Hall Effect Magnetic Sensor

[R65_Mark]

Not sure if this is a ride report or a technical commentary.  Rode 2,137 miles on my ’84 R65 last year without losing a beat.  Quick warm up ride this weekend to get ready for an oil change and I’m stranded without a spark ten miles from home.  Call to a buddy had one favor used up but me home on a trailer about an hour later.  Diagnostics tonight pinpointed the hall effect trigger as the problem and a Honeywell 2AV54 replacement ordered from Newark a few minutes later ($18.53 plus shipping).  Hopefully should be back up and running next weekend.  Given I’m digging into the ignition system, from reading on the subject in the past few days it seems like I should take the time and re-seat the ignition amplifier on fresh conductive paste as preventative maintenance.  I installed a Dyna coil and new plugs/leads mid last year, so I think I’ve got everything covered, but if anyone has any other ideas about what might be worth inspecting or replacing while I’m digging around, please let me know.

Mark

[Bob_Roller]

Hopefully, Monte (montmil) will see this, and correct me, if I'm wrong .

There is a quick test to check if the Hall sensor is at fault, or the ICU .

I believe you remove the electrical connector going to the 'bean can', then place a wire between the power supply from the ICU, to the output wire of the Hall sensor .

Don't know the wire colors, so don't try it until you find out for certain what they are .

If the ICU is good, the spark plugs will have a spark jump across the spark plug, when to touch the wires with the jumper .

You're bypassing the Hall sensor with this test .

I don't know if this is the test you performed or not .

[R65_Mark]

Bob,

I’m happy that I’ve nailed the hall effect sensor as the culprit.  Did the disconnect battery, front cover off, disconnect hall sensor, spare spark plug connected to one lead and grounded well, reconnect battery, ignition on, put center (trigger) pin on connector under front cover to ground a few times and got a nice fat spark at the plug every time – bad hall effect sensor confirmed.  Trick looks like it is going to be getting the old sensor off its mounting in the ignition sensor/can and the new one on and spliced in to the old wiring.

Mark

[Bob_Roller]

From the looks of it, the new sensor needs to be riveted to it's plate, there's two stubs of metal coming off of the sensor, that goes into mating holes on it's mounting plate, these need to be mushroomed on the bottom side to secure the sensor .

Don't know if the material is soft iron or not, I have two sensors, and I know the material is magnetic .

[R65_Mark]

I'll take pictures of the new sensor, what I find when I get into the can, and how I end up fixing things.  The joys of spring.

Mark

[nhmaf]

This thread on the airheads website has good tips on taking the bean can apart, etc.
http://www.airheads.org/content/view/271/98/

I've also found other scrapings on the net on replacing the sensor - I think that there are a couple variations of the bean can which can make the process a bit tricker/easier to deal with.

[darrylri]

Then, there is this thread in our own forum:

http://www.bmwr65.org/cgi-bin/yabb2/YaBB.pl?num=1262401521

[montmil]

One helpful tip regarding disassembly; it is not necessary to remove the wrapped wire coil and anchor pin on the drive dog end of the bean-o can to replace the Hall sensor and refresh the internals. to

I have overhauled three bean cans within the past year, replacing Hall sensors bought from Newark Electronics. The rivet "legs" of the new Hall sensor can be peened as conventional rivets. A sturdy bench vice with an anvil, a tapered punch and a ball peen hammer will get the job done properly. Note that the hole in the sensor plate is slightly larger than the rivet leg diameter. This is a good thing...

I found that it was helpful to begin forming the shop head of only one of the rivet legs, then stop when you can still move the sensor by using firm pressure with your fingers. Temporarily refit the metal shield. Check the position of the Hall sensor. You do not want sensor/shield contact. The 'window" in the shield is the key.

After confirming your approval of the sensor position with the single rivet lightly snugged down, remove the shield and proceed to form the shop head on the second rivet/leg. Check shield clearance again. Return to the first rivet and complete forming the finished shop head.

You should be good to go with the new Hall sensor.

The mechanical advance mechanism is the anachronism of BMW's bean can design. A modern electronic ignition trigger with an old school ignition advance. OK, it works fairly well...  Clean off all the surface rusting. The weight pivots need lubricant but it is my understanding from lengthy research that the 'floor" of the advance plate should not be lubed as the nylon feet are designed to function without being greased up. A very light wipe of a non-evaporative lube is what I use. Keeps rust at bay.

Hope this assists you with your adventure.   Monte

[R65_Mark]

Monte,

Thanks for the tips.

I’m going to take my time and do the general clean up you suggest.  The sensor will be here tomorrow, so I’ll make a start after work.  Unless I hit a snag, should be back up and running during the day on Saturday at the latest.  I’ll report back with any news/questions.

Mark

[R65_Mark]

All,

Sensor arrived today.

The first two pictures here show the new part.  The third picture is from Darryl Richman’s post on this same topic earlier this year and is what I expect to find when I dive in deeper.

Here’s my question for Monte, Darryl and others that have been here before.  What have you done to splice the new sensor wires to the old harness?  There does not look like much room for anything exotic.  I’m thinking cut and solder wires together with a heat shrink wrapper.  Any other thoughts or guidance appreciated.

Thanks.

Mark

[montmil]

Here’s my question for Monte, Darryl and others that have been here before.  What have you done to splice the new sensor wires to the old harness?  There does not look like much room for anything exotic.  I’m thinking cut and solder wires together with a heat shrink wrapper.  Any other thoughts or guidance appreciated.

Stagger the wire splices so as not to end up with a big ol' lump under the heat shrink tubing.

Solder is the best technique as long as you do not allow excess molten metal to wick up the wire and make it overly inflexible. Radio Shack has a small heat sink tool that's helpful. Also at the Shack, you'll find tiny 22 ga bare metal crimp connectors if you choose to go that route.

Shrink tubing on each wire, then a larger bit over all. Big Tip #1: Remember to add the tubing prior to joining old and new wires. All my wire joints are outside the bean can shell. Big Tip #2: Avoid crushing the bean can wire leads between the front cover / timing chain case when refitting said front cover. Good way to ruin a fairly simple job.

Monte

[darrylri]

I actually made my wire connections inside the bean can so as not to compromise the nice sheath that runs from the can to the plug.  I staggered them a bit, but I needed to preserve as much room as possible at one end so I could keep the heat shrink tubing away from the soldered area.  I did not shrink them all together.  It was a fussy job, and I was very thankful for my little Horror Fright "third hand" tool with its too alligator clips and a magnifying glass.

[montmil]

I actually made my wire connections inside the bean can ...

Yep, that's doable but a true test of one's patience and soldering skills. Bending the soldered replacement wires to lay neatly without operational interference inside the can is also something to consider. Properly done, the bean can's wire sheath looks "factory" and the concourse judges will be none the wiser. Either way gets you back on the road.   Monte

[darrylri]

Monte, if you'll have a glance at my "free bike", you'll know that concours judging is far from my mind on this.  

However, the weather protection that the stock sheath provides looks much better than what I could achieve with shrink tubing, and that is what drove my decision.  

I was very careful to make sure that the wires inside the can stayed routed through the clip (visible in the photo), so that they will be out of the way of the spinning umbrella/shield.

Probably it doesn't matter which way you do it, so long as you do a workmanlike job.

[R65_Mark]

Monte & Darryl,

Thanks for the ideas.

I’ll make my mind up which way to go when I tackle this on the weekend.

If I remember, I’ll take some before and after pictures and let you judge the outcome!

Mark