oddball RFI problem

I recently installed an EI FP5L fuel flow instrument in my Tri-Pacer. I
would notice that at irregular intervals, it would reset it self. I'd catch
the green Remaining LED flashing as it does after a power on self test. I
could never get it to fail in flight. I pulled and prodded all the wiring,
replaced the circuit breaker, tapped on the instrument, etc etc. Talking to
EI, the mentioned the possibility of RFI getting into the thing. I keyed the
radio on all the frequencies that I normally use and again it didn't fail.

Next I returned the instrument---you guessed it no defect found!

Well when I got it back I hooked a battery charger to the plane and turned
the master switch on and let it cook for a while. After a time of trying all
different combinations of stuff turned on and off to try to make the fuel
monitor reset, I finally found if the radio is keyed WITH the strobe on,
bingo !! the fuel flow resets.!!!! The strobe is mounted on the vertical
stabilizer. All grounds are run to a good common ground with the exception
of the strobe which uses the airframe as a return.

I haven't checked the SWR on the radio yet, but for the life of me I can't
figure out why turning the strobe on makes the damn thing reset when keying
the radio.

Although it did it in the hangar, apparently this isn't 100% true in flight
since the strobe is always on, and the thing resets intermittently.

Anyone have an idea??

Thanks

Paul
N1431A

You have an RF leak from a com coaxial cable connector
somewhere. Probably a bad ground at the connector. This is a common
problem with electronic voltage regulators, as they are sensitive to
much smaller currents than mechanical regulators and will misread
interference as an overvoltage condition, and will drop the alternator
offline when the mike is keyed if there's any RF leakage.
Look at the connector at the antenna. Condensation in the
aircraft ceiling will corrode it.

Dan

Does anybody have a clue what this feller is saying or why he is saying it?
Why would the cable connector itself be more likely to have a failure than a
break in the coax shield itself? What does an electronic voltage regulator
have to do with it? Why wouldn't it burp with straight RF rather than
requiring the strobe to be on?

My hit is that both the strobe AND the transmitter are getting into the
engine analyzer, but neither of them of themselves are of a level to cause
the problem. HOwever, added together they rise above the trigger point.
I'd start looking at a way to put a filter on the A+ supply lead going from
the battery bus to the analyzer and see if this doesn't cure the problem.
Rat Shack sells "alternator whine filters" that may do the trick. At least
they are inexpensive and an easy try.

Jim



wrote in message
oups.com...
You have an RF leak from a com coaxial cable connector
somewhere. Probably a bad ground at the connector. This is a common
problem with electronic voltage regulators, as they are sensitive to
much smaller currents than mechanical regulators and will misread
interference as an overvoltage condition, and will drop the alternator
offline when the mike is keyed if there's any RF leakage.
Look at the connector at the antenna. Condensation in the
aircraft ceiling will corrode it.

Dan

Ditto on what Jim said... it sounds like a conducted susceptiblity
problem, not a radiated susceptbility problem. That means that
transient noise is conducting into the analyzer via the power lead and
upsetting it, not via radio waves through the air. A decent PI filer
should fix it, and the one that Jim mentioned is probably sufficient...

Dean

On 15 May 2006 21:09:38 -0700, wrote:

Ditto on what Jim said... it sounds like a conducted susceptiblity
problem, not a radiated susceptbility problem. That means that
transient noise is conducting into the analyzer via the power lead and
upsetting it, not via radio waves through the air. A decent PI filer
should fix it, and the one that Jim mentioned is probably sufficient...

Dean

yeah and couldnt it be just a simple load problem.
if the power supply was marginal, the load of the strobe added to the
increased power drain of transmitting could just bring the supply down
far enough to cause the unit to reboot (for want of a better word)

I've seen a fire water bomber that periodically lost all electrics
because the start up drawdown of a 12 volt dc pump motor tripped the
circuit breakers. had the chopper pilot's attention centre stage until
someone put a multimeter on it and identified the cause. instantaneous
25 amp draw through a 15 amp circuit breaker :-)

doesnt the guy's comment that he tested with a battery charger hooked
up sound a warning bell? dicky battery that needs replacing or dicky
generator or dicky regulator?? of the three I'd guess battery.

(dicky - australian slang for something that works erratically)
Stealth Pilot

Voltage dropout is a type of conducted susceptibility noise. The
length of the dropout determines whether it is just a transient, or a
true brownout.

Avionics systems are supposed to be able to handle short term dropouts
associated with bus switching, etc. Maybe this one is poorly
designed...

Dean

On 16 May 2006 11:30:22 -0700, wrote:

Voltage dropout is a type of conducted susceptibility noise. The
length of the dropout determines whether it is just a transient, or a
true brownout.

Avionics systems are supposed to be able to handle short term dropouts
associated with bus switching, etc. Maybe this one is poorly
designed...

Dean

the guy needs to put a fast sampling volt meter on it and redo the
sequence.
the chopper outages were only identified when the data from the meter
was uploaded to a pc and looked through with an excel spreadsheet.
you never would have picked it while looking at the meter display.

Stealth Pilot

Got out to the hangar with my watt meter to see if I had a high VSWR that
might be putting RF where it wasn't wanted. When I turned the slug to the
reflected position, the meter reading was so low I could barely read it.
That eliminated that possibility.

I next found out that I could not duplicate the problem. The only difference
this time was that the top cover over the instrument panel had been removed
for access. I called EI in Bend Oregon for some tech assistance.

Their take on it, was that the garbage was getting in via the ground lead!!
They said to shorten it and the trouble would be gone. Since they make the
instrument and have seen this before, I followed their instructions, pulled
the ground lead off the buss and made a separate ground as close to the
instrument as possible. I did fudge a bit and wound the ground lead through
a ferrite bead right where it exits the instrument case. They kind of
ignored my comments that the problem only occurred with the radio was keyed
AND the strobe was on.

If this does it great. If not I think my next attack might be to get a
ferrite bead on the power to the strobe up on top of the vertical stabilizer
and filter the input to the fuel flow instrument. I have a skylight in my
PA22 and looking up through it while fiddling with the wiring I see the
antenna is perhaps 24 inches from the instrument. I guess not too many
fabric airplanes have fuel flow instruments installed. :-(


Thanks Jim for the tip on the hash being additive and hitting a threshold in
the instrument.

Cheers:

Paul
N1431A

"Tri-Pacer" wrote in message
...

Got out to the hangar with my watt meter to see if I had a high VSWR that
might be putting RF where it wasn't wanted. When I turned the slug to the
reflected position, the meter reading was so low I could barely read it.
That eliminated that possibility.

I next found out that I could not duplicate the problem. The only
difference
this time was that the top cover over the instrument panel had been
removed
for access. I called EI in Bend Oregon for some tech assistance.

Their take on it, was that the garbage was getting in via the ground
lead!!
They said to shorten it and the trouble would be gone. Since they make the
instrument and have seen this before, I followed their instructions,
pulled
the ground lead off the buss and made a separate ground as close to the
instrument as possible. I did fudge a bit and wound the ground lead
through
a ferrite bead right where it exits the instrument case. They kind of
ignored my comments that the problem only occurred with the radio was
keyed
AND the strobe was on.

If this does it great. If not I think my next attack might be to get a
ferrite bead on the power to the strobe up on top of the vertical
stabilizer
and filter the input to the fuel flow instrument. I have a skylight in my
PA22 and looking up through it while fiddling with the wiring I see the
antenna is perhaps 24 inches from the instrument. I guess not too many
fabric airplanes have fuel flow instruments installed. :-(


Thanks Jim for the tip on the hash being additive and hitting a threshold
in
the instrument.

Cheers:

Paul
N1431A


Why would you want to add inductance back into the ground lead that you just
shortened (eg reduced inductance and resistance). Ditch the ferrite bead in
the ground lead. If you really want to use it put it in the power lead.
Good Luck
Jim Stockton