Lost Spark while riding '78 XS650

[Melnic]

Back in town and its cold today but I took the bike for a 5 mile ride through the residential neighborhood up the street and back down so if it died on me, I could coast home. We have some open areas so I did ramp it up to 40mph about 6 times. So far so good so when I have more time I'll take it for a longer ride and maybe where warmer gear too

 

[dave_a]

I just found the major problem to my low voltage to the coil was the fuse holder. When it was making consistent contact it was down 2.5v fom battery voltage. For added fun its issues were intermittent. Replaced with a good quality blade fuse.
It's not like these bikes vibrate or anything!

 

[Melnic]

I just found the major problem to my low voltage to the coil was the fuse holder. When it was making consistent contact it was down 2.5v fom battery voltage. For added fun its issues were intermittent. Replaced with a good quality blade fuse.
It's not like these bikes vibrate or anything!

I am having a hard time recalling if I fiddled with the fuse when I found the floating one. My fuse holder is a new assembly though put on last year. I was thinking months ago switching to blade since these tiny fuses are not that common. what blade holder did you use? The one I used for my XJ650 is I think too tall to mount on top of the battery cover on the XS

 

[dave_a]

I had a Hella marine fuse over in the boating department I confiscated.
FYI - There is an older thread that talks about Chinesium blade fuses tha pass way more current than the fuse rating. The takeaway was use a known good brand quality fuse.

 

[Melnic]

so, back to the drawing board.
Took a ride today after getting back in town.
Died on me a couple miles from the house and about 1/2 mile on the 45mph road.
Died and would not immediately start. Let it sit for about 2 minutes then it started and I got far enough into the neighborhood where it was uphill from my house. It died. I pulled the spare plug I had and no spark from either side. Walked/coasted home. It did not start immediately when I got it home. I looked at the battery and it was 12.4V w/ key switch off. It was 12.65 when I left the house. I have a volt meter tied to the brown wire at the handlebars and when it died it was running at about 14v. When I was trying to start it, it was lower. I don't get a very good reading when its not running on that meter for some reason. its not tied directly to the battery, but to the key switch brown wire. It does tell me pretty good that the battery is charging.

I jumped on another bike and went for a 45 minute ride.

When I got back , with the battery sitting its voltage creeped back from sitting, It got spark and started.

Boom, I had just realized I had a battery tester. I ran the tester and it failed the battery. It dropped to 8.95v on starting. I did order a new AGM battery but its not here yet.

I think I'll also double check the rectifier using the diode test. I did have to replace that last year.

I may just do some testing with a variable power supply this week. I"m just curious how low of a supply voltage the coils need in order to make a spark.

But what I don't get is I thought that a weak battery would just make it hard to start, not cut out while running.
I can get it not starting cause the battery voltage drops below the necessary voltage to let the ignition spark, but again not cut out.

 

[dave_a]

My bike had a dead cell in the battery.
It also wasn't charging very well - maybe 13v. Not 14+. With a fresh charge I could kick start it, and it would run ok for a while. Putting around in a parking lot after a 30 min ride (little or no charge to the batt) it would start to misfire. I put the turn signal on and the engine really started to miss. Had to rev it up much higher for it to clear out.
SO - we're running a total loss system under 3000, meaning the battery is giving up its power/ state of charge. If we start with a bad battery, we have a very small reserve. Ride around under 3k rpm, we can cross the line into not enough "juice" for the coils.

 

[Melnic]

That is what perplexes me. It cuts out while I"m speeding up at at 3.5-4K Rpm I ran 5 miles in the neighborhood at about 3K and 30mph and it did not cut out. I will see if my helmet cam catches my volt meter but I swear I did not see it drop out. I'll need to wire my handle bar volt meter direct to the battery and put it on a switch. I hope its as simple as the battery but still wonder if something else is not up to spec that makes it more sensitive..

 

[Jan_P]

My take on that is that the spark is weak and the mixture gets richer changing to mid range Jetting 3-5 --. 4 k and
extinguishes the flame

Is Rotor Slip Ring Resistance measured ?

 

[Jim]

I hope its as simple as the battery but still wonder if something else is not up to spec that makes it more sensitive..

I think the battery's worth a shot... after all, it's your money, not mine.

Seriously though, what's causing it affects both cylinders.... obviously, otherwise one cylinder would get you home.
Each cyl. has it's own complete ignition system, and I know from experience one will get you home when the other dies.
Same with the carbs.... independent to both cylinders.
That puts it down to electrical that affects both ignitions... or fuel delivery.

At least a new battery will eliminate that possible cause.

Just curious, when it dies does it just up and quit, like turning off a switch?
Or does it buck, pop and fart... and generally act like you're running out of gas?

 

[Melnic]

I think the battery's worth a shot... after all, it's your money, not mine.

Seriously though, what's causing it affects both cylinders.... obviously, otherwise one cylinder would get you home.
Each cyl. has it's own complete ignition system, and I know from experience one will get you home when the other dies.
Same with the carbs.... independent to both cylinders.
That puts it down to electrical that affects both ignitions... or fuel delivery.

At least a new battery will eliminate that possible cause.

Just curious, when it dies does it just up and quit, like turning off a switch?
Or does it buck, pop and fart... and generally act like you're running out of gas?

I'll see if I can pull the video from sunday.
It dropped off for fraction of a second a few times, then sudden and full drop like I shut off the ignition switch.
Would not start, then gave it a couple minutes then it started. next time it died, it was bam, all of the sudden so as you say, its both at the same time.
I had cleaned the ignition switch before and I swear when it was failing previously, I measured 12V at the points which means the switch was working.
I could also bypass the ignition kill switch and wire in a separate one for a test as well. Just ordered a handlebar mounted switch from amazon so I can try that this week. I have bullet assortment so I can easily run a fuse to the battery, then to the switch then to the 2 12V connectors on each coil. I could even make a Y connector so that I can run both the standard kill switch plus the new switch and turn on the new switch if it dies.

As an Engineer, I want to really find out THE issue. Curiosity has the best of me on this one. I like solving mysteries and documenting it.

I have touched nearly everything in the circuit previously.
I have change points
pulled points disk out fully
New caps on both plugs
Pulled ignition switch to clean contacts
Pulled kill switch to clean contacts
cleaned bullet plugs.
New battery last fall (cheap one that is now failing battery test0
rebuilt rotor
replaced with the solid state regulator from the AMC car
replaced rectifier (It passed the diode tests yesterday)

 

[dave_a]

I didn't catch the year of your bike. But I had to adjust my regulator (see 5Twins excellent DIY), to obtain proper alternator voltage at the battery. That after installing a new battery on my 74.

 

[Jim]

I could even make a Y connector so that I can run both the standard kill switch plus the new switch and turn on the new switch if it dies.

That's how I'd play it... direct power from the battery and a switch on the bars.

 

[Melnic]

I

I didn't catch the year of your bike. But I had to adjust my regulator (see 5Twins excellent DIY), to obtain proper alternator voltage at the battery. That after installing a new battery on my 74.

Its a 78 and I removed the stock regulator (that has the voltage adjustment screw) and put in a solid state one that is not adjustable.

 

[Melnic]

That's how I'd play it... direct power from the battery and a switch on the bars.

sure will make it easier to get home if that voltage feed is the issue

 

[Melnic]

so, No new battery yet but I did do some data gathering and going to spell out some things that should be known by many here but may not be known by all (anyone correct a mistake I have and I'll edit)

So, Battery goes to the ignition switch then that output (brown) goes to the Handle bar Volt Meter and to the kill switch.
Out of the Kill switch comes the red with white stripe wires that go to a plug in the headlight, then to wires to the harness that then goes to the + side of the each of the coil inputs. = side of each coil input goes to cooresponding point. Other end of point goes to ground.
Spark activates when the cam OPENS the points
When the the points are closed, 12V is across the coil (points are grounding the - side of the coil).
Transformers (the coils) work on AC or in better terms a change in voltage. The change from 12V across the coil to 0V across the coil produces One impulse high voltage on the secondary (secondary is sent to the plug).

Wires and connectors have a resistance. In simplest terms, Thicker wires have less resistance (of the same metal type and stranding). Good connectors have lower resistance and poor connectors/connections or dirty ones have a higher resistance. Resistance is generally bad in wires but manufacturers didn't use heavy gauge wires to save $ and connectors as they age increase in resistance as they get corrosion or the crimps loosen.

So, as said, I have a handlebar volt meter at the Brown wire that is AFTER the ignition switch. but before the Kill switch.
I hooked up a variable power supply instead of the battery. This PS is strong enough (5A) to run the power but not start the starter motor.
I wanted to see how low I could get the Battery voltage and still get a spark at the plugs.
Here is a simplified drawing of the circuit.

I started by measuring the voltage at the Points with Both points open, Left closed/right open , and Left open/right Closed.
Supply Voltage 12.0V
Both Open:
Supply 12.0V 2.75A
Vhb= 11.0V
VLP = 11.28
VRP = 11.29

Left Point Closed
Supply 12.0V 4.79A
Vhb 10.4V
VLP = 0.028V
VRP = 9.13V

Right Point Closed
Supply 12.0V 4.77A
Vhb 10.4V
VLP 9.18V
VRP 0.025V

So from this, we yield that with points both open (no current going thru coils), Bike is drawing 2.75A (Tail light, indicator bulbs, Rotor coil).
Each coil static current draw is approx 2A. We also know that from measuring the open points voltage of the opposite points, we know the voltage hitting the input to the coil (9.13V) so there is a 2.87V drop in the switches, connectors and wiring from the 2A coil draw. (I don't like that!)
That 2.87V drop is from all the connections, wires and switches. between the battery and the Coils.

Next thing I did was adjust the Battery voltage at the power supply lower and lower to see how low of a Battery voltage I could get and still get sparks at the Plugs. I hooked up my spare plug and rotated the points cam so that one Point was open and the points under test was closed. I then used my finger to manually open the points under test and look for a spark on that plug. This told me how low of a battery voltage will still yield a spark. In each case, the Current went down (as the supply voltage went down)

Left Points
10.0V Battery consistent spark
9.5V Intermittent spark
9.0V weak intermittent spark

Right Points
10.0V Battery consistent spark
9.5V Consistent Spark
9.0V Intermittent Spark.

So, as we can see even a pretty low Battery voltage will still give a spark.
I looked back at my video and did some frame freezing and when I lost ignition and the bike was still in gear and motor/alternator spinning, voltages were at 13V+ at the handle bars.

I have a hard time believing the voltage at the battery is dropping too low to give a spark.
I'm more thinking Intermittent connection in the feed voltage to the coils in the harness/connectors or a ground.
Bypass switch to run battery direct to coils will tell us what is going on. If that makes it recover, its the harness. If not, its a ground.
What grounds could be bad? I don't think Battery to frame, cause if that happened the Handle bar meter would go dark.
I cleaned up the Condenser to frame ground.
Points disk to Engine contact? that would be unusual as there are so many contact points for that.

At a few amps, 18ga wire won't be dropping a volt in the distance that the wires are. Those voltage drops are from connections.

 

[Melnic]

Video from over a week ago.
The first time it cut out, you can hear it "blip" off twice before it cut out then I pulled in clutch to coast to the driveway.
2nd time I left the clutch out a little longer but it still did not recover.
It cut out on me 3 times on that trip and I pushed it up the hill to the top of my neighbor hood to push it home.

 

[Jim]

It's a long shot, but you could try running with the gas cap loose. Maybe not venting. It's not unheard of on these bikes. That said, I still think it's electrical and you're on the right track.

About all I can add to your info above is you're not taking/factoring dwell into your coil tests, so those numbers might not be accurate for real time operation. Other than that, agree to all.

 

[Melnic]

I agree that dwell is important and the voltage measurements may not be true to how low of a voltage I can go. I did take oscilloscope captures of the points. you can see how the voltages at the points spike and are not so clean while running. These scope pics are with the bike idling with the current battery. Its on the rising edge that the spark occurs. The voltage fluctuation while at 12V is from the opposite points opening/closing.
I"m going to start making wires to hook up the bypass switch and fuse to the coil power.

Yes, the day that the bike would not start and I found the spare floating fuse, I had opened up the gas cap. That is however not something that causes this symptom but more a float flow issue that is more like hesitation and slight bogging and not a total kill of the ignition. Not had gas cap venting issue before, but definitely have had fuel flow issues on another bike as well as some RC gas powered airplanes.

.

 

[Melnic]

SOB!!!!

I went out to put the battery back in and start to measure up the wires for the bypass switch. and then thought to myself, what was I missing from my simplified drawing. The fuse., yes, the fuse box and the connector that I mentioned I moved and it started to work last week. Well,
Look what I saw. I took a pic of what I saw, then with very little force, it pulled right out.
With that electrical tape there, I think someone had messed with this before. One of the worse things you can do with an electrical plug is make a tight wrapping that pulls on the outer connections.

Screw driver pointing to the red wire with white stripes that goes to the Ignition fuse.

 

[dave_a]

From the video it sounded like you hit the kill switch (which you didn't). Did I hear that correctly?
Most electrical components will have to cool off to function again. So I'd guess coils & condenser are ok, because your bike started right up again in the driveway didn't it?
I'd suggest taking your tank off, unplug the coils, plug your VOM into the harness side of the 2 wire coil brown. Ground the VOM negative. lead, turn the ignition on and start wiggling wires from the battery into the headlight shell, to the handlebar switch. When the VOM drops...bingo.
My 2 cents. Good luck