Alternator Troubleshooting

[ralay]

We've had an intermittent alternator problem for the last 2 years that has been driving us batty.  I know it's pretty difficult to successfully diagnose anything over the internet, but I'm posting it anyways, because we're really dumbfounded at this point.

Here's some background:
Engine: Yanmar 3QM30H, starts fine, runs fine
Batteries: new this year AGM marine batteries, work fine in every regard, are charged fine by other sources (ex~ solar)
Alternator: Yanmar 35A, internally regulated, dedicated ground wire

Problem: Most of the time the alternator won't put out anything (voltage or amperage).  I keep getting ready to toss it, but then it will randomly wake up and start putting out 35 A for 5-10 minutes, which makes we worried that we have some other problem a new alternator might fix.

What I've already done/probably eliminated:

Installation: the thing worked perfectly for the first 6 months after I installed it, so presumably it's wired up right.

Batteries: We've bought new batteries since the problem started and it's had no effect. 

Wiring:  The alternator output shares 1/0 wires with the starter from the battery to the starter, then has it's own 2G cable for ~1ft from starter to alternator.  It also has a 2g ground cable ~2ft long.  I've removed and polished every connection.  It made no difference.  The alternator has one wire for excitation.  This circuit is as follows:
batt > key > warning lamp > alternator > gnd 
I've also removed, polished, and tightened every connection here to no effect. 

If I measure the voltage between the alternator output post and ground or the excitation wire and ground (with key on), the voltage is exactly the same as what I read across the battery terminals.

Furthermore, I've measured the voltage across the alternator output post and the ground post with the engine running and it's still exactly battery voltage. 

This leads me to believe that the problem is not a bad connection/high resistance in the wires.

What's Left?

So the alternator itself could be toast.  I had it tested at an auto electric shop and they told me it was toast (but they also told me it was externally regulated, so I dunno about "experts".)  It's also frustrating that it occasionally works at it's full capacity for short bursts.  What could have failed inside that is only broken part of the time?

The the other possibility is that there is a belt problem.  The belt never squeals, though I do often hear a buzz.  I tested the alignment of the alternator and the crankshaft pulley with a dowel and put some washers in to get the alternator lined up just right.  Unfortunately, it doesn't have a tach connection or I could use the tach output to help decide if it's slipping.  We have a Yanmar belt on there now, which has a different bevel than a regular 3/8" belt.  We have it pretty freaking tight.  It's really hard to imagine this being the problem, but I'm ready to be surprised.   

The last funny bit of diagnostic evidence, I've noticed is that the alternator warning lamp doesn't turn on when key is turned (but before the engine is started - we have a push button start).  The oil pressure lamp lights, but not the alternator lamp.  The lamp isn't burned out.  It'll light if you ground it and there is 12V on the far side of it.  But it doesn't come on when the alternator isn't working and it does flicker a little just before the alternator resurrects itself. 

Anyone have any ideas?  Any further tests?  I don't want to drop $130 on a new alternator only to find out we've got some other problem.  The only other thing I can think of to try is to get a laser tach and shoot the crankcase pulley and alternator pulley and calculate whether there's slippage.

[Frank]

I had a similar problem a few years back and paid for a new one...did the same.
In my case....although the wire 'looked good'....it was frayed and nearly severed under the 'wrap' that went around the wire to connecter.....so out of sight.
Not saying that is your issue.....just giving you one more thig to check.

Also....too tight on the belt is hard on the bushings...

My 1.5 cents
about all this is worth  :)

[ralay]

I've thought about that.  But the fact that I'm not reading 14+V at the alternator terminals makes me think we've got other problems. 

I ordered a laser tach ($14) on amazon prime today and had it shipped ahead to a friendly face in Norfolk.  I'll also probably order an aftermarket alternator (the Yanmar alternator is $900 for a 35A internally reg alternator!).  If the problem is slippage, the second alternator will be our spare.  If it's not slippage, I'll put the new one on and try to crack open the other one and to see what the heck is going on. 

Our solar panels are doing a good job of meeting our demand, but I want the alternator working before we go offshore.  Never know when you're going to need to be running the bilge pumps.

[Norman]

From the voltage readings and charge light tests, most problems are eliminated.

Your problem is inside the alternator.  The circuit from the light goes to the regulator and then to the field winding.  There should be a path through the excitation winding to ground.


1st, if the brush is worn or its tension spring is corroded and weak, there will be no circuit for excitation, but intermittently contact may take place.  The brush is on the end away from the belt, and is usually held in with a small screw or bolt.  Some alternators have two brushes.  If this is the problem, take the brush or brushes to a shop or auto supply store and see if they can find a match.  Take the alternator too, and check fit.  Otherwise, order the correct brushes.

2nd, If the wire from the terminal to the regulator is loose, corroded, or broken, the same symptoms will be present.  You need to take the end off to get to the internal wire to check this problem, and THE BRUSHES MUST BE OUT TO DO THIS!  Scratch a deep line across each end bell and the adjacent metal before disassembling to assure that the parts go together properly aligned.  When you reassemble, put a drop of thread lock or fingernail polish on the through bolt threads to prevent loosening from vibration.

Least likely, the regulator electronics have an intermittent failure, and it needs replaced.

I hope that this provides the answer to your charging issues.

The AC windings are good, as are the diodes.

[ralay]

Norman, Thanks for the ideas.  If I have a spare, I'll definitely try to get it apart for diagnosis, though I may need to wait a few days til we get to Norfolk.  Is it possible to measure the continuity/resistance between the lamp terminal and the ground terminal when the alternator isn't excited?  Or do I have to have the case open? 

[Norman]

There should be continuity from that terminal to ground!

Have you pulled out the brushes yet?  They should be easy to do, but if the back of the alternator is inaccessible, naturally, the alternator must come off.

[CharlieJ]

Not saying that is your issue.....just giving you one more thig to check.

Also....too tight on the belt is hard on the bushings...

and she will, for sure. The lady is VERY thourgh

[ralay]

I just tested the continuity between the lamp and ground terminals - open.  That would explain why the warning light doesn't work and why the alternator isn't working.  I can't just pull the brushes out the back.  The case has to be opened to get at them.  The last time I tried, I couldn't separate the halves.  I think there is a press fit into the rear bearing or something.  May have to devise a method or just apply more force (Woody is my "enforcer" in the absence of mechanical advantage.).  I don't need the alternator mounted to run the engine (water pump is camshaft-driven) so maybe I can wrestle with it on our way up the ditch. 

[Owly055]

The idiot light cannot go to ground or it will be on all the time, weather the alternator is charging or not.  It will connect to one of the windings in the alternator in a way that it will have ground when the alternator is not charging.    Alternators are 3 phase AC units, usually star wound, and where all 3 stator windings join at a common point is a common place to attach the idiot light.  It can also attach at the positive brush terminal.  At this point it will see a ground when the alternator is not charging, but not when it is charging.  You have a positive and a negative brush, and your regulator may regulate the field by controlling either the positive or negative.  In the past, most American alternators regulated on the positive side.... now it's anybody's guess unless you take it apart and study it out.     Obviously your diodes are OK, or you would never get any significant output. The diodes convert the 3 phase AC to DC, and are a fairly common failure point.   If this is an internal regulator alternator as you say, I would buy a regulator and replace it.   Remember when you take the alternator apart that the stator (windings) which are wrapped around the laminated steel section between the forward and rearward half of the alternator stays with the rear.  The rear of the rotor shaft comes out of the rear bearing, which on most alternators is a needle bearing, so there should be virtually no resistance to coming apart.   The most likely scenario for an alternator not wanting to split is that the stator assembly is releasing from the rear half of the alternator, rather than the front half.  This leaves the output wires attached, and holding the alternator from coming apart.    The forward portion should separate from the rear, leaving everything in the rear.  If the laminated steel portion (stator) is trying to come with the forward portion, you need to insert a screwdriver or something and pry it loose so it stays with the back of the alternator.  Some European and Japanese alternators allow brush access from the rear, or even incorporate the brushes into the regulator assembly which can all be removed from the rear.

                                            H.W.

[ralay]

Stator windings with the rear. Gotcha.  Still can't budge it.  It's a dual foot mount and I even got some threaded rod and two nuts to try to force the two feet apart while I pried on the other side.  No dice.  I should be able to separate the halves without disassembling the pulley, fan, etc., right?  The rotor isn't anchored to the rear except by bearings, right?

[ralay]

Took pulley, fan, front case off.  Rotor is NOT budging.  Definitely not just needle bearings back there.  Must be pressed into back of case.  How the heck do I pull it out of there? 

[Norman]

The brushes lock the rear cover on.  The commutator has slip rings between raised insulation barriers, and if forced, either the brush housings or those insulator will be broken.

The above is true on MOST alternators, yours may be an exception.  I have ruined an alternator that way many years ago.

Look again for a screw or bolt that can be removed on the cover.  They are not always clearly identifiable as brush retainers.  If your internal regulator is removable from the outside, it may have the brush holders integral with it.  On a marine unit, there may be a water tight cover on the brush holder, so be creative in deciding what may need to come off!

If you take the brushes out and pull the end case off, there Is usually a small hole to insert a paper clip to keep the brushes lifted as you put the end case back on, then the pin is pulled out to drop the brushes down on the commutator.

If you are still working on it when you get up in the Bay, I can try to solve the problem!

In the mean time, solar is great, isn't it?

I have serviced Ford and GM alternators, and nearly all had externally replaceable brushes.

Norman

[Norman]

PS!  The idiot light circuit provides current from the battery, thru the ignition switch, regulator, brush, field winding, brush, to ground.

If any part of that circuit is open, no excitation and no ground for the light.

That is why I deduced that the problem is in the series circuit inside the case.  You had determined that 12 volt battery reached the regulator terminal, thru th idiot light.  Your information was very complete, as was your external trouble shooting.

The current thru the light provides the initial field to start generating.  Alternators do not self excite, but must have external input to start. 

Norman

[Owly055]

     I have disassembled and overhauled hundreds of alternators and never once seen a raised barrier as you describe between the slip rings.... not once!   I of course cannot speak for the specific alternator here as I don't even know what make it is... I'm assuming a Hitachi most likely.   What is of concern here is that if a ball bearing is in the rear of the alternator and the bearing comes out of the case with the shaft, it may not pass the brushes without damage.
     I do not doubt that an alternator such as you describe does exist somewhere, there are a number foreign makes I have not worked on.    Any alternator built this way will have brush holders that can be removed from the rear like Bosch does, where the brush holder is integrated with the regulator.  As the brushes on the alternator in question apparently cannot be removed without dismantling the alternator this cannot be the case.
     I'd appreciate it if you would name a brand of alternator with a barrier such as you describe...... I'd like to see look up a blow apart of it.

                              H.W.

The brushes lock the rear cover on.  The commutator has slip rings between raised insulation barriers, and if forced, either the brush housings or those insulator will be broken.

The above is true on MOST alternators, yours may be an exception.  I have ruined an alternator that way many years ago.

Look again for a screw or bolt that can be removed on the cover.  They are not always clearly identifiable as brush retainers.  If your internal regulator is removable from the outside, it may have the brush holders integral with it.  On a marine unit, there may be a water tight cover on the brush holder, so be creative in deciding what may need to come off!

If you take the brushes out and pull the end case off, there Is usually a small hole to insert a paper clip to keep the brushes lifted as you put the end case back on, then the pin is pulled out to drop the brushes down on the commutator.

If you are still working on it when you get up in the Bay, I can try to solve the problem!

In the mean time, solar is great, isn't it?

I have serviced Ford and GM alternators, and nearly all had externally replaceable brushes.

Norman

[Owly055]

Took pulley, fan, front case off.  Rotor is NOT budging.  Definitely not just needle bearings back there.  Must be pressed into back of case.  How the heck do I pull it out of there?

Presumably you have no give whatsoever, so it has to be locked into the bearing...............   The front cover should slide off with the pulley and fan removed........the shaft should slide easily through the bearing in the front half............but that will get you nowhere at all.   The fact that is doesn't suggests considerable electrolysis  between the shaft and the bearings. 

Alternators that have a ball bearing in the rear are not particularly common....... at least not American ones.  If there is a rear ball bearing, there often is a steel plug in the rear covering the bearing, rather than just a blind cast pocket.   I've seen this on a number of foreign alternators.    Just shooting in the dark here, but if there is such a plug, I would try to remove it.   Engineers do some strange things, and it is just possible someone designed it with a ball bearing that in pushed into place from the outside of the case, and put a snap ring on the end of the shaft....... I've never seen this, but that doesn't mean that nobody ever did it.    I've worked on various kinds of machinery all my life, and if there is one thing I've learned it's that the capacity of engineers to make things stupidly complex knows no bounds..........

                                 H.W.

[Norman]

Owly055 is right on the mechanicals of recent alternators, I just checked the 2 on my bench, and those on my Mustang.

I pulled one Motorcraft apart, and it has needle bearings in back, and the slip rings are the same height as the insulation, but there is a slot between the two that the brush can drop down in coming out and trap the rotor in place.  This may be the problem I had years ago.

The big puzzle for Rachel is why the shaft will not come out.  Still, the only reason that I can visualize is the brushes catching on something.

Obviously, since it turns in service, if it is not brush related, it must be some sort of retainer on the outboard end of the shaft, but I have never seen that.  Avoiding patent infringement creates many weird designs.

H. W., clearly you have much more experience with automotive alternators than I.  I only work on my cars and a few friends.  In 50 years, I have only taken a car or truck that I or my sons owned to a garage for electrical repair once.  I keep my cars for more than 100,000 miles in most cases, so have done a lot of alternator brush replacements.  My Mustang is an 83, GMC truck 2000, and Fusion is 2011.

Fortunately, the engine can be run without the alternator!

Norman

[ralay]

Took the alternator to an auto electric shop in Great Bridge and had a fellow pull the rotor.  It was a bearing pressed into the case.  He told me it just took a lot of banging to get it out.  Brushes were worn down.  For $20, he put the rotor on a lathe and smoothed the slip rings and gave me 2 brushes, though they're not exactly the right size, so I may have to run back in the morning.  I noticed the solder for the brushes is REALLY hard to melt.  I have a small butane soldering iron that will melt regular 60/40 solder like nothing.  Is this stuff high temp solder?  Can I solder new brushes in with regular solder?

[Owly055]

Took the alternator to an auto electric shop in Great Bridge and had a fellow pull the rotor.  It was a bearing pressed into the case.  He told me it just took a lot of banging to get it out.  Brushes were worn down.  For $20, he put the rotor on a lathe and smoothed the slip rings and gave me 2 brushes, though they're not exactly the right size, so I may have to run back in the morning.  I noticed the solder for the brushes is REALLY hard to melt.  I have a small butane soldering iron that will melt regular 60/40 solder like nothing.  Is this stuff high temp solder?  Can I solder new brushes in with regular solder?

Alternator brush leads do not get particularly hot, so 60/40 is OK.  Be careful not to allow the leads which are often a kind of braided copper to wick up solder beyond the attachment point.  This will make them brittle.  Use as little solder as possible.  More is NOT better.  A soldering gun is best in my opinion.   A the hotter the tip, the less it heats the adjacent area before the solder melts, and the less the tendency for solder to wick.   Solder follows heat.

I've rarely ever had to solder brushes in an alternator.  Most have ends that attach without soldering.   Is this a Bosch or Hitachi?

Starters are the place where you cannot get away with ordinary solder on brushes, but fortunately most do not require soldering, Ford being the most common exception, and they need to be silver soldered due to the temperatures involved.   

                                         H.W.

[ralay]

It's a Hitachi.  I've seen the type of brushes you're talking about but these ain't them.  There's a plastic/resin assembly that the regulator and brushes are mounted on.  There are two tiny little tabs that the brushes were soldered to and 4 tiny little tabs the regulator was soldered to.  I clipped the brush I couldn't desolder and used a mini file to remove the old solder.  All in all, not a service-friendly alternator.  Every tutorial I've read had some other design easier to get into and swap.  But our engine takes an alternator with a 3.15" dual foot, so this is a hat we're working with unless I buy a $700 Balmar.

[Owly055]

It's a Hitachi.  I've seen the type of brushes you're talking about but these ain't them.  There's a plastic/resin assembly that the regulator and brushes are mounted on.  There are two tiny little tabs that the brushes were soldered to and 4 tiny little tabs the regulator was soldered to.  I clipped the brush I couldn't desolder and used a mini file to remove the old solder.  All in all, not a service-friendly alternator.  Every tutorial I've read had some other design easier to get into and swap.  But our engine takes an alternator with a 3.15" dual foot, so this is a hat we're working with unless I buy a $700 Balmar.

     There is nothing wrong with Hitachi alternators, but I find Nippondenso electronics to be the most reliable in the industry.   the last two cars I've had went over 250K miles with Nippondenso alternators and starters, and were still on the originals when I sold them.  The engines were also all original and still running like a top and not burning any significant oil.  I run only the best oil and change it every 2500 to 3000 miles.  This modern stuff lasts a surprisingly long time.   
     As a long distance cruiser, it is probably best to buy brand new when it comes to electrical components.  Your engine can be critical in a number of circumstances such as fighting current or staying off a lee shore, not to mention the importance of keeping the house batteries up so your electronics work.  We've come to depend on things like AIS and radar, GPS, chart plotters, SSB, and our laptop computers and tablets.  It's good to have a sextant and know how to use it, but even then you need an ultra reliable time piece and paper charts which many people don't carry anymore.   Just imagine setting out on the transpac and losing all electronics, then trying to hit Lihui, or the eastbound leg of the ARC shooting for Horta without your trusty GPS.   I would hope that most people take sextant fixes and use dead reckoning and compare it to their GPS fixes, but I suspect few actually do.  We've become very dependent on technology.   I grew up long before this technology came along and learned to navigate with a non-topo map and a compass on cross country (off trail) hikes in the Cascades.   Now I seldom see anybody in the back country without GPS, and a cell phone if not an emergency Sat phone, and often some sort of computer, and few if any of them ever leave the trail system.........It's a different world, and one I find a bit absurd.    Long distance sailing it just makes sense to use whatever technology is available.  I wouldn't even consider setting out across the Atlantic without GPS, probably at least two of them, and at least an AIS if not also a radar, and some form of electronic charts if only my Ipad.  I would also insist on having at least 3 ways to charge the house batteries.... engine, solar, wind or tow generator.   I want an engine that starts and runs reliably and replacement parts for critical systems.

                                         H.W.
.