UPDATE #3 -- Headset Noise

[Jonathan Goodish]

I put a voltmeter on the alternator output and measured between 32-35 AC
volts at moderate load and moderate RPM. Looks like some diodes have
decided to give up.

Does anyone have any info on the Plane-Power alternators? There is a
PMA'd replacement for my aircraft, and I will go that direction if there
are no big disadvantages.



Thanks,
JKG

[David Lesher]

Jonathan Goodish > writes:

>I put a voltmeter on the alternator output and measured between 32-35 AC
>volts at moderate load and moderate RPM. Looks like some diodes have
>decided to give up.

Maybe. Voltmeters are a really bad way to look at how much non-sine AC
ripple there is on a DC bus. Unless it's a known-good "True RMS" meter
with AC-only coupling, it will lie.

I've lost track; do you have audible whine in the audio? Does it
change with load?



--
A host is a host from coast to
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

[Jonathan Goodish]

In article >,
David Lesher > wrote:
> >I put a voltmeter on the alternator output and measured between 32-35 AC
> >volts at moderate load and moderate RPM. Looks like some diodes have
> >decided to give up.
>
> Maybe. Voltmeters are a really bad way to look at how much non-sine AC
> ripple there is on a DC bus. Unless it's a known-good "True RMS" meter
> with AC-only coupling, it will lie.
>
> I've lost track; do you have audible whine in the audio? Does it
> change with load?

David,

I suspect that the meter isn't going to be 32 volts AC off the mark,
even though it isn't a True RMS meter. The recommended threshold for AC
voltage from the alternator is anywhere from 0.8 to 1.0 volts, based on
the documentation I've read and my discussions with an engineer at
ElectroSystems.

However, I have alternator whine in the audio which changes with RPM and
load. I've tried just about everything else to get rid of the whine, so
this is about the last thing I know to do.



JKG

[Nathan Young]

On Wed, 19 Apr 2006 12:06:37 -0400, Jonathan Goodish
> wrote:

>In article >,
> David Lesher > wrote:
>> >I put a voltmeter on the alternator output and measured between 32-35 AC
>> >volts at moderate load and moderate RPM. Looks like some diodes have
>> >decided to give up.
>>
>> Maybe. Voltmeters are a really bad way to look at how much non-sine AC
>> ripple there is on a DC bus. Unless it's a known-good "True RMS" meter
>> with AC-only coupling, it will lie.
>>
>> I've lost track; do you have audible whine in the audio? Does it
>> change with load?
>
>David,
>
>I suspect that the meter isn't going to be 32 volts AC off the mark,
>even though it isn't a True RMS meter. The recommended threshold for AC
>voltage from the alternator is anywhere from 0.8 to 1.0 volts, based on
>the documentation I've read and my discussions with an engineer at
>ElectroSystems.
>
>However, I have alternator whine in the audio which changes with RPM and
>load. I've tried just about everything else to get rid of the whine, so
>this is about the last thing I know to do.

Does your plane have a split master switch? If so, shut down the
alternator while flying and see if whine goes away.

-Nathan

[Jonathan Goodish]

In article >,
Nathan Young > wrote:
> >However, I have alternator whine in the audio which changes with RPM and
> >load. I've tried just about everything else to get rid of the whine, so
> >this is about the last thing I know to do.
>
> Does your plane have a split master switch? If so, shut down the
> alternator while flying and see if whine goes away.


Yep, it goes away completely.

[mikem]

Jonathan Goodish wrote:
> I put a voltmeter on the alternator output and measured between 32-35 AC
> volts at moderate load and moderate RPM. Looks like some diodes have
> decided to give up.

Do you have access to an oscilloscope? That will tell you the "truth"
about what
is happening on the bus. Go look at the waveforms I posted in your
earlier thread...

Is your A/C 14V or 28V. If your A/C is 28V, and your meter doesn't
block dc, then the 32V reading might sort of make sense?

If I were to hook my true-RMS Fluke 77 DVM to my 14V Cessna, under
charge, the ripple would be less than 0.25Vac.

[Jonathan Goodish]

In article . com>,
"mikem" > wrote:
> Do you have access to an oscilloscope? That will tell you the "truth"
> about what
> is happening on the bus. Go look at the waveforms I posted in your
> earlier thread...
>
> Is your A/C 14V or 28V. If your A/C is 28V, and your meter doesn't
> block dc, then the 32V reading might sort of make sense?
>
> If I were to hook my true-RMS Fluke 77 DVM to my 14V Cessna, under
> charge, the ripple would be less than 0.25Vac.

Mike,

Unfortunately, I do not have access to an oscilloscope. I've begged and
pleaded but can't find anyone local who has one. As a result, I would
be taking a risk by replacing the alternator, but it's cheaper than
buying an oscilloscope. :-)

My A/C is 14V. My meter is a Greenlee, and it does block DC.



JKG

[mikem]

I have no explanation for the 32V reading.

Here is another idea. Take the alternator off the airplane, and take it
to an automotive shop that has an alternator test bench. For example,
the Pep Boys store about 1/2mi from my house has one. They can spin it
up, apply field current through a rheostat, apply a load, check its
output current and ripple. If it has a blown diode, they should be able
to tell you. You will have to make up some story about why they cant
sell you an "exchange"

Take your core, and either have it repaired, o/h, or trade your core
for one that is already rebuilt by a FAA repair station. If there is
nothing wrong with yours, you can work with your mechanic to reinstall
it on the aircraft.

I have had the Ford Prestolite alternator in my 182 develop "open
diode" ripple. In my case, the one of the stator wires became
disconnected from the diode stack due to vibration. It was "repaired".
I got charged about $100, and it was returned with a yellow tag..

[Jonathan Goodish]

In article om>,
"mikem" > wrote:

> I have no explanation for the 32V reading.


I had an opportunity to poke around some more this evening.

Despite Greenlee's insistence that my meter blocks DC when measuring AC
volts (and vice versa), that appears to not be the case. The meter
reads 26 volts AC when measured on the straight battery--no alternator.
It appears to read AC correctly, and DC correctly, but not AC ripple on
a DC bus.

I have a 20 year old Beckman meter that was out of calibration, but I
calibrated for AC and DC volts against the new Greenlee. The Beckman
appears to block DC when reading AC volts.

Both meters show around 12.8v at the battery, and almost exactly 14.0
volts DC when the alternator is charging, both at the battery and the
bus. I've determined that the Greenlee's AC reading on the DC bus is
highly inaccurate, but the Beckman shows almost zero AC volts (0.008) on
the DC bus--I assume because the battery is absorbing the ripple.
However, at the alternator, the Beckman shows approximately 0.30 volts
AC, which increases to a maximum of 0.65 volts AC under full load and
max static RPM. These look more like reasonable numbers.

Could the voltage regulator be at fault? Load seems to amplify the
noise, but the bus voltage is almost exactly 14.0 volts, regardless of
load or RPM. I would think if the regulator was failing, I'd see some
fluctuation, especially under full load. Could I be missing something?

I know that it may very well still be a ground loop. There are some
oddities that I still can't explain, however, like the intermittent
static/crackling when the copilot keys the PTT and transmits. The
landing light causes occasional crackle in the audio, but a corroded
connector may be contributing to that problem, not sure.



JKG

[David Lesher]

"mikem" > writes:


>Jonathan Goodish wrote:
>> I put a voltmeter on the alternator output and measured between 32-35 AC
>> volts at moderate load and moderate RPM. Looks like some diodes have
>> decided to give up.

>Do you have access to an oscilloscope? That will tell you the "truth"
>about what
>is happening on the bus. Go look at the waveforms I posted in your
>earlier thread...

What He Said. Can you finagle any scope at all? That, and a major
load will reveal bad diodes.

Also, look at mikem's diagrams. Note how he put values in for all
the ground connections. Maybe he'll do them again with 'bad ground'
values.

Poor Ground Shall Cause Grief.


--
A host is a host from coast to
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

[Travis Marlatte]

It's not the regulator.

I'm sure that most of the time, what mikem and Jonathan are saying is true.
But my symptoms and yours sound exactly alike. I checked connections and
grounds then had the alternator rebuilt. The rebuild cost about $275. $450
total with labor. No more whine. If this were a $10,000 rebuild, I would
have spent more time diagnosing. As it was, I'm glad to spend the money and
be able to fly with confidence.

--
-------------------------------
Travis
Lake N3094P
PWK

"Jonathan Goodish" > wrote in message
...
> In article om>,
> "mikem" > wrote:
>
>> I have no explanation for the 32V reading.
>
>
> I had an opportunity to poke around some more this evening.
>
> Despite Greenlee's insistence that my meter blocks DC when measuring AC
> volts (and vice versa), that appears to not be the case. The meter
> reads 26 volts AC when measured on the straight battery--no alternator.
> It appears to read AC correctly, and DC correctly, but not AC ripple on
> a DC bus.
>
> I have a 20 year old Beckman meter that was out of calibration, but I
> calibrated for AC and DC volts against the new Greenlee. The Beckman
> appears to block DC when reading AC volts.
>
> Both meters show around 12.8v at the battery, and almost exactly 14.0
> volts DC when the alternator is charging, both at the battery and the
> bus. I've determined that the Greenlee's AC reading on the DC bus is
> highly inaccurate, but the Beckman shows almost zero AC volts (0.008) on
> the DC bus--I assume because the battery is absorbing the ripple.
> However, at the alternator, the Beckman shows approximately 0.30 volts
> AC, which increases to a maximum of 0.65 volts AC under full load and
> max static RPM. These look more like reasonable numbers.
>
> Could the voltage regulator be at fault? Load seems to amplify the
> noise, but the bus voltage is almost exactly 14.0 volts, regardless of
> load or RPM. I would think if the regulator was failing, I'd see some
> fluctuation, especially under full load. Could I be missing something?
>
> I know that it may very well still be a ground loop. There are some
> oddities that I still can't explain, however, like the intermittent
> static/crackling when the copilot keys the PTT and transmits. The
> landing light causes occasional crackle in the audio, but a corroded
> connector may be contributing to that problem, not sure.
>
>
>
> JKG

[Jonathan Goodish]

In article >,
David Lesher > wrote:

> "mikem" > writes:
>
>
> >Jonathan Goodish wrote:
> >> I put a voltmeter on the alternator output and measured between 32-35 AC
> >> volts at moderate load and moderate RPM. Looks like some diodes have
> >> decided to give up.
>
> >Do you have access to an oscilloscope? That will tell you the "truth"
> >about what
> >is happening on the bus. Go look at the waveforms I posted in your
> >earlier thread...
>
> What He Said. Can you finagle any scope at all? That, and a major
> load will reveal bad diodes.
>
> Also, look at mikem's diagrams. Note how he put values in for all
> the ground connections. Maybe he'll do them again with 'bad ground'
> values.
>
> Poor Ground Shall Cause Grief.

I just don't think that I have poor grounds. I've checked just about
every ground I could find--alt to engine, engine to aiframe, battery to
airframe, avionics, intercom, etc. They're all good, no corrosion. I
cleaned up the big ones before reassembly.

Ground loop could be an issue, but I'm still not 100% convinced that the
alternator isn't at fault. Unfortunately, I've not been able to find
anyone who has a scope, for I know that it would tell more conclusively
what is happening.



JKG

[john smith]

> Ground loop could be an issue, but I'm still not 100% convinced that the
> alternator isn't at fault. Unfortunately, I've not been able to find
> anyone who has a scope, for I know that it would tell more conclusively
> what is happening.

Do you know any amateur radio operators?
One of them will either have one or know someone who has one.

[mikem]

Make sure you go back and read the stuff I recently posted in your
UPDATE 2 thread

[[email protected]]

Then you just need to put a good alternator noise filter in line with
the +12v power line to your audio intercom. You've got alternator whine
riding along the power line into your intercom system and it's being
amplified into your headphones.

Take a look at the Lonestar model 122253-10A filter.

http://www.sacskyranch.com/acatalog/Sacramento_Sky_Ranch_noise_filters_65.html

I used to fly a Cherokee that had one of those skinny Terra radios in
it, that would pick up alternator whine really bad, and the Lonestar
filter placed inline with the radio's power wire fixed it up perfectly.
The Lonestar filter was only about $80 back then, and now costs about
$110.

[John_F]

A quick way to tell a bad diode is the use of a clip on amp meter that
will read AC and DC amps. Only the Hall sensor ones will do this. Load
the output of the alternator to at least 50% of rated output and
measure the AC and DC amps. If the AC amps exceed 15% of the DC amps
you have a bad diode or stator winding. This works on all 3 phase DC
output alternators.
John

On Wed, 19 Apr 2006 22:13:56 -0400, Jonathan Goodish
> wrote:

>In article om>,
> "mikem" > wrote:
>
>> I have no explanation for the 32V reading.
>
>
>I had an opportunity to poke around some more this evening.
>
>Despite Greenlee's insistence that my meter blocks DC when measuring AC
>volts (and vice versa), that appears to not be the case. The meter
>reads 26 volts AC when measured on the straight battery--no alternator.
>It appears to read AC correctly, and DC correctly, but not AC ripple on
>a DC bus.
>
>I have a 20 year old Beckman meter that was out of calibration, but I
>calibrated for AC and DC volts against the new Greenlee. The Beckman
>appears to block DC when reading AC volts.
>
>Both meters show around 12.8v at the battery, and almost exactly 14.0
>volts DC when the alternator is charging, both at the battery and the
>bus. I've determined that the Greenlee's AC reading on the DC bus is
>highly inaccurate, but the Beckman shows almost zero AC volts (0.008) on
>the DC bus--I assume because the battery is absorbing the ripple.
>However, at the alternator, the Beckman shows approximately 0.30 volts
>AC, which increases to a maximum of 0.65 volts AC under full load and
>max static RPM. These look more like reasonable numbers.
>
>Could the voltage regulator be at fault? Load seems to amplify the
>noise, but the bus voltage is almost exactly 14.0 volts, regardless of
>load or RPM. I would think if the regulator was failing, I'd see some
>fluctuation, especially under full load. Could I be missing something?
>
>I know that it may very well still be a ground loop. There are some
>oddities that I still can't explain, however, like the intermittent
>static/crackling when the copilot keys the PTT and transmits. The
>landing light causes occasional crackle in the audio, but a corroded
>connector may be contributing to that problem, not sure.
>
>
>
>JKG