81 xs650 dies while cruising.

[LeftyKash]

Still having problems with the bike.
Tried a new rotor- I thought this would be it. when the new one came in, the magnet was notably stronger than the old one, even though the old one had a magnet glued onto it.

I tried a new R/R, same as the original one- doesn't go above 3.5 or 3.8

At this point I went for a ride- about 10 miles in, I went to change lanes and the bike cut in and out with my turn signals. Every time the light came on, the bike started running rough, when it came off, the bike would do ok. I turned around to head home, but after another couple of miles, the bike wouldn't go over 3k rpm. My bike was still showing 13.5vdc at the battery, the second voltmeter installed between the red/white wire and ground at the coil was showing less voltage, like 11vdc or so. it was moving up and down with the turn signal.

When I finally got home, I put it on a timing light and I could see the timing going crazy when it should be advancing. Instead of advancing the light was showing anywhere from slightly retarded to full advance. Idle was fine.

I tried ordering two different CDIs to do the gonzo CDI. I'll include pictures of the little harness extender that I made for it. (in the pictures you can see the reserve lighting unit jumped out, it looks like I've got the headlight signal running to ground, but it's not)
both showed up with floating grounds (high resistance between top left and bottom right pin)
both would run just fine at idle, but would not advance.
Tried switching the polarity between pins 1 and 4 didn't fix the advance, didn't take notes on whether or not the bike started.
i tried a 5k resistor between the advance green/white wire and the ground wire, same results.

I checked grounding between the handlebars, headlight bucket and frame and the showed some resistance (with the bike off)- my meter was acting a little strange actually, showing a high <1K resistance and low (2-3ohms) sporadically, but I temporarily ran jumpers and it didn't change the engine advance.

I wanted to bring up the voiltage at the coil, since it seemed a bit low, I tried jumping the key switch. That didn't do anything.

I'm going to take my new battery in to be checked where I bought it, and maybe order a regular TCI ignition off ebay and a mike's xs wiring harness.
Since I can't get it to fail any diagnostics, I'm not really sure what to do with this thing. I'm getting pretty frustrated, to be honest. I did notice what appeared to be some holes in the epoxy on the stator, but I don't know if that means anything. I'll include pictures.

 

[Jan_P]

Have not read all posts but that is something to work on

At this point I went for a ride- about 10 miles in, I went to change lanes and the bike cut in and out with my turn signals. Every time the light came on, the bike started running rough, when it came off, the bike would do ok

The reason it cuts off I do believe is to little power on the ignition Box or System ..
Either the flasher circuit shorts to ground or the extra consumption lowers it below what the TCI Needs

Here where I live I would disconnect the Turn signal circuit -- Using hand signals ; out in the traffic for a while
It was legal not sure if it still is
But in town I still do it so I don't forget cancel it .

Not read it but a Ground is the first thing to check is ground clean and With some grease .. Very Important.

Please clarify --> I tried a new R/R, same as the original one- doesn't go above 3.5 or 3.8
Is that 13.5 and 13.8 ? Volts

Please Measure actual charging voltage across Battery at some rpm .. 13.5 May be battery voltage and no Charging taking place
Then do the same -- lights on Measure across the battery can also look if the light varies with rpm
Then turn the flasher on lights on and Measure again. across the battery

 

[LeftyKash]

Have not read all posts but that is something to work on

At this point I went for a ride- about 10 miles in, I went to change lanes and the bike cut in and out with my turn signals. Every time the light came on, the bike started running rough, when it came off, the bike would do ok

The reason it cuts off I do believe is to little power on the ignition Box or System ..
Either the flasher circuit shorts to ground or the extra consumption lowers it below what the TCI Needs
I'm with you there. I didn't mention it above- but I I see the same 'cutting out' if I toggle the hi-beam switch.
Here where I live I would disconnect the Turn signal circuit -- Using hand signals ; out in the traffic for a while
It was legal not sure if it still is
But in town I still do it so I don't forget cancel it .
I can go for a run and try this. I'll just pull the turn signal fuse.
Not read it but a Ground is the first thing to check is ground clean and With some grease .. Very Important.
The battery ground is clean and greased. The engine ground on the top right side of the cylinder looks OK. The ground at the ignition coil is clean and greased. Any other ones that I'm missing?
Please clarify --> I tried a new R/R, same as the original one- doesn't go above 3.5 or 3.8
Is that 13.5 and 13.8 ? Volts
Battery voltage is usually 12.9 when i'm starting the bike. when I rev the bike up it goes to 13.5 with the original R/R 13.8 with the replacement one.

Please Measure actual charging voltage across Battery at some rpm .. 13.5 May be battery voltage and no Charging taking place
Then do the same -- lights on Measure across the battery can also look if the light varies with rpm
With headlight on the the headlight gets a bit brighter as t he battery goes from 12.9-ish at idle to 13.5 when revved
Then turn the flasher on lights on and Measure again. across the battery

Will do

Next steps:
go for a ride with turn signal fuse disconnected, see what happens.
measure voltage at battery with flasher on.
I have been wondering if a fault in the TCI is dragging down the voltage for the bike, but I'm not really sure if there's a good way to check for that.

 

[Jan_P]

Will do

Next steps:
go for a ride with turn signal fuse disconnected, see what happens.
measure voltage at battery with flasher on.
I have been wondering if a fault in the TCI is dragging down the voltage for the bike, but I'm not really sure if there's a good way to check for that.

If this was me I would measure first ..saving a Pushing the bike home if not charging.
I cannot say for sure but the weak charging giving the ignition problems is common
No matter what Ignition.
TCI is difficult to test for an average person .No Service instruction for that

# 17 there

http://www.xs650.com/threads/1983-xs650sk-points-location.60010/

 

[LeftyKash]

If this was me I would measure first ..saving a Pushing the bike home if not charging.
I cannot say for sure but the weak charging giving the ignition problems is common
No matter what Ignition.
TCI is difficult to test for an average person .No Service instruction for that

# 17 there

http://www.xs650.com/threads/1983-xs650sk-points-location.60010/

A friend of mine also suggested putting an Ammeter between the regulator and the battery to see if the charging system is putting out enough amps- The thinking was that I could turn on the signals or the high beams and see if the Ammeter reacts. I ordered a cheap 30A automotive unit. Looking at the 81 650SH schematics, I think it would go between on the red wire coming out of the regulator- not sure though.

 

[Jan_P]

Not sure but I believe it will work either current is going into the battery charging ... or out of it discharging
Why not place it near the fuse there is an opening possible ..if it is a temporary test
Old British bikes had Ampere meter.
like below
Not heard of it used on XS 650 s where Voltmeter is common
please read more fex there

http://www.xs650.com/threads/wiring-in-a-volt-meter.6245/

Perhaps the reason being that an Amperemeter is in line in the wiring
And if it breaks .it might stop the transfer ( Is it open Circuit in English ) whereas a Voltmeter sits parallel.
If it breaks the circuit can still work

 

[xjwmx]

Don't know where this thread has wandered, but the original post is describing the classic TCI failure. If you were to look at the solder joints with a magnifier you'd see some faint rings around the component leads. The cure is to just resolder every joint on the board, adding a tiny bit of solder to each to make sure it doesn't turn into a cold joint. It only takes 15 minutes. You can solder right through the coating on the board. Decades of vibration causes the joints to crack.

 

[LeftyKash]

I checked the TCI visually and didn't see anything. I also checked the diodes with a meter and they came back good. When I tried to start the bike to check the turn signals it wouldn't start. The harness and TCI that I ordered arrived, so I'm putting that in now and hoping that fixes everything. I've got an ammeter, which I may throw on the bike if I'm not seeing 14.5VDC with the new wiring. I'll still planning on checking the turn signal load on bike too.

 

[LeftyKash]

I've got the new harness in now. I found the Mike's XS harness for the '81 had the blue/black wire for the headlight/safety relay tied into the wire for the tail lights. This had the headlight always on, and tied into the main fuse instead of just the headlight fuse. I corrected mine, but buyer beware! Also- not sure where the speedo/tach illumination lights wires go on this new harness- they don't light up, and aren't tied to the blue/black wire like they should be.

When I started the bike without the headlight mounted, everything was good- showed 14.4 volts when revving the bike! awesome!
I put the headlight back on to go for a test run and the bike wouldn't go over 2 or 3k. I put a timing light on it, and saw no ignition advance. Didn't write down the voltage showing with the headlight on, but I think it was over 14. Not sure where to go from here.
I did some amp checks:
Flashers = .2A off 4.6A on
Brake = 3.2 A (3.8A on main 20A fuse, 7.2A with brake on)
Headlight 3.08A
Ignition circuit .3A

My initial problem was the bike dying briefly and starting right back up, while travelling on the freeway.
Now my problem is that the bike won't revv over 3k.

Here's what I've done:
New ignition coil/caps/plugs
new rotor from Mike's XS, had a noticeably stronger magnet.
new regulator rectifier (still only showed 13.8v)
New harness/used TCI ignition (now showing 14.4v)
What should I even check? There's not much left! - I may try putting a magnet on the new rotor, because I see where that's helped people in other threads. My old rotor definitely had a weaker magnet than the new one, but my thinking is that if the new rotor is thinner or has weaker shielding between the charging current and the magnet than this may help.

 

[Jan_P]

Thinking out loud --does it run on both cylinders. ??
if one cylinder falls it can be problem revving up it needs to be done with a soft touch .

Charging appears be there.
How does the strobe light function when the rev is below that max 3 k
both sides
Does both headers get warm.
Checking ground ,,, and perhaps new plug wires. Connected with some grease.
A video perhaps.???

Again when lights off the bike runs ..please measure what voltage across battery
Turn the light on .please measure what voltage across battery see if it drops under 12 V and sabotages the TCI operation
because of a short in the light circuit.

 

[5twins]

My buddy's '80 SPII has the same "headlight always on" issue using a MikesXS harness as yours did. Don't know about the speedo/tach lights, we'll need to check that. Could you go into a little more detail on exactly where you found the blue/black wire tied into the blue tail light wire? This may be my buddy's problem too.

 

[LeftyKash]

Sure. I found the 'bad' splice above the left side airbox. I'll attach a picture of the location where you can separate the harness. I didn't get a picture of my repaired harness, but here's what I did.

The 2 blue/black wires get spliced together. They connect the safety relay up to the RLU.
The solid blue wires all get spliced in a second group. The one up front comes from the keyswitch and heads out to the three coming from the back.

I couldn't find any low resistance connection to the gauge lights and it didn't look like they came on- I think they showed 4.6ohms to the keyswitch maybe. I wasn't taking notes. Someday I'll unravel more of the harness and see what's going on there.

I emailed Mike's XS to let them know they had a bad harness or a supplier problem.

On an unrelated note, what's a good material for taping up these harnesses? The tape used seemed a little less 'gummy'/slippery then regular electrical tape. I just wire tied mine every few inches as I was unravelling the tape, but that won't work for the long term.

 

[5twins]

Thank you for that, very detailed. I'll refer my buddy to this thread. We'd like to retrofit a headlight on/off switch to his bike but can't do so with the headlight on all the time.

 

[LeftyKash]

Did a few checks tonight, maybe got it running with the old rotor. Hopefully I'll get to take a test ride tomorrow.

my notes:
Both cylinders definitely firing. Checked by spraying with a water bottle.
I measured voltage at two different places simultaneously.
Keyswitch:
brown wire coming out of the keyswitch connector and left turn signal ground connector,and once at the

battery:
connected positive to negative terminals

Keyswitch on, bike not running. Right off the trickle charger;
11.9at key
12.4at battery

No headlight, bike runs just fine.
no headlight, idle
13.9v battery

no headlight 4k rpm
14.4key
14 battery

With headlight, bike now only stumbles at 2-3k
headlight, idle
key 12.8
11.5 battery

headlight 4k rpm
Key 14.5
Battery 12v
Let off the throttle and battery settled to a higher idle voltage than before. shows like 13-13.2

I'll attatch videos of rolling on the throttle. Today with headlight the bike stumbles and takes off instead of hitting a hard stop at 2-3k. it runs just fine with no headlight. I think the improved 'with headlight' performance is because i had the bike on the charger all night. Before my last test i had been troubleshooting the new wiring harness, so the battery was a little lower (but not much).
After doing all this work, i removed the new Mike's xs rotor to put on a magnet. My brother suggested putting the o.g. rotor (that the p.o. had put a magnet on previously) just to see what it does, since I've changed so much since starting this project. I did, and the bike worked just fine, even with the headlight on. I saw the charging voltage climb too. Didn't write down the values though.

It's funny because the o.g. rotor has a weaker magnet, even though there's a glued on one. Perhaps just the fact that it's closer to the pickup is all that matters in this case.

 

[Jan_P]

Thank you Sir

I guess your 20 A fuse is loose --- weak springs there ????
grind and grease ..

If you get 14.4 V you have charging
Not knowing your wiring but that should be Impossible I believe
As a a first assumption that is wrong
headlight 4k rpm
Key 14.5
Battery 12v

At no situation can there be a higher voltage somewhere in the system than across battery via main fuse as far as I know
Please look at below
From battery there goes a wire the small one to the 20 A Main fuse that goes directly to the regulator rectifier

as far as I know if the charging system puts out 12 V or more I always find that measuring across the battery
Is must be wrong otherwise there can be lower voltage and current across the main fuse than in the system
Normally it is a Top --Down system Alternator output is max regulated into the battery and the consumers in the system causes a voltage drop
with zero at the minus on battery

Please check that wire and that the fuse is sitting properly
+ Battery ground service
Please check the connection of the headlamp figure out why it has more voltage than at the battery is it connected right ?
If that does not help I would disconnect the thick wire going to the starter from battery and kick start see if that helps
Please come back

 

[LeftyKash]

Yeah, I thought it was pretty weird. Here's a couple of thoughts as to why it might read that way: Perhaps the cheap voltmeter I was using at the keyswitch is out of calibration. The meter I used on the battery is calibrated annually, so it should be just fine.
My other thought is that I need to measure between the battery positive and ground instead of between positive and negative terminals.
Either way I'll check it tonight.

The springs on the fuse box main fuse are pretty firm- it's a new box that came with the new wiring harness.

 

[LeftyKash]

I tried again with the meter between the battery leads, and between battery + and ground. I got good readings this time. Not sure how I buggered it up last time.
13V @ batttrery with bike off
1400RPM Key 13.1 Battery 13.3
~1800RPM Key 13.5 Battery 13.6
3500RPM Key 14.4 Battery 14.4
4000RPM Key 14.3 Battery 14.4

I rode about 30miles, and the voltage was pretty solid at 14.4 if I had any movement going. I have a second gauge temp installed at the ignition coil and it shows 13.8 solidly while cruising.I did notice a little 'pause' in the ignition when I turned my turn signal on, on the freeway. Only with the first turn signal pulse. when this happens, I'm still showing 14.5V on the keyswitch and 13.8 at the coil.
After sitting with the bike braked and signal on at stop lights, the bike will cut in and out a little bit with the turn signals. Not really badly, but it's there.
Keyswitch meter shows 12.3 idle, 11.9-12 with brake plus turn at idle

When I got home I quickly checked draw on the turn signals at the fuse.
Single turn signal is 1ish amp on the meter.

At this point I'm pretty convinced that it's noise from the alternator interfering with the pickup coil signal. I put some LED taillight and license plate lights on the bike to reduce the load. I'll probably do LED turn signals and headlight in the next couple of weeks. I'm thinking this will get rid of the last little issue. I'll probably resolve the gauge lights messed up in the new harness over the winter, unless it really starts to bug me!

Since I now have a 'spare' rotor I'm debating selling it online, or maybe getting a heat-resistant rare earth magnet to glue on it and keep it laying around.

 

[Team Junk]

This really sounds like a ground issue. Only thing that is common with the TCI, headlight and turn signal circuits.

Keyswitch meter shows 12.3 idle, 11.9-12 with brake plus turn at idle

I think 11.9 is low to run the TCI. Perhaps take a look at the stator.

I tried ordering two different CDIs to do the gonzo CDI. I'll include pictures of the little harness extender that I made for it. (in the pictures you can see the reserve lighting unit jumped out, it looks like I've got the headlight signal running to ground, but it's not)
both showed up with floating grounds (high resistance between top left and bottom right pin)
both would run just fine at idle, but would not advance.
Tried switching the polarity between pins 1 and 4 didn't fix the advance, didn't take notes on whether or not the bike started.
i tried a 5k resistor between the advance green/white wire and the ground wire, same results.

Maybe take a look at the pickup windings ?

I rang out the ignition coil, with the bike cold.

Don't think I've heard that except from my Ma Bell buddy's

 

[LeftyKash]

I'm with you on the grounds, but I'm not sure where to check. I tested everything to the ground strap on the battery.
Headlight bucket to battery strap is good.
harness ground at frame/coil is good to the battery strap
engine strap to frame up under the tank is good

This really sounds like a ground issue. Only thing that is common with the TCI, headlight and turn signal circuits.
I think 11.9 is low to run the TCI. Perhaps take a look at the stator.

I'm not sure what to check other than the resistance and voltage checks I did when the bike was using the old harness. It passed them all, but I can check again. I've already thrown so many parts at this thing, I'm hesitant to just buy a stator without a 'failed' test.
6/22/21
stator voltage check, between the wires. 11.2v 11.5 v 11.5v @1800rpm OK.
stator resistance check, .6ohm .6ohm .7ohm OK.
I'll check the brush voltage too.

Maybe take a look at the pickup windings ?

They tested ok, the wiring looks ok (what i can see of it) and they show .750k ohms on both sides to black. seems ok. The face of the pickup coil looks ok too.

Don't think I've heard that except from my Ma Bell buddy's

Well what do you call it when your meter calls to tell you that your circuit is complete?
I worked for a while a technician in an automotive manufacturing facility where everyone used that interchangeably with doing any kind of voltage or resistance check.

 

[Jan_P]

I am a bit worried about that also.

Keyswitch meter shows 12.3 idle, 11.9-12 with brake plus turn at idle

Mr TJ is concerned about ground ..and that is the perhaps most common problem
I have a substandard wiring and have ground drawn directly to minus on battery via an extra wire.
For testing one option for testing could be to draw both power and ground from the box
To battery .. Not permanently there just for testing and then removing since the parts might not stand that all the time
For ascertain that is right
Charging appears to be there and battery is tested right ?? not to old.???
Are we sure it is both sides or one side only ?

I am at --- the TCI has not the right physics ,,power / ground