Magneto experts needed ('68 Johnson 9.5hp top cyl intermittent spark)

[Joe_the_boatman]

I picked up a fairly clean 1968 Johnson 9.5hp yesterday, and after adding fresh gas and putting it in a test tank, she started right up. Well, sort of. Ran rough, raw fuel was coming out the exhaust, and then died when throttled down. I checked spark (inline spark tester, air gap) and got spark on bottom cylinder but none on top. Ran a minute or two more just for grins, and then lost spark on both cylinders! Packed it up and called it a day.

Today, pulled flywheel to check magneto. It was all super clean, points looked clean, no cracked coils, wires in good shape, all connections tight.

Primary resistance on each coil (to spark plug boot) measured 4.16kΩ & 4.17kΩ, which told me wires and (probably) coils were OK. Swapped condensers, top still had no spark (but at least it ran, the bottom had spark again), so condensers are OK. Cleaned points with 200, then 500 grit sandpaper, set gap to 0.020".

(As an aside - Fluke's "Troubleshooting Outboard Motor Magneto Ignitions" guide says "In any event do not use sandpaper. The grit will imbed itself in the contact metal, rendering the points useless." While leeroysramblings.com says "Best way to clean points is to...Use a strip of 400 grit (black) sandpaper... and polish them shiny bright." I went with Leeroy on this one, but if they're ruined, I'm ready to replace them anyhow).

Back to the issue.... Once points were cleaned, resistance across contacts measured:
Points open: ~1.0Ω.
Points closed: 0.0-0.1Ω

Inline spark checker showed spark on both cylinders. Attached wires to plugs and it ran great (both cylinders firing) for about 30 seconds, then sputtered (running on one cylinder) and died when throttle reduced to idle. Raw fuel in exhaust. Checked spark again, and top cylinder again had no spark. Dang!

Swapped spark plugs and coils, the only parts left between the two systems that haven't been swapped. Cleaned points again and set to 0.020". Meter showed 0.0-0.1Ω again with points closed. Double checked all connections and buttoned her back up.

Started it up, but still no spark on top cylinder. So, no change.

What am I missing here? I swapped everything from the running cylinder to the dead one (except plug wires and points, which tested OK with a meter), and still have a dead cylinder. I don't like throwing new parts at undiagnosed ignition problems, but don't know what else to do here. I could probably replace everything for around $100, but that's what I paid for the whole engine.

(PS - Before someone suggests it, I didn't get the plug wires mixed up - one has a metal tag stamped "TOP").

So, what next?

Video, running on one cylinder, for grins:

 

[racerone]

Check plug wires where they take a 90 degree bend under the magneto plate.

 

[Crosbyman]

Back to the issue.... Once points were cleaned, resistance across contacts measured:
Points open: ~1.0Ω.
Points closed: 0.0-0.1Ω

point open should be infinity not 1 ohms

also check primary wire to the points ....for chaffing against the rotating shaft
as to wires if you are going to refurbish the electricals you should install new plug wires (metallic core only 7mm)

Engine $100 + $100 parts is cheap investment for years of pleasure to come up to you

 

[Crosbyman]

btw disconnect the kill switch wires (2 clips) while testing

 

[McGR]

Breaker points will measure infinite with the coils disconnected. With the coils connected the primary winding is in parallel with the points. So, with the coils in the circuit, about 1Ω is normal when the points open.

 

[Crosbyman]

obviously but not how I test things . at minimum it is best to simply disconnect the coil's ground wire which may be easier than fiddling with the point screw but that would leave the condenser connected which personally I prefer to be disconnected.

imho ...all components should be individualized (isolated) incl. condensers to measure their characteristics.

I know some meters are precise enough to measure the solid ground from the point contact. The condenser is also in parallel and could be shorted to ground . it needs to be tested individually hopefully at high voltage (250-300v) to measure any internal leakage.

 

[Crosbyman]

Leeroy's Ramblngs is coming soon

We are doing some maintenance on our site. Please come back later.

www.leeroysramblings.com

Leeroy's Ramblngs is coming soon

We are doing some maintenance on our site. Please come back later.

www.leeroysramblings.com

 

[Joe_the_boatman]

Breaker points will measure infinite with the coils disconnected. With the coils connected the primary winding is in parallel with the points. So, with the coils in the circuit, about 1Ω is normal when the points open.

Exactly - I should've clarified that this was with the coils connected. With coils out, open circuit across contacts.

 

[racerone]

Sorry----These magnetos were installed on Johnson / Evinrude from about 1950 to the late 1980's.----Very reliable and easy to trouble shoot.

 

[Joe_the_boatman]

obviously but not how I test things . ...

imho ...all components should be individualized (isolated) incl. condensers to measure their characteristics.

I know some meters are precise enough to measure the solid ground from the point contact. The condenser is also in parallel and could be shorted to ground . it needs to be tested individually hopefully at high voltage (250-300v) to measure any internal leakage.

I agree with all that. The 1 ohm across open contacts stuck in my head as the reference I used when comparing readings across the freshly cleaned point contacts. Each time they were cleaned, I didn't necessarily disconnect everything.

My meter has capacitance capability, though I've never used it. Since the problem didn't follow the capacitor after swapping them, I figured it to be good. It'd still be good to have some data on it though.

 

[Joe_the_boatman]

Sorry----These magnetos were installed on Johnson / Evinrude from about 1950 to the late 1980's.----Very reliable and easy to trouble shoot.

Yeah I agree. And I'll look at the 90° bend under the plate today as you suggested earlier.

It just bugs me that the plug wire ohms out as it's supposed to, so if the minute voltage from the meter is enough to make continuity, you'd think the 10k+ volts wouldn't have a problem.

I have plug wire, crimper tool, and boots from other jobs. I'm not necessarily being lazy in initially not replacing them, just don't want to risk more work if something else broke when putting in new wires (or lose a screw or something stupid like that.)

After sleeping on it last night, I'm convinced the wires are giving me problems somehow. Most likely shorting to ground somewhere, through cracked insulation (although the exposed wire is nice and flexible). These 9.5's vibrate like crazy; probably rubbed through the insulation somewhere.

 

[Joe_the_boatman]

Leeroy's Ramblngs is coming soon

We are doing some maintenance on our site. Please come back later.

www.leeroysramblings.com

Leeroy's Ramblngs is coming soon

We are doing some maintenance on our site. Please come back later.

www.leeroysramblings.com

Thanks for those. Already read through Leeroys' stuff (even printed it), but haven't watched that video yet.

 

[McGR]

obviously but not how I test things . at minimum it is best to simply disconnect the coil's ground wire which may be easier than fiddling with the point screw but that would leave the condenser connected which personally I prefer to be disconnected.

imho ...all components should be individualized (isolated) incl. condensers to measure their characteristics.

I know some meters are precise enough to measure the solid ground from the point contact. The condenser is also in parallel and could be shorted to ground . it needs to be tested individually hopefully at high voltage (250-300v) to measure any internal leakage.

I agree. Just wanted to clarify why the OP was seeing about 1Ω with the points open.

 

[Crosbyman]

These 9.5's vibrate like crazy; probably rubbed through the insulation somewhere.
your model should have vibration dampeners on each side if not maybe kit would help

btw I worked on a 4hp last winter same magneto basically...... one cyl not firing
drove me nuts swapping parts coils etc... then Eureka... fire on both plugs never did find the problem

so... don't worry just keep at it.

testing capacitance with a meter is a good start but....catch up on MR.Mohat's articles Part 1-4

Magazine Articles | Western Reserve Chapter

see Leroy' ...
Vibration Dampener : This powerhead does a lot of shaking when running at a trolling speed. In 1967 it came out with a dampener system, which involved some metal brackets with fiber washers and springs. There was one mounted on the starboard side being attached by two of the carburetor base bolts. The port side unit was buried up under the recoil starter. These were attached to the motor and with the other end to the inner part of the side skirts. It appears that this was their cure, without changing any other parts.



A reader of these articles found a New Old Stock retrofit damper kit and installed them on his motor. He said it was a much easier job than he had anticipated. There is adequate room to slip them in on the starboard side, and also on the port side after the starter recoil unit is removed, without other major parts removal. The only real issue was where to drill the mounting holes in the shrouds. Fortunately, the retrofit kit came with OMC templates for this. Having used it once, he does not believe that it is truly necessary, as even after being exceedingly careful in using the template to locate the holes, they (the holes) looked to be almost 1/8" out of the ideal position. He actually thinks one could do as well to pre-assemble the two units, lightly bolt them in place and then carefully estimate the places to drill.

 

[Joe_the_boatman]

Thanks everyone for your comments. Replaced both plug wires and she purrs like a kitten now. Racerone called it spot on - insulation was rubbed through at the bend when the wire passes through the plate. I should've planned on changing those in the first place. It was about a three hour job; had to remove the armature plate, which meant disconnecting the throttle linkage and carb/intake, going slowly to not drop any hardware into the lower cowl (AKA the bottomless pit), which would've really slowed things down.

Next is impeller change and lower unit oil change, and then she'll be ready for the lake.

 

[Crosbyman]

Another happy ending thanks for your feedback !