Generator PTO Generator - Lights Flicker

[teg33]

You are correct on my wiring situation, #6 to #12. This was a temporary setup to test the generator. The lights flicker worse at a light load (4.1A one side, 6.7A other side) which shouldn't be affected by this wire size.

Now it may be possible that the small wire is preventing the problem from going away completely at my higher test load (about 29A on each side) which includes resistive and motor loads.

I also learned from Tiger that this smaller size generator does not have an active voltage regulator, it relies on capacitor regulation. We have tried changing the capacitors with no improvement.

Can anyone answer my earilier "governor" question? What type of governor is on the B2400 and how well does it regulate the engine/PTO speed at various loads? Could it allow a fluctuation up and down (oscillation) in the PTO speed as it regulates?

Thanks

 

[teg33]

Can anyone explain the Kubota governor?

What type of governor is on the B2400 and how well does it regulate the engine/PTO speed at various loads? Could it allow a fluctuation up and down (oscillation) in the PTO speed as it regulates?

Thanks

 

[SnowRidge]

I know nothing in particular about Kubota governors, but I'll bet it's a mechanical flyball device. In any event, if the lights were flickering because of changing frequency (RPM), you should be able to hear it. If as you say, the frequency meter is correct (and steady), then that too would pretty much rule out generator RPM as an issue.

So that leaves voltage sags/surges. You said the voltmeter reads normal and you have put an o'scope on the line and can see nothing wrong. That doesn't add up.

If there is any kind of voltage or frequency instability of a magnitude great enough to be visible with a light bulb, it should be easily be visible on the scope, so why isn't it. I guess we need to know a little more about how you are checking the output of the generator.

Are measuring the voltage at the light bulb socket? What kind of meter are you using? What are you using to read THD? What kind of o'scope are you using? What sweep and sync settings are you using? What powers it? Not the circuit under test, I hope.

Working theories without knowing more:

1. The generator has crappy voltage regulation, especially under light loads, and it is hunting. You are seeing the result.

2. There is a resistive leakage path at the transfer switch or elsewhere, and you are seeing phase differences between the generator and the utility power.

 

[Soundguy]

I rather think the pto shaft not equaling a constant velocity joint is / may be a factor.

My capacitive regulation on my northstar 12.5kw genny is fine with a single 100w lamp plugged in. course.. my genny is mounted on a cary-all, and the tractor pto runs straight out to the genny input.. no angle, and no bends for the pto yokes to make.. etc... perhaps this is a bigger issue than it would seem to be?

Also makes me wonder if the pto type is having a factor.. I.E. a tractor that has a built in ORC may not suffer as much as one that doesn't when looking at the pto angle and the velocity of the shaft.. that is.. if the shaft is speeding up du to an angle.. it may not have to 'slow' down on the downside of the rotation if the tractor has an ORC, as the shaft will be able to overspeed.. whereas a tractor with a non live/non independent pto may lead/lag the pto shaft making that oscilation more noticeable?.. just a theory..

Soundguy

Soundguy

 

[SnowRidge]

If it was due to angular differences on the PTO, wouldn't the flicker have to be at a rate of N(PTO_RPM)/60, where N is the number of velocity changes per revolution. It seems like N would be either 1 or 2, but I haven't thought it out. Assuming a 540 PTO, it should mean a steady flicker rate of either 9 or 18 HZ. A rate like that is observable and could be computed with the aid of a stop watch.

If the calculated rate was 9 or an even multiple of 9, it would tend to confirm the theory.

Edit: I'm not thinking straight, and I oughta quit while I'm behind -- as my wife likes to tell me. A 9 HZ rate may be observable, but it isn't countable without test equipment, at least not accurately. /forums/images/graemlins/crazy.gif /forums/images/graemlins/blush.gif /forums/images/graemlins/crazy.gif

 

[Soundguy]

Got to have some fast thumbs.. but yeah.. that's the idea. Also to throw in the mix .. I had been thinking about genny poles and 1800rpm vs 3600 rpm.. but after thinking.. it doesn't muddy it that much, as if the pto angular problems are happening.. then it will still be a multiple that you could track.

Soundguy

 

[Dutch445]

When you say the lights "Flicker",
can you describe it more accurately?
Is it a constant flicker, every few seconds,

or once a minute,

and once it's doing it, does it stay constant
until increased load is put on the generator?

and i'm way over my head here,, but thought I'd throw
this in...
is there anyway to eliminate that 50' lead and get the
generator closer to your transfer switch?
/forums/images/graemlins/confused.gif

 

[teg33]

Sorry I missed all this input but had a long day. Here's some more...

- I ruled out the PTO shaft alignment, checked it again and everything is straight as an arrow.

- I tried to attach a video clip but couldn't create a useful file less than 100k. The problem is more of a pulsating than a flicker, I'm estimating 3 pulses per second worst case with light load. If this were a governor caused fluctuation, I gotta believe you could hear the engine difference, which I do not.

- Measurement tools were a Tektronix TDS3054B Digital Scope and Fluke 83B Power Quality Analyzer, both with calibration stickers and powered off house current. Probes connected at transfer switch before the house subpanel. I'll attach a sample plot or two of the worst case.

- I could eliminate the 50' lead BUT I did connect a gas powered Coleman through this same connection and had no problem with flicker. To me, that should rule out the lead unless it is somehow interacting differently with the PTO gen.

Thanks again

 

[teg33]

The worst case plot with a larger time scale. Note what appears to be some amplitude modulation on the waveform.

 

[Soundguy]

Wow.. is that a brush or brushless unit?

The first pic showed what? 63.x hz,.. that's way unacceptable. Nice waveform too -not-.. reminds me of modified.. I tested my northstar output with a freq counter and graphing fluke vom.. came out with a 'perfect to the eye' sine wave.. as far as the pixels on the scope looked... and a rock steady 60hz, Heck.. measuring my house line... I got an occasional 60/61 reading, and more voltage fluctuations... House ran between 114 and 130 depending on what was running. Genny ran pretty near 119/120v...

Soundguy