SunGold Power SP6548 Flickering lights after new install

[Quattrohead]

Have not. Didnt really see the need since they only flicker when under DC power from the 6548. After seeing the issue disappear on another inverter I am scratching my head. The lights do no flicker on grid, do not flicker on gen, do not flicker on another inverter wired directly to them.

I am leaning towards using the LV6548 type units in my new build but flickering light's will do my head in LOL.
Maybe I should have a separate LF transformer type system for delicate loads. The Growatt 6000/12000 type units are a beast.

 

[nofear72784]

Right now we are back on grid as the wife can't take the flickering. Really need to get to the bottom of it.

 

[Lilikoi_Heavy_Industries]

Really need to get to the bottom of it.

Try calling the person who sold you the inverter then?

 

[nofear72784]

Yep, sent them videos of the behavior and awaiting a response.

 

[nofear72784]

Afraid I am going to be told it's a "characteristic" of the HF inverter since I see tons of posts about this on Growatts and no one seems to have figured out a fix.

 

[SparkyJJO]

An oscope could tell what's going on with the AC waveform.

 

[nofear72784]

An oscope could tell what's going on with the AC waveform.

On order, should be here tomorrow.

 

[SpongeboB Sinewave]

Would love to see this waveform of several AC cycles to see what the deal is.

No reason that HF inverters can't be steady

 

[nofear72784]

I will be sure to put it up. My pet peeve is dead threads that have no ending. Fixed or not the info will be here. No way I can be the only person that has this issue.

 

[SparkyJJO]

Please do a comparison of the grid waveform too for a reference point.

 

[nofear72784]

Got bored and went back to messing with things. Checked all the AC and DC wiring and connections again, nothing there. Did find that the hertz output of the dimmer at minimum on the inverter is half of what the grid measured but, the voltage was the same. Not sure how they work but that is at least something.

 

[shopman]

What model of Reliance transfer switch do you have? Your discussion about wiring the neutral is confusing as the neutral in the transfer switch should be nothing more than a pass-thru of what would be the neutral coming from a generator to the neutral in your main panel. It's usually just a piece of wire going from the input socket to the neutral bonding point of your main panel.

 

[Shimmy]

Got bored and went back to messing with things. Checked all the AC and DC wiring and connections again, nothing there. Did find that the hertz output of the dimmer at minimum on the inverter is half of what the grid measured but, the voltage was the same. Not sure how they work but that is at least something.

Sounds like the dimmer is not tracking the waveform properly-- not really "true sine wave." Would be good to find out the V(THD) for the inverter output, but the scope read will be interesting.

 

[nofear72784]

What model of Reliance transfer switch do you have? Your discussion about wiring the neutral is confusing as the neutral in the transfer switch should be nothing more than a pass-thru of what would be the neutral coming from a generator to the neutral in your main panel. It's usually just a piece of wire going from the input socket to the neutral bonding point of your main panel

That's correct. It is simply a pass though to the main panel. The neutral ground bond screw in the inverter was removed since I am using a generator transfer panel and not a true sub panel. The neutral and grnd are already bonded on the house side whether the inverter is in bypass or battery mode.

 

[nofear72784]

Sounds like the dimmer is not tracking the waveform properly-- not really "true sine wave." Would be good to find out the V(THD) for the inverter output, but the scope read will be interesting.

It will be interesting. They do not give any specs on THD I read through the manual and online material several times now.

 

[upnorthandpersonal]

When you increased the load to 2kW, did you do this with a purely resistive load? I've noticed some inverters have trouble with LEDs because of their power factor, which is often at or below 0.4 - so very reactive. You could try to add one non-led bulb in your fixture, or just add a purely resistive load on the inverter (coffee maker would do) and see if the flicker gets less/goes away.

 

[nofear72784]

Just gave it a try with a 1200w space heater, no change in the flickering. Adding an incan bulb to places like an office with just 1 fixture kinda stinks. Point is to save energy not use more to help the inverter figure things out. If my other cheap pure sine inverters run these loads directly and don't have an issue this certainly should. The more I mess around the more I am leaning towards a faulty unit or a bad inverter design. Anxiously and nervously awaiting the delivery of a oscilloscope. Will be a nightmare to send this back and wait weeks for them to verify and send another.

 

[shopman]

That's correct. It is simply a pass though to the main panel. The neutral ground bond screw in the inverter was removed since I am using a generator transfer panel and not a true sub panel. The neutral and grnd are already bonded on the house side whether the inverter is in bypass or battery mode.

Ok. That's the correct way to wire it. Any time one of these inverters comes up where you have to remove a screw, I question as to whether or not it actually removes the neutral/ground bond. You might want to verify that with a meter.

I'm running a similar setup with a Reliance switch, 8kw low frequency Growatt and Lutron dimmers with Cree floods. We have absolutely no flickering with this setup. The only flickering I ever see is with the cheap LED bulbs from Home Depot we use in a few locations, and that flickering is always due to temporary loads on the system(refrigerator starting, washing machine tumbling, etc.). It will be interesting to see what your scope reveals.

 

[nofear72784]

Unfortunately the flickering is constant. I could deal with flickering when loads change. Not a bad idea, I can do that

 

[nofear72784]

1.45 megaohms between neutral and ground on the output. I would say it's unbonded