When do you replace the alternator?

[AlleyCat67]

When I overhauled the engine on the Tiger a few years ago, the Aerotech alternator had 700 hours. The A&P said that was near end-of-life and I should replace it ASAP. Of course, my next A&P disagreed and said I should just keep it until incipient failure. That was 400 hours ago and (knock on wood) it's still running without issue. So, when do you all recommend replacing the alternator? And what would be an early indications of failure (other than a major discharge & quick landing)?

 

[Dan Thomas]

In aircraft, the alternator brushes wear faster than in a car because that alternator is running near its redline all the time in cruise, and the brushes get the worst of it. When the brushes get worn enough they pop out of their holder and then the brush springs chew up the rotor slip rings and trash the alternator. You end up buying a new alternator.

For this reason, manufacturers like Cessna recommend alternator brush inspection every 500 hours. If you have that done, and brushes are replaced when they need it, you will eliminate 90% of alternator failures. The rest of the problems mostly have to do with frayed and deteriorating alternator wiring. Regulator failures are rare, and they'll often start acting up to give some warning. Annual inspections are supposed to catch crappy wiring, but I've found so much of it that I know for a fact that too many mechanics are just not looking for stuff. They hold to the bankrupt philosophy that says "If it ain't broke, why fix it?" That doesn't work well in aviation.

Alternator brushes are cheap. Alternators aren't. Alternator failure at a bad time, like at night or in IMC, is a very bad thing. There is a constant stream of posts on POA detailing the alternator failures experienced by members, and it's so totally unnecessary.

The answers are found in the maintenance manuals. When all else fails, read the directions. From a Cessna manual:



See that 7 in the Special Inspection Item column? Here:




Every 500 hours. You're at 1100 hours. I used to find the brushes worn about 30 or 40% at 500, and they'd be bad enough at 1000 hours that they'd never make it to 1500, so they'd get replaced. They're cheap. It's the labor to take the alternator out, open it, inspect stuff and change the brushes. But it's cheaper than a trashed alternator, which costs a lot more money and you still have the labor to contend with, along with the risk factor. Just having the thing die on you at some away-from-home airport can cost a pile of money and time.

While the alternator is apart the bearings can get checked, too. To the mechanics here: DON'T put a lot of grease in that rear bearing. It will squeeze out when the rotor shaft goes back in and it will get on the brushes and slip rings and make a resistive sludge that will limit field current and the alternator's performance will be really weak. Kelly/Hartzell and others have repeatedly made this mistake for years. Exchange alternators that fail within a few hours.

Note, too, in the first image, the 100-hour (annual) requirement to check the alternator wiring (connections). That's the other thing that often doesn't get done.

 

[Bell206]

when do you all recommend replacing the alternator?

To add to the above, most components/accessories will have recommended mx schedules. By following those you run a better chance at getting the max life out of the item. Here's another reference ICA that takes you up to 2000hrs/12 year on a different type alternator.
http://www.tcmlink.com/pdf2/sb11-2.pdf

 

[thomasdr72]

Pretty much all of your accessories should opened/cleaned/serviced each 500 hours, sooner if you don't fly much (as per table above). If you get into this habit, the parts cost is usually minimal. Labor is a bit more. Failures should become nearly non-existent...

V/r,

 

[AlleyCat67]

Good information all. I will dig out the maintenance manual for a consult but otherwise plan on a servicing.

 

[Dan Thomas]

Good information all. I will dig out the maintenance manual for a consult but otherwise plan on a servicing.

You might also see what times are on some other components. Magnetos and vacuum pumps are also commonly-ignored items that can cause plenty of grief when they're run to failure. Bendix (now CMI) magnetos are a 400-hour item. Slicks are 500 hours. Vacuum pumps, as per Cessna, are a 500-hour item, but if you get a Rapco or Tempest pump you'll have a vane wear inspection provision that lets you run the pump until its vanes reach the minimum safe length instead of just replacing it at 500 hours to be safe. First check is typically at 600 hours and then every 100 or annual after that. Takes three minutes. Gets a lot more hours out of your pump while avoiding failure. Vacuum pump failure is almost always due to the vanes getting too short and cocking and breaking the carbon rotor. A few pumps fail because the mechanic, while spraying solvent over the engine to wash it, gets solvent into the pump drive. It gets into the pump and gums up the carbon stuff in there and it seizes and breaks. The Tempest pump, at least, has a neat little seal to prevent that.

There's also as six-year limit on the plastic vacuum pump drive shear coupling. It's there to purposely fail if the pump seizes up good and proper. You don't want the accessory drive gearing damaged, so the drive coupling shears. It ages with time and heat and so on and can fail even if the pump is OK, hence the six-year recommended replacement.

 

[guzziguy]

Personally, I'd like to congratulate the OP for having an aircraft in such great shape that random failures on other components of his aircraft don't occur and he is now forced to go looking for things to fix!
Well done! I wish I had your problems!

 

[AlleyCat67]

Heh. It's taken two years to deal with (and pay for) all the deferred maintenance from the previous owner. But at least I'm now on to stuff that's wearing out because of MY flying rather than his.

 

[Huckster79]

Thank you for a great thread. My alternator is coming g out this annual to replace the tach cable seal. I am going to look into having the brushes replaced and such while it’s out anyway

 

[Domenick]

...Alternator brushes are cheap. Alternators aren't.
...
While the alternator is apart the bearings can get checked, too. ...

Got sources for brushes and bearings?
Specifically for a PA-28-161: Chrysler 4111810 FCH

While I'm here I'll ask the same question about my starter: PrestoLite MZ4222R

 

[nrpetersen]

And the front mount of belt driven alternators on 4 cylinder Lycomings can be expected to give trouble. The resonant frequency of the alternator on the mount I tested was about 80 Hz which would be excited by the engine's second harmonic at 2400 rpm. The whole engine squirms about it's CG at that frequency. The vibration environment on the front of these engines is tough.

 

[Dan Thomas]

And the front mount of belt driven alternators on 4 cylinder Lycomings can be expected to give trouble. The resonant frequency of the alternator on the mount I tested was about 80 Hz which would be excited by the engine's second harmonic at 2400 rpm. The whole engine squirms about it's CG at that frequency. The vibration environment on the front of these engines is tough.

Yup. The U-bracket that holds the alternator to the engine can crack and break in the bend radii. It's a tight bend for a thick piece of steel, so they bend it while annealed and then heat-treat it. So now it's a bit brittle, and they often don't smooth the edges before bending it, so small cracks are likely already there when it's new. They also don't chamfer the bolt holes, so they break there too. These were one of the for-sure items to check every 100 hours.

 

[Dan Thomas]

Got sources for brushes and bearings?
Specifically for a PA-28-161: Chrysler 4111810 FCH

While I'm here I'll ask the same question about my starter: PrestoLite MZ4222R

That stuff can be had at Aviall (now Boeing). The big paper catalog I had before I retired can be found online here: https://catalogs.shop.boeing.com/Aviall-Product-Catalog-and-Source-Book/C/
Go to the Engine Accessories and you'll find the parts breakdowns starting on page 342.

IIRC that alternator is a Motorola unit.

From https://www.aircraftspruce.com/catalog/pnpages/07-01504.php I get this:

Q: What parts are included in 90-3057 HARTZELL ALTERNATOR O/H KIT FOR 4111810R?
Part number 90-3057 consist of 1 each of the following:
74314 BRUSH ASSEMBLY, NEG, ES4314 x 1
X-4130S BEARING, REAR x 1
74312 BRUSH ASSEMBLY,POSITIVE,ES4312 x 1
74129 NUT & WASHER ASSY(19-24)ES4129 x 1
79401 X-4612, BEARING, BALL x 1
74309 LOCK TAB - THRU BOLT ES4309 x 3
74310 BOLT, 10-32, HEX HEAD, ES4310 x 3
74336 CAPACITOR ASSEMBLY, ES4336 x 1
74337 RECTIFIER, POSITIVE, ES4337 x 1
74338 RECTIFIER, NEGATIVE, ES4338 x 1

I don't know if AS&S sells brushes alone. The ES numbers are Electrosystems numbers, available as parts from Aviall. You can find the part numbers in that Aviall catalog and try AS&S if you want.



Or just try Aviall.

 

[nrpetersen]

Yup. The U-bracket that holds the alternator to the engine can crack and break in the bend radii. It's a tight bend for a thick piece of steel, so they bend it while annealed and then heat-treat it. So now it's a bit brittle, and they often don't smooth the edges before bending it, so small cracks are likely already there when it's new. They also don't chamfer the bolt holes, so they break there too. These were one of the for-sure items to check every 100 hours.

That wasn't the alternator design I encountered. This one (what's it from?) is much worse. What are they thinking? Short bolts, directly threaded into the crankcase, relying on safety wire to keep from coming apart, and only two fasteners co-linear (should be at least 3 in a triangle arrangement) so that the bracket can't resist vibration in the engine's roll axis. A future AD waiting to happen.

This would have been an early Harley Davidson approach except it would have to be chrome plated!

 

[Domenick]

That stuff can be had at Aviall (now Boeing). The big paper catalog I had before I retired can be found online here: https://catalogs.shop.boeing.com/Aviall-Product-Catalog-and-Source-Book/C/
Go to the Engine Accessories and you'll find the parts breakdowns starting on page 342....

Thanks!

 

[kyleb]

Note, too, in the first image, the 100-hour (annual) requirement to check the alternator wiring (connections). That's the other thing that often doesn't get done.

Yep. That on is on my "inspect every time the cowl is off" list. I had the lug on the alternator cable fail at the next startup after I returned from Oshkosh one year. I was lucky that it didn't fail on the way to or from the show. It would have been pretty inconvenient...

 

[YKA]

In aircraft, the alternator brushes wear faster than in a car because that alternator is running near its redline all the time in cruise, and the brushes get the worst of it. When the brushes get worn enough they pop out of their holder and then the brush springs chew up the rotor slip rings and trash the alternator. You end up buying a new alternator.

For this reason, manufacturers like Cessna recommend alternator brush inspection every 500 hours. If you have that done, and brushes are replaced when they need it, you will eliminate 90% of alternator failures. The rest of the problems mostly have to do with frayed and deteriorating alternator wiring. Regulator failures are rare, and they'll often start acting up to give some warning. Annual inspections are supposed to catch crappy wiring, but I've found so much of it that I know for a fact that too many mechanics are just not looking for stuff. They hold to the bankrupt philosophy that says "If it ain't broke, why fix it?" That doesn't work well in aviation.

Alternator brushes are cheap. Alternators aren't. Alternator failure at a bad time, like at night or in IMC, is a very bad thing. There is a constant stream of posts on POA detailing the alternator failures experienced by members, and it's so totally unnecessary.

The answers are found in the maintenance manuals. When all else fails, read the directions. From a Cessna manual:

View attachment 93192

See that 7 in the Special Inspection Item column? Here:

View attachment 93193


Every 500 hours. You're at 1100 hours. I used to find the brushes worn about 30 or 40% at 500, and they'd be bad enough at 1000 hours that they'd never make it to 1500, so they'd get replaced. They're cheap. It's the labor to take the alternator out, open it, inspect stuff and change the brushes. But it's cheaper than a trashed alternator, which costs a lot more money and you still have the labor to contend with, along with the risk factor. Just having the thing die on you at some away-from-home airport can cost a pile of money and time.

While the alternator is apart the bearings can get checked, too. To the mechanics here: DON'T put a lot of grease in that rear bearing. It will squeeze out when the rotor shaft goes back in and it will get on the brushes and slip rings and make a resistive sludge that will limit field current and the alternator's performance will be really weak. Kelly/Hartzell and others have repeatedly made this mistake for years. Exchange alternators that fail within a few hours.

Note, too, in the first image, the 100-hour (annual) requirement to check the alternator wiring (connections). That's the other thing that often doesn't get done.

I concur. 1000 hours is all, they are easy to replace, so I have done the brushes myself several times, and done them on a friends plane three times. Not a hard job at all.

 

[PaulMillner]

And the front mount of belt driven alternators on 4 cylinder Lycomings can be expected to give trouble. The resonant frequency of the alternator on the mount I tested was about 80 Hz which would be excited by the engine's second harmonic at 2400 rpm. The whole engine squirms about it's CG at that frequency. The vibration environment on the front of these engines is tough.

Concur! We did some vibration testing with the DSS folks a while ago now... and the alternator was vibrating away, like several ips (inches per second). While balancing the alternator is one option (where's National now that we need them?!) Lycoming offers two different u-shaped mount brackets, thick and thin... sometimes changing to the other one sufficiently detunes the system to arrest the resonance problem.

I'm also a great advocate for alternator brush R&R on a 500 hour basis. Remember, too, that the brushes arc more at altitude, due to poor insulating properties of thinner air. It used to be that aviation brushes were of different materials and had different style connectors, but that distinction may have gone away with time, at least for my Ford-Aeromotive alternator.

 

[Dan Thomas]

If you're having persistent problems with cracking or loosening alternator brackets you should check to see that all the correct parts are installed and secure for your model. For instance, the O-320-A has stuff like this:



You're looking at it from the bottom. #1 is is an alternator bracket; in some other models the bracket is a little different, or a single bolt is used to hold the alternator instead of the two shown here (#17 and #21). Some models have #3 for fore-aft bracing. All of them should have #14, a strut that runs from under the alternator bolt over to a lug on the starter. All STC'd starters should also have that lug. That strut stops the bracket rocking on its pad on the engine and absorbs the thrust on the bracket created by belt tension and alternator torque reaction. #6 sets the belt tension and holds the outer side of the alternator still. I have seen some really sad homemade versions of this, especially on Continentals. If all of this stuff is the right stuff and it properly attached and torqued, everything should work well.

Not all bracket bolts are drilled-head for lockwire. Some use the locking tab #2; you bend the rounded sides up against the head flats after torquing.

The big Continentals (470, 520) have a variety of mounting systems for the alternator, and those parts MUST be installed as a set. Older versions had shock-mounted alternators with the adjusting arm also shock-mounted. Using the newer non-shock-mounted bracketry with the old shock-mounted arm, or the new rigid arm with an old shock-mounted bracket, is asking for endless trouble.

 

[nrpetersen]

The loosening torque for a #2 locktab is just as bad as safety wire. If bolts 5 were long and with nuts, there would be preload. Having three would be even better. Being only two and short & with direct threads into aluminum it is a marginal design worthy of a Harley.