it all apart in frustration again, any pointers on where to start? Anyone had similar?

Thanks

ToGGoT

Hi,
I would really recommend running through the troubleshooting chart in the info section, a lot of members worked on that to help refine it and it works well because it leads someone with a problem through a logical testing process with a series of hold and test points, does it do this or does it do that? if it does that, then move to this step etc.
Regards,
K-BIKE

It looks like a problem that I encountered few years ago.
It was the FI computer plug not tightly pushed in.
In this case, you would expect that the injection would not work at all, but it's the opposite. The fuel injectors were constantly open.

When did this happened?
After replacing the battery. (same as you)

Best thing to do (after checking the FI computer plug) is to unplug one injector and check if you have a solid ground or if it's pulsing. You will have to plug a LED to see if it's flashing. A multimeter won't be of any use for this test.

CF

I had a similar problem. Flooded engine etc. I did not go as far as unplugging the fuel pump etc, but seemed the same symptoms.
Unplugged temp sensor off the stub behind the radiator.

Do the checks as said above but also fit new plugs, it only takes a small amount of raw fuel to destroy the plug. I have tried all the cleaning and torch things and never had much success with the K,

Toggot,

if you want me to pop round on my K that won't be a problem at all. If needed would be able to swap parts around to prove a suspect part etc, as plymouth isnt that big your bound to be only a couple of miles away!

Good on you twincarb.
Regards,
K-BIKE

+1

good work twincarb

Thankyou Twincarb, thats very kind!

I'll keep you posted how i get on. I actually do have a spare fuel computer somewhere in the loft - finding it might be another story though.

Is there a quick way to test a fuel computer to see if its working?

ToGGoT

The easy answer is no there is no easy way... That's because you would have to be able to put signals in that the computer is seeing.

The only simple way is to try it on another machine or try a known working one on the bike.

The injectors have a permanent live to them if I recall correctly and it is the earth that is switched on and off... If you have a permanent earth on the circuit all it will do is allow fuel to pass through the injectors.

The other possibility is that the injectors have seized open but I don't think that is likely as for the amount of fuel your suggesting it would suggest all 4 are bad.

I am sure that its possible to check that the computer is pulsing the earth which I think could be done by disconnecting all the injectors electrical conections and checking it with a scope or analogue meter you would expect the meter needle to pulse... a scope would probably show it better...

Someone will correct me if i am wrong i hope...

Unpluging the power of one injector is enough. Remember, the 4 injectors are opening and closing at the same time.
If you have a 12v LED, just plug it instead of the injector. If the LED is a standard 3.3v you need a resistor to drop the voltage.

You know what? I reckon i have some LED's and some 470 resistors in my bits box...

Will keep this updated.

ToGGoT

good news there then toggot, if your able to measure the resistance of each of the injectors as well and show the readings on here it would be handy... In Berts trouble shooting guide it says they should be 4 ohms.... however on various other posts it says 16 ohms... maybe a typo there...

I have just checked all of mine as a comparison and they are 16 ohms.... I wonder if that is my fuel economy problem.

and Bert has just posted as well.... 4 is divisable by 16 and there are 4 injectors so can see it could be easy to get the head thinking that way esp when it's to cold to head out....

If the resistance is miles off I am guessing that would suggest the injector wouldn't close but wouldn't be a definitive answer....

Which then gets me thinking if all 4 injectors are out of the engine but connected to the harness if the ignition is on you could see fuel leaking from them..... (No smoking and no pressing the starter button at this point!!!)

I think you are right, the injectors should be 16 ohms.

At the bottom of the FI page, I have the following note...



When I wrote this troubleshooting guide, the bike was in the shed not accessible for any testing (You cannot do anything when the temperature is around -20 ).
It maybe time to correct some of the mistakes (or typos).

All four injectors read 16 ohms.

With an LED, and 470 ohm resistor connected to the lead from injector #4, the led comes on solid the second you press the starter, and stays on until you release the button - you then hear a click (presumably the FI relay) a few seconds later, and the LED dims.

This behaviour is NOT shown with the ECU unplugged - The LED does not light at all.

I cant immediately source my spare ECU - i think it might be at work, so will have to dig it out tomorrow. Unless anyone has any further ideas?

Thanks for the help so far - looks like we're onto something!

ToGGoT

ToGGoT wrote:All four injectors read 16 ohms.

With an LED, and 470 ohm resistor connected to the lead from injector #4, the led comes on solid the second you press the starter, and stays on until you release the button - you then hear a click (presumably the FI relay) a few seconds later, and the LED dims.
ToGGoT

Does the LED dims or start blinking?

Edit:

Sorry, that wasnt clear at all - when the FI relay clicks off, the LED goes out. At no point does the LED lose brightness, or flicker.

ToGGoT

If the EFI computer was sending the right signal you would expect the LED to have flashed. So it is pointing back to the EFI at the moment....

If you can't find your one give me a shout and I will happily bring mine around to confirm that yours isn't behaving!

Well, I still cant find my spare FI Computer, but, having spoken to my local garage, I decided i wasn't going to lose anything from cracking open the existing one.

It was quite damp. [/Understatement]

Having dried it out on the radiator, there is quite a bit of the white crusting from the inside of the lid having been deposited on the board by the moisture evaporating, and one corner is a bit brown where presumably the liquid has pooled.

My brother has offered to take it to work tomorrow and soak it in their ethanol tank, followed by the hot cupboard, which had worked for a few mobile phones... So nothing to lose at this point really...

Will update this thread with results.

ToGGoT

At least your deff getting there then... lets hope that the components haven't fried on the board.

I do wonder if the controller itself is used on some of the BMW cars of the 80's as well as our bikes....

i think you found it .....good luck !

even better, get a new (working ) secondhand one ....

Well, she only bloody starts!


Was very hesitant, and i ended up with choke off, holding it on partial throttle to get it to fire, but i think i can put this down to a lot of unburned fuel in the exhaust, and having sat for a few months.

Am really quite chuffed with the BMW basic circuitboard - it has survived surprisingly well bearing in mind the amount of water in there!

Ran very smoky for about 10 minutes before it started to clear out, but again, this is fairly normal for this bike!

Thanks everyone - will add this all to the experience pool!

ToGGoT

Nice one.... best we see you back out there soon the weather today was ideal for riding.... Have you got any ideas how the water leaked into the controller in the first place?

The most common way for water to get in is by using a pressure washer.
Pressure washer + Motorcycle = Disaster

Haha! Definitely no chance of that! I am not a "washer".

Probably just the winter months sat out under a dodgy cover. I have a better cover now, so hoping it wont happen again.

Anyway, MOT this morning. Failed on a front tyre, and that's it, so pretty chuffed. Discovered on the way up the lane that the speedo didn't work - Another excuse to take the instruments to bits i suppose!

Took them off when i got home, had the back off, and poured (And i really did pour - all over my desk) a LOT of brown water out. That would explain the blacking to the gear indicator then.

Its all in bits sat on the radiator at the moment, and will have a good clean - the seal was totally shot, so i'm not surprised it leaked...

ToGGoT

ToGGoT wrote:
the seal was totally shot, so i'm not surprised it leaked...

ToGGoT

I always replace mine with an O'ring (I am not sure about the section, but I think it's 3/32"). They are cheap.

I just made an O ring from 3mm nitrile and it works fine but I think the 3/32 would be better as its a bit smaller.