First failure

[gismo]

He had a family function to go to in the afternoon. I'm not sure if he plans to just drop it until Monday or continue tomorrow. Hopefully the latter.

One can hope.

Thanks. According to my mechanic's earlier logic, since the deficit is still load dependent it would have to be the new alternator, but obviously that isn't necessarily the case here. I think he might be suspecting the alternator, since he's consulting with T&W for troubleshooting help, but I'm not sure.

Sounds to me he's still just guessing, not diagnosing.

That part he doesn't agree with. He says that 13.4 is plenty to charge the battery and so is just fine. Anything more, I gather, he believes would cook the battery.

In this he is very, very wrong. There is no doubt a voltage range specified in your airplane's service manual (he does have that I hope) and I'm certain 13.4 is outside that range. Technically that makes the airplane unairworthy, at least for night or IFR flight.

Yeah, I had the same thought, though I have no reason to suspect him of dishonesty and he did say that the rotor tested bad. That's all he said, I was planning to ask him if they had given further details the next time I talk to him.

More likely misunderstanding or lack of knowledge than dishonesty.

It was that, plus change the tires and brake pads, and now the electrical problem. All of that is done, except he has yet to add the oil. The electrical problem is the stickler since he clearly does not know how to troubleshoot it on his own. That's the part that worries me.

I will describe your strategy to him and see what he says.

I suggest you print out the charging system guide I posted a link to earlier in this thread and provide him a copy. Be sure to point out the troubleshooting procedure at the end.

And FWIW, you could add the oil yourself.

 

[Skip Miller]

An oil change on a single takes multiple days????

No, getting the A&P to do the oil change takes multiple days.

Remaining culprits are the new alternator (unlikely but possible),

Check with Kareem about Kelly quality.....

-Skip

 

[azure]

The new alternator is NOT a Kelly, at least not any more if it ever was, as it's been rebuilt by a different company that I've never heard of, called (I think) T&W Electrical.

And Lance, I've read through that article and I'm not sure what to make of that troubleshooting guide. The commentator (Nuckolls?) disputes much of what the author writes and doesn't seem to think the guide would be of much use to a novice technician. I take it you disagree? Would you follow the procedure as given, or would you modify it?

 

[Steve]

Keep in mind a Piper electrical system isn't necessarily the same as a Cessna electrical system.

 

[azure]

I spent about an hour and a half with my mechanic last night, standing on the brakes while he checked voltages in about a half dozen different spots with his digital VOM and also measured current draw with an inductive ammeter as I tried different loads. Here is what we found:

1. With minimal load (panel lights & TC only), the JPI reads 14.0 volts.

2. Adding all the avionics brings that down to about 13.8.

3. Adding strobes and nav lights does very little.

4. Adding the beacon brings it down to 13.3-13.4.

5. Adding the pitot heat brings it down to 12.3-12.4.

6. The pitot heat only draws about 5 amps. The beacon draws about 10.

7. A single reading check on the JPI showed its battery voltage to be accurate for that reading.

8. The alternator output with everything on except the beacon, including the pitot heat is 12.8 volts (battery voltage according to the JPI was 12.5).

9. "Everything" excludes the landing and taxi lights, which are still disconnected for replacement.

This is all pretty much consistent with what I observed in flight for the six months I've owned the airplane (except that I never had any reason to run the pitot heat). So I'm satisfied that I did in fact have an alternator failure and the new alternator is fine. However, the battery voltage response to load is not good, and probably never has been. My mechanic checked resistances across the master and all the circuit breakers except the 60 amp alternator CB (due to rat's nest wiring under the panel) and couldn't find anything, but the system response under load doesn't lie. According to my mechanic, the behavior is the same with the old voltage regulator and with the one that came with the new alternator, so the regulator is unlikely to be the culprit. Also, the field voltage did not drop abnormally under load. Since the battery checks out fine (capacity up to spec), there must be some resistive loss somewhere in the alternator wiring or the CB. We both suspect the alt CB and he will untangle the rat's nest and check it, and probably replace the CB.

Other than that, his recommendations are:

1. Get a newer, electronic beacon (either LED or another low-current design).

2. Until I'm ready to do that, if I'm in IMC and need the pitot heat on, switch off the beacon.

3. Since the battery is charging fine except when both the beacon and pitot heat are on, he says it's airworthy and to fly it, but consider redoing all the wiring sometime in the future.

Comments?

edit: It's a little surprising to me that the pitot heat, which only draws 5 amps, brings down the bus voltage so much more than the beacon, which draws twice as much. More than surprising, it doesn't make sense to me. What am I missing?

 

[gismo]

I spent about an hour and a half with my mechanic last night, standing on the brakes while he checked voltages in about a half dozen different spots with his digital VOM and also measured current draw with an inductive ammeter as I tried different loads. Here is what we found:

1. With minimal load (panel lights & TC only), the JPI reads 14.0 volts.

2. Adding all the avionics brings that down to about 13.8.

3. Adding strobes and nav lights does very little.

4. Adding the beacon brings it down to 13.3-13.4.

5. Adding the pitot heat brings it down to 12.3-12.4.

6. The pitot heat only draws about 5 amps. The beacon draws about 10.

7. A single reading check on the JPI showed its battery voltage to be accurate for that reading.

8. The alternator output with everything on except the beacon, including the pitot heat is 12.8 volts (battery voltage according to the JPI was 12.5).

9. "Everything" excludes the landing and taxi lights, which are still disconnected for replacement.

This is all pretty much consistent with what I observed in flight for the six months I've owned the airplane (except that I never had any reason to run the pitot heat). So I'm satisfied that I did in fact have an alternator failure and the new alternator is fine. However, the battery voltage response to load is not good, and probably never has been. My mechanic checked resistances across the master and all the circuit breakers except the 60 amp alternator CB (due to rat's nest wiring under the panel) and couldn't find anything, but the system response under load doesn't lie. According to my mechanic, the behavior is the same with the old voltage regulator and with the one that came with the new alternator, so the regulator is unlikely to be the culprit. Also, the field voltage did not drop abnormally under load. Since the battery checks out fine (capacity up to spec), there must be some resistive loss somewhere in the alternator wiring or the CB. We both suspect the alt CB and he will untangle the rat's nest and check it, and probably replace the CB.

Other than that, his recommendations are:

1. Get a newer, electronic beacon (either LED or another low-current design).

2. Until I'm ready to do that, if I'm in IMC and need the pitot heat on, switch off the beacon.

3. Since the battery is charging fine except when both the beacon and pitot heat are on, he says it's airworthy and to fly it, but consider redoing all the wiring sometime in the future.

Comments?

First, unless your mechanic has a 4 wire ohmmeter, resistance measurements won't be useful for finding the source of your problem as there's no way with a conventional meter to detect the kind of "high resistance" that could be a cause. Assuming the alternator is producing the proper voltage (we'll get to that in a bit) you're getting around 2 volts drop at something like 20 amps which would mean a tenth ohm in the circuit. The best you can do with a conventional meter is about a half ohm and even that is tricky to get right. But there's no need to measure resistance directly, the proper technique is to measure the voltage drop along the path in question. For example you should measure the voltage across the 60A CB with enough loads on to drop the bus a volt or two. If the CB is the problem the "missing" voltage will appear across it although I suspect that's extremely unlikely. If the resistance of the CB was high enough to drop 2 volts at 25 amps it would catch on fire from the 50watts of power being dissipated.

But first I'd check to see what the voltage is on the "Bat" terminal of the alternator when the bus voltage is being pulled down by the beacon and pitot heat. If it's not up at 14v or higher there's no point in looking for high resistance CBs etc. And if the voltage at the alternator is low as I suspect the next thing you do is check the field voltage under the same conditions. If the field voltage is more than a volt or so less than the bus voltage there's a problem in the regulator or field wiring but if the field voltage is good the alternator is bad or the belt is slipping.

Another simple method for determining alternator health is to disconnect the regulator from the field terminal on the alternator and temporarily run a wire from the field directly to the battery (preferably with a 5A fuse in the line). Start the engine carefully keeping the RPM as low as possible and turn on several loads. Then slowly raise the RPM while monitoring the bus voltage which should rise with the RPM. If you can reach normal voltage (≈14v) at or below 1800 RPM the alternator is OK, otherwise it's bad (or the belt is slipping).

I've already posted a number of times how important it is to have someone who knows how to troubleshoot alternator systems when you have an unobvious problem like this. It's your money you are wasting but regardless of how much you like this mechanic or how good his engine/airframe skills are it's really time for an electrical expert IMO.

edit: It's a little surprising to me that the pitot heat, which only draws 5 amps, brings down the bus voltage so much more than the beacon, which draws twice as much. More than surprising, it doesn't make sense to me. What am I missing?

The current draw from a pitot heater (or almost any other electrical heater) starts out a lot higher than the nominal current flow once the element has reached operating temperature. I wouldn't be surprised to see your pitot heat pulling 10 amps initially, decreasing to 5 after 30-60 seconds. Coutd this be the source of your confusion?

Another factor is the "bootstrapping" nature of the regulator's ability to supply current to the field. As the bus voltage drops the regulator will increase the field voltage to precipitate more alternator output. But if something is causing the bus voltage to drop as you add loads that very drop in bus voltage limits what the regulator can deliver to the field so in your case the field current is actually dropping along with the bus voltage and that causes a further drop in both. This vicious circle ends when the voltage drops far enough that the resistive loads (lamps, motors, etc) pass less current due to the lower voltage fed to them or when the battery starts to take over and supplies some of the current. IOW it may be the sequence in which you applied the loads rather than the particular current of each load that determined which load had the greatest effect on bus voltage.

 

[Dan Thomas]

The first failure probably had to do with the name on the Alternator. I've had several Kelly alternators (and starters and regulators...) fail. The alternators had far too much grease in the rear bearing and it gets flung out and ends up on the rotor's slip rings, where it makes a tough sludge from graphite from the brushes. That increases field circuit resistance and the alternator can't keep up because the field can't get enough current no matter what the regulator tries to feed it. I think that's the "bad rotor" issue and it can be fixed just by cleaning the slip rings and replacing the brushes and cleaning out the bearing and using just a spot of the right stuff in it.

I have a small chunk of plywood with four screws in it and four long wires, one from each screw, that connect to airframe ground, alternator BAT terminal, alternator FLD terminal, and one that can be connected to others like the regulator's A terminal if the wire between it and the alt BAT terminal is suspected, or to the reg F terminal if that wire might be a problem. Sitting in the cockpit with that chunk of wood and a meter, one can measure voltages and see what the alternator and regulator are up to. Typical field voltage is only around 5 volts while running, increasing to battery voltage if the load is really heavy and the RPM low. The field voltage should drop as RPM rises as long as the load doesn't change. If it doesn't, the regulator is bad.

Dan

 

[azure]

First, unless your mechanic has a 4 wire ohmmeter, resistance measurements won't be useful for finding the source of your problem as there's no way with a conventional meter to detect the kind of "high resistance" that could be a cause. Assuming the alternator is producing the proper voltage (we'll get to that in a bit) you're getting around 2 volts drop at something like 20 amps which would mean a tenth ohm in the circuit. The best you can do with a conventional meter is about a half ohm and even that is tricky to get right.

I didn't look carefully at his meter (it looks a lot like the digital Fluke meter my Dad had), but he claims he can measure resistances in the 0.1-0.2 ohm range. But, as you say...

But there's no need to measure resistance directly, the proper technique is to measure the voltage drop along the path in question. For example you should measure the voltage across the 60A CB with enough loads on to drop the bus a volt or two. If the CB is the problem the "missing" voltage will appear across it although I suspect that's extremely unlikely. If the resistance of the CB was high enough to drop 2 volts at 25 amps it would catch on fire from the 50watts of power being dissipated.

Right, and that's actually what I told him he should do, directly measure the voltage across it under load. He said he couldn't as of yet because of the wiring mess. So we're trying to do the best we can in the meantime. If there's any point in doing that once he's got the mess untangled, I will insist on it. However, based on what you say next:

But first I'd check to see what the voltage is on the "Bat" terminal of the alternator when the bus voltage is being pulled down by the beacon and pitot heat. If it's not up at 14v or higher there's no point in looking for high resistance CBs etc.

As I posted, it's 12.8v with pitot heat alone pulling it down. So that's not a normal response of an alternator to a load pushing 40 amps?* If not, then it's clearly a problem with the alternator or regulator, and...

*No, I take that back. With the beacon off the total load can't be more than 30A. I don't know a thing about alternators, but my guess is that's definitely NOT normal.

And if the voltage at the alternator is low as I suspect the next thing you do is check the field voltage under the same conditions. If the field voltage is more than a volt or so less than the bus voltage there's a problem in the regulator or field wiring but if the field voltage is good the alternator is bad or the belt is slipping.

He says the field voltage is good, and this is alternator number two. Also, he says the behavior is the same regardless of which of the two regulators he has at his disposal are installed. So it sounds like everything points to belt slippage. I will suggest he check it, if he hasn't already.

I've already posted a number of times how important it is to have someone who knows how to troubleshoot alternator systems when you have an unobvious problem like this. It's your money you are wasting but regardless of how much you like this mechanic or how good his engine/airframe skills are it's really time for an electrical expert IMO.

Lance, it's not a matter of liking this mechanic -- though I do like him, it's getting my electrical system back in good health that's important. If someone reading this knows of a good aircraft electrics expert in this area I would hire him in a minute. I simply don't know anyone who fits that bill. There is certainly no one like that on field, and my guy has another couple of hours of work at least before the plane is even flyable.

The current draw from a pitot heater (or almost any other electrical heater) starts out a lot higher than the nominal current flow once the element has reached operating temperature. I wouldn't be surprised to see your pitot heat pulling 10 amps initially, decreasing to 5 after 30-60 seconds. Coutd this be the source of your confusion?

If it was initial draw wouldn't the drop in bus voltage due to turning on the pitot heat tend to reverse itself after a while as the draw decreased? Even turning off the beacon (10 amps according to my mechanic's inductive ammeter -- this is what I can't figure out) does not bring the battery voltage back up any higher than about 12.5v as long as the pitot heat is running. What does help to speed the recovery after turning off the pitot heat is increasing the rpm.

 

[Capt. Geoffrey Thorpe]

I have a small chunk of plywood with four screws in it and four long wires, one from each screw, that connect to airframe ground, alternator BAT terminal, alternator FLD terminal, and one that can be connected to others like the regulator's A terminal if the wire between it and the alt BAT terminal is suspected, or to the reg F terminal if that wire might be a problem.

It's a circuit board?

 

[Steve]

Is the Master switch original? That can be the culprit if it develops high resistance due to contact surface corrosion or loss of tension. The higher resistance can result in lower measured bus voltage and overall higher current draw. I've had that and an overvoltage relay both do that in a C177B at different times.

Just a suggestion.

 

[Steve]

It's a circuit board?

 

[azure]

@Dan: I hesitate to even mention this, but when I asked my mechanic again about the old alternator last night he said it was definitely the stator, not the rotor, and that he had misspoken before.

I also just spoke to him and he swears the alternator belt is not slipping. His plan now is to have my old alternator rebuilt by T&W (warranty work on the new alternator) and shipped out. In the meantime we'll (hopefully, wx permitting tonight) recheck the field and alternator output voltage under load with the new voltage regulator.

And now I've got to finish grading labs or I won't be able to be there to stand on the brakes.

 

[azure]

Is the Master switch original? That can be the culprit if it develops high resistance due to contact surface corrosion or loss of tension. The higher resistance can result in lower measured bus voltage and overall higher current draw. I've had that and an overvoltage relay both do that in a C177B at different times.

Just a suggestion.

Thanks. I think he said he tested it, but I will make sure tonight that we check it under load, and check the ALT side as well.

(But if I understand the PDF that Lance posted, my mechanic's check with the engine off would have been the best testing conditions of all, provided he checked the ALT side of the master -- no alternator output, so the regulator would source the maximum current possible from the battery through that switch to the field circuit. I should think that if there were corrosion, he would have found voltage across the switch. So yes I'll make sure it's been done, but my guess is it has -- unless I'm missing something.)

 

[gkainz]

I have a small chunk of plywood with four screws in it and four long wires, one from each screw, that connect to airframe ground, alternator BAT terminal, alternator FLD terminal, and one that can be connected to others like the regulator's A terminal if the wire between it and the alt BAT terminal is suspected, or to the reg F terminal if that wire might be a problem. Sitting in the cockpit with that chunk of wood and a meter, one can measure voltages and see what the alternator and regulator are up to. Typical field voltage is only around 5 volts while running, increasing to battery voltage if the load is really heavy and the RPM low. The field voltage should drop as RPM rises as long as the load doesn't change. If it doesn't, the regulator is bad.

Dan

It's a circuit board?

a PCB ... Plywood Circuit Board

 

[gismo]

I didn't look carefully at his meter (it looks a lot like the digital Fluke meter my Dad had), but he claims he can measure resistances in the 0.1-0.2 ohm range. But, as you say...

Many multimeters have 0.1 ohm resolution but that doesn't mean you can accurately measure resistance that accurately. The meter leads themselves along with the banana plugs likely have .1 to .3 ohms in the loop and that will vary as does the contact from the probe to the terminal being probed. A four wire ohmmeter uses two wires to apply a current and two separate wires to measure the resulting voltage. The current used for this is also often quite a bit higher than what's used for traditional resistance measurement with a digital multimeter. But I digress...

Right, and that's actually what I told him he should do, directly measure the voltage across it under load. He said he couldn't as of yet because of the wiring mess. So we're trying to do the best we can in the meantime. If there's any point in doing that once he's got the mess untangled, I will insist on it. However, based on what you say next:

As I posted, it's 12.8v with pitot heat alone pulling it down. So that's not a normal response of an alternator to a load pushing 40 amps?* If not, then it's clearly a problem with the alternator or regulator, and...

*No, I take that back. With the beacon off the total load can't be more than 30A. I don't know a thing about alternators, but my guess is that's definitely NOT normal.

Your guess is correct.

He says the field voltage is good, and this is alternator number two. Also, he says the behavior is the same regardless of which of the two regulators he has at his disposal are installed. So it sounds like everything points to belt slippage. I will suggest he check it, if he hasn't already.

I couldn't tell from your earlier post that you were measuring at the bat terminal of the alternator. But with only 12.8 on the alternator and something close to that on the field it's ether the alternator or the belt. That said, to get the alternator's true output you have to reference the "Bat" terminal to the alternator's case (ground) and the field must be measured between the two field terminals (if there are two, otherwise reference the alternator case). Also FWIW, some alternators connect one end of the field internally to a source of voltage but I don't believe that configuration is used on airplanes. A wiring diagram would make that obvious.

Lance, it's not a matter of liking this mechanic -- though I do like him, it's getting my electrical system back in good health that's important. If someone reading this knows of a good aircraft electrics expert in this area I would hire him in a minute. I simply don't know anyone who fits that bill. There is certainly no one like that on field, and my guy has another couple of hours of work at least before the plane is even flyable.

I understand. Sometimes you have to play the hand you're dealt. At least he's trying to figure it out instead of just replacing everything in sight.

If it was initial draw wouldn't the drop in bus voltage due to turning on the pitot heat tend to reverse itself after a while as the draw decreased? Even turning off the beacon (10 amps according to my mechanic's inductive ammeter -- this is what I can't figure out) does not bring the battery voltage back up any higher than about 12.5v as long as the pitot heat is running. What does help to speed the recovery after turning off the pitot heat is increasing the rpm.

That does actually sound like belt slippage. It's not uncommon for a belt to drive properly until the load gets high then start to slip. And once it starts slipping it keeps slipping until the load diminishes greatly. Raising the RPM would lower the force on the belt but it also allows the alternator to produce more output with the same field current so the fact that the voltage picks up with RPM isn't conclusive.

Two indications of belt slippage are shiny driving surfaces on the belt and the ability to turn the alternator pulley by hand without the engine turning. The latter requires quite a bit of hand strength though.

 

[azure]

Well, one thing is for sure it takes more than just the pitot heat to pull the bus voltage down (solid at 14.0v). It takes more than pitot heat plus beacon to really do it (13.8). There seems to be a critical current draw beyond which the alternator simply can't put out enough electrons. That current draw is reached if the nav lights alone are added, but that leads to a very slow drop that seems to stabilize at 13.1. Turning on the avionics causes a quick drop into the 12.x's. With pitot heat, beacon, nav lights, strobes, and avionics, the bus voltage stabilizes at about 12.0v.

This is with a different voltage regulator than we were using last night (also a different alternator than the original, which is going back for rebuild tomorrow). The voltage at the alternator end of the field wire as well as at the regulator end under maximum load was roughly 12.0v. The BAT terminal of the alternator read 12.3v (the bus voltage was 12.1 when that reading was taken). I'm not sure where he was referencing from. BTW there is only one field wire.

edit: He also used a jumper to bypass the master switch and fed 12vDC directly from the battery to the alternator field terminal. This made no difference: under maximum load, the bus voltage was still ~12v. So I don't think the master can be implicated.

With a normal night VMC load (beacon, navs, strobes, and avionics), the bus voltage hovers at around 13.5-13.6. That's not significantly different from where it was with the old alternator before it failed (13.4-13.5). I don't know for sure, but would guess at this point that adding the pitot heat to all of that would have done the same thing as it does now.

There is no sign of belt slippage, and he even tightened the belt to be sure. Tightening made no difference to the alternator output. He also demonstrated that turning the alternator pulley turns the prop.

So where does that leave us? Alternator #2 is weak to almost exactly the same degree as alternator #1 (before failure)? That seems extremely unlikely to me. It does seem that something is sapping the output of any alternator that's installed in the airplane so that it's barely adequate under normal loads and can be easily overwhelmed. My mechanic admits that he doesn't have a clue and although he is trying to diagnose it instead of shotgun it, from what he tells me shotgunning is the only thing he really knows to do at this point, and the only reason he's not doing it is because I've said I won't pay for that. But he's clearly out of his depth.

Lance, thanks for all your input and patience. I think I need to find a local electrics expert. A friend told me this evening that Bill @ Huron Avionics would be a good source and I will be calling him first thing tomorrow morning.

 

[DutchessFlier]

Liz: I am so sorry that you are going through this stuff...frustrating as all heck, and can get very expensive! (I know from having that total electrical failure last spring)

I know very little about these electrical systems. But, from reading all of the posts here, and your experiences, is it a correct assumption that somewhere in the system, apart from what you have already tested, something is drawing down the output and leaving less available where it should be? Is this indicative of a ground fault someplace thats draining electrons away, no matter what the output of the alternator? Can you isolate systems in the plane to troubleshoot where the fault is?

 

[azure]

Thanks Bruce. Yes it's very frustrating, and expensive. The old alternator had definitely failed, so I was going to have to replace it no matter who I went to. But now I've probably racked up 5-6 hours in labor charges that led nowhere, all because my mechanic really doesn't know what he's doing when it comes to diagnosing electrical problems. Last night I told him to finish installing the Whelens and check them out and change the brake pads (not sure why he hasn't done that yet) and get ready to just zip the cowl.

As far as isolating systems, well we've already done that in a sense. He's measured the current draw of each of the circuits separately. The only one that's a little surprising is the beacon which seems to draw from a low of 6 to a high of 10 amps during its cycle. I'll ask Bill McKelvey this morning if a partial "short" to ground on the beacon circuit could act that way, but the problem is that adding all the known loads doesn't come close to the rated output of the alternator. And usually shorts anywhere trip a CB and that isn't happening.

I didn't know you'd had a TOTAL electrical failure?! That would be quite scary! Where were you when that happened? Luckily I was pretty close to home when my alternator failed -- if this had happened last week coming back from TVC at night, I could have been stranded somewhere in the north woods! So if there's something wrong that overtaxed the first alternator to the point of failing, I want to get it fixed before it does the same thing to this one.

 

[bbchien]

My total Electrrical Failure was induced by the intallation of my MX20 display. That moved the Northstar M3 down four inches in the panel. At the time of the M3 install the manual called out for a NACA duct to cool the stack, which was plumbed in.

Unfortuantely at the time of the reinstal, they had changed the spec, calling out for a fan. With the M3 moved down the panel, moisutre that made it rhought the duct's turn ended up flooding the stack.

IMC 12,000 at night over Cape Girardeau MO, in late November, at OAT ~0F resulted in the panel getting flooded. The floor, too. The breakers, The smell definitely was electrical.

OFF went the master and the fun began.

 

[azure]

Well I just got off the phone with Bill @ Huron Avionics. Unfortunately he doesn't know of ANY aircraft electrics experts in this area and would be doing pretty much what my mechanic is doing. I think if anyone would know someone, he would.

He also said that T&W alternators are not known for reliability either, not much if at all better than Kelly Aerospace.

So yes, his only thought is that alternator #2 is probably bad too. His suggestion was to find a "known good alternator" and see what it does in my airplane.

 

[Dan Thomas]

It's a circuit board?

Nope, just a place to terminate the wires safely so I can use the voltmeter while running the engine.

Dan

Edit: I see some of you thing my redneck Plywood Circuit Board is funny. But I can diagnose the problems that you pay big bucks to other folks to fix. Shotgunning is an unfortunate affair and costs a lot of owners a lot of money. When I was in the air brake remanufacturing industry, at least three-quarters of the control valve, compressor and booster cores had absolutely nothing wrong with them; the heavy-duty mechanics didn't know what they were doing and had no clue as to how the systems worked. Their only solution was to keep replacing stuff until the problem went away. It led to a lot of phony warranty claims and made us put an 800-number card into every package so the mechanic could call us on the sly and get the answers he needed. If Kelly and some of the other rebuilders were smart they'd do the same thing, but they can't even seem to get their production quality right, never mind have someone who has real answers.

Dan

 

[gismo]

Well, one thing is for sure it takes more than just the pitot heat to pull the bus voltage down (solid at 14.0v). It takes more than pitot heat plus beacon to really do it (13.8). There seems to be a critical current draw beyond which the alternator simply can't put out enough electrons. That current draw is reached if the nav lights alone are added, but that leads to a very slow drop that seems to stabilize at 13.1. Turning on the avionics causes a quick drop into the 12.x's. With pitot heat, beacon, nav lights, strobes, and avionics, the bus voltage stabilizes at about 12.0v.

This is with a different voltage regulator than we were using last night (also a different alternator than the original, which is going back for rebuild tomorrow). The voltage at the alternator end of the field wire as well as at the regulator end under maximum load was roughly 12.0v. The BAT terminal of the alternator read 12.3v (the bus voltage was 12.1 when that reading was taken). I'm not sure where he was referencing from. BTW there is only one field wire.

edit: He also used a jumper to bypass the master switch and fed 12vDC directly from the battery to the alternator field terminal. This made no difference: under maximum load, the bus voltage was still ~12v. So I don't think the master can be implicated.

With a normal night VMC load (beacon, navs, strobes, and avionics), the bus voltage hovers at around 13.5-13.6. That's not significantly different from where it was with the old alternator before it failed (13.4-13.5). I don't know for sure, but would guess at this point that adding the pitot heat to all of that would have done the same thing as it does now.

There is no sign of belt slippage, and he even tightened the belt to be sure. Tightening made no difference to the alternator output. He also demonstrated that turning the alternator pulley turns the prop.

So where does that leave us? Alternator #2 is weak to almost exactly the same degree as alternator #1 (before failure)? That seems extremely unlikely to me. It does seem that something is sapping the output of any alternator that's installed in the airplane so that it's barely adequate under normal loads and can be easily overwhelmed. My mechanic admits that he doesn't have a clue and although he is trying to diagnose it instead of shotgun it, from what he tells me shotgunning is the only thing he really knows to do at this point, and the only reason he's not doing it is because I've said I won't pay for that. But he's clearly out of his depth.

Lance, thanks for all your input and patience. I think I need to find a local electrics expert. A friend told me this evening that Bill @ Huron Avionics would be a good source and I will be calling him first thing tomorrow morning.

While it's possible that there's an unexpected load that causes the total load to exceed the alternator's max output I'd say that is extremely unlikely as such a load would quickly deplete the battery whenever the master was on and the engine not running or with the alternator switched off. I'd say it's far more plausible that your rebuilt alternator is defective and from what you posted that's about the only two possibilities. If you want to be certain that the alternator is defective before replacing it again (I would), all your mechanic needs to do is measure the current flowing in the wire connected to the "Bat" terminal on the alternator when the bus voltage is low with the engine running. The easiest way to do that is using a clamp on DC ammeter but lacking that, you could try disconnecting the wire and inserting an automotive type ammeter or equivalent if your mechanic has one lying around. I'll bet you're going to find that the "new" alternator isn't putting out anywhere near it's rated current (make sure you have at least 1800 RPM for the test).

 

[Capt. Geoffrey Thorpe]

Edit: I see some of you thing my redneck Plywood Circuit Board is funny.

I thought it was a good pun.

But I can diagnose the problems that you pay big bucks to other folks to fix.

Indeed.

I hope you didn't think that I was diss'n your good idea. If so, I apologize.

 

[gismo]

As far as isolating systems, well we've already done that in a sense. He's measured the current draw of each of the circuits separately. The only one that's a little surprising is the beacon which seems to draw from a low of 6 to a high of 10 amps during its cycle. I'll ask Bill McKelvey this morning if a partial "short" to ground on the beacon circuit could act that way, but the problem is that adding all the known loads doesn't come close to the rated output of the alternator. And usually shorts anywhere trip a CB and that isn't happening.

6-10 amps doesn't sound all that high for a beacon if it's of the rotating type. The variation and higher current is probably an indication that the beacon's mechanical drive is binding and could use some TLC and lubrication.

I didn't know you'd had a TOTAL electrical failure?! That would be quite scary! Where were you when that happened? Luckily I was pretty close to home when my alternator failed -- if this had happened last week coming back from TVC at night, I could have been stranded somewhere in the north woods! So if there's something wrong that overtaxed the first alternator to the point of failing, I want to get it fixed before it does the same thing to this one.

A handheld com radio (with external antenna connection and headset adapter) plus a portable GPS can go a long ways towards mitigating almost any electrical or avionics failure. Like Dr Chien, I fly an airplane with two of most things including engines and alternators. But most non-glass panel twins still have a single electrical bus that's common to all loads and if something happens to that or you get smoke when it's powered you've got yourself a total electrical failure, so I've taken my own advice WRT a handheld com and portable GPS. So far I've never needed the comm in flight but I did get to use the portable GPS one time when my panel mount unit's screen went black over Lake Michigan on my way to Bellaire.

You can also get there if you are so dumb you manage to take off and fly around for a while with both alternators shut off like I did on my second flight in the airplane (that's what low time in type will give you) but at least then the fix is pretty simple (turn them on) and with my extensive troubleshooting skills I was able to diagnose the problem (stupid pilot error) by the time the third electrical device quit working due to the low voltage.

 

[gismo]

I thought it was a good pun.



Indeed.

I hope you didn't think that I was diss'n your good idea. If so, I apologize.

I thought it was great idea too but I still had to chuckle when I pictured it in my mind.

 

[azure]

While it's possible that there's an unexpected load that causes the total load to exceed the alternator's max output I'd say that is extremely unlikely as such a load would quickly deplete the battery whenever the master was on and the engine not running or with the alternator switched off. I'd say it's far more plausible that your rebuilt alternator is defective and from what you posted that's about the only two possibilities. If you want to be certain that the alternator is defective before replacing it again (I would), all your mechanic needs to do is measure the current flowing in the wire connected to the "Bat" terminal on the alternator when the bus voltage is low with the engine running. The easiest way to do that is using a clamp on DC ammeter but lacking that, you could try disconnecting the wire and inserting an automotive type ammeter or equivalent if your mechanic has one lying around. I'll bet you're going to find that the "new" alternator isn't putting out anywhere near it's rated current (make sure you have at least 1800 RPM for the test).

Yep, I was thinking last night that's the one thing we haven't directly measured, the total current the alternator is putting out when it's struggling. My mechanic has an inductive ammeter that he was using to measure the current drawn by each device, so why not the current going through the alternator output wire? But I'm not sure it's worth paying his labor cost for that test at this point. The original alternator is going back to T&W to be rebuilt under their warranty for the new alternator. So it's not going to cost me anything to try it after it's rebuilt other than the labor of doing the swap. If it can handle the load, great. If not, then we'll check its output current.

My question now is what shops turn out good rebuilt alternators? According to Bill, it's possible with both Kelly and T&W to go through multiple bad alternators before finding a good one. Since we're already committed to T&W I guess it's really a question for the future.

 

[DutchessFlier]

Okay:

Everyone hits me with the same thing about using the Plane Power alternators vs anyone elses...BTW when I did have the total failure, they replaced it with a Kelly alternator, but I am not gonna comment on anything in my plane right now!!

Dr. Bruce and I have the same first names and last name initials, so were gonna have to work this out somehow.... I can sign my name as my nickname: Moose, but I dunno maybe that will work.

The failure: due to the primary lead shorting out on the front baffle of the plane just where it exits the alternator: fried the whole system, inflight, we lost every electrical instrument, the flaps, nav/comms, the whole enchalada. Mind you, this was the absolute first flight in the plane the afternoon that I bought it,and after a prebuy, from a tech who I knew and selected to do the inspection. You never know! Got to 44N in a hurry, landed nordo, noflapo, and hooked up with an awesome A+P on the field to do the repairs. We ended up basically rebuilding the whole system from the alternator back, all the wiring, the Regulator/Master switch/cleaned and rewired the main bus, and replaced the alternator and field pop-out breakers with push-pull ones (which I really prefer to have in the plane). This failure led me to research how I could shed as much amp draw in the plane as possible, which brought me to the LED's and the rest as they say is history...and a lot of AU's!!!

 

[bbchien]

Yep, I was thinking last night that's the one thing we haven't directly measured, the total current the alternator is putting out when it's struggling. My mechanic has an inductive ammeter that he was using to measure the current drawn by each device, so why not the current going through the alternator output wire? But I'm not sure it's worth paying his labor cost for that test at this point. The original alternator is going back to T&W to be rebuilt under their warranty for the new alternator. So it's not going to cost me anything to try it after it's rebuilt other than the labor of doing the swap. If it can handle the load, great. If not, then we'll check its output current.

My question now is what shops turn out good rebuilt alternators? According to Bill, it's possible with both Kelly and T&W to go through multiple bad alternators before finding a good one. Since we're already committed to T&W I guess it's really a question for the future.

I keep spare alternator. It's a Kelly which I have sent out to precision Airmotive for re-OH with the yellow Kelly tag in place. It's the only way to go.

 

[azure]

I keep spare alternator. It's a Kelly which I have sent out to precision Airmotive for re-OH with the yellow Kelly tag in place. It's the only way to go.

Actually there's another way to go -- install a standby alternator in the plane. I've been thinking about doing that since this happened. I know at least one Cardinaleer got a 337 signed off for this. But I don't know how much red tape was involved. I also suspect it was expensive.

Okay I've got two more names now from the two Bruces: Precision Airmotive and Plane Power. Thanks guys. My nightmare now though, is how many T&W alternators I'll have to go through (under warranty, so it's time and frustration not AMUs, but still...) before they send me a good one. The next one will be my original, a Kelly, now rebuilt by T&W. Any bets that all they'll do is rebuild the shot stator and return it to its pre-failure state -- unable to put out its rated amperage? Do these shops even bench test their rebuilds for rated current??

 

[Henning]

So where does that leave us? Alternator #2 is weak to almost exactly the same degree as alternator #1 (before failure)? That seems extremely unlikely to me.

Not me. I've gotten 7, yes 7 alternators before before I got a good one. Never assume that because you changed a part that you now have a good part. I'm not saying that's the issue, just not to discount it. If it's from Kelly, it's not even in the realm of unlikely actually.

 

[azure]

Not me. I've gotten 7, yes 7 alternators before before I got a good one. Never assume that because you changed a part that you now have a good part. I'm not saying that's the issue, just not to discount it. If it's from Kelly, it's not even in the realm of unlikely actually.

Yeah Henning, I've been persuaded by a really good avionics tech... and it's not just Kelly, it seems T&W is pretty bad too. See my last post. They don't have any more in stock so they're going to take my original Kelly and rebuild it, probably today, and turn it around (they promised a one-day turnaround). I'm not optimistic that they'll do anything but fix what failed (claimedly, the stator). The thing was apparently bad from the day I bought the plane.

I expect my plane to be down for the long haul now as we try one T&W alternator after another. Unless I want to throw the money I sank into this replacement out the window and go with a different company. This, of course, is exactly what I was afraid of from the start.

 

[gismo]

Not me. I've gotten 7, yes 7 alternators before before I got a good one. Never assume that because you changed a part that you now have a good part. I'm not saying that's the issue, just not to discount it. If it's from Kelly, it's not even in the realm of unlikely actually.

The potential for a defective "new" electrical part is one of the main reasons (the other being wasted time and money) that the common practice of shotgunning (replacing parts without diagnosis until the problem is solved) can be so ineffective. Not only do you risk replacing one bad part with another one, it's also too likely you'll replace a good part with a bad one and at that point you have two problems that mask each other's symptoms.

 

[Capt. Geoffrey Thorpe]

The potential for a defective "new" electrical part is one of the main reasons (the other being wasted time and money) that the common practice of shotgunning (replacing parts without diagnosis until the problem is solved) can be so ineffective.

http://www.youtube.com/watch?v=iMs9NudasVI

 

[Dan Thomas]

The next one will be my original, a Kelly, now rebuilt by T&W. Any bets that all they'll do is rebuild the shot stator and return it to its pre-failure state -- unable to put out its rated amperage? Do these shops even bench test their rebuilds for rated current??

The stator is nothing more than a bunch of heavy copper wire wound into an iron core in the shape of a ring. There is so little that can go wrong with it. A short between windings is possible, or a broken wire from the stator to its diodes, and there are only three wires. If it's shorted it's probably junk, and if it's a broken wire it might be fixable. You might have a shorted diode, which will bleed off much of the current generated.

Dan

 

[N801BH]

Ok guys and gals.... Someone remind me what the actual function of a FAA yellow tag is ?

With comments coming from repair shops saying " X,Y Z brand are known for bad parts right out fo the box then one has to ask,
How can a FAA approved facility sell poor quality yellow tagged items ?
You guys and gals are paying premium price for crappy stuff that could potentially kill someone.

I know if it were me I would be taking any yellowtagged part that was bad out of the box right to the closest FSDO, setting it on the desk of the regional inspector and asking ,,,,,, What the F^@# over ?


I mean why have a certified process when in reality it is just smoke and mirrors ?

jus my .02 cents worth, now,I feel better.

Ben.

 

[azure]

I just got off the phone with my mechanic. He says T&W claims that the alternator we sent to them -- my original alternator, the Kelly that suddenly failed to put out enough to even charge the battery two weeks ago -- is perfectly good and putting out its 70 amps rated current.

Now we're in a total quandary and have no idea what to try next. Obviously we're not going to put that one back in my airplane! But what else could possibly be wrong?

Here were my mechanic's thoughts:

1. Belt slippage could still be a possibility -- he suggested that he could sand alternator pulley a little. I hesitate to do this for fear that it might accelerate belt wear. Is this a realistic concern?

2. Another thought he had is that my electrical cabling, which is bundled together in salmon pink tape, might have some hidden fraying internally. This makes no sense to me. He measured 12.5 volts max, under peak load, at the output terminal of the alternator. I can't think of any way that frayed cable downstream could account for that. Besides, even at maximum load there is not much drop between the alternator output terminal and the bus -- at most 0.5 volts. He did measure the actual output wire current with his inductive ammeter under maximum load and found ~25 A.

Any thoughts from the gallery would be more than welcome! I am going crazy here trying to decide what to tell him to try next. My instincts tell me it has to be either the alternator is bad or the belt is slipping since the alternator is not putting out its rated current and that chasing bad wiring downstream is a waste of time. If these alternators are really good then it has to be the belt. But are they good? If T&W turns out one bad alternator after another, can I really trust their bench testing? Yet if they say they got 70A out of that alternator, it's hard to imagine how that can be in error.

Redux for those who haven't read the rest of the thread:

1. With original alternator (a Kelly) I never saw bus voltage higher than 13.4v in flight. Occasionally dropped to ~11.5 or so on the ground.

2. Charging system failed in flight two weeks ago -- bus voltage dropped to 11.5v, then down to 10.2 by the time I landed.

3. Alternator replaced with a T&W rebuild. With new alternator, ground testing shows same bus voltage (13.5) under same load conditions as those under which the old alternator behaved similarly. Under maximum load (pitot heat, avionics, all lights except landing), bus voltage drops to 12v.

4. Alternator output voltage at maximum load is 12.3 to 12.5v, actual current draw is 25A (I calculate a resistance of <=0.02 ohm between alternator output terminal and the bus).

5. Turning the alternator pulley turns the prop.

Going crazy here... help needed... thanks.

 

[Dan Thomas]

I just got off the phone with my mechanic. He says T&W claims that the alternator we sent to them -- my original alternator, the Kelly that suddenly failed to put out enough to even charge the battery two weeks ago -- is perfectly good and putting out its 70 amps rated current.

Now we're in a total quandary and have no idea what to try next. Obviously we're not going to put that one back in my airplane! But what else could possibly be wrong?

Here were my mechanic's thoughts:

1. Belt slippage could still be a possibility -- he suggested that he could sand alternator pulley a little. I hesitate to do this for fear that it might accelerate belt wear. Is this a realistic concern?

2. Another thought he had is that my electrical cabling, which is bundled together in salmon pink tape, might have some hidden fraying internally. This makes no sense to me. He measured 12.5 volts max, under peak load, at the output terminal of the alternator. I can't think of any way that frayed cable downstream could account for that. Besides, even at maximum load there is not much drop between the alternator output terminal and the bus -- at most 0.5 volts. He did measure the actual output wire current with his inductive ammeter under maximum load and found ~25 A.

Any thoughts from the gallery would be more than welcome! I am going crazy here trying to decide what to tell him to try next. My instincts tell me it has to be either the alternator is bad or the belt is slipping since the alternator is not putting out its rated current and that chasing bad wiring downstream is a waste of time. If these alternators are really good then it has to be the belt. But are they good? If T&W turns out one bad alternator after another, can I really trust their bench testing? Yet if they say they got 70A out of that alternator, it's hard to imagine how that can be in error.

Redux for those who haven't read the rest of the thread:

1. With original alternator (a Kelly) I never saw bus voltage higher than 13.4v in flight. Occasionally dropped to ~11.5 or so on the ground.

2. Charging system failed in flight two weeks ago -- bus voltage dropped to 11.5v, then down to 10.2 by the time I landed.

3. Alternator replaced with a T&W rebuild. With new alternator, ground testing shows same bus voltage (13.5) under same load conditions as those under which the old alternator behaved similarly. Under maximum load (pitot heat, avionics, all lights except landing), bus voltage drops to 12v.

4. Alternator output voltage at maximum load is 12.3 to 12.5v, actual current draw is 25A (I calculate a resistance of <=0.02 ohm between alternator output terminal and the bus).

5. Turning the alternator pulley turns the prop.

Going crazy here... help needed... thanks.

Corroded crimp terminals at the alternator output and/or alternator breaker input. All of these resistances create heat that can sometimes be seen as darkened terminal insulators. The ammeter connections can also suffer. As they get hot, terminals conduct less and less and oxidize faster. Breaker that's closed but not conducting; their contacts oxidize and the resistance goes up and sometimes the heat created will make the braker pop. How old, in years, is this airplane?

Dan

 

[flyingcheesehead]

2. Another thought he had is that my electrical cabling, which is bundled together in salmon pink tape, might have some hidden fraying internally. This makes no sense to me. He measured 12.5 volts max, under peak load, at the output terminal of the alternator. I can't think of any way that frayed cable downstream could account for that.

If there's a short somewhere downstream, it can draw down the voltage at the alternator if you're nearing or at the peak output, or the regulator isn't working properly.

 

[azure]

Corroded crimp terminals at the alternator output and/or alternator breaker input. All of these resistances create heat that can sometimes be seen as darkened terminal insulators. The ammeter connections can also suffer. As they get hot, terminals conduct less and less and oxidize faster. Breaker that's closed but not conducting; their contacts oxidize and the resistance goes up and sometimes the heat created will make the braker pop. How old, in years, is this airplane?

Dan

The airplane is 36 years old...

I'll have to ask my mechanic exactly where he measured the alternator output voltage. Where exactly are the crimp terminals? I thought those would be part of the alternator itself. My understanding is that he measured the output voltage AT the alternator.

How could corrosion at the alt CB affect the voltage at the alternator output itself? Wouldn't that lower only the bus voltage?

The alternator output voltage goes low under load. Even with the voltage regulator and the master switch bypassed (alternator field terminal hotwired to 12v). I keep coming back to that and can't see a way around the conclusion that the alternator is not putting out what it should be.

 

[azure]

If there's a short somewhere downstream, it can draw down the voltage at the alternator if you're nearing or at the peak output, or the regulator isn't working properly.

Surely, but we were nowhere near peak output. Measured current was 25A. Also, bypassing the regulator (hotwiring the alternator field terminal to battery voltage, 12v) made no difference at all.