Mysterious R65 behaviour.. totally stumped.

[ambrose78]

Hi Folks,

First up hello all. I'm Pete. I live in Wollongong in Australia.  I became an R65 rider at the start of the year when I fell in love with a black '81 model when looking for my first bike.

So far the bike has been great - have riden about 2000km's and also passed the rider test with flying colours.  

This morning however when returning home after about an hour long ride on the highway the bike conked out on me (Thankfully whilst coming down the hill on the way home - if a breakdown is going to happen that is the best place for it).

The best way to describe what happened, I was coasting down a rather long hill and turned off into my street (also a rather long down hill), when trying to apply a little more power to speed up after slowing for the corner there was nothing there kinda like no fuel however I have an 70-80% full tank.

So I flicked the tank to reserve position in case of a blockage to the fuel line - no difference

Further experimentation:
clutch in    - engine stalls.
clutch out  -a lot of backfiring and some rather impressive flames shooting out the exhaust and no power.

After getting the bike inside I check the fuel lines - all okay - plenty of fuel flowing through those, also smells of fuel in the garage after cranking the engine (along with flames shooting out the exhaust).

Pulled the spark plugs out - no signs of being fowled or indeed flooding.

Then noticed a rather strange thing, with the ignition on and the plug grounded (no cranking with starter motor) - there are intermittent sparks flying across plug gap when the sparks fly the Tacho needle also pulses(!?). The frequency can be as quickly as 2-3 per second or as long as every 3 seconds.

The tacho has always pulsed ever since I owned the bike.

Anyone else experienced this before?.
Could the timing chain be shot?
Could it be the ICU? Or a seemingly unrelated diode somewhere in the mysterious BMW circuitry?

Any thoughts appreciated.

Cheers
Pete
:-?

[Bob_Roller]

Welcome aboard,

I would first start looking for a loose electrical connection .

Yeah, I know that doesn't narrow it down a lot.

Try wiggling the ignition key around to see if it has any effect.

About the only way you're going to get a spark across the sparkplug, is if the is an interruption of power on the ignition circuit, causing the magnetic field around the coil to 'collapse', and inducing an electrical charge in the secondary windings of the coil.

With the engine not running, the Hall sensor in the ignition canister is not sensing anything, so I'm going to have to say a loose connection somewhere.

Do you have any other electrical issues ?

[nhmaf]

The flames coming out the exhaust indicate that the bike is pumping air/fuel out the exhaust which isn't getting burned, so unless something catastrophic happened to your carbs, like a jet needle falling out - which seems improbable - the bike isn't getting regular plug firing sparks.

This could be a bad ignition coil if it still has the original coil, as the 81-83 years are notorious for the dreaded "crack-o-matic" coil problems - several on this forum have had this issue in the past year or two.  Often with the cracked coil syndrome, the bike runs OK until one has a moist/damp day, and then it is dead, but if cracked or burned through they can exhibit other weird symptoms of erratic firing.   Another possibility is the ICU, which fires the coil based on the crank position sensor signal - if the wires to the sensor are loose then the ICU will fire the coil intermittently.  If it has been rather warm where you are, it might be that the ICU is overheating - the back of the ICU has a metal plate in it, which is supposed to be in intimate contact with a silvery aluminium alloy heatsink that it bolts to.  If the heatsink paste has all dried out it might not be able to dissipate enough heat into the alloy heatsink are get too warm and misbehave.   Or you may have a loose connection, bad wire in the wiring between the ICU and the coil.

If you haven't ventured under the fuel tank yet, this might be a good opportunity for you to examine the condition of the coil and the ICU under there.   this thread will give you some idea of what to look for as far as the coil goes- it could possibly have cracks or little holes were an electrical arc can short to the frame.   There are other threads on this somewhere in the tech section, but here is a link to one from me that may help - it talks about the coil and the ICU on my bike:

http://www.bmwr65.org/cgi-bin/yabb2/YaBB.pl?num=1195434246/3#3

[Justin B.]

I would suspect something in the area of the ignition pickup (hall sensor) and associated wiring or ICU since these spurious pulses also cause a twitch at the tach.  It's my thinking that the coil is being told to fire indicated by the twitch in the tach needle.

[larstorders]

If you think the ecu is dicky (and it sounds rather like it is) a replacement can be found on most any MkIII Volkswagon Golf or similar vintage Audi. They clip straight on. i have one myself   . Check 'Snowbum's ' site for more details and help with boxers. Good luck and congrats on joining the R65 owners. Especially with a well mannered R65 which waits until it gets you home before falling sick.

[vestandpants]

trying not to repeat what's been said, I agree with the above, Ive had problems with the coils in the past especially when its damp causing loss of power and erratic behaviour.

[ambrose78]

Thanks for all the advice.

pulled my coil off - its a newish looking Bosch jobbie which the p/n ends in 203 which thankfully rules out the crack o matic coil problem.

I'm going to find a new ICU today (and heatsink paste)  and fingers crossed thats the problem.

Cheers
p.

[ambrose78]

New ICU now in. success - The tacho doesn't flick and the erroneous sparks and tacho flicks have stopped, but the wee beastie is still not running. I have tested the trigger wire on the harness according to the info on this page  http://bmwmotorcycletech.info/Ignition.htm and the icu is working to plan - short the pin and spark fires everytime.

Anyway it now looks like the sensor CAN is the problem cause when assembled and cranking there is no spark. So priced a replacement one from BMW = 1.5 legs so looks like will be rebuilding it. I'll take some photo's as I go.


Additionally when doing the testing inside the cover I accidentally managed to short the diode board with the engine casing (doh!) and now the charge light doesn't come on. I have changed the light bulb but still doesn't work so looks like I will be looking for a new diode board.

Does anyone know if there are diode boards that are more resistant to accidental shorts? Perhaps the omega one?

Cheers
P

[steve hawkins]

I do not think you should be shorting you electronic ignition.  They do not like it.

Diode boards also do not like to be shorted - sounds like you have toasted it.  

Be careful you might be causing more new problems than solving old ones.

Rule of thumb is to disconnect your battery and test connectivity, rather than fiddling with a live system - at least as far as the ignition is concerned.

Mine is a 1979 version has the old points bean can system where you can short and test the spark.  Post 81's it is not good practise.

[nhmaf]

Hmmm,
One of the "mantras" that we all recite when working under the front cover and the diode board is to always first disconnect the battery to avoid shorts - I know, too late for that now but take solace in that many of us have "been there and done that" before, too.    None of the diode boards can handle being shorted out under 'live' conditions, though you may find that some of the aftermarket boards are a little less pricey and have better heatsinks on them than the BMW one.   It is possible that you may have only 1 or 2 blown diodes that could be repaired, but you need to be handy with a multimeter and soldering irons.

If you or one of your friends are handy with electrical stuff you can build a little tester gizmo with which you can verify whether the bean can/ignition trigger is working properly before tearing it apart or paying some significant money to have someone repair it (new ones are really pricey).  The gizmo could be built for a few dollars.

HEre is a link to a very long thread about one of our member's working on this problem, and one the later pages of the thread are some schematics of the gizmo - it is rather simple but might save you some $$ and aggravation.
http://www.bmwr65.org/cgi-bin/yabb2/YaBB.pl?num=1223843560/47#47

[larstorders]

Ooops. Unfortunately, old Volkswagons are not a ready source of replacement diode boards. The boards can be rebuilt though.
Hall Effect triggers are available in the guts of old Volkswagons. Rebuilding the 'Bean can' is an easy job for an experienced sparks, darned fiddly for the rest of us. I've looked at it, but not attempted.
Good luck

[ambrose78]

my wee beastie runs again!!!


Thanks everyone for their input. Ended up doing the  $19 Jaycar sensor (P/N zd1900) fix following the instructions on this page http://www.deathstar.org/~flash/ig_trig.html . The actual replacement which really isn't too hard as long as you have a decent utensil to push the pin out of the shaft. It does pay to be very methodical noting where all the spacers and washers. I took me about 3 hours working reasonably slowly to do.

hehe thankfully the charge light problem fixed itself..... well turned out that in hast I had forgotten to replug in the regulator after having to remove it to put the ICU in. .

Anyway off for some riding

Thanks again

[nhmaf]

Good on ya, Aussie Pete !

Taking your time and being methodical is what it is all about.  These machines are not so complex - which is a good thing, but sometimes in their simplicity they can be confusing, too !

I tried to follow the link in your post but it comes up  404 - not found?

[MrRiden]

Added an "l" to the end and it werks
http://www.deathstar.org/~flash/ig_trig.html

[Rob Valdez 79 R65]

I fixed AussiePete's broken URL, just to save other readers' some trouble.


And I am anal...