Starter Motor Failure
Our starter motor failed in mysterious circumstances at Eucla, half way across the Nullarbor Plain. We'd stopped to use the facilities at the Eucla Roadhouse and afterwards, the starter motor refused to function.
The solenoid on the starter motor had been giving intermittent trouble for a couple of years and usually tapping the start relay connection was enough to get it going, but not this time. I replaced the solenoid but the motor was still faulty so we eventually assembled a working one out of spare parts and found that a replacement armature was also required to get it going.
Further analysis on the faulty components was obviously required after we arrived home. This is it.
Solenoid Failure
In Eucla I took the solenoid off and checked it electrically. Not surprisingly one of the windings had failed internally.
Since these are not repairable items (at least, not without a lot of effort and desperation) we ordered a new one which took a couple of days to be delivered from Perth. In the event that didn't fix the starter problem and we had to wait until we could bump start the Oka and travel on to Coolgardie, without switching the engine off, where I could fix or replace the motor.
The second, concurrent, problem was a faulty armature, which turned out to be a winding fault, rather than excess commutator wear (which would have been simple to fix). Why and how we could have 2 problems which occurred at the same time is still a mystery, unless the solenoid was in one of its in-between failure periods and it was actually the armature which failed first. I guess we'll never know but it did delay us for a few days, luckily right outside the Eucla Motel.
Opening up the Solenoid
The starter solenoid is assembled into a steel cylinder, the top edge of which is peened over during the manufacturing process to hold the contactor plate in place, so it can't be easily be disassembled.
Electrical operation of a starter solenoid is explained in this useful article.
I first unsoldered the winding wires from the terminal posts (there are 2 wires on one terminal) using a solder sucker and desoldering braid to remove the solder. The thinner wire came completely out indicating a broken or fused wire in the hold-in winding. Then I cut through the skin of the cylinder near the contactor using a hacksaw and removed the internal components.
Where I cut the solenoid cylinder
The windings are wound around a plastic former on a steel base plate which is held in place by a large circlip.
After removing the circlip and sliding out the coil assembly, it was immediately obvious that there had been a burning event and both windings were burnt black at the terminal end. Although the hold-in coil had failed, it was unclear whether the thicker pull-in winding had overheated, causing the thinner hold-in winding to fail, or vice versa.
Solenoid Repair Possibilities |
The solenoid is a fairly simple electrical device and if necessary it could be rewound and reassembled by cutting out the burnt sections of wire (the insulating enamel may have been damaged by heat causing further short-circuits) and rewinding them. The windings both have around 100 turns on them so a few turns missing would not affect operation appreciably. The cut off cylinder could be repaired using a hose clip, tie wraps or similar to hold the contactor plate in place.
Pull-in winding after removal of the hold-in winding |
Although the solenoid could be repaired if the situation was really desperate enough, it's probably a better choice to carry a spare at around $100 and save all the potential angst.
Contents of the solenoid "To reassemble, simply reverse the disassembly process" |
Causes of Failure
There was no obvious mechanical or electrical reason for the winding to fail, but possible causes are:
- Prolonged engagement of the starter motor, allowing the windings to heat up to the point where the insulation breaks down, causing a partial short circuit and leading to eventual winding failure, or
- Heat from the exhaust system which runs nearby, aggravating the internal self heating of the solenoid windings and leading to the same result. This is most likely to happen when restarting an already hot engine (which coincidentally is exactly what happened to us in Eucla), or
- Shit just happens when you least expect it.
The solenoid on our Oka is very close to the exhaust |
Using Iskra starters (and possibly other makes), there are 2 frame types with different locations for the starter solenoid:
- on the side of the motor (which is close to the exhaust pipe on our Oka), or
- below the motor where it's more exposed to impact damage (but no more so than the engine sump would be).
There is the merest possibility of the front drive shaft or its capture hoop touching the solenoid in the lower position at maximum axle articulation, but that seems most unlikely to happen in practice as the bump stops/airbags would have to completely fail to do their job.
Solenoid located on the side of the motor, like ours, Iskra IM527
The only difference between the motor types is the position of the solenoid and therefore the battery cable runs.
My preference now is for the lower location since impact damage would be rare and obvious, rather than the side location (which is where ours is currently) where the solenoid is exposed to continuous exhaust heat, so I shall be changing mine for the lower type and fortunately I have acquired a spare frame of this type. For starters with the solenoid near the exhaust, I recommend fitting a heat shield of shiny aluminium, or even reflective tape, between them to reduce the effects of radiant heat. |
Armature Problem
Well I've now done some analysis of the faulty armature.
I started by buffing up the commutator to ensure that the brushes would seat properly and eliminate that as a problem. I didn't have a lathe big enough to hold the armature so I made a simple stand by clamping the bearings on each end in small vices clamped to the workbench. I then turned the armature by hand and used a flat file to dress the commutator. [I did try using a cordless drill and belt to turn the armature but too many hands were needed to hold it steady and it only took about 10 minutes anyway].
Rudimentary lathe assembled to dress the commutator |
Once the armature was smooth and flat, I rebuilt the motor and tested it on a car battery using jumper leads.
The motor still didn't turn but a whiff of smoke indicated where the problem was: 2 adjacent segments are presumably shorted internally and a burn mark was evident on the end of the commutator where the smoke appeared.
The burn mark where the smoke appeared from. |
To confirm that this was not caused by internal short circuits in the area of the brush plate, field coil connections and/or fixing bolts, I removed the end cover and held the motor in a vice while applying voltage.
The motor still didn't turn but it put up quite a light show to confirm there is indeed an internal problem with the armature, irrespective of its rotational position.
A light show but no movement, and I was wearing eye protection. |
Conclusion
So the armature would seem to be completely cactus and probably not worth repairing. But it also confirms that our problem in Eucla was indeed caused by 2 concurrent problems, the failed solenoid and an armature fault, although I now suspect that the armature fault was the primary show stopper.
The fixed field coils seem to be very robust so should be more reliable than rotating/sparking parts:
- They don't move,
- They're made of thick, solid copper bars,
- I have a spare set.
Brushes can be replaced and I have a spare set, plus brush plate.
Solenoids can be replaced (I now have a spare), bypassed or even fixed on the side of the road but not the armature.
So the question now is whether to acquire a new or s/h (spare) armature or a whole new (spare) starter motor or get this one repaired.
Or do nothing and hope the existing one will last for ever.
