Help GSF250 Bandit - ELECTRIC BLUES

Hi Guys,

Thought I would share with you my adventures with a $500 1994 GSF50 Bandit that I bought in a non running state because it "needed a new CDI".

The CDI was only the beginning of the issues, so I thought I'd share some of the learnings I had during this project. I'm armed with a parts fiche in Japanese, a roughly translated and brief description of some service data, and the workshop manual for a similar bike (GSX250F).

Upon picking up the bike, I took it home and got to work. I had already ordered a new CDI which was soon due, but more on that later.


image hosting free

DRAINING THE TANK
First I drained the tank, discovering what a Vacuum-operated petcock was in the process.
(Petcock which requires vacuum pressure to allow flow when set to 'On' or 'Reserve')
The petcock also had a setting "PRI".
Google told me that this was 'Prime', used to refill the carb bowls should they be empty for some reason.
It bypasses the vacuum lock which is in place for On and Reserve. This means that the engine doesn't need to be turning (thus generating the vacuum) for fuel to flow.

I discovered, however, that the vacuum bypass on Prime was not functioning!
(Set petcock to prime, but no fuel flows from the hose leading from petcock to carbs)

So, I create my own vacuum by connecting a length of clear PVC hose to the spigot at the back of the petcock, sucked on it and then crimped said hose with some needle nose pliers and a shifter.



Fiddly, yet effective - tank successfully drained!

More to come very soon.

 

CHECKING ELECTRICS

With the New CDI arrived, I plugged it in and gave it a hopeful turn over before realising that it is for the wrong model. I bought wrong ass CDI like a goddamn idiot. 32900-11D00 instead of 32900-11D10. This cost me a good amount of time screwing around with the wrong part.
I was a bit confused when the CDI I was replacing had a different serial number to the original!
This was hugely frustrating for me, and I've since ordered a newly made unit from RegulatorRectifier.com
I decided that the cost of the unit (~$320 with postage) was offset by the fact that it has new parts, rather than being a 20yr old box god knows how close to death.

I also opened up the old CDI just to have a peek... may even consider having it repaired by a solderer.



While waiting for that...

Next step was reconfirming what I did before buying the bike; checking for a spark. Battery is flat, so jumped off car to perform tests. (I know, I know.)
To check spark, I pulled the leads one by one and grounded while hitting the starter motor - No Spark. *!*!*!

I was making sure to only do one lead at a time. Given it is a wasted spark system, figured that each pair (1-4, 2-3) completed a single circuit.

To eliminate the Spark Plugs did a screwdriver test: still no spark. *!*!*!

At this point, I had read about a few people diagnosing electrical faults, and coming up with issues from Safety Switches and bad batteries. I hooked up a trickle charger overnight. It gets to the 5th of 6 charging stages, and holds a steady voltage now.

*!*!*! I've since learned that this is a super bad idea. You can destroy your coils if the spark plug isn't properly earthed. I've purchased a spark plug tester that seems to work by inductance - similar to a timing light.

 

CHECKING THE SAFETY SWITCHES

Safety switches on this bike operate on the ignition coil circuit (not the starter motor - why don't they

do that? so much easier to test!)

PRO TIP: I wish I had realised this far earlier, but to check to see if the safety switches are preventing spark, just check that the ignition coil circuit is closed.
Unplugged the positive wire from the coil, and used a multimeter to check if DC voltage is flowing from this wire to ground (or check for continuity between this wire at the coil, traced back to the CDI.)
Then check that the return wire to CDI (Negative wire) from coil has continuity back to CDI.
My coils have a mark on each terminal for + and - as indicated in pic (can't see it but location shown).



When I ran through this test the first time, it didn't work and I eventually discovered that the bikes Side stand switch wasn't actually working.
I confirmed this by tracing the wires from the side stand up to a connector and then testing continuity when the side stand was up, and again when down. The result was the same both times: something was wrong. I ended up simply removing the switch altogether, and twisted the wires together to close the circuit.


By now I know that the wiring to the Ignition Coil is good, and the safety switches aren't screwing me.

 

CHECKING THE SIGNAL GENERATOR

This (well actually everything else as well) was new territory.
At a bit of a loss as to how to check this.

I get continuity from the wires at the signal coil back to CDI.
I do get a change in the voltage across the two pickup terminals when I turn over the starter. (red and blue arrows)
Whether it's sufficient, I dunno. The Multimeter doesn't refresh fast enough to give an accurate idea of the pulse or actual voltage.



Also, this part of the bike is really cool. I like it. Dunno why. Anyway.

 

TESTING IGNITION CIRCUIT

After this, I really kinda lost track of what to do. So I started randomly testing different things, using either of the two CDIs and trying to get a spark (or simply smoking cigarettes, drinking beer, scratching my head and staring at the bike - I did this WAY more).

During this time I got really weird results. Most strangely, I would get a single delayed spark on one coil (2-3) consistently exactly 4 seconds after stopping the starter motor. This seemed to point to a CDI fault (Imagine: Coil is charged, CDI never gives signal to fire, Coil dumps charge after safe interval)
When testing the coils primary circuit wires, I found that 2-3 was getting a steady 10V, while 1-4 was getting none.
So all things seemed to point to 1-4 being bad and 2-3 being good.
I checked and rechecked all of the wires in the ignition circuit for continuity, writing down the results till I filled an A4 page with my horrible ass handwriting.



While checking the ignition coil voltage against new CDI, old CDI and no CDI, I checked all combinations on both coils to be diligent.

I then realised that Coil 2-3 was getting voltage across the primary circuit even when a CDI wasn't plugged in...

Long story shortened slightly, I found that the negative/return wire for the coil was earthing somewhere.
Only got it to show up once with the multimeter set to megaohms.
I ended up splicing in another wire on top of the loom, and left the earthed wire in place but disconnected.

I have now found and corrected two things (side stand safety and earthed Coil-CDI wire) that would have prevented a spark.
Do I get a spark now?
Nope.

 

CHECKING THE IGNITION COILS

I've established that the ignition coil primary circuit wires have continuity to the CDI.
I checked the resistance on the Primary (Pin to Pin) and Secondary windings (Plug Cap to Plug Cap) of each coil.

Coil 1-4 Coil 2-3 Serviceable Range
Primary: 4.2 Ohms 4.2 Ohms 2.8-4 Ohms
Secondary: 17K* Ohms 17K* Ohms 13K-18K Ohms

*Originally the secondary circuit was reading 25K Ohms, but I later discovered that there is a 10k Ohm resistor in the spark cap, and I was testing through one spark cap on each side.

So the Coils seem to be .2ohms out of spec... doesn't seem too bad? Should I care?

 

CHECKING THE REGULATOR RECTIFIER

I had to read a few different resources to work this one out, but managed to get results after some time.

I am currently reading a consistent 1.9 million ohms (fucken heaps) in the direction that should show infinite resistance.
(Positive lead to B/W wire (ground circuit) and Negative Lead to the three black alternator leads)
(Negative lead to B/R wire (battery circuit) and Positive Lead to the three black alternator leads)

I am also reading infinite resistance in the direction which should have had 3000 ohms (or 1000 Ohms - 10,000 Ohms, depending on who you ask)

So the resistance is super high and in the wrong direction. I also checked to make sure that the Multimeter leads are plugged into the right holes into the Multimeter body.

This one has me really stumped right now. Maybe the bike got struck by lightning?! (joking... but only sort of)
Any suggestions?

 

Thanks guys!

I used the diode function as well, all reading pretty close to each other but still apparently reversed?(.54x)

I figure the diode function tests the resistance also, but instead assigns an arbitrary value around 0.5 and uses that as a basis for comparison?

The wrong direction thing is still bugging me, but for now I'll disregard the reg/rec as suggested.

Probably wait until I get the CDI and go from there.

Anything else anyone might suggest checking?

 

Hey guys, these two diagrams seem to suggest that because I am getting voltage to the ignition coil (&/or CDI - here 'ignitor') then the ignition coil must be live, suggesting that the safety switches (side stand, kill switch, ignition switch, neutral switch) are all in the state required to get a spark.

I decided this based on tracing a complete circuit between battery (Left side) and 'to ignitor/ignition coil' (right side) of the complex diagram.

The second diagram seems to suggest that the engine stop switch is between CDI and coil, but also send to say the same.



I have no background in electrical anything, so I could be assuming incorrectly!

Earth strap looks to be on the rearmost right hand bolt of the starter motor; this shall be the next port of call!

 

Hey murdo, to check the engine >frame earth, would i just clean the engine mounting bolts and mating surfaces? Or might there be a wire as well or something?

 

Stu that's awesome. I found an across one which is on the very edge of legible; it actually hurts your eyes trying to read it. This is way better!

Murdo interesting idea on the additional earth lead, will give that a shot also.

Maelstrom Actually the CDI was $2500, its a super special one. One of the 0's rubbed off! I'll sell to you for 20% off??