Fire!

[GBW]

A warning to all!

I am cheating a little as my boat only spent a year on a canal (The Gloucester-Sharpness) and is now sea borne but this forum is invaluable for information on my BMC 1.5 engine.

A salutary warning.

 

Sailing from Milford Haven to Worms Head under power (no wind) to await the tide, when 8 miles off I went below to update the log.  Checking the engine for water temperature and oil pressure I was greeted by flames.

These were extensive but succumbed to CO2 and water.

Although yet to be confirmed, the cause seemed to have been the alternator output connection to the starter solenoid.  This, as in cars, is a direct connection perhaps for convenience.

I was unable to stop the engine with the stop solenoid (an unrelated fault referenced previously) but the CO2 finally did so.  I could not subsequently restart.

My suspicion is that a loose connection at this point caused resistive heat eventually sufficient to ignite adjacent flammable cables and hoses.

The solenoid is wrecked but the motor may be ok.  My on board manual tells me that the two cannot be separated.  Back home, looking at my spare, I believe this to be incorrect.  Time will tell!

I return to the boat tomorrow (in Tenby after a tow from the lifeboat) and will update.

 

What do others do to detect fire?  There was surprisingly little smoke.

 

[Tony Brooks]

I have never found a BMC 1.5 starter where you could not separate the whole solenoid from the starter, but I suspect it is just the plastic cap that needs changing. However, I have seen caps that are swaged in place rather than held in place two screws. Any chance of a photo.

[MtB]

27 minutes ago, GBW said:

What do others do to detect fire?  There was surprisingly little smoke.

 

On canal boats, nothing is routinely installed for fire detection beyond perhaps a £20 smoke alarm.

 

It's a non-problem for us as engine fires are 1) very rare, and 2) canal boats are rarely more than about 60ft from dry land. 

 

Sea boats must be a different kettle of fish though. 

 

 

Edited June 11, 2023 by MtB

[Tony Brooks]

40 minutes ago, GBW said:

What do others do to detect fire?

 

Basically, keep an eye on cable runs, connection and fuel systems. As Mike says, engine fires are very rare on well maintained boats.

[GBW]

Access on sea-going boats varies.  I sailed with someone on his 47 foot ketch and become convinced that just after the keel was laid, the bed were installed, the engine mounted on them and then the hull was completed around it!  Access was virtually nil and the engine needed to be lifted for repair.

On the other hand, some boats have engine spaces all around them.  Mine lies in between but, inevitably, the oil filter and starter motor are down the "wrong" side!

Fire on boats in general is very rare but when the engine is "below", evidence takes longer to reach the helmsman. When we arrived at Tenby dock, two fire officers were waiting.  They performed a pyrotechnic inspection of the engine space using a hand held device.  Perhaps such a device exists for a permanently mounted operation.  Temperature sensors are generally fitted but normally on the cooling system.  I would never (previously) have considered one at the starter motor.  I may change my view!

In the many cars I have possessed I can't ever remember checking the connection to the starter motor.

Incidentally, that was the second "loose nut" experience on the trip.  The first involved a through bulkhead connection where 5 volts were lost from the (different) alternator to the batteries due to a loose nut.  It is now a loctite nut.

25 minutes ago, Tony Brooks said:

I have never found a BMC 1.5 starter where you could not separate the whole solenoid from the starter, but I suspect it is just the plastic cap that needs changing. However, I have seen caps that are swaged in place rather than held in place two screws. Any chance of a photo.

 

The spare starter I have is in working order and I am reluctant to wreck its usefulness by an unsuccessful dismantling.  The access to the solenoid on the boat is relatively simple.  Two of the three bolts securing the starter are visible and accessible.  The third is below and difficult, although I seem to remember in the dim and distant removing and replacing it for some reason.

Visible in the photo is an internal rubber shroud.  extraction of the operating armature is now reluctant.  Perhaps I'll try pulling the motor armature next.  Would removing the operating pivot pin be helpful?

 

 

[Tony Brooks]

No, certainly not. If you take it out you will then need to reset it and that is not so easy.

 

My guess is that it is the rubber shroud causing the problem, and I am sure that is a marine embellishment. normally, the body of the solenoid pulls straight off the solenoid armature with no rubber shroud in place. I can't see why the body will not just pull backwards, leaving the shroud and armature in place.

 

Could the threads on the studs be binding on the rubber shroud where it passes through the shroud's metal plate?

 

I suspect the shroud is hooked over the pivot arm so that is why it will not come out.

 

I think your plastic cap does use screws to secure it, but you do need to get the correct replacement.

[GBW]

The shroud I referred to is internal - black in the photo.

Using a bit more courage, the body did pull off.

The tag connections to the coil will need unsoldering I think to allow removal of the cap.

However, with a bit of luck and using the invaluable advice from you and MtB, I might get away with swapping just the body.  There is no reason to think the motor has been damaged.

This seems to be the cap;-

https://www.ebay.co.uk/itm/183653472588?_trkparms=amclksrc%3DITM%26aid%3D777008%26algo%3DPERSONAL.TOPIC%26ao%3D1%26asc%3D20220705100511%26meid%3D53ad4aa2c3cc47f39555afaaccf1a579%26pid%3D101524%26rk%3D1%26rkt%3D1%26itm%3D183653472588%26pmt%3D1%26noa%3D1%26pg%3D2380057%26algv%3DRecentlyViewedItemsV2&_trksid=p2380057.c101524.m146925&_trkparms=pageci%3A1333ea0d-084c-11ee-b214-a2477a6c0310|parentrq%3Aaa3d6e431880ab8f2efb20ddfffd74df|iid%3A1

 

Thanks to both of you.

[Tony Brooks]

44 minutes ago, GBW said:

The shroud I referred to is internal - black in the photo.

Using a bit more courage, the body did pull off.

The tag connections to the coil will need unsoldering I think to allow removal of the cap.

However, with a bit of luck and using the invaluable advice from you and MtB, I might get away with swapping just the body.  There is no reason to think the motor has been damaged.

 

 

No reason at all for the actual motor to be damaged.

 

Yes, you do have to unsolder the tags, and you need to ensure the cap goes on in exactly the same way, otherwise it will not work and "machine gun".

 

Whilst I can't say it is OK to swap bodies, I know that I would try it before spending money.

 

Oh, and back to the fire question. Most of the forum members boats are made of metal, steel, so an engine fire is far less serious than on a GRP boat. you can get heat activated engine room fire extinguishers that you can fit, but your fire would probably reignite once they/it runs out. An ammeter and voltmeter may well have given you a clue that all was not well down below.

[GBW]

The ammeter is below (and the cockpit mounted repeater not yet fitted) but all that would show, I think, would be intermittent charging with a flickering needle (which I would probably ignore!).

I have a bunch of clip on thermostats, penny sized, intended for coolant temperature monitoring but one adjacent to the starter may well have been effective.

[Tony Brooks]

21 minutes ago, GBW said:

The ammeter is below (and the cockpit mounted repeater not yet fitted) but all that would show, I think, would be intermittent charging with a flickering needle (which I would probably ignore!).

I have a bunch of clip on thermostats, penny sized, intended for coolant temperature monitoring but one adjacent to the starter may well have been effective.

 

But the voltmeter would be jumping up and down as well. In fact, it could be reading many tens of volts. It depends upon if it was heating cause by a loose connection or a short circuit. If any instrument starts doing unexpected things, it needs investigation.

Edited June 11, 2023 by Tony Brooks

[GBW]

Repeating an ammeter in the cockpit (times 2 as I have two alternators) implies heavy cables with quite long runs.  My intention is to measure the volts from across the existing cables with micro-ammeters.  I have done this to sense the glow plug current and it is a boon.  Fsd means four glow plugs drawing current, half full scale, two etc.  The problem is finding meters with moving coil mechanisms as they all seem to be thermal (at least the ones I have found - it isn't obvious until you use them).

However, I have found some cheap (i.e.Chinese!) meters on ebay fsd 50 microamps which will do the trick.  Needs calibrating of course.

The same meters will do for volts as well.  I am a firm believe in monitoring - information aids diagnosis!

 

The thermostats I referred to are as these;- https://www.ebay.co.uk/itm/254292043897

 

All these things take time!

 

 

 

Edited June 11, 2023 by GBW

[Tony Brooks]

1 hour ago, GBW said:

Repeating an ammeter in the cockpit (times 2 as I have two alternators) implies heavy cables with quite long runs.  My intention is to measure the volts from across the existing cables with micro-ammeters.  I have done this to sense the glow plug current and it is a boon.  Fsd means four glow plugs drawing current, half full scale, two etc.  The problem is finding meters with moving coil mechanisms as they all seem to be thermal (at least the ones I have found - it isn't obvious until you use them).

However, I have found some cheap (i.e.Chinese!) meters on ebay fsd 50 microamps which will do the trick.  Needs calibrating of course.

The same meters will do for volts as well.  I am a firm believe in monitoring - information aids diagnosis!

 

The thermostats I referred to are as these;- https://www.ebay.co.uk/itm/254292043897

 

All these things take time!

 

 

 

 

If you think that you need long cables then you have the wrong sort of ammeter - a cheap one or an old CAV one with the shunt on the back. you can fit a shunt in the main cable and use thin wires to join the shunt to the meter. As the meter is really a mV meter with a dial showing amps you could use bell wore apart from the BSS and such like.

 

And no, an ammeter that uses thick cables will almost certainly be an inaccurate moving iron type - hence the low price.

 

Nowadays  a better option is a digital one like one of the so called battery monitors as long as you ignore the % charged read out and anything related to it.

 

 

[GBW]

I think that is what I just posted!

"My intention is to measure the volts from across the existing cables with micro-ammeters. " 

My digital ammeters are by Nasa but live below.  It is a sailing boat (sorry!).

[Bee]

Crikey! That's not good for the heart rate!

[Tony Brooks]

5 minutes ago, GBW said:

I think that is what I just posted!

"My intention is to measure the volts from across the existing cables with micro-ammeters. " 

My digital ammeters are by Nasa but live below.  It is a sailing boat (sorry!).

 

Except they will still be millivolt meters, not micro-ammeters. You can get meters with a sender that clamps around the cable, but not sure how accurate they are. If you are not interested in an accurate numerical readout, which for amps you would probably not be, one of these may be simpler than trying to calibrate for volt drop.

 

https://www.amazon.co.uk/KETOTEK-Voltmeter-AC80-300V-Inductor-Industry/dp/B07M5RT6VQ/ref=sxin_16_pa_sp_search_thematic_sspa?content-id=amzn1.sym.f4b968f1-42c0-4265-bdde-a765c88ff214%3Aamzn1.sym.f4b968f1-42c0-4265-bdde-a765c88ff214&cv_ct_cx=ammeter&keywords=ammeter&pd_rd_i=B07M5RT6VQ&pd_rd_r=5bb16b93-da84-4f7e-8056-03f7a7406d62&pd_rd_w=yJOQS&pd_rd_wg=UboMp&pf_rd_p=f4b968f1-42c0-4265-bdde-a765c88ff214&pf_rd_r=7NBRWBSMS6H2GA0ZZT98&qid=1686498108&sbo=RZvfv%2F%2FHxDF%2BO5021pAnSA%3D%3D&sr=1-4-1c12e6fc-61d1-41ee-8e02-a6ca5f2da604-spons&sp_csd=d2lkZ2V0TmFtZT1zcF9zZWFyY2hfdGhlbWF0aWM&psc=1

[GBW]

No - sorry,  They are micro ammeters.  But seriously, is there any difference?

I believe a "voltmeter" measures the current through a resistor which is sized to give the correct deflection for the voltage applied.  

(I used to build multimeters!)

[Tony Brooks]

31 minutes ago, GBW said:

No - sorry,  They are micro ammeters.  But seriously, is there any difference?

I believe a "voltmeter" measures the current through a resistor which is sized to give the correct deflection for the voltage applied.  

(I used to build multimeters!)

 

If you really did build multimeters, you would know that analogue ammeters measure the volt drop across a range resistor (meter coil in parallel with resistor) while the voltmeter has the range resistor connected in series. One could argue that a moving coil multimeter is using current to move the coil and I suppose it is, but that current is proportional to the voltage so which is it measuring? The current or the voltage. Things get more difficult when we move to digital meters because the A to D chip must be reacting to voltage.

 

No point in continuing this because if you are measuring volt drop the meter is a voltmeter, whatever it's dial claims.

[GBW]

You're right!  The current is what generates the magnetism that moves the needle and the voltage generates the current.

Lets leave it there!

Overnight, I had the thought that the alternator output is actually connected to the battery not via the solenoid terminal but via another route.  If that is the case, my initial diagnosis is wrong and I am now worried that the cause is more serious.

I return to the boat today and will report back.

[Tony Brooks]

We need to consider where the heat came from to initiate combustion. My suspicion is that it was a cable or cables that was on fire and that may have burned/melted the plastic solenoid cap. The supply for the engine electrics is probably taken from the main solenoid positive terminal. So to generate enough heat to ignite the cable insulation, we need a short circuit of some type because as soon as the engine starts there should only be the instrument and possibly bilge blower current passing through the terminals. I doubt that would be enough to overheat the battery cable sufficiently. I acknowledge sea boats may be wired differently, especially if it uses a 1,2,both,off switch for battery charging. So how could such a high current flow occur.

 

If the solenoid jammed in contact, then the motor would continue to run and that might heat the solenoid contact enough to weld them together. In such a scenario, the sprag clutch on the pinion could well eventually seize up, and then the engine would drive the armature which would promptly fly apart, and you would have one massive short circuit that would ignite the solenoid cover and probably the main battery cable close to the terminals. Although not common, this is a known issue.

 

If this is what happened, I would have expected a noticeable loss of engine power for a short while as the engine accelerated the armature.

 

I think you may be well advised to take the motor off for internal inspection.

[GBW]

The mystery deepens. Returned to the boat. 
As expected, there is only one connection to the solenoid - the battery. The cable is securely attached to the solenoid stud. However, the stud is not attached to the solenoid. This is probably the result of the fire. The end cap is brittle and partly disintegrated. There is no obvious sign of a shorting mechanism. Is it possible the cap may have become carbonised and progressively increasingly conductive permitting a short to frame. It took  around five hours to remove the starter due to bad choice of bolts. 

[Tony Brooks]

On the BMC1.5 there have been a number of different solenoid configurations. The earliest had the solenoid hanging down and on some boats within a few mm of the engine drip tray base or bilge. If the bilge water level got to the solenoid, the plastic cap would do exactly as yours has done, but the damage may have started years ago. If this was the case for yours then there is a good chance the actual motor is fine.

 

Later ones have the solenoid stuck upwards at about 45 degrees to the horizontal. The hanging down ones caused so many problems on the fleet that we changed the lot to the 45 degree version. This can or could be done with a new drive end plate.

 

If you remove the oil filter from the two bolts that secure the filter head to the block, you can usually get a 3/8 sq drive socket and  extension onto those bolts.  I will deny I ever said this but on the fleet the starters were only held on by the two easier to access bolts. You do need to ensure they are always tight.

[GBW]

There were three bolts but not what I would have fitted. One top one was easy, the bottom one was accessible with a long extension on a socket and undid surprisingly easily (after numerous attempts and trips in the bilge (water). However, the third top one was easy of access but impossible to apply any sort of spanner due to the proximity of the casting.  The nut was eventually removed after drilling it’ll split it. I think socket head screws will be a better replacement. 
I will test the starter on the “bench” before replacing it. 

[GBW]

Noting Tony's observation on the location of the solenoid, I fitted the spare replacement as it was a "topside" solenoid.  Incidentally, it should be possible to remove the solenoid whilst in-situ which is not the case with the "downside" one.

The replaced starter started the engine without problems and I eventually made it back to Cardiff on three tides.

The problem of accessibility to the nuts nestling against the casting could be solved by introducing a bush under the nut head (or even a snack of washers.

The cause of the fire remains unknown.  The disintegrated solenoid showed no signs of the contacts being welded; indeed, as explained earlier, the battery lead contact remained connected to the lead.

Many thanks for all the advice and especially to Tenby Harbour Master who spent a couple of frustrating hours failing to remove the same nut that defeated me.

 

The thought of the solenoid operating with the engine running is horrifying.  Is it possible the starter would disintegrate probably running faster once the engine speeded up.  Goodbye starter ring?

Edited June 17, 2023 by GBW
Addition.

[Tony Brooks]

11 minutes ago, GBW said:

The thought of the solenoid operating with the engine running is horrifying.  Is it possible the starter would disintegrate probably running faster once the engine speeded up.  Goodbye starter ring?

 

They have a freewheel between armature shaft and pinion, but I have doubts how long it would last if it stuck in mesh and the engine was kept running. I think it would resolve into either the freewheel bursting (I have seen that) or the armature and commutator flying apart, I have seen that as well. I don't think that I have seen the ring gear suffer significant damage from such an event.

[GBW]

Today I met someone who used to work for Lotus.  I explained my experience.  He asked "Was it a bakelite cap on the solenoid?".

He agreed that crazing of the surface had probably caused electrical tracking through the cap sufficient to ignite the adjacent cables.