Are these batteries being cooked?

williaty · March 2013

The charger is a Victron Energy Phoenix Multiplus Compact 24/2000/50 120V. The batteries are (soon to be replaced with a better sized choice) 4 UPG Universal Power UB12500 50Ah AGM batteries wired 2 serial/2 parallel. I finally have enough of the system up and running to start doing some testing of it. Tonight, I ran the first major discharge/recharge cycle with logging of the current and voltage data. I've already caught one thing I don't like how the charger is running it and I have fixed that. However, I'm not sure about the Absorb Stage termination. I had somehow gotten the wrong impression and thought that the Victron chargers were capable of terminating Absorb based on Ending Amps. As far as I can tell, that's not true. It appears that you have an option for a fixed Absorb time and an Absorb time that is proportional to Bulk with a max time limit. Right now, the charger is configured to make Absorb proportional to bulk. The Bulk stage on this run took about 5 minutes and the charger stayed in Absorb stage for what looks like about 180 minutes. I am concerned that the charger might be keeping the batteries in Absorb for too long and cooking them, since I've read that AGMs don't like to be over-Absorbed. Below, I've included a graph of the last portion of the Absorb stage's current flow into the battery. Since the bank is 100Ah, it passed the 1% current that I've heard bandied about for Ending Amps at 19:08 but the Absorb stage didn't terminate until 20:56. So it hung in Absorb stage for almost 2 hours after the charge current had dropped to less than 1% of the bank capacity. On the other hand, you can see that the current into the bank did continue to decrease, albeit very slowly at the end, right up until the moment that Absorb terminated. I've also read that the charge flow into the bank is another way to determine proper termination for Absorb.

So what do you think? Are these batteries getting cooked in Absorb for too long or is this about right?

Cariboocoot · March 2013

Re: Are these batteries being cooked?

I think you're right: the Absorb stage is too long. The batteries probably aren't damaged after only one such cycle, though.

Unfortunately I don't have a manual for that particular charger. Perhaps you should set the Absorb time to a maximum of 1 hour. The Bulk:Absorb ratio is normally 1:1, so what's happening here sounds a bit odd to say the least. Does it have a minimum Absorb time? Some controllers do, and it is to compensate for a condition where the Bulk time is very short (like 5 minutes) and wouldn't do much for the Absorb 'clock'. That last hour of Absorb looks quite useless to me.

BB. · March 2013

Re: Are these batteries being cooked?

You could probably take 1-2 hours off the timer here and still be OK... I am not sure about AGM, but for flooded cell we are looking for >90% SOC a couple times a week as a normal "full charge" regime. In your case you are looking at around 1-2 AH of extra stored energy in the bank--or less than 2% additional SOC points. Pretty much a "don't care" amount for a daily charge--And probably only something I would suggest doing once or a few times a month (more like an extended absorb time used to "lightly" equalize AGM batteries).

My two cents anyway.

-Bill

williaty · March 2013

Re: Are these batteries being cooked?

Here is a link to the manual for this charger.

Another document on their site claims that the Adaptive Charge program sets Absorb time equal to 5x the Bulk time. The minimum setting for the max Absorb duration is 1 hour. As far as I can see, there's no way to set anything about the Absorb timer other than the maximum.

Cariboocoot · March 2013

Re: Are these batteries being cooked?

Not impressive.
There is a mention of it being able to be configured by computer, but it does not go in to detail as to whether this gives you greater flexibility (as with the Morningstar units) or not.
The only choices for Absorb time all seem to equate to "too long for your batteries".
You do have a very small battery bank @ 100 Amp hours so you could expand it, but that doesn't excuse the lack of flexibility in Victron's design.

williaty · March 2013

Re: Are these batteries being cooked?

The most annoying thing is that one of the reasons I bought the Victron stuff is that I thought it would terminate off of ending amps. I'm not sure if I was mistaken in that belief or if one of their other products I read about DOES terminate off of EA and when I switched to deciding I wanted an I/C I just assumed it'd be the same.

The computer gives you more controls compared to the DIP switches, but that's what I've been talking about in terms of settings already.

I'm already planning to change the batteries out to a 220Ah bank. Hopefully, that'll be accomplished within the next 2-3 months.

Cariboocoot · March 2013

Re: Are these batteries being cooked?

If you crank the Absorb max time back to 1 hour does it work? Or does it still do 5X Bulk? That should be only 25 minutes if the batteries Bulk up in only 5.

williaty · March 2013

Re: Are these batteries being cooked?

Well, turns out there's a little more going on in that first run than I was aware of at the time. The BatterySafe mode was activated. When the voltage in Bulk reaches 28.8V, it starts limiting dV/dt in order to prevent something or the other. Looking at the logged data, I determined that the Bulk-Absorb transition occurred when the current started dropping. With it in BatterySafe mode, it probably considered Bulk to last longer than I did, which makes some of the numbers make sense. So I need to do another test where I disable BatterySafe and then limit the Absorb time to 1 hour and see what the current looks like at termination of Absorb. The system is new enough I'm not willing to leave it alone while it runs. I'm hoping later today after some errands I'll have the time to turn some electrons into heat and then go through another charge cycle. Hopefully, the results from that will look better and I'll post back late tonight or sometime tomorrow.

williaty · March 2013

Re: Are these batteries being cooked?

OK, I ran another discharge/charge cycle today. I used a 1500W electric heater to draw the battery bank down to around a 30% DoD and then charged it back up. The first chart is just the whole test run.

Now, on to just the charging portion. For this run, I deselected the BatterySafe profile modification, which makes it run a traditional Bulk to Absorb transition. Additionally, I reduced the maximum Absorb time to 2 hours. I found an old document from Victron claiming that their Adaptive profile sets Absorb time equal to 5 times the Bulk time. Ignore the sawtooth shapes in the current trace early in the charging, that's me messing around with the shore power limit trying to figure out how it actually worked.

Summary:
1) Charging began at 19:59
2) Bulk to Absorb transition happened at 20:27. Total 28 minutes of Bulk.
4) Charging current dropped below 1% of C at 21:41 after 80 minutes of Absorb
3) Absorb terminated at 22:27. 120 minutes of Absorb total.

So, according to Victron's old document, the 28 minutes of Bulk would have triggered 140 minutes of Absorb. The Absorb termination instead hit the max life timer at 120 minutes. However, Absorb should have terminated 40 minutes sooner, at a total of 80 minutes of Absorb.

My questions for you guys that know more than I do:
1) Should I go ahead and lower the max Absorb timer to 90 minutes?
2) This was to a 30% DoD at a draw of 0.7C. Is anything (deeper DoD, faster draw, etc) potentially going to cause the Absorb to need to be longer than what I saw in this test?

BB. · March 2013

Re: Are these batteries being cooked?

Williaty,

How will you be using the system? Daily use, emergency backup, weekend, etc.?

There is not a lot of current flowing at the end of the cycle (a good thing). Does the battery get warm? Do the battery caps get warm (typically where the catalysts are located).

It appears that the catalysts have a life time AH limit (eventually the catalyst breaks down from recombination of gases). Trying to figure out how long/how often the battery charges vs that extra long charging tail.

-Bill

williaty · March 2013

Re: Are these batteries being cooked?

The answer is somewhere between daily use and emergency backup and I'm not entirely sure yet where the balance will be. The purpose of the system is to provide continuous power to a pump. I the context of the grid going down, the system obviously qualifies as an emergency backup and will be recharged by generator and solar. However, on a daily basis, the pump draws enough inrush current that it's repeatedly popping the breaker it's attached to. So I'll be using the PowerAssist feature on the Multiplus to help get the pump started without popping the breaker all the time. I haven't yet gotten to the point where I feel comfortable enough with the system to attach it to the pump, plus I'm waiting on the last couple of parts. So I don't yet know exactly how much draw from the batteries starting the pump will cause. I don't know if it'll stay in float, go back to Absorb, or even go back to Bulk. My hope is that it'll stay in float. I really don't think it'll go back to Bulk. Time will tell.

The batteries don't even get warm enough to feel like they're above room temperature even during the 70A draw and immediate 33A recharge. I need to get some thermocouples and a display to actually monitor everything since it's too small of a difference to feel with my hand.

BB. · March 2013

Re: Are these batteries being cooked?

If the charger does not do this hours long absorb multiple times a day--Then I would not worry too much.

Is the float voltage at the battery mfg'rs recommend setting?

-Bill

williaty · March 2013

Re: Are these batteries being cooked?

Battery manufacturer recommends 2.3-2.35V/cell, it's right at 2.325V/cell. There's another feature of the charger I've been wondering about. The charger has a feature called Storage Mode. After 24 hours of un-interrupted float, it drops to 2.2V/cell (hardcoded, not adjustable). Once a week, it pops back up to the Absorb setpoint for an hour before dropping back to 2.2V/cell for another week. Would I be better off with that feature on or off?

BB. · March 2013

Re: Are these batteries being cooked?

Does not sound like a problem... AGMs can set for 2-6 months between charging--So a very low voltage float should be fine.

UPS Systems with AGM batteries tend to cook the batteries and they need replacement every one to two years--from what little I have seen.

If you are not cycling the batteries (by 10-20%) every day but letting them float for days/weeks on end--I would be more worried about over charging damaging the batteries (IMHO).

-Bill

williaty · March 2013

Re: Are these batteries being cooked?

BB. wrote: »

If you are not cycling the batteries (by 10-20%) every day but letting them float for days/weeks on end--I would be more worried about over charging damaging the batteries (IMHO).

Over-charging by simply being on float for so long, or overcharging because they've spent too long on Absorb?

BB. · March 2013

Re: Are these batteries being cooked?

I don't know--But I believe it is because they keep the charging voltages high so the batteries are at 100% capacity for the next power failure. Remember that a UPS system may run for 15-30 minutes and expect the systems to be shut down or a backup generator to be up in 5 minutes. So, if the battery is not fully charged, the system may "die" before the loads are shed/shared.

With our usual design rules of thumbs--We assume the battery is >90% for "full charge" and that "missing" stored energy is not going to be a problem. And avoiding the 100% charge every day / 24x7 100% charging, the batteries should last several times longer.

-Bill

Are these batteries being cooked?

Comments

Categories