Question about a Mixed Technology Battery Bank

[nhsnm]

I have a question that I thought I would present to you all.

I currently have a 1.5KW Panel Array, feeding a Classic 150, with a 48V Flooded Lead Acid battery bank and a Magnum inverter.

I have found myself with an interesting opportunity in that I am able to pickup a very large AGM battery bank for free (I have to pay a core fee of $200 for them).  These batteries have been used to support a 75KV UPS for a data center and are about 2.5 years old. 

While these "new" batteries are more than enough to support my current system (even with planned upgrades) I would like to keep both banks active.  I am planning on doubling the size of my panel array so I was pondering the design for this and would like a bit of advice.

Currently my thought was to buy a new Charge Controller (probably a 250) to maintain the new AGM bank.  With both battery banks in parallel for the input to the inverter.  But I am concerned that the new bank has significant higher capacity and am thinking that this might have issues.  An alternate solution was to place the Flooded batteries into a "standby" mode and keep them isolated from the supply or even giving them another inverter and manage both systems completely separate. 

So with that all said.  Anyone have any advice, thoughts, comments or concerns?

[Westbranch]

Keep them separate and with individual CC's. 
The 'best' charge parameters for these 2 types are too divergent . 
AGMs need very specific charger settings vs FLA, which can be much more tolerant of non optimal settings... You could kill your AGMs very quickly.

[vtmaps]

Currently my thought was to buy a new Charge Controller (probably a 250) to maintain the new AGM bank.  With both battery banks in parallel for the input to the inverter.

If both batteries are in parallel then both controllers are in parallel. As westbranch mentioned, that's not a good idea.   Why would you buy a Classic 250 (rather than a Classic 150)?

--vtMaps

[nhsnm]

Thank you for the input.  Pretty much confirming what I was thinking.  My concern is that I am using the Magnum BMK battery monitor automatically switching the load to Line voltage when my batteries reach 80% charge and switching back to battery / solar at 100% charge.  So if I have bank 1 FLA at 330AH and Bank 2 at 527AH AGM the BMK will have trouble identifying the 80% charge state. 


I was thinking the 250 so I could run the voltage levels up dropping the current therefore reducing line loss.  To be honest I really have not put a lot of thought into what CC to get next.  It was pretty much off the cuff. 

As a small update the "new" battery bank is 30 - 5 year old 12v 88AH AGM batteries that have only been in standby mode for a Data center UPS.  They were all tested a couple of months ago and tested good with a standby life span of 10 years.  The bonus is that I only have to pay the core charge on the batteries and the enclosure, battery cables, breaker and inverter cables are all free.  This effectively triples the size of my battery bank.  Now if I could just figure out how to move the 75KVA ups I would be totally set.  LOL

[Westbranch]

75KVA ups

A Liebert?

What voltage do you intend?  Those are pretty small Ah batteries so @ 48V you could use 12 @ 3x4 for a stable chargeable bank. Now what to do with the other 18?

[nhsnm]

Yes a leibert 75KVA ups.  If only I had a fork truck.

I know they are pretty small but since they are free I figure I could just build a bank of 7 parallel strings of 4 in series.  Leaving me with 2 extra.  I know that there are certain flaws with this design and the purist in me is screaming right now.  But since in the end I will have a 527 AH bank with all associated hardware enclosure etc for just $200 it is worth it.

Anyone want a bank of 8  1 year old 6 volt 330AH FLA batteries?  Just kidding.  I think I will keep my FLAs

[mike90045]

What is the warranty lifetime of those AGM batteries.   I'd thought ~7 years was the limit for AGM.   Lot of work for just 2 years of marginal life.

[nhsnm]

The warranty on these batteries is 10 Years standby (which is all they have ever been other than for an occasional 1 - 2 minute run for testing) or 266 100% discharges.  So given that I am assuming a 3 - 5 year effective life. 

In addition in reality 24 of the batteries are only 2 years old but I am basing the age of the bank on the oldest batteries (which I might not even use).

[dgd]

The warranty on these batteries is 10 Years standby (which is all they have ever been other than for an occasional 1 - 2 minute run for testing) or 266 100% discharges.  So given that I am assuming a 3 - 5 year effective life. 

In addition in reality 24 of the batteries are only 2 years old but I am basing the age of the bank on the oldest batteries (which I might not even use).

10 year standby life does not really work as a life measure for batteries in a working RE system.
Expecting  a 3-5 year effective life may be very optimistic considering that you want to configure them as seven parallel strings of 4 in series. Once they get into a decent charge/discharge cycle for a few months I think things could get very unbalanced leading to damaged cells.
Then I can imagine you spending significant time constantly measuring voltages and monitoring currents, ok if you enjoy this but otherwise a discouraging use of time..
So I would suggest saving your $200 else try to make a profit on selling them 

dgd

[nhsnm]

I hear you on that.  However I figure the worst case scenario is that I put them under load and they start having issues.  In that case I will still have a $5k battery enclosure and a good  number of batteries for cores.  Best case I get a few years our of them.  Either way I am not going to loose any money (unless you count the potential profit of selling them).  I am more interested in the experiment and seeing what happens.  (I love to tinker with high voltages lol)

[Westbranch]

OK then I feel you  should charge them all to 48v and then first build a bank of 3 strings of 48v and learn on/from them.  When you kill them via experiment you can replace them and try again.  I sort of did that with a dieing battery at the start. I did learn a lot \i would not have otherwise...

[nhsnm]

I really like that idea.  ESP since a few are older. I can expirement with them then pull them out and leave the 2 year olds for production.  Nice idea.