Alternator getting very hot

[rgriffiths]

I know that this is a common question but I have tried everything I can think of.

 

It all started when I replaced the 2 domestic batteries. The system has a 200 watt solar panel with an MPPT controller and two alternators - one domestic and one for the starter. The problem is with the domestic charging and these are the symptoms:

 

• Two brand new leisure batteries installed.
• The domestic system is charging at around 13.3 volts
• The alternator gets very hot when charging - too hot to touch for more than a second or two.
• I have had the alternator tested (twice) and everything is fine. On the test bed at the alternator repair place it does not get hot at all but I am not sure of that is significant.
• Back on the boat if I remove the +ve connection from either of the two batteries (so that it is only charging one of them) it jumps to 14.1v. Connect them both then it drops back to 13.3v.
• The batteries drop from 12.9v to 12v quite during the day which is strange given the solar panel and decent sunlight we are getting at the moment. They drop to about 11 over night.
• I have changed the controller just in case there was a problem there.
• Controller was reading 5 to 6 amps charging today.
• There is not a great drain on the system other than the the fridge.
• No shorting in the system that I can find.

 

I realise that this is vague but I am hoping someone might give me some more things to test and more than anything give me some idea why an alternator which has test fine is getting so hot.

 

As ever all help gratefully received.

[Alan de Enfield]

Most likely - One or both of your batteries has an internal short and is demanding so much power it is dragging down the voltage, and, generating a huge amount of heat in the alternator.

 

1) Feel the batteries - is one warm / hot ?

2) Look at the batteries - is one of them distorted with any side / top / bottom distorted and 'blown' outwards.

3) Smell - is there a bad egg smell  ?

4) Gas - get your Co meter and put it in the engine room, does it sound ?

 

Possible :

You say "Brand new batteries" who connected them up ?

Have you connected the batteries up in series (so they become a 24v bank) and trying to charge with a 12v alternator ?

 

 

 

Edited August 3, 2019 by Alan de Enfield

[rgriffiths]

Thanks Alan. I was unsure about one of the new batteries and so swapped it for another new one but symptoms seem to be the same. Maybe both were duff. I will check as you suggest and maybe swap the second one too. 

[MtB]

I second the faulty battery suggestion.

 

The alternator will only be getting hot if it is chucking out maximum current for along periods, and as the battery is holding the charge voltage down to 13.3v all the time this must be what is happening. 14.4v would be the expected charge voltage after an hour or two even without the solar.

 

Do you have an ammeter? What is the charging current?

 

Edit to add:

 

Ah you said charge current 6A. This IS a puzzle then... Possibly you have a massive load on the alternator not going through the ammeter shunt. I had something like this recently. The heater plugs of the engine had jammed ON ,drawing 40A 24/7.

 

 

 

 

Edited August 3, 2019 by Mike the Boilerman

[nicknorman]

It sounds to me like a diode fault in the domestic alternator. Clearly it isn’t feasible that there is a faulty battery if with either battery in circuit things are OK, with both it is not.

 

Faulty diodes cause large AC currents to flow in the alternator which make it hot without actually charging the battery much.

 

The only evidence against my diagnosis is that the alternator repair shop declared them OK. Twice, but presumably the same shop twice? Maybe try a different alternator shop.

 

electricity is invisible and so without the correct measurement tools you cannot know what is going on. Get a DC clamp meter  which will show you if the alternator is indeed supplying a lot of dc current or not, and when you set it to ac current, whether it is supplying a lot of that (which it shouldn’t be).

 

With diode failure is is possible that current is back-flowing into the alternator when the engine isn’t running. You could eliminate this by disconnecting the fat (charging) wire from the alternator at the battery end and see if the solar charges the batteries better / batteries hold their charge during the day better.

 

[Tony Brooks]

7 hours ago, nicknorman said:

It sounds to me like a diode fault in the domestic alternator. Clearly it isn’t feasible that there is a faulty battery if with either battery in circuit things are OK, with both it is not.

 

Faulty diodes cause large AC currents to flow in the alternator which make it hot without actually charging the battery much.

 

The only evidence against my diagnosis is that the alternator repair shop declared them OK. Twice, but presumably the same shop twice? Maybe try a different alternator shop.

 

electricity is invisible and so without the correct measurement tools you cannot know what is going on. Get a DC clamp meter  which will show you if the alternator is indeed supplying a lot of dc current or not, and when you set it to ac current, whether it is supplying a lot of that (which it shouldn’t be).

 

With diode failure is is possible that current is back-flowing into the alternator when the engine isn’t running. You could eliminate this by disconnecting the fat (charging) wire from the alternator at the battery end and see if the solar charges the batteries better / batteries hold their charge during the day better.

 

Agree, most likely an alternator fault but we do need definitive current and voltage readings for the battery bank concerned. Without being told any measurements given could be from either or both banks.

[Chewbacka]

8 hours ago, nicknorman said:

It sounds to me like a diode fault in the domestic alternator. Clearly it isn’t feasible that there is a faulty battery if with either battery in circuit things are OK, with both it is not.

 

Faulty diodes cause large AC currents to flow in the alternator which make it hot without actually charging the battery much.

 

The only evidence against my diagnosis is that the alternator repair shop declared them OK. Twice, but presumably the same shop twice? Maybe try a different alternator shop.

 

electricity is invisible and so without the correct measurement tools you cannot know what is going on. Get a DC clamp meter  which will show you if the alternator is indeed supplying a lot of dc current or not, and when you set it to ac current, whether it is supplying a lot of that (which it shouldn’t be).

 

With diode failure is is possible that current is back-flowing into the alternator when the engine isn’t running. You could eliminate this by disconnecting the fat (charging) wire from the alternator at the battery end and see if the solar charges the batteries better / batteries hold their charge during the day better.

 

With an ac generator which is then rectified to dc the out put is not smooth, difficult to explain, but with a transformer the output is 50Hz AC after going through a bridge rectifier the output is dc but it goes from 0v to full volts 100 times per second.  With a power supply it then has smoothing capacitors to give a more constant voltage.

With a vehicle alternator it is (I think) a three phase rectified output without smoothing capacitors so there will be a significant amount of ripple, especially at full load though a good battery will be smoothing this a lot.  So getting to my point, I would expect that if you set the clamp meter to AC you will read this ripple but I don’t know how much as I have never measured it on mine.

So an AC reading may not mean faulty alternator, or am I wrong??

[rgriffiths]

Thanks again. The alternator has been thoroughly tested and I am confident it is fine. Diode tests are all good too. It is a 70 amp alterntor ans is outputing around 40 amps on idle. The batteries are in parallel. 

 

Will check it again on Monday and report back. 

[Alan de Enfield]

1 minute ago, rgriffiths said:

Thanks again. The alternator has been thoroughly tested and I am confident it is fine. Diode tests are all good too. It is a 70 amp alterntor ans is outputing around 40 amps on idle. The batteries are in parallel. 

 

Will check it again on Monday and report back. 

Then it can (pretty much) only be a battery fault, or a very heavy load for a lengthy period.

 

The suggestion of 'stuck' on Heater plugs would do it (MtB example) 

[MtB]

14 minutes ago, Alan de Enfield said:

The suggestion of 'stuck' on Heater plugs would do it (MtB example) 

 

Just to expand on this, the relay switching them on and off welded its contacts together. 

 

 

[Dr Bob]

10 hours ago, rgriffiths said:

I know that this is a common question but I have tried everything I can think of.

 

It all started when I replaced the 2 domestic batteries. The system has a 200 watt solar panel with an MPPT controller and two alternators - one domestic and one for the starter. The problem is with the domestic charging and these are the symptoms:

 

• Two brand new leisure batteries installed.
• The domestic system is charging at around 13.3 volts
• The alternator gets very hot when charging - too hot to touch for more than a second or two.
• I have had the alternator tested (twice) and everything is fine. On the test bed at the alternator repair place it does not get hot at all but I am not sure of that is significant.
• Back on the boat if I remove the +ve connection from either of the two batteries (so that it is only charging one of them) it jumps to 14.1v. Connect them both then it drops back to 13.3v.
• The batteries drop from 12.9v to 12v quite during the day which is strange given the solar panel and decent sunlight we are getting at the moment. They drop to about 11 over night.
• I have changed the controller just in case there was a problem there.
• Controller was reading 5 to 6 amps charging today.
• There is not a great drain on the system other than the the fridge.
• No shorting in the system that I can find.

 

I realise that this is vague but I am hoping someone might give me some more things to test and more than anything give me some idea why an alternator which has test fine is getting so hot.

 

As ever all help gratefully received.

Where are you?

There maybe someone close who has a D.C. Clamp meter.

The numbers you give are strange. Can you give more details on voltage vs time?

For example you say they start at 12.9v. When? Then they drop to 12. Is that 12.00v? Then down to 11v. Is that 11.00v? If so the batteries have very little charge after a night. So how are you getting back to 12.9v if it's only charging at 13.3v. Are you using a battery charger to get back up there?

 

You really need access to a D.C. Clamp meter to see what is draining the batteries .....and an infrequent red thermometer to check the temp of the alternator. Difficult to tell by hand as they do get to 90 deg C.

 

editted to add.......pressed post too soon!

Edited August 4, 2019 by Dr Bob

[dmr]

9 hours ago, nicknorman said:

It sounds to me like a diode fault in the domestic alternator. Clearly it isn’t feasible that there is a faulty battery if with either battery in circuit things are OK, with both it is not.

 

Faulty diodes cause large AC currents to flow in the alternator which make it hot without actually charging the battery much.

 

The only evidence against my diagnosis is that the alternator repair shop declared them OK. Twice, but presumably the same shop twice? Maybe try a different alternator shop.

 

electricity is invisible and so without the correct measurement tools you cannot know what is going on. Get a DC clamp meter  which will show you if the alternator is indeed supplying a lot of dc current or not, and when you set it to ac current, whether it is supplying a lot of that (which it shouldn’t be).

 

With diode failure is is possible that current is back-flowing into the alternator when the engine isn’t running. You could eliminate this by disconnecting the fat (charging) wire from the alternator at the battery end and see if the solar charges the batteries better / batteries hold their charge during the day better.

 

 

Is that actually true? Surely each phase sees an AC current so losing a diode means that one phase has no current so there is less current flow/heat in the alternator. There will be a bigger AC component to the battery and through a tine of bit of wire in the alternator after the regulator.

 

Maybe because the OP has now decent batteries the alternator is just working harder for longer to charge them?

 

...............Dave

[Keeping Up]

4 minutes ago, dmr said:

 

Is that actually true? Surely each phase sees an AC current so losing a diode means that one phase has no current so there is less current flow/heat in the alternator. There will be a bigger AC component to the battery and through a tine of bit of wire in the alternator after the regulator.

 

Thats true if the diode has failed as open circuit, but what if it has failed as short circuit? That phase will then spend two thirds of its time taking current from the others

[Dr Bob]

3 minutes ago, dmr said:

 

 

Maybe because the OP has now decent batteries the alternator is just working harder for longer to charge them?

 

...............Dave

Yes.

 

The voltage numbers given seem to suggest a high load taking the down fro 12.7v to 11.?? Overnight which on 2 new batteries is 150Ahrs plus.......so not consistent with just a fridge on. Is the alternator just fighting to get the charge in?

 

The OP doesn't mention if he has shore power. Was that 13.3v measurement taken when the bank was at 11 volts? Interesting to see the voltage If the batteries can be 'half' charged using a battery charger. 

 

More info needed.

[dmr]

2 minutes ago, Keeping Up said:

Thats true if the diode has failed as open circuit, but what if it has failed as short circuit? That phase will then spend two thirds of its time taking current from the others

Ooooh I need to think about that, though I think short circuit failures are very rare.

 

...............Dave

[nicknorman]

9 minutes ago, dmr said:

 

Is that actually true? Surely each phase sees an AC current so losing a diode means that one phase has no current so there is less current flow/heat in the alternator. There will be a bigger AC component to the battery and through a tine of bit of wire in the alternator after the regulator.

 

Maybe because the OP has now decent batteries the alternator is just working harder for longer to charge them?

 

...............Dave

As KU says, diodes can fail short circuit. But of course even if it fails open circuit, the regulator will supply max field current to try to get the voltage up. More significantly I think, power dissipated by the resistive element of the stators is I squared R. So if the current is not evenly distributed within the 3 phases (other 2 phases having to work harder to compensate for failed phase), the I squared element means that the overall heating is a lot more for the same total current.

[MtB]

22 minutes ago, Dr Bob said:

Difficult to tell by hand as they do get to 90 deg C.

 

 

And to expand on this, it is a phenomenon that the hottest temperature the human hand can bear to hold is about 60c for most people. Above that we are forced to let go of the hot object or surface.

 

Consequently it is very hard for us to judge temps above 60c by touching them. One wy to tell if it is above 100c is to spit of one's finger then touch the surface. If it sizzles it is above 100c but doesn't hurt if you are quick!

 

 

[dmr]

Or    the field diodes have gone??? I think I have known these to fail short circuit, so the alternator is draining the batteries, and getting warm all the time? I think it should still charge ok though????  

Too busy boating to think too much about leccy!

 

..............Dave

4 minutes ago, nicknorman said:

As KU says, diodes can fail short circuit. But of course even if it fails open circuit, the regulator will supply max field current to try to get the voltage up. More significantly I think, power dissipated by the resistive element of the stators is I squared R. So if the current is not evenly distributed within the 3 phases (other 2 phases having to work harder to compensate for failed phase), the I squared element means that the overall heating is a lot more for the same total current.

I suppose we could calculate this if we really wanted to, but Im going boating.

 

................Dave

[Alan de Enfield]

11 hours ago, rgriffiths said:

I have had the alternator tested (twice) and everything is fine. On the test bed at the alternator repair place it does not get hot at all but I am not sure of that is significant.

I realise that it is all speculation at the moment, but surely, if the alternator has been tested (twice) by a reputable 'alternator repair place' you would have expected it to have been checked thoroughly before saying it is OK.

 

The fact it does not 'get hot' at the 'alternator test place' would lead me to think it is something related to the boat.

 

Is the immersion heater on and powered by the inverter ? (you can admit it - I've done it)

Have you looked / smelt / felt the batteries ?

Have you got a DC clamp meter yet ?

Edited August 4, 2019 by Alan de Enfield

[Keeping Up]

To me one telling observation is that the charging is at 13.3v rising to 14.1v when one battery is disconnected. That sounds like two very flat batteries to me, with the alternator working at it's maximum current into 2 batteries and probably at or near maximum into just one. My own 170 amp alternator would be producing 185 amps into 13.3 volts, if my BM2 ammeter is to be believed. So perhaps it just needs a long long time to charge them properly, during which time it will indeed get very very hot.

[Tony Brooks]

2 hours ago, Chewbacka said:

With an ac generator which is then rectified to dc the out put is not smooth, difficult to explain, but with a transformer the output is 50Hz AC after going through a bridge rectifier the output is dc but it goes from 0v to full volts 100 times per second.  With a power supply it then has smoothing capacitors to give a more constant voltage.

With a vehicle alternator it is (I think) a three phase rectified output without smoothing capacitors so there will be a significant amount of ripple, especially at full load though a good battery will be smoothing this a lot.  So getting to my point, I would expect that if you set the clamp meter to AC you will read this ripple but I don’t know how much as I have never measured it on mine.

So an AC reading may not mean faulty alternator, or am I wrong??

To try to answer your question.

 

When you look at the battery voltage during charging on an oscilloscope you will see the ripple but I have never measured the ripple voltage. Going by the height of the ripple over the charging voltage I can say its not that much, only a volt or two above charging voltage. Where looking at the ripple is useful is to see if a phase has gone down for some reason. The missing hump is then easy to spot.. However I very much doubt the OP has access to a scope.

 

 

9 minutes ago, Keeping Up said:

To me one telling observation is that the charging is at 13.3v rising to 14.1v when one battery is disconnected. That sounds like two very flat batteries to me, with the alternator working at it's maximum current into 2 batteries and probably at or near maximum into just one. My own 170 amp alternator would be producing 185 amps into 13.3 volts, if my BM2 ammeter is to be believed. So perhaps it just needs a long long time to charge them properly, during which time it will indeed get very very hot.

That is perfectly possible but to be sure we still need the current and voltage readings. Those will prove the faulty   item be it just very flat batteries.

[ditchcrawler]

12 hours ago, rgriffiths said:

 

• I have changed the controller just in case there was a problem there.
• Controller was reading 5 to 6 amps charging today.
• There is not a great drain on the system other than the the fridge.
• No shorting in the system that I can find.

 

I realise that this is vague but I am hoping someone might give me some more things to test and more than anything give me some idea why an alternator which has test fine is getting so hot.

 

As ever all help gratefully received.

Controler, is this an alternator controller of a solar controller and is the 5 to 6 amps whats coming from the alternator or solar. Your fridge will probably use about one battery full of power in 1 day a 110 ah battery is only good for about 55Ah. If your batteries are going down to 11 volts they are very flat and will take quite a bit of recharging. 200 watts of solar at best angle to the sun and good sun will only give a maximum of  14 amps and you will be nowhere near 100% efficiency so I am not surprised at 6 amps. If the batteries are that flat it will take your solar probably 24 to 30 hours of sunshine  without you taking a single Ah out. Your alternator is probably pumping out its maximum current  trying to recharge your batteries. If your alternator is something like 70 Amp then it will be dissipating something like a Kw of heat.

[cuthound]

2 hours ago, Tony Brooks said:

To try to answer your question.

 

When you look at the battery voltage during charging on an oscilloscope you will see the ripple but I have never measured the ripple voltage. Going by the height of the ripple over the charging voltage I can say its not that much, only a volt or two above charging voltage. Where looking at the ripple is useful is to see if a phase has gone down for some reason. The missing hump is then easy to spot.. However I very much doubt the OP has access to a scope.

 

 

That is perfectly possible but to be sure we still need the current and voltage readings. Those will prove the faulty   item be it just very flat batteries.

 

^^^^ This.

 

Batteries are very good at smoothing out ac, because they act like a capacitor. The ripple is only visible on a 'scope and will be pretty small in amplitude. 

 

If the OP has measured charge and rest voltage and current, a d inspected the batteries for signs of shorts and found none, then it must be flat batteries.

[dmr]

This is a waveform of charge current

 

the ripple is the switching of the regulator, dunno if all alternators are similar. If we zoom in then the ripple from the phases of the alternator are visible.

 

This is my old long since gone domestic alternator and I think the Adverc was connected.

 

........Dave

 

 

 

 

[Chewbacka]

Looks like you have a 5amp ripple, this would probably show on your clamp meter on the ac setting, so be careful if diagnosing an alternator with an ac clamp meter.  It may be just fine.