AIMs inverter beeps and causes voltage spikes

[mberding]

Let's set up the situation:

1. A charger is running (could be solar, DC-DC, or shore power) and is outputting 14.something volts (generally 14.2 or 14.4 depending on charger)
2. All batteries are in over-voltage protection mode (because they're sucky grade B cells and no amount of top-balancing will ever bring them in line)
3. A load is placed on the inverter. Could be a mini fridge, heat gun, anything with a couple hundred watts.

What happens next is very interesting. As soon as the load is placed on the inverter (AIMS 3000 watt), while the batteries are in over-voltage protection mode, the system voltage spikes. It's brief, but can jump as high as 19 volts, as I'm watching it on my volt meter. The inverter beeps, but does continue functioning and runs the load. There's also a max-air fan that beeps and complains of high voltage too.

This situation does NOT happen if the batteries are NOT in over-voltage protection mode. Then system acts perfectly normally.

My working theory as an uneducated electrical-engineer wanna-be is that the inverter is taking power from the batteries to run the load, but it takes more than it needs. So it tries to relinquish some, but there's nowhere for it to go since the batteries are in over-voltage protection mode. So then the system voltage spikes.

Can anyone give a better theory, or otherwise prove I'm crazy?

 

[Tomthumb62]

I had something like this happen between my battery and charge controller.

The battery, as it got close to full (14.0v), would trigger an over-volt protection and the Charge MOS would turn off. This would cause the charge controller to spike as high as 19v (Renogy Rover 30A).

Now, I followed @sunshine_eggo 's eggscelent advice on how to slowly bring the battery into better balance.

I forget the exact details, but it went something like this:

Use my Victron IP22 12/30 (single output) 120v charger in "power supply" mode. I set the power supply for 14.4v and charged the battery and looked to see if this triggered an over-volt protection. It did, so I tried again at 14.2, then 14.0, then 13.9 then 13.8. 13.8v didn't trigger a protection, so I kept it there for another few days (12v 200Ah Powerurus). Then I waited a couple of days after than and upped it to 13.9v. No protection triggered. Good! Then 14.0v, again no triggered protection. I quit I think around 14.1 or 14.2v, I forget, now.

So my experience shows you CAN balance these kinds of batteries, it just takes some patience and a bench power supply or a charger that can has a power supply mode like the Victron IP22.

Now I would still get voltage spikes in the Renogy SCC if I set the absorb too high like 14.4v. Now a few weeks after that, I also replaced the Renogy SCC with a Victron Smartsolar 100/30 (similar specs, way better quality) and have had zero voltage spikes since then, regardless of whether I set the Victron to 14.0 or 14.6v. The Victron seems to be smart enough to not create voltage spikes.

 

[mberding]

The battery, as it got close to full (14.0v), would trigger an over-volt protection and the Charge MOS would turn off. This would cause the charge controller to spike as high as 19v (Renogy Rover 30A).

This makes sense when used with a charge controller, since it's designed to put power into a battery. My question here is, am I crazy thinking the inverter is sending it's excess 12v power back from the inverter to the batteries when it decides it doesn't need it?

With regards to top-balancing the battery: that would likely work in most situations, but these cells are really bad, grade B- cells or something like that. Really terrible. Did a proper top-balance before creating the packs and they ended up testing at around half capacity (as measured by a Victron shunt). I don't hold out any hope they're ever going to be recoverable.

 

[sunshine_eggo]

Hello again.

#1 is pretty much guaranteeing this.

If something is sending a charge to the battery, and it engages OVP, you will get a voltage spike.

If your inverter is on shore power, it is acting as a charger. A load will cause a battery voltage drop thus engaging the charger. This could re-engage OVP and trigger a spike.

If OVP is active, and a load causes the voltage to drop, but it doesn't drop below the release threshold and stays in OVP, a charger will react to the voltage drop and will cause a voltage spike.

Do you get this behavior if you fully charge the battery, disconnect from all charge sources, wait 60 seconds and then apply the load?

 

[Tomthumb62]

This makes sense when used with a charge controller, since it's designed to put power into a battery. My question here is, am I crazy thinking the inverter is sending it's excess 12v power back from the inverter to the batteries when it decides it doesn't need it?

I don't know.

What I did learn that with cheap devices, comes cheap other things. In my case, I am unsure whether those 19v spikes ever made it to the actual battery (I don't have a history tab on my BMS or shunt in the system). Since you are dealing with low-quality cells and inverter, it's hard to say if that extra voltage is making it's way to the battery or not.

I wasted my time and money on low-quality Renogy stuff in the beginning. Wish I had only started with Victron to begin with, because none of the Victron devices has given me headaches or voltage spikes.

 

[mberding]

Hello again.

#1 is pretty much guaranteeing this.

If something is sending a charge to the battery, and it engages OVP, you will get a voltage spike.

It's only spiking right now if the inverter has some kind of draw. Otherwise the system seems to be sitting perfectly happen at 14.-something volts.

If your inverter is on shore power, it is acting as a charger. A load will cause a battery voltage drop thus engaging the charger. This could re-engage OVP and trigger a spike.

To clarify, this inverter is very simple and doesn't have a shore power input concept. It's just a 12v DC to 120v AC inverter. There's a separate shore power charger in the system.

If OVP is active, and a load causes the voltage to drop, but it doesn't drop below the release threshold and stays in OVP, a charger will react to the voltage drop and will cause a voltage spike.

This sounds like exactly what's going on.

Do you get this behavior if you fully charge the battery, disconnect from all charge sources, wait 60 seconds and then apply the load?

Excellent diagnostic question and the answer is: No. System works perfectly fine when OVP isn't in play.

This system (a friend's van) used to have BattleBorns in it. Then these crappy batteries got installed, and issues started happening. This is why I've been so pro-BattleBorn lately. Even at 14.4 volts, the batteries were NOT in OVP, and the whole system worked smoother.

And yes, I'd love to just adjust all the charging sources to stop charging at 13.8 volts, but it's cheap equipment that doesn't have those options.

 

[mberding]

I wasted my time and money on low-quality Renogy stuff in the beginning. Wish I had only started with Victron to begin with, because none of the Victron devices has given me headaches or voltage spikes.

Agreed, I'm all about the Victron equipment for every application and system these days. The cheap stuff just isn't worth it.

 

[sunshine_eggo]

I don't know.

What I did learn that with cheap devices, comes cheap other things. In my case, I am unsure whether those 19v spikes ever made it to the actual battery (I don't have a history tab on my BMS or shunt in the system). Since you are dealing with low-quality cells and inverter, it's hard to say if that extra voltage is making it's way to the battery or not.

By definition, the BMS and all battery attached equipment sees the spikes, but the battery cells themselves don't.

 

[sunshine_eggo]

It's only spiking right now if the inverter has some kind of draw. Otherwise the system seems to be sitting perfectly happen at 14.-something volts.

With the voltage essentially stable, the charger is just supplying either 0A or micro-A and is thus not impacting voltage.

To clarify, this inverter is very simple and doesn't have a shore power input concept. It's just a 12v DC to 120v AC inverter. There's a separate shore power charger in the system.

If that's the case, then simply turning the charger off before engaging the load should suffice. No 60 second wait.

This sounds like exactly what's going on.

Excellent diagnostic question and the answer is: No. System works perfectly fine when OVP isn't in play.

This system (a friend's van) used to have BattleBorns in it. Then these crappy batteries got installed, and issues started happening. This is why I've been so pro-BattleBorn lately. Even at 14.4 volts, the batteries were NOT in OVP, and the whole system worked smoother.

And yes, I'd love to just adjust all the charging sources to stop charging at 13.8 volts, but it's cheap equipment that doesn't have those options.

If they have any canned options, GEL often has a 14.0-14.2V. If an upgrade is an option, the Powermax PM3-LK units can be configured as a constant voltage source, or a 3 phase charger with pot adjustable bulk voltage.

PM3 12V Series - PowerMax Converters in 12, 24 & 48 Volts

AC to DC Power Converters can be used for RV Converters, Car Audio Power Supplies and Battery Charging for 12, 24 & 48V - Lithium Compatible

powermaxconverters.com

 

[mberding]

With the voltage essentially stable, the charger is just supplying either 0A or micro-A and is thus not impacting voltage.

I guess I'm just surprised that a charging source would spike so high so quickly. Seeing 19v (even if just for an instant) is really not expected when the charger is supposed to be putting out 14.4

If that's the case, then simply turning the charger off before engaging the load should suffice. No 60 second wait.

Agreed, in theory, but if your solar charge controller is always running unless you manually throw the switch to disconnect the panels, every day, multiple times a day, it's not the best.

If they have any canned options, GEL often has a 14.0-14.2V. If an upgrade is an option, the Powermax PM3-LK units can be configured as a constant voltage source, or a 3 phase charger with pot adjustable bulk voltage.

Unfortunately these cells are so bad that they go over voltage at like 13.5 or something. Haven't done enough diagnosis with them yet to see exactly where it is, but it's pretty far south of 14v.

And with all these bad cells, I was trying to just add a nice "normal" battery to NOT be in OVP at 14.4 and avoid the whole situation. That's what lead to the LiTime threads I've been posting.

Thanks for your help sunshine_eggo, it's nice to have someone to bounce this stuff off of so it's not solely rattling around in my head.

 

[Partimewages]

Have you tried doing a top balance? You can do it with a $50-$75 bench top charger.

 

[mberding]

Have you tried doing a top balance? You can do it with a $50-$75 bench top charger.

The cells were top balanced before being assembled with a desktop charger. They're exceedingly crappy cells though and won't maintain balance through even a single cycle, so OVP kicks in way too early, all the time.

 

[sunshine_eggo]

I guess I'm just surprised that a charging source would spike so high so quickly. Seeing 19v (even if just for an instant) is really not expected when the charger is supposed to be putting out 14.4

This should not be a full time thing. Any charger should drop to 13.5V-ish to float the battery. Holding LFP at 14.4V is a great way to damage it long term.

Agreed, in theory, but if your solar charge controller is always running unless you manually throw the switch to disconnect the panels, every day, multiple times a day, it's not the best.

A solar charge controller follows a charge algorithm and should float the battery at ~13.5V. Most MPPT would interpret the abrupt drop to 0A at or above 14.4V as, "holy shit, this battery is full" and drop to float.

Unfortunately these cells are so bad that they go over voltage at like 13.5 or something. Haven't done enough diagnosis with them yet to see exactly where it is, but it's pretty far south of 14v.

So there's already a potential 0.9V difference... let's say this shitty battery is as high resistance as a lead acid... 4mΩ. 0.9V/.004Ω = 225A. yep. The instant the BMS allows a charge, the charger immediately sends the lower of 1) its rated output current or 2) 225A.

Are you still surprised there's a voltage spike when the charger instantly outputs its max?

And with all these bad cells, I was trying to just add a nice "normal" battery to NOT be in OVP at 14.4 and avoid the whole situation. That's what lead to the LiTime threads I've been posting.

new inexpensive batteries are imbalanced... they consistently correct over the course of a few days/weeks.

Thanks for your help sunshine_eggo, it's nice to have someone to bounce this stuff off of so it's not solely rattling around in my head.

When I'm not listening to other people's rattling, I'm deafened by my own!

 

[mberding]

This should not be a full time thing. Any charger should drop to 13.5V-ish to float the battery. Holding LFP at 14.4V is a great way to damage it long term.

Agreed, and since this user only occasionally gets shore power, it'll never be long term. The DC-DC and Solar charge controllers are a bit smarter and will drop their voltages a bit.

A solar charge controller follows a charge algorithm and should float the battery at ~13.5V. Most MPPT would interpret the abrupt drop to 0A at or above 14.4V as, "holy shit, this battery is full" and drop to float.

I think these batteries are so bad that they're in OVP at 13.5-ish. When the inverter kicks in, the voltage drops more, and I'm sure the solar CC then thinks it should go back to bulk. But, it's a cheap solar CC, so who knows.

So there's already a potential 0.9V difference... let's say this shitty battery is as high resistance as a lead acid... 4mΩ. 0.9V/.004Ω = 225A. yep. The instant the BMS allows a charge, the charger immediately sends the lower of 1) its rated output current or 2) 225A.

Are you still surprised there's a voltage spike when the charger instantly outputs its max?

It's quite the series of events:

1. Batteries are in OVP
2. Charger is at 14.4
3. Inverter draws power, voltage drops to probably low 13's as that's where the batteries are actually at
4. Batteries provide power, but don't release their OVP mode yet
5. Charger ramps up charge rate
6. Inverter draw subsides
7. Charger is still at high rate and batteries don't accept power
8. Results in power spike
9. Charger then sees spike and reduces output
10. Batteries may or may not release OVP depending on how much power was sucked out of them

What gets me is how fast this whole process happens.

new inexpensive batteries are imbalanced... they consistently correct over the course of a few days/weeks.

These new batteries need a whole sheet of instructions on how to get them properly working, or some videos or something. Possibly a new video topic for Will. "How to top-balance your cheap new batteries so you can avoid crazy OVP situations"

When I'm not listening to other people's rattling, I'm deafened by my own!

Happy to be of service!

 

[sunshine_eggo]

Agreed, and since this user only occasionally gets shore power, it'll never be long term. The DC-DC and Solar charge controllers are a bit smarter and will drop their voltages a bit.

If speed of charging isn't an issue, a charger set for 13.6V will eventually top off an LFP battery. Would make the spikes less frequent and severe.

I think these batteries are so bad that they're in OVP at 13.5-ish. When the inverter kicks in, the voltage drops more, and I'm sure the solar CC then thinks it should go back to bulk. But, it's a cheap solar CC, so who knows.

Ouch!

It's quite the series of events:

1. Batteries are in OVP
2. Charger is at 14.4
3. Inverter draws power, voltage drops to probably low 13's as that's where the batteries are actually at
4. Batteries provide power, but don't release their OVP mode yet
5. Charger ramps up charge rate
6. Inverter draw subsides
7. Charger is still at high rate and batteries don't accept power
8. Results in power spike
9. Charger then sees spike and reduces output
10. Batteries may or may not release OVP depending on how much power was sucked out of them

What gets me is how fast this whole process happens.

Yep... within milliseconds of OVP release.

#10 is happening each time you get a spike, but it may be immediately reinstated.

These new batteries need a whole sheet of instructions on how to get them properly working, or some videos or something. Possibly a new video topic for Will. "How to top-balance your cheap new batteries so you can avoid crazy OVP situations"

Find the OVP voltage and lower bulk/absorption/boost to something below that but above 13.8V and float 13.6V foat for the first few weeks of regular use. Once you can attain 14.4V w/o OVP, set 14.0-14.4V bulk/boost/absorption and 13.5V float.