Voltage Regulator or Alternator

A

aiu0462

Filing Flight Plan
Joined
Nov 13, 2008
Messages
25
Location
Arizona
Display Name
Display name:
Steve
New Concord battery, volts to battery showing 13.3, it used to be 13.7, this reading is flying, not on ground. Volts sometime creep up to 13.7, then slowly varies going down to as low as 12.8. All readings are from my JPI. After start, amp meter goes to 19 then slowly back to zero and stable, no flickering. Everything seems Norman and without the JPI, I would not know there were any changes going on.
Where do I start?
 
Visually inspect all battery, master relay, and alternator connections for corrosion. Always a good thing to do on these systems and it's a cheap starting point.
 
Last edited:
Norman......yup, spell check sucks when you don't proof.
 
Look up the phrase "galloping alternator".

Here's a response from Bob Nuckolls, who's a specialist in electrical system design for aircraft. I had the same issue 2 years ago.

My question:

I had the opportunity to do a 7+ hour X/C a week or two ago and it was driven into my head that I have a soft ~3 CPS noise in my headsets and the voltmeter and ammeter readings are fluctuating at the same time and frequency as the noise. If I turn off the alternator, the noise goes silent and the ammeter and voltmeter stabilize too.

The ammeter shunt is measuring battery charge or discharge and the voltmeter comes off of the buss downstream of the diode. Both gauges are analog, and the needles are bouncing up and down by a half volt (13.5-14V) and a half amp or thereabouts. The alternator is the B&C 40 amp unit and the regulator is the Ford unit recommended 15 years ago when the airplane was built. All of the components have been in service for 14 years and have approximately 950 hours of use on them.

Any suggestions on rectifying this condition?

Thanks,

Kyle Boatright​

Bob's response:

The voltage regulator sense (sensor?) voltage shares
a path with alternator field current in this
regulator. This condition is well known in
architectures of this type and is often called
the 'galloping ammeter'. In the older Cessnas,
I recommend replacing everything from the bus
bar to the regulator which would include breaker,
alternator switch and wires.

Doing any ONE thing might 'cure' the problem but
it's the sum-total of environmentally driven resistance
creep that finally stacks up to cause the instability.
You may cure it with one replacement but only by
replacing ALL will you get back to as-new condition.
 
Start by cleaning all your grounds, battery to ground cable, ground cable to airframe, airframe to engine, alternator to engine cradle, regulator to airframe etc. etc.If you have an old mechanical regulator, try cleaning he contact points with a point file. If those don't do it, then start thinking about replacing parts.

Cheap and easy first.
 
It's NEVER the voltage regulator. It's often the battery.
 
He says he's got a new Concorde.
 
Do you know how to use a volt meter ?
If yes, then this is what you do. A voltmeter measure the "potential difference" between two points. So the "potential difference" between the neg and pos post is 12.6 volts on a 12 v battery. Easy so far right ? So, with many many years specializing in electrical trouble shooting this is what I've learned - you can measure resistance with an ohm meter, amps with an ammeter (I had one of those easy peasy clamp on things) and you can even draw nice little pictures on a "silly scope" screen which I found to be completely unnecessary in 95% of the work I've done.

So, without any further ado - what you want to measure the potential difference of is one end of a wire to the other end. Do this while its "working" or in other words - electrically on. This is the equivalent of a resistance and voltage drop test in one swoop. You should not see any more than 0.2v ideally. In absence of a voltmeter, I've found many a bad connection in my mechanic career by just using my hands, as a corroded wire or connection will produce heat.

The ohm meter will only power the wire with a 9v battery inside the multi meter. You can actually pull a good ohm reading on a bad connection as the connection doesn't heat up. When a connection heats up the resistance also goes up. That's why you need to do the voltage drop while the circuit is hot.

Take it for what it's worth. There are many ways to skin a cat. In my 20 years of wrench bending the voltage drop test became my most used tool in my tool box for good reason.

Or you can just pull the ol shotgun out and replace every part it hits ;)
 
jaybee said:
Do you know how to use a volt meter ?
If yes, then this is what you do. A voltmeter measure the "potential difference" between two points. So the "potential difference" between the neg and pos post is 12.6 volts on a 12 v battery. Easy so far right ? So, with many many years specializing in electrical trouble shooting this is what I've learned - you can measure resistance with an ohm meter, amps with an ammeter (I had one of those easy peasy clamp on things) and you can even draw nice little pictures on a "silly scope" screen which I found to be completely unnecessary in 95% of the work I've done.

So, without any further ado - what you want to measure the potential difference of is one end of a wire to the other end. Do this while its "working" or in other words - electrically on. This is the equivalent of a resistance and voltage drop test in one swoop. You should not see any more than 0.2v ideally. In absence of a voltmeter, I've found many a bad connection in my mechanic career by just using my hands, as a corroded wire or connection will produce heat.

The ohm meter will only power the wire with a 9v battery inside the multi meter. You can actually pull a good ohm reading on a bad connection as the connection doesn't heat up. When a connection heats up the resistance also goes up. That's why you need to do the voltage drop while the circuit is hot.

Take it for what it's worth. There are many ways to skin a cat. In my 20 years of wrench bending the voltage drop test became my most used tool in my tool box for good reason.

Or you can just pull the ol shotgun out and replace every part it hits ;)
Click to expand...
what you are describing is known as a voltage drop test,
should only be done while circuit is under load.
 
jaybee said:
ummm... yeaa.... that's what I said, a couple of times in fact...
Click to expand...
I'm wondering, isn't voltage regulation a function of the voltage regulator? when the voltage varies isn't that the fault of the voltage regulator? a high resistance connection seldom corrects itself or varies the resistance unless it is loose.
 
Tom-D said:
I'm wondering, isn't voltage regulation a function of the voltage regulator? when the voltage varies isn't that the fault of the voltage regulator? a high resistance connection seldom corrects itself or varies the resistance unless it is loose.
Click to expand...

1 - Yes
2 - possibly
3 - could vary depending on loading and amount of wattage its producing

Anyone could sit here and make any amount of guesses that could possibly be right.

I supplied a tried and true trouble shooting technique to ensure an accurate diagnosis, not a guess on what it may be.
 
If the former battery was "weak," that is, if the cranking was poor, it may have been instead a bad connection between the battery and starter, including all the contactor and cable terminals, or the contacts in the contactors themselves. That's where the voltage drop test comes in real handy before spending the money on expensive stuff.
 
Last edited:
Thanks for all the help guys, I'm working my way through your help. Currently checking all connections, cleaning then reinstalling. Yes I know how to se a volt meter and will be starting that shortly.
I report back when it is fixed.
Thanks again to all for you ideas and help.
Steve
 
jaybee said:
Anyone could sit here and make any amount of guesses that could possibly be right.
Click to expand...
And that is why I don't try to trouble shoot over the internet.
jaybee said:
I supplied a tried and true trouble shooting technique to ensure an accurate diagnosis, not a guess on what it may be.
Click to expand...
One that might not be needed when the proper test procedures were tried first.
 
jaybee said:
"Might not" ?

Keep making guesses.

If you want to help the dude, then tell him what your "proper test procedures" are.
Click to expand...
his A&P should be the one to do that.
 
Okay guys,,,,thanks for all the help and ideas!!

Jaybee, thanks for taking the time giving me all the info,,,invaluable and I even made sense of it.

I traced ALL the wires and connections, removed, cleaned and reinstalled, then did the Jaybee testing.

Bottom line, I ran the engine all looked good,,,,13.6 volts and steady. Flew and all stayed the same. I'm not worried about a .1 difference, could be a JPI error, or at least that's what I am writing it off too.

Thanks again guys for all the help, even those who said I should proof read before posting.

PS.....I did not proof this.

Steve
 
weirdjim said:
Wow. I was taught that the only metal with a positive resistance coefficient of temperature is tungsten. I guess I was taught wrong. :frown2:

Jim
Click to expand...
Yes, you were taught wrong. It happens.
 
Clark1961 said:
Yes, you were taught wrong. It happens.
Click to expand...

Ah, grasshopper, you make the same assumption that Edison made when he tried to make a copper filament lightbulb. Suggest you might want to read up on the resistance coefficient of temperature for copper near the incandescent point and see why Edison's experiments with copper filaments didn't pan out. Tungsten is the only metal known that maintains positive TC all the way to and beyond incandescence.

Jim
 
There are almost always exceptions. Tungsten filament in a gaseous (other than air) envelope isn't relevant to then conversation. Light bulbs are resistive loads, which would be a bad thing for most conductors.
 
GlennAB1 said:
There are almost always exceptions. Tungsten filament in a gaseous (other than air) envelope isn't relevant to then conversation. Light bulbs are resistive loads, which would be a bad thing for most conductors.
Click to expand...
Tungsten filaments are not by any stretch of the imagination in a gaseous envelope unless you consider the halogen bulbs that put halogen gas under more than atmospheric pressure inside to keep the nasty oxygen stuff from coming in. Otherwise they are in as best as we can create it a VACUUM envelope. And no, they are not resistive but temperature dependent resistive.

Discussion over. Time to start hacking on TBO or some other misunderstood concept.

Jim
 
weirdjim said:
Ah, grasshopper, you make the same assumption that Edison made when he tried to make a copper filament lightbulb. Suggest you might want to read up on the resistance coefficient of temperature for copper near the incandescent point and see why Edison's experiments with copper filaments didn't pan out. Tungsten is the only metal known that maintains positive TC all the way to and beyond incandescence.

Jim
Click to expand...
As I said, you were taught wrong. Circuit design at incandescent temperatures is a really bad idea.

Yes I have seen bus bars glow. It is a bad situation and things are either going to be dark soon or very bright.
 
weirdjim said:
Tungsten filaments are not by any stretch of the imagination in a gaseous envelope unless you consider the halogen bulbs that put halogen gas under more than atmospheric pressure inside to keep the nasty oxygen stuff from coming in. Otherwise they are in as best as we can create it a VACUUM envelope. And no, they are not resistive but temperature dependent resistive.

Discussion over. Time to start hacking on TBO or some other misunderstood concept.

Jim
Click to expand...

Aargh. I should know better than to get between two people arguing on the internet, but I'm stupid...

Incandescent lights are very rarely under a vacuum. They are usually filled with argon, but there are other gas mixtures. Even halogens only contain a small amount of (usually) bromine in an argon/krypton environment.

Also, all resistive loads respond to some degree to temperature. Saying a resistance is temperature-dependent is redundant.

Reference (among many): http://www.lamptech.co.uk/Documents/IN Gases.htm
 
My experience with connectors and heat is that they get oxidized a bit and start micro arcing (is that a term?). That generates heat and more oxidation as well as carbon which increases the resistance. It's not that the material's resistance increases, just that the connection gets worse due to oxidation and carbon. And gets hotter.

I've removed a 20A 2 pole breaker that had a loose wire. One side of the breaker was melted away and it never kicked because the current was never over it's rating. (Cutler Hammer Square D clone, back in the late 70s' early 80's).
 
weirdjim said:
Wow. I was taught that the only metal with a positive resistance coefficient of temperature is tungsten. I guess I was taught wrong. :frown2:

Jim
Click to expand...

FAA Aviation Maintenance Technician - General 8083-20 Chapter 10 Page 19 said:
4. The fourth major factor influencing the resistance of a conductor is temperature. Although some substances, such as carbon, show a decrease in resistance as the ambient (surrounding) temperature increases, most materials used as conductors increase in resistance as temperature increases. The resistance of a few alloys, such as constantan and Manganin™, change very little as the temperature changes. The amount of increase in the resistance of a 1 ohm sample of a conductor, per degree rise in temperature above 0° Centigrade (C), the assumed standard, is called the temperature coefficient of resistance. For each metal, this is a different value; for example, for copper the value is approximately 0.00427 ohm. Thus, a copper wire having a resistance of 50 ohms at a temperature of 0 °C will have an increase in resistance of 50 × 0.00427, or 0.214 ohm, for each degree rise in temperature above 0 °C. The temperature coefficient of resistance must be considered where there is an appreciable change in temperature of a conductor during operation. Charts listing the temperature coefficient of resistance for different materials are available. Figure 10-41 shows a table for “resistivity” of some common electric conductors. The resistance of a material is determined by four properties: material, length, area, and temperature.
Click to expand...

Might want to address your concerns to the FAA ?
 
aiu0462 said:
Okay guys,,,,thanks for all the help and ideas!!

Jaybee, thanks for taking the time giving me all the info,,,invaluable and I even made sense of it.

I traced ALL the wires and connections, removed, cleaned and reinstalled, then did the Jaybee testing.

Bottom line, I ran the engine all looked good,,,,13.6 volts and steady. Flew and all stayed the same. I'm not worried about a .1 difference, could be a JPI error, or at least that's what I am writing it off too.

Thanks again guys for all the help, even those who said I should proof read before posting.

PS.....I did not proof this.

Steve
Click to expand...
Steve, glad what I wrote was helpful as that was all I had hoped to do - provide help with basic techniques based on basic principles.
 
You must log in or register to reply here.
Back
Top