Voltage variation

On Sep 23, 1:31 pm, Jay Honeck wrote:
I still haven't had time to work with my A&P to remove my (newly)
flakey JPI EDM-700 unit. (As you may recall, some of the LED segments
that make up the number display are going in and out.)

Someone here suggested that it might be due to low voltage, possibly
indicating an alternator problem. I've started watching the voltage,
and have noticed something odd:

The JPI has its own built-in voltage indicator. It will usually read
13.4 or 13.5 volts. We also have a stand-alone, panel-mounted digital
voltage/amperage meter, mounted a few inches below the JPI. It
usually reads a few tenths of a volt different than the JPI. (I can't
remember if it's up or down...)

a) Is this normal? Why would they read different voltages?

b) What voltage is normal? Is 13.4 too low?

Depends, is the landing light on? Few GA planes can maintain 14v with
the landing light on.

-Robert

Roy Smith wrote:
In article ,
wrote:
If the JPI is coming out soon, it might be interesting to pull the
other meter as well. Connect both of them to a 12 V battery sitting
on a table - with the same gauge and lengths of wire - and see what
they read.

The bench test sounds fine, but the same gauge and lengths of wire part
is kind of silly.

Quite possibly. However, if I _don't_ say that, somebody will try this
test with jumper cables for one meter and two 500 foot spools of 30 gauge
wire for the other meter, and then write me an indignant email when they
buy a new meter and its reading still doesn't match the other one.

As for a reference voltage source, any reasonable DVM you can buy at
Radio Shack fo $20 will be more than accurate enough for this kind of
work.

Precision is cheap. Accuracy costs a little more.

Matt Roberds

Anytime you are looking for electrical problems, you need to
keep in mind Ohm's Law. Voltage is equal to resistance times Current.
Reading voltages at various places in the airplane gives you little
information other than that the conductor is not entirely
disconnected, unless you remember that current flow affects voltage
drop.
I once forgot this principle and spent money that didn't need
spending. The strobe on the tail wasn't working, so I checked to see
that there was power available to it by pulling the feed wire off,
measuring the voltage there, and finding battery voltage. OK. Must be
the strobe's power supply shot. Bought another ($500) and installed
it. It didn't work, either. Then I remembered: E=IxR. With the strobe
turned on (but not working, of course) I took voltage measurements at
the bus (OK), at the circuit breaker's output input terminal (OK), at
the breaker's output terminal (not OK!) and realized that the breaker
was internally defective. There was full battery voltage there with
the power supply disconnected, because the meter requires only a few
microamps to drive it and the hundred ohms or so in the breaker's
corroded contacts wasn't enough to drop the voltage to it. Connecting
the power supply introduced a much lower resistance, increased the
current demand to two or three amps, and caused a huge voltage drop so
that the strobe was dead. The airplane needed a $15 breaker, not a
$500 power supply.

Dan

On Sep 24, 9:03 am, wrote:
Anytime you are looking for electrical problems, you need to
keep in mind Ohm's Law. Voltage is equal to resistance times Current.
Reading voltages at various places in the airplane gives you little
information other than that the conductor is not entirely
disconnected, unless you remember that current flow affects voltage
drop.
I once forgot this principle and spent money that didn't need
spending. The strobe on the tail wasn't working, so I checked to see
that there was power available to it by pulling the feed wire off,
measuring the voltage there, and finding battery voltage. OK. Must be
the strobe's power supply shot. Bought another ($500) and installed
it. It didn't work, either. Then I remembered: E=IxR. With the strobe
turned on (but not working, of course) I took voltage measurements at
the bus (OK), at the circuit breaker's output input terminal (OK), at
the breaker's output terminal (not OK!) and realized that the breaker
was internally defective. There was full battery voltage there with
the power supply disconnected, because the meter requires only a few
microamps to drive it and the hundred ohms or so in the breaker's
corroded contacts wasn't enough to drop the voltage to it. Connecting
the power supply introduced a much lower resistance, increased the
current demand to two or three amps, and caused a huge voltage drop so
that the strobe was dead. The airplane needed a $15 breaker, not a
$500 power supply.

Dan

I should add: Many batteries get replaced because the engine
doesn't crank very well. You need to take voltage drop measurements
across the master and starter solenoids while cranking; you'll often
find that their contacts are burned and introducing resistance with
the large current flow. Resistance checks mean nothing on these
things; even a quarter ohm will cost a bunch of voltage, and with
large current flows heat is generated, which increases the resistance
further. Same goes for cable and terminal crimps and connections.

Dan

writes:


I should add: Many batteries get replaced because the engine
doesn't crank very well. You need to take voltage drop measurements
across the master and starter solenoids while cranking; you'll often
find that their contacts are burned and introducing resistance with
the large current flow. Resistance checks mean nothing on these
things; even a quarter ohm will cost a bunch of voltage, and with
large current flows heat is generated, which increases the resistance
further. Same goes for cable and terminal crimps and connections.

Yes. I measure, for example:

a) On the battery term.
b) On the lug
c) On the wire in the lug.(That's why you have a sharp pointy probe.)
....

The hassle is, you have to do this under load, which usually means
"cranking" which means various dangerous parts are moving...


--
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

Thinking more about Jay's undervoltage issue...

I can't see why he can't do all the testing with the engine off. That's
LOTS safer. Yes, he'll drain the battery slightly but we're talking only
a few minutes here.

He'll need a DVM. In my experience, even really cheap [as in $5 at Harbor
Fright] DVMs seem to be surprizingly close. (But you can always test it.
Ask your avionics shop to compare to their lab meter at 13v and see.)


Then make a simple schematic:

JPI

1-----------2breaker3---bus4----[master contactor]5-------6Battery+

and measure two ways; from each point to a known clean ground, and
from one to another, such as 2-3, 3-5 etc.

Each technique has its own advantages and disadvantages. "To ground"
will be hard to see small drops; 'across' is twice as many good
connections to make in tight places.

But what you are looking for is: How much does the voltage drop from the
battery along the way? With no/little load; the answer should be 0, no
drop, even if there are bad connections. Under load, the voltage drop
will be more visible.

So the first thing needed is this: What is the voltage at 6, the battery
post? And 1, where the other device connects? If the JPI (and the master
relay) are the only thing on, and it's more than one volt of drop; I'd look
deeper.

If the battery is much less than 13.8v, with the master off; I'd wonder how
it cranks the plane.

--
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

David Lesher wrote:
I can't see why he can't do all the testing with the engine off. That's
LOTS safer. Yes, he'll drain the battery slightly but we're talking only
a few minutes here.

The test results with engine on vs. engine off probably depend somewhat
on where the battery is. If it's under the cowl, then engine on or off
probably won't make that much difference - all of the battery-recharge
current will be mostly under the cowl and won't affect the readings
much. If the battery is back in the tail someplace, then I would expect
more difference between engine on and engine off - there will be several
amps flowing from the generator/alternator back to the battery,
especially if the engine has just been started.

Then make a simple schematic:

JPI

1-----------2breaker3---bus4----[master contactor]5-------6Battery+

and measure two ways; from each point to a known clean ground, and
from one to another, such as 2-3, 3-5 etc.

The only thing I would add to this is to test both sides of everything,
like:

JPI1---2breaker3---4bus5---6[master contactor]7---8Battery+

This helps show if you've got bad contacts in the master contactor or on
the bus.

"To ground" will be hard to see small drops; 'across' is twice as many
good connections to make in tight places.

Tip: Get some heat-shrink tubing and cover up all but the last 0.1-0.2"
(2-4 mm) of the meter probe tips. This helps prevent shorts when poking
around in tight places. You can also do this with electrical tape, but
the heat-shrink stays put better.

If the battery is much less than 13.8v, with the master off; I'd wonder
how it cranks the plane.

I'd expect the voltage on a fully-charged battery to be something like
12.6 to 12.8 V. 12.3 or 12.4 might still crank the engine, but slowly.
12.0 or 12.1 is totally discharged.

Disclaimer: This is based on experience with ground vehicles and
equipment. I don't have an A&P; I don't even have a TG&Y. Your
mileage may vary.

Matt Roberds

My JPI routinely runs 13.4 - 13.5 in flight after initial charging at
13.6 - 13.8, no problems with display.

Paul kgyy wrote:
My JPI routinely runs 13.4 - 13.5 in flight after initial charging at
13.6 - 13.8, no problems with display.

Mine is similar. Initial charge voltage is ~14.0 volts, drops to ~13.6
as charge rate decreases after start. It has never made sense to me, but
I assume it is a characteristic of the voltage regulator. I believe it
is a solid-state version, replaced several years ago.

So far I've had no problems with my JPI 700, display or otherwise. It
has been installed since 1998 in my Cessna R172K. Flown about 400 tach
hours since that time.