Inverter capacitor charging spark!

[Maast]

About 1 amp hour per month. If the cells have hit LVDC at 2.5, 1 AH will destroy the cells.

This type of configuration would never be UL listed. The entire load side will always be energized; of course I'd never touch stuff without first checking ... that's not the point. A fault condition should remove all potential.

Will it work? Sure. I'm not OK with this configuration. Some are, some aren't. That's why this thread isn't so cut and dry.

I prefer giving people education and options than to say "just do the following" - that's why this thread isn't only 2 posts - a question and an answer.

Fair enough, it's your system your building so you have to be happy with it. I'm just pointing out a much simpler solution, one I'm personally comfortable with. Though I do have to say invoking the UL is a bit specious.

Some amplifying information, say you have a 'leaky' system and after a half-hour your system has stabilized at 80% of charge, that's ~40 volts giving a ~10 voltage difference through a 10kohm resistor and so your current would be 0.00102 amps. 1 amp hour would be 41 days.

1 amp hour out of a 48v 100ah battery bank with a (IMO more reasonable LVD) of 2.8v/cell would drain your battery down to 2.792v/cell. To reach 2.5v thats 153 days.

So I'll admit I did overstate how long it'd take as I hadnt actually crunched the numbers - 2ah would take 306 days if you had a 100AH bank.

Though I would hope that if somebody is going to leave a system unattended for that long they'd disconnect the batteries.

I'm thinking the best way for a person to find what their capacitor/system self-discharge rate is to do a bit of trial-and-error and see what their voltage stabilizes at and adjust the resistance up or down to where they want their precharge voltage to be held at, then take the value of the resistance and run it through a online calculator to find out what their 'leakage' current is to see if they're comfortable with it.

Here's one that I just found that'd help.

 

[cinergi]

Fair enough, it's your system your building so you have to be happy with it. I'm just pointing out a much simpler solution, one I'm personally comfortable with. Though I do have to say invoking the UL is a bit specious.

I don't mean to be, but some of us are building whole-house systems or RV systems. I want people to evaluate the safety of their systems, should, e.g., a firefighter get involved (which is the reason I invoked UL listing; I'm thinking about the rapid shutdown requirements).

1 amp hour out of a 48v 100ah battery bank with a (IMO more reasonable LVD) of 2.8v/cell would drain your battery down to 2.792v/cell. To reach 2.5v thats 153 days.

Can you help us understand how you got there? I don't think my 280's would have 1Ah left in them between 2.8 and 2.5 based on my memory from capacity testing them.

Either way, it illustrates that people have to think about their cutoff voltage in terms of "what if the system faulted and I was away for X period of time"? It's entirely possible I'd want to leave stuff in my fridge/freezer while I travel for a month so I leave everything on.

 

[Maast]

Can you help us understand how you got there? I don't think my 280's would have 1Ah left in them between 2.8 and 2.5 based on my memory from capacity testing them.

I found a graph of testing of a 18650 in mah per volt, then I scaled up from there. .........(edited out, bad data, mixed up watt hours and amp-hours)

Edit, disregard second half of statement, been spending the afternoon doing too much hopping from watt-hours to amp-hours. Though it'd worth it to pop one of my 280ah cells on the tester to actually see how many ah-per-volt the current crop of cells do.

 

[HaldorEE]

This seems like a nice way to avoid the spark / surge when connecting the battery to the inverter, but wouldn’t this same precharge be needed when the BMS automatically reconnects the battery?

What I mean is, if LVD is triggered overnight when discharging the battery and the BMS disconnects, won’t a similar surge / spark occur when solar charging begins in the morning and the BMS reconnects the battery to the inverter?

I don't believe my BMS will automatically reconnect if it ever disconnects due to low cell voltage. I am using a common port BMS and a disconnect event unplugs everything, charge sources as well as loads.

My system doesn't depend on the BMS disconnecting during normal operation. The charging high voltage cutoff is controlled by the Victron BMV-712 and the discharge low voltage cutoff is controlled by the inverter (Victron Multiplus) and a Victron Battery protect for DC loads.

The BMS primary role is to protect me if the cells get seriously unbalanced in which case I will rebalance the cells before using them again, or as a last ditch backup to the Victron equipment in the event of a malfunction.

So if my BMS ever disconnects, I am going to first figure out what went wrong. Fix it and only then manually reconnect the battery pack.

 

[HaldorEE]

About 1 amp hour per month. If the cells have hit LVDC at 2.5, 1 AH will destroy the cells.

This type of configuration would never be UL listed. The entire load side will always be energized; of course I'd never touch stuff without first checking ... that's not the point. A fault condition should remove all potential.

Will it work? Sure. I'm not OK with this configuration. Some are, some aren't. That's why this thread isn't so cut and dry.

I prefer giving people education and options than to say "just do the following" - that's why this thread isn't only 2 posts - a question and an answer.

That is exactly why I put a push button in series with the precharge resistor. Can't leave it on.

I plan on loading and unloading my battery pack for the surface of the sun season (summer). Don't want to cook the lithium batteries in my RV when it is not in use during the Phoenix summer.

I have a battery connection procedure and it is pretty simple.

• Turn off the inverter.
• Turn off the battery circuit breaker.
• Plug in the battery pack.
• Push and hold the button for one second.
• Turn on the battery circuit breaker.
• Turn on the inverter.

Battery disconnect procedure is mostly just to make sure I leave things in the mode they need to be prior to reconnecting the batteries:

• Turn off the inverter.
• Turn off the battery circuit breaker.
• Unplug the battery pack.

And the 3 position switch is not fool proof. You still have to turn it to the precharge position before plugging in your batteries.

Manual, is manual. I have a 150A circuit breaker I am also using for a disconnect. That let me get rid of a separate battery disconnect switch in my system.

 

[fafrd]

I found a graph of testing of a 18650 in mah per volt, then I scaled up from there. .........(edited out, bad data, mixed up watt hours and amp-hours)

Edit, disregard second half of statement, been spending the afternoon doing too much hopping from watt-hours to amp-hours. Though it'd worth it to pop one of my 280ah cells on the tester to actually see how many ah-per-volt the current crop of cells do.

I’ve got this data for a several of my 280Ah cells.

1 cell ran for 0.86h @ 2.65A between 2.792V and 2.532V, meaning ~2.28Ah or 0.08% of rated capacity.

A 2nd cell ran for a total of 2.27Ah between 2.793V and 2.525V at similar low current levels under 4A.

And a 3rd cell being discharged at 10A delivered an additional 2.33Ah between 2.8V and 2.5V.

EVE’s discharge curve indicates a steeper fall-off at lower discharge levels, so if you were discharging at 0.5C / 140A rather than my very gentle discharge rate of ~0.01-0.04C, you’d likely have more Ah between 2.8V and 2.5V than the ~2.3Ah I measured...

 

[FilterGuy]

That is exactly why I put a push button in series with the precharge resistor. Can't leave it on.

Yup. I struggled with the various scenarios between using a pushbutton vs the Perko Switch.

Push-button: Can't leave it on but can forget to use it.
Perko Switch: Can leave it in precharge but can't forget to use it.

I decided to go the perko rout but designed in a PTC resistor that will limit the current if the switch is left in precharge. Even if the user does not pause on the pre-charge position, the capacitors get charged enough. That worked well, but I after I built the system I discovered the PTC was not really needed. If the switch is left in the pre-charge position the inverter comes on but immediatly goes into Low-Voltage disconnect mode. The onlything running is the internal logic circuits and it only takes a little current so the resisters in the pre-charge don't heat up.

 

[HaldorEE]

Yup. I struggled with the various scenarios between using a pushbutton vs the Perko Switch.

Push-button: Can't leave it on but can forget to use it.
Perko Switch: Can leave it in precharge but can't forget to use it.

I decided to go the perko rout but designed in a PTC resistor that will limit the current if the switch is left in precharge. Even if the user does not pause on the pre-charge position, the capacitors get charged enough. That worked well, but I after I built the system I discovered the PTC was not really needed. If the switch is left in the pre-charge position the inverter comes on but immediatly goes into Low-Voltage disconnect mode. The onlything running is the internal logic circuits and it only takes a little current so the resisters in the pre-charge don't heat up.

That is what I figured also. The 8W, my 82 ohm resistor is going to deliver is not enough current to actually turn the inverter ON.

 

[Sanwizard]

I have the same 50-ohm resistor (as seen in mrnatural22 photo )connected in series with a push-button switch. These two are then connected across the terminals (in parallel) of my 100 amp circuit breaker that is in series with the positive lead of my battery.

Before I switch on my breaker, I hold the push button switch for about 5 seconds then let go of the button and switch the breaker on.

I don't ever see any sparks anymore and when testing the resistor current when pushing the button, there is very little and prevent that spark. - It works for me!

Can you share where you bought the pushbutton? Whats the amps and volts max for it? I have 580ah at 48V, and am wondering if I need a high amp button just in case.

 

[Hedges]

Can you share where you bought the pushbutton? Whats the amps and volts max for it? I have 580ah at 48V, and am wondering if I need a high amp button just in case.

48V, 50 ohm?
You might need a switch that can carry what, 1000 mA?

 

[Jim Burrow]

anything can work as a switch. For example, a doorbell button will work. The 50-ohm resistor drops the current so low that any tiny switch can carry the voltage and current. In reality, all the voltage is across the resistor because the switch acts like a short. - it is a short circuit with zero voltage across it. Note that I don't close the circuit breaker until I have pushed the push button to neutralize the spack. - Sorry, if I lost you. I'm a retired electronic engineer and many times I use terminology that isn't normally used in electricity terminology and in this group.

 

[Sanwizard]

anything can work as a switch. For example, a doorbell button will work. The 50-ohm resistor drops the current so low that any tiny switch can carry the voltage and current. In reality, all the voltage is across the resistor because the switch acts like a short. - it is a short circuit with zero voltage across it. Note that I don't close the circuit breaker until I have pushed the push button to neutralize the spack. - Sorry, if I lost you. I'm a retired electronic engineer and many times I use terminology that isn't normally used in electricity terminology and in this group.

No worries! My brother is an EE, so I am used to it! Ohms law baby!

 

[OzSolar]

My 24v/5000w Victron Quattro would not not even try to turn on unless I pre-charged it or had two batteries in parallel. A single 24V/200AH LifePo4 went from 26.3 v to 5.5. The manufacturer was the first person to mention that the BMS was going into overcurrent mode and that I needed to pre-charge the inverter. The BMS did appear to immediately reset without any issues.

I still find it odd that the neither the inverter or battery documentation mentions ANYTHING about needing to pre-charge the inverter. I also found it odd that my battery supplier was not aware of this. In fact they were in the process of having me return the battery but thankfully the factory got back with us almost immediately. Kudos to Tony at BestGo Battery by the way. Of course I never hook up a live wire to a battery, I always use a breaker to avoid the spark but now I know its imperative to preventing the BMS from going into a fault mode. I think the inverter manufacturers should be supplying pre-charge circuits but at a minimum at least warn us of this.

 

[Supervstech]

My 24v/5000w Victron Quattro would not not even try to turn on unless I pre-charged it or had two batteries in parallel. A single 24V/200AH LifePo4 went from 26.3 v to 5.5. The manufacturer was the first person to mention that the BMS was going into overcurrent mode and that I needed to pre-charge the inverter. The BMS did appear to immediately reset without any issues.

I still find it odd that the neither the inverter or battery documentation mentions ANYTHING about needing to pre-charge the inverter. I also found it odd that my battery supplier was not aware of this. In fact they were in the process of having me return the battery but thankfully the factory got back with us almost immediately. Kudos to Tony at BestGo Battery by the way. Of course I never hook up a live wire to a battery, I always use a breaker to avoid the spark but now I know its imperative to preventing the BMS from going into a fault mode. I think the inverter manufacturers should be supplying pre-charge circuits but at a minimum at least warn us of this.

I don’t know... but I think the battery supplier needs to have this in their documentation... the BMS is going into fault, they need to have a warning...

 

[jtaon001]

Well, yeah, a bit OCD.... BUT...

I have a 100A switch between the battery and the inverter, so no touching wires to pre-charge. Just connect your wires and throw the switch.

Wouldn't there still be a spark after you switch it on since there's no resistor? Just curious because this seems a lot simpler than routing a toggle switch with one side a resistor for pre-charge and another side main power haha

 

[Supervstech]

Wouldn't there still be a spark after you switch it on since there's no resistor? Just curious because this seems a lot simpler than routing a toggle switch with one side a resistor for pre-charge and another side main power haha

This makes it so the spark is inside the switch... if the switch is not DC arc rated, it will not survive many inverter power ups.
NOTHING will prevent the spark, except a precharge resistor in some form. The switch just makes it so you don’t see the spark.

 

[DW SD]

In my case, my RV, all the 12V DC loads are tied via heavy busbar, including the inverter. This includes various computers, like for the slide controller and a diesel - fired heater, which at battery startup draws 15 amps in its purge mode.

If I add a pre-charge resistor and all the items power up, will it still charge the inverter capacitors?

I suppose I could add another heavy switch to isolate all of the other DC loads. Would this be necessary or do the capacitors charge first?

Trying to avoid Rube Goldberg complication :D
Thank you,

Doug

 

[FilterGuy]

If the voltage is high enough to start your devices, then the capacitors are probably charged enough to not be a problem.

 

[Hedges]

The only way a precharge resistor will work is if charging only capacitors, no DC loads. Or, if resistor is small enough to power the loads (total current times resistor value "V = IR" is about 1V or less.

Precharge is only needed for capacitors, which present about zero ohms initially. Just a switch or breaker feeding inverter (which has large capacitance) needs precharge resistor, not your other loads.

 

[FilterGuy]

The only way a precharge resistor will work is if charging only capacitors, no DC loads. Or, if resistor is small enough to power the loads (total current times resistor value "V = IR" is about 1V or less.

Precharge is only needed for capacitors, which present about zero ohms initially. Just a switch or breaker feeding inverter (which has large capacitance) needs precharge resistor, not your other loads.

That is not my experience..... I have built systems with the pre-charge on the DC loads and the Inverter.... it worked fine. (Keep in mind that I always make my pre-charge resistor a very low value. Clearly, this would not work well if you have huge DC loads but smaller loads are ok.

Having said that.... my typical build will have the precharge on the inverter only.