120VAC inverter spiking around 190V, LiFePO4 jumping from 13.5VDC to 96VDC

newuserdude · October 2017

Equipment:

- 12V 300Ah LiFePO4 rated@100A continuous
- PUGU 12VDC to 120VAC 3000W Pure Sine inverter (cheap eBay brand)
- 4/0AWG wire (giant stuff)

Problem:

- When I try to run a high load appliance, like a 1500W microwave (microwave is rated for 120VAC @60hz, it is NOT rated for 120-240VAC):
- Inverter is outputting ~190VAC under load according to my multimeter (Craftsman 82140). I've read something about RMS...not sure if that has anything to do with this.
- Battery is jumping from 13.5VDC to ~96VDC! I measured at the battery terminals where I had the inverter connected before and during load.

I know I'm only rated to run ~1100W continuous, but the same thing happens when I run the microwave at 50% power and when I plug in my 1800W induction cooktop and don't even turn it on (I know the inverter outputs ~190VAC, but I'm not sure if the battery is also jumping from 13.5VDC to ~96VDC when running the microwave at 50%)

I've tried another of the exact same inverter, as well as a 2500W Cobra Modified Sine inverter and all THREE inverters are showing output around 190VAC with my multimeter. I did try another multimeter on one of the inverters and got the same numbers.

What else should I check?

mcgivor · October 2017

A simple calculation for root mean square "RMS", is to multiply the indicated voltage by 0.7071 which in the case of 190V is ~135V, but it's not as simple as that, inductive loads can create havoc with inverters, especially the cheaper variety, the waveform may not be as pure as the name indicates, having spikes particularly at the peak, or crest, where the transition from rising voltage to dropping voltage occurs, both the positive and negative cycles Inductive loads such as a microwave or induction cook tops, cause the current to lag the voltage, being out of phase, may confuse the switching which generates the waveform, without actually checking with an ociloscope it's impossible to speculate what is actually occurring in your particular case. The wattage of the microwave will actually be higher than the name plate suggests due to power factor, additionally more than often the 50% power represents a 50% duty cycle, or full power in cycles.

The voltage reading of ~96V at the battery is not a true reading, the impedance of a digital multimeter is too high to pull down phantom feedback from the inverter, grounding the negative of the input and one leg of the output, to create a neutral would probably eliminate these eronious readings, do you have your system configured as such?

I've included a link about RMS and a pdf about inverter wiring, hope this information helps.
http://www.electronics-tutorials.ws/accircuits/rms-voltage.html

newuserdude · October 2017

THANK YOU. No, the inverter is not grounded. It doesn't have a ground connection. I'm installing it in a van.

Should I simply connect the battery and inverter negative terminals to the chassis?

The diagram shows the battery negative going to ground and the inverter having a dedicated ground connection also going to ground. My inverter only has positive and negative, no dedicated ground connection.

The induction cooktop creates the same ~190VAC readings on the inverter when the induction cooktop isn't even turned on. It's 1800W and the inverter says it draws 350W when it's simply plugged in and not even turned on. I'm guessing it's charging up the cooktops magnets or capacitors or something, and will stop drawing when the cooktop is 'full'?

I'm going to try plugging in a lightbulb and seeing if it gets brighter or dimmer when running the microwave. That should tell me if the multimeter is being accurate.

newuserdude · October 2017

I just noticed it also shows the wires getting grounded on the output side of the inverter as well...

Does that mean all three of those things need to be connected to my van's chassis?

1. Battery negative
2. Inverter on the battery side (my inverter doesn't have a dedicated ground, do I just ground the negative to the chassis?)
3. Inverter on the output side (connect each run of 120VAC cable to chassis, too?)

mike95490 · October 2017

peel the PSW sticker off the inverter, now you have a cheap MSW inverter.

with real 120vac, you have close to 190v peak, and RMS voltage is 120v. but cheap inverters put out garbage waveforms that mess things up

newuserdude · October 2017

Below is the best theory I've got so far.

@mcgivor Do you know how to ground in a vehicle? I have a few questions, maybe you could answer them quickly:

1. Do the 12VDC battery, AC inverter, and AC breaker box (or each AC appliance) all get grounded? I thought I read something about not grounding AC and DC together to the chassis. And I thought I read something about how in a vehicle, on the AC side, you only need to worry about the positive and negative and that the 'ground' connection isn't actually doing anything - something like just needing to ground the negative side of the AC and ignoring the 'ground' connection.

2. If the inverter doesn't have a dedicated ground connection, would I just run the negative to the chassis?

3. Is it possible that my crappy 3000W pure sine inverter could be tested (perhaps with an ociloscope) so I can get proof that it's safe to use? I'm already basically planning on replacing it with a much better unit to hopefully resolve all of these issues, but I'm curious if there's a way to truly test the inverter and determine whether it's safe to use and what at loads it can actually produce a pure sine wave before things get 'distorted' or are no longer safe/correct.

"A simple calculation for root mean square "RMS", is to multiply the indicated voltage by 0.7071 which in the case of 190V is ~135V, but it's not as simple as that, inductive loads can create havoc with inverters, especially the cheaper variety, the waveform may not be as pure as the name indicates, having spikes particularly at the peak, or crest, where the transition from rising voltage to dropping voltage occurs, both the positive and negative cycles Inductive loads such as a microwave or induction cook tops, cause the current to lag the voltage, being out of phase, may confuse the switching which generates the waveform, without actually checking with an ociloscope it's impossible to speculate what is actually occurring in your particular case. The wattage of the microwave will actually be higher than the name plate suggests due to power factor, additionally more than often the 50% power represents a 50% duty cycle, or full power in cycles.

The voltage reading of ~96V at the battery is not a true reading, the impedance of a digital multimeter is too high to pull down phantom feedback from the inverter, grounding the negative of the input and one leg of the output, to create a neutral would probably eliminate these eronious readings, do you have your system configured as such? "

120VAC inverter spiking around 190V, LiFePO4 jumping from 13.5VDC to 96VDC

Comments

Categories