12V lights flicker on generator

[joebedford]

I know there is another thread here called "Voltage fluctuation" and I suspect I'm getting some of the same effects. However, this is an Onan 5500 built-in that didn't used to do this.

Is there something that I can test to see what the problem is?

Is there something that wears in the Onan that could cause this? My unit only has about 200 hours.

It still does it with load (air conditioner) but all my electronic stuff works fine. Doesn't do it on shore power so it's not the converter (WFCO 100 amp). State of battery charge doesn't affect things.

[enblethen]

Thanks for the update!

[joebedford]

Just to close this thread: today was the 1st so I did my generator exercise: no 12V light flicker! That loose ac wire in the generator junction box was the culprit.

Park power was 126V. Generator no load was 119V steady. Turned on both a/c's and voltage went UP to 122. Only measured one leg.

DC voltage stayed at 13.4 throughout this exercise.

[joebedford]

The only other AC device we used on the trip was the microwave and it's on the other leg. Hence only the 12V was flickering.

[time2roll]

Odd that nothing else was flickering.
Wire looks very clean for a bad connection.

[enblethen]

Looks as though you may have found your problem!

[joebedford]

I took the suggestion to look for loose connections (it was the easiest) and I found this:





I gave that wire nut a little tug and look what happened!

I'm willing to bet serious money (as much as a dollar) that this is the cause of flickering. That black wire is one of the hots from the genset.

I'll let you know for sure when I do its monthly exercise in a few days.

[enblethen]

I do not think a loose connection would cause the issue.
I said early on that the voltage regulator maybe suspect.
Standard electrical voltage tests with a good quality meter should be enough.
OP may even hear the genset surging.

[MEXICOWANDERER]

WFCO makes for a GREAT power supply when it is denuded of it's case and air is passed across those vertical ceramic resistors. But like it or not, the WFCO ain't the Rolls & Surrette of power supplies. Lambda is...

When a generator is CAUSING the problem, the cause must be fixed, not the effect.

My system did not have flickering lamp effects, merely too much champipple on the output. I do not want the DC to DC boosters, buckers and inverter subjected to ripple. Check out the LIFELINE BATTERY PDF manual and read Concorde's opinion about PULSE CHARGING and ripple. Ripple does a great job softening cured plate paste. I might have a tad of experience in that area...

Batteries offer a specific farad capacitance which cannot address frequencies above a certain level. Ripple and any other frequency anomaly needs to be TUNED OUT with selected capacitance.

[Gdetrailer]

enblethen wrote:
That is why OP needs to check both the AC and DC systems to isolate where the problem is.
Yes, it could be the converter failing, but I hate to see parts being installed without verifying what the problem is.

You didn't tell the OP just how to go about "testing"..

There are literally thousands of ways to "test" electrical systems.. Not all can be interpreted easily separately..

Volt meter?

Analog?

Digital?

Waveform analysis?

Scope?

Distortion?

Noise?

Frequency?

Even using a light bulb on the gen can be rather difficult to determine if the gen has a "loose wire"..

I simply do not believe the gen has a "loose wire" running around. A loose wire will cause a lot of other "issues" mainly one of intermittent OR COMPLETE LACK of 120V output.

The generator (especially a built in" is THE MOST EXPENSIVE PART TO BLAME IN THE SYSTEM. Built in gens are not cheap, nor easy to fix and if you start fumbling round inside one most likely WILL BREAK SOMETHING EXPENSIVE..

If you still believe the OPs gen has a "loose connection" then perhaps you should "volunteer" your own gen and give them some personal help..

Replacing a WFCO converter is actually NOT just throwing parts at it, it is AN UPGRADE in converters. WFCO converters are not really renowned as being hearty or "best" and often fail to get into bulk charge mode..

So, really the OP needs to try a light bulb on the output, it should be pretty stable in brightness but please note, IT WILL VARY with the engine RPM so if engine is not running smooth it will vary.. Light bulb also CAN vary in brightness due to the AVR MAKING CORRECTIONS.

In short no matter how good of a generator a light bulb is going to reveal SOME variations in voltage output.

The converter SHOULD BE ABLE TO TOLERATE voltage as low as 108V and as high as 135V WITH NO VARIATION IN OUTPUT (IE RIPPLE)..

The converter SHOULD also be able to operate at a MIN 50-60HZ (in fact the input stage of a switching converter is a bridge diode followed by a 470 uf to 1,000 uf filter cap).. This means technically you CAN feed a converter DC voltages if it is at least one half the expected AC peak voltage at it should work (IE 70-80V DC)!

Mex's capacitor suggestion may or may not work and IF it does work it is covering up a lousy converter by filtering and buffering some of the ripple..

But keep in mind that MANY IF NOT ALL old school linear converters relied on the BATTERY to regulate and filter the DC.. And it worked.. so I am very doubtful that tossing a large value capacitor at it is the right thing to do..

[MEXICOWANDERER]

Well, lessee, I resolved the problem with my WFCO with a 5K 25WVDC electrolytic cap.

But this needs to have the source of the problem corrected not a band aid.

Does Radio Shack still sell stuff in stores? An AC and a DC analog meter should be easy to find. I love digital for resolution. I love analog for oscillation up to the point of the capacity of the meter to respond. I use the hell out of analog. Why? I pry off the lens, and make a red mark at the target value.

With analog a person does not need to INTERPRET what he sees. You have to READ a digital gauge. Your brain has to PROCESS abstract numbers. With analog it's a matter of physical relationships - needle versus mark. Deviation is a signal to define the warning with digital if you wish. Compare the two for a day. Input thirty glances at each type of meter. Then decide which is faster, easier, and least demanding. I am not going to monitor a meter that needs interpolation in order to be useful when monitoring.

Digital has it's place and is not substitutable. Analog has it's place and is not substitutable. Many expensive meters combine BOTH functions in a single gauge. Very common in process environments where a misread gauge can kill you.

[enblethen]

That is why OP needs to check both the AC and DC systems to isolate where the problem is.
Yes, it could be the converter failing, but I hate to see parts being installed without verifying what the problem is.

[Gdetrailer]

enblethen wrote:
You should test to see where the loose connection could be.
Need to test 12 volt DC and 120 volt AC systems.
I am still leaning toward an issue with the genset output as it doesn't do it on shore power.

Can't convince anyone that the problem has nothing to do with a loose connection of the gen..

Please READ THIS POST where the OP SOLVED their "flicker" of LEDs (LEDS act faster than a filament)by REPLACING THE CONVERTER.

The LEDS ONLY FLICKERED ON GENERATOR POWER, NOT ON SHORE POWER.

If you REALLY want to test the generator output you should plug in a 120V INCADESCENT LIGHT BULB INTO THE GENERATOR and COMPARE the 120V bulb against the 12V bulb running off the converter. My bet is the 120V bulb won't be "flickering" at the exact time if at all..

[enblethen]

You should test to see where the loose connection could be.
Need to test 12 volt DC and 120 volt AC systems.
I am still leaning toward an issue with the genset output as it doesn't do it on shore power.

[Gdetrailer]

joebedford wrote:
I have mostly halogen lights. They flicker quickly for minutes at a time.

Why would you think it's the WFCO when it doesn't do this on shore power?

Someone recently posted the very same issue you have.

They also had a WFCO converter..

There was many round and round discussions about adding a capacitor to the output..

The OP actually solved the problem by replacing the WFCO with a Progressive Dynamics converter.

I do not think a capacitor would have solved the issue anyways.

Generators do not provide 100% rock solid AC voltages nor is the frequency exactly 60hz. Slight changes in voltage or frequency can conceivably cause a switching converter to also "follow" those changes.

WFCOs do seem to have a weak spot in output voltage regulation when it comes to operation on a generator.

Your choice is to put up with it or replace the converter.