O/c alternators

[Graham_Wright]

It is well reported that to run an alternator without a load will cause a voltage run-away with dire consequences.

What constitutes an adequate load?
A big fat battery, a 6w indicator lamp, an LED, a 15volt Zener?
What period is required to cause damage? Is the break caused by load switching which may be milliseconds enough?

 

[Jcorstorphine]

Not sure you are on the right track here. An alternator does not require a load such as lamps or fans etc, what it does need is to be connected to a battery which helps to regulate the voltage down to arround 14 (ish) volts. Its more a case of an alternator will not cope with an open circuit (no battery) as the voltage will rise to a point where the internal diodes will be trashed.

To put it in perspective, if I start the diesel engine in my boat, the alternator does not require to provide any power at all to any part of the boat or the engine as I have removed the fuel solenoid and all the lights etc are off. What it does do is recharge the battery.

However if I was to switch off the battery switch with the engine running, the alternator volts would rise without the battery being connected and I would be off to the repair shop to get new diodes fitted.

 

[Graham_Wright]

Thanks, but surely the open circuit voltage will be limited by a resistive load in accordance with Ohms law? Certainly a dead short will limit the o/c volts!

 

[VicMallows]

It's quite a good question, and at this time of night I would not trust myself to give an answer (in case I get it wrong!). I will ponder. But what's so difficult connecting a battery?

Vic

 

[Graham_Wright]

No problem connecting just worried about the times when it is inadvertently disconnected hence the reference to duration of disconnection

 

[bdsweeting]

I think a quick flash-over when changing switches may be OK (don't quote me!) but the alternator needs to be connected to the battery so that it knows what voltage to generate. If the field wire is not connected then the alternator will just keep putting out volts in increasing numbers 2, 10, 20, 50, 100, 200, etc until the output diodes can no longer cope with the power output.

Basically - don't 'inadvertently' turn it off for too long.

 

[William_H]

Most alternators require a battery source to provide intial excitation via the field or regulatorand (opften) ignition light wire. The power produced internally feeds 3 small diodes which then provide power for the excitation the voltage from the battery then only provides a sense of the battery (also output) voltage. So if the field /battery wire is disconnected once it is up and running there is plenty of power for the field but the regulator senses no voltage so tries to increase the output. ( the power from the 3 field diodes goes to the top of the field (rotating) coil while the regulator connects the bottom of the field coil to ground to complete the current path. When voltage at the battery gets too high it imoses a resistance in this return path of the field coil current throttling back the output)
So a bit depends on whether the sensing of the voltage is done at the alternator output (not normal) or if it is done from the battery via the generator (ignition) light. The latter causing problems. Of courser without a battery connected to soak up the excessive output voltage the output voltage could rise quite high. With a battery connected but no sense wire the large current into the battery would cause a volt drop in the alternator output windings so that the voltage would rise only at the rate the battery voltage can rise until it boils. Please note there are manyb exceptions to this general arrangement.
The diodes can only fail through excessive current and overheating or through exceeding the reverse voltage capability of the diode. With no connection the main output diodes should be ok but the field supply diodes wil have a high current and high reverse voltage as they power the field. Having said that I think most power diodes have a reverse voltage capability of at least 200volts. Newer alternators may have even more capable modern diodes. Any damage from excess reverse voltage can happen in milliseconds as it is not a heating failure but a voltage flash over. So I guess it is best not to risk the damage however don't be suroprised if you get no damage. From one with little practical experience but a lot of theory hope it sheds some light on the situation will

 

[gjeffery]

One "improvement" that I am contemplating before launching, is to insert an Over Current Trip in the main battery cable. The downside would be that any inadvertant trip, while the engine is running, would cause the alternator to go open circuit.

Would anyone who has this arrangement share experiences?

 

[Graham_Wright]

That was a comprehensive reply and, when I have time, I'll draw me a piccy and I might even understand it!
It seems that it is the battery volts which dictate the situation not the load. However, if there is no excitation, there should be no output? If I remember correctly, the residual magnetism can be sufficient to "kick start" the alternator anyway and we would then have a runaway possibility.
So the original question stands;- if there is a resistive load, what will determine the alternator output volts limit?
The post about the breaker is very valid. The way round seems to be to have some sort of permanently connected load straight across the battery (such I am suggesting).

 

[VicS]

The electrical section of the Reading College course on boat maintenance is well worth a read for anyone working on their boat electrics. Quite a comprehensive bit on alternators. http://www.reading-college.ac.uk/marine/

 

[oldharry]

[ QUOTE ]
I think a quick flash-over when changing switches may be OK (don't quote me!)

[/ QUOTE ]

Sorry, but that is not quite right. Disconnecting the load from a running alternator causes a big voltage leap. known as a 'spike' (because of the way it appears when voltage level is displayed graphically.) It is this spike that overloads and destroys the diodes.

Alternators will run quite happily open circuit - simply because there is no current flow to do damage to anything.

Bad connections causing momentary interruptions to the output circuit, resulting in spikes, is the single most common cause of premature failure of an alternator.

 

[Ships_Cat]

Agree on the running open circuit being ok.

And with respect to the spike bit, I would think that the previous suggestion of using some other load to absorb spikes if a breaker is used between alternator and battery (or simply during switching) is fraught with risk. It would be difficult to incorporate a load with anything resembling the capacitance of the batteries for absorbing spikes (in my view the load has to be capacitive, not simply resistive as the lamp suggestion assumes). Is not territory I would be inclined to explore myself.

John

 

[Graham_Wright]

So is an RC combination ok?
This post has produced a fascinating disparity of suggestions but no answers anyone is prepared to bet their life on. Yet it must be of concern for anyone who has a switch in their battery feed.

 

[Ships_Cat]

Hi Graham

Yes, and the battery is, of course, also an RC combination. I could not comment on what component sizes one would need, but I suspect the capacitor might have to be quite large.

On small vessels ie most pleasure vessels, my view is that the alternator should not be able to be accidently isolated from the battery (for spike reasons, not open circuit running reasons). Taking our own case as an example, we have seperate alternators for the cranking battery and the service batteries permanently connected; no switches at all so no accidental spikes. Just need to remember that when working near/with the alternator it is live unless the battery is disconnected!

John

 

[pvb]

Switch in the battery feed....

Not sure whether this is just an academic argument or a real practical question. If the latter, I'd offer the following thoughts.

First, alternator output should be wired directly to the batteries, not via a switch. If you wire directly, there's no real chance of the alternator becoming open circuit. For safety, the wire should be fused, next to the battery, with a fuse rating higher than the max output of the alternator. If you have 2 battery banks (starting and domestic), the feed can either be through a diode splitter, or via a VSR (voltage sensitive relay). If you use a diode splitter, you ideally need a battery-sensed alternator or a "smart" regulator to overcome the 0.7volt drop across the diodes.

Secondly, if you have a 1-2-Both switch, replace it with 2 simple on-off switches - one for the starter battery; one for the domestic battery(ies). Much less opportunity for things to go wrong.

 

[fireball]

We have two seperate battery switches, turn one on, the other off - in that order ...

If you wanted to add in an Over Current Trip then unless you wanted to protect the charging current then just put the trip in AFTER the connection to the altenator - I would suspect you would need to do this unless your going for a big enough trip for the starter motor - in which case your electronic kit would melt before tripping it...

 

[Ships_Cat]

Re: Switch in the battery feed....

I am sure we are on pretty common ground pvb, so the purpose of the following is not to disagree with you, but I personally am a little wary of fusing at the batteries on small vessels ie most pleasure vessels, especially the cranking battery for engine availability reasons.

In my view there is a quandry in choosing the fuse capacity as if one sizes it to protect the cables it will not in most instances prevent fires which are most commonly caused by arcing at currents below the ultimate cable current capacity, not by the rare occurance of a hard welding together short between the positive and negative cables (not talking about downstream of fusing/breakers on the services switchboard here, of course).

If one then looks at sizing it for higher current draw arcs, then one has to consider that one does not want it blowing during cranking, or even if a windlass is off the same battery (which usually use modified cranking motors but are in a service where they can commonly draw locked rotor currents - but in any event should have its own breaker obviously rated well below any battery fuse rating).

Against that, for the cranking battery and the case where the engine is not an isolated negative one (ie all metal, including the alternator chassis, is insulated from the negative of the sensors, etc) then a real risk of fire exists in the cranking circuit especially at the cranking motor.

In the case of the alternator to battery cabling a current source exists at both ends of the cabling (alternator one end, battery the other) so in fact both ends of the cable should be fused, if at at all, if one is worried about an intermediate length short between them through chafe or whatever.

I feel it is a difficult decision which will vary from boat to boat but my preference is not to fuse the cranking battery and not so adament about the service ones. My preference is to design a boat where it is pretty much unlikely to have any positive to negative short on the alternator/batteries side of the switchboard (for service batteries) and engine (for cranking battery) by using an isolated negative engine and alternators, and (especially for metal boats) not grounding the negative to the hull/AC earth or other metals ie the DC system is completely isolated from the rest of the boat. Obviously, for an existing boat one is stuck with what one has, including the quality of the existing wiring job, and any fusing decision has to take that circumstance into account.

I have seen posts pointing out that cars are often now so fused, but cars are not boats and cranking (especially) and other system reliabilities are not so important to them. They also do not have potential very high current draw items such as windlasses hung off the battery either. I would also suspect that cars, with their bodies at DC negative potential represent a far greater fire risk through shorts in the alternator/battery/cranking circuits than any plastic, wooden or non grounded negative metal boat does.

A very important consideration for safety reasons with the larger boats I am involved with is their ability to recover from a dead ship ie loss of electrical systems. On those there are opportunities for multiple battery banks, generators with their own cranking batteries, and even hand crankable generators to initiate recovery of power. However, those things are generally not relevant to a small pleasure boat but the same safety concerns still exist (or though I concede much more so for a single engined power boat rather than a sail boat which, assuming wind, has sails as alternative propulsion).

I appreciate that all the above swings no way or the other but I hope I have given some food for thought as to whether fusing the batteries should be an automatic decision or not.

Regards

John

 

[halcyon]

Little bedtime story, when we supplied Westerly with switch gear we used to fit an integral split charge relay system. On one yacht the two battery isolator switches were on the front face of the chart table seat. Sliding in or out in oilies could catch a switch, red Hella flag type.
This yacht leaves Jersey under engine, engine charges it's battery and brings in relay paralleling the domestic bank and putting on charge. Navigator decides they had cleared danger and were on course for the Alderney Race, so they set sails.
All going fine, so they turn off the engine, blackness, ( it's a night passage ) they lose all cabinn lights, nav lights. all nav equipment, vhf the lot. They are now in the Race, start the engine, engine dead, switch panel dead as a dodo. So semi panic sets in, why no electric, they had it, so what do you do, you turn in back on having noticed that both switches were off.
The navigator getting in and out of the chart table seat had knocked both off, why had they power ?
Thet had been under engine, with the split charge relay engaged, when the switches were turned off, the capacitive / resistive electrical load on the boat was enough to maintain regulation. Thus all loads were supplied direct from the alternator, and the voltage was high enough to keep the relay engaged. When they turned the engine off, they lost feed to the electrical system and all when dead.

So yes you can run a alternator with the batteries isolated, why is to complicated at this hour. But if you put a current trip in the battery feed you could well have the same situation and the alternator keeps feeding the system, and what ever the fault was.

Brian

 

[William_H]

A fuse or circuit breaker is only effective in protecting wiring from over current melting the insulation causing a fire. An overcurrent breaker for an alternator would not protect the alternator or battery form large current over a period long enough to boil batteries or melt altetrnator as this current is quite acceptable short term recharging a battery after starting but is too much long term. Battery switching IMHO should be by separate switches not those 1,2,both switches. Most authorities in my case the club safety inspection requires that all loads be disconnedcted when leaving the boat. There are a few exceptions off course. So an arrangement where you switch on on arrival at the boat and off on departure would minimise chances of inadvertant switching off with engine running. I would go for a large switch for engine battery to engine. A small switch for services battery to services and a VSR to ensure charging of the services battery. I would have a jumper lead for times when you need services battery to help the engine battery or a separate well gua4rded switch. In aviation you can get a plastic guard to cover a critical switch, you must open the guard to work the switch which gives a little more protection. In boating you can just make the switch remote and hidden. regards will

 

[pvb]

Re: Switch in the battery feed....

Yes, John, there's a variety of pros and cons involved. My advice was that alternator output should be wired direct to the batteries, and that this wire should be fused next to the battery. You've suggested that the wire might also be fused at the alternator end - that's a bit "belt & braces", I'm more concerned about cables which would be live when the boat's not being used.

I didn't mention fusing the cranking battery, although I do believe in this (and I've done it on my boat). Your concept of a low current arc causing a fire is intriguing, and I wonder whether that's what happens in practice with heavy duty battery cables. The only time I've accidentally shorted a battery cable, the sparks would have lit up a small town! You're right to say that choosing the right fuse capacity is important. Good quality "Mega" fuses have quite a slow-blow characteristic and so are ideal for this purpose; I use a 150A fuse on my starter battery.