Lithium Iron Phosphate (LiFePO4) vs Lead Acid Batteries - Industrial Application & Consideration 1

[Instrument1352]

Hi,
we are replacing an old 30KVAh UPS system for an offshore oil installation. We have been given options of:
- Lithium Iron Phosphate (LiFePO4) Batteries; as well as
- Valve Regulated Lead Acid Batteries - maintenance free (the same as existing ones)
I am aware of the advantages of the LiFePO4 (as published online or on paper) but would like to know first hand experiences using them or probably the reason one has decided not to use them!

My concerns about LiFePO4 batteries would be:

1- Any special charging requirements. Will a typical industrial UPS (e.g. Cutler Hammer, Gutor, APM, etc.)
2- Safety and reliability. I know they are much more reliable than Lithium Ion batteries but will they be more reliable than Lead Acid batteries? In particular a risk of fire cannot be tolerated in our application.

Appreciate if you could share your experiences and thoughts on this.

 

[ScottyUK]

I wouldn't pick VRLA for a critical UPS battery. I'm sorry I can't comment on LiPo's in this application.

Any reason why you've disregarded the flooded cell technologies? The newer recombination cells require less frequent topping up than plain vented cells, but retain most of the benefits of reliability and long life and have low rates of hydrogen evolution.

 

[itsmoked]

I gotta disagree with Scotty on this one.
We use VLRA batteries exclusively with UPSs. They work great with no complaints. Use quality ones of course, Trojan or something similar. Stay far away from Interstate.

I would use flooded only on a huge application like substations or power plants. Skip all the equalizing hassles and electrolyte spills and thrills the end result being shortened lifespan unless there's someone actually tasked with battery maintenance.

As for LiFePO4 there's no question they're good batteries. Safest of the Li family. They cost a hefty premium over VLRA batteries. They have a whole lot more discharge cycles in them but in an emergency UPS is that even a factor? How many times in 5 years does the UPS dip below 50% battery capacity? Once? Never? If you use the UPS frequently then I'd use LiFePO4. If you use it only in emergencies, rarely, I'd stick with the VLRA.

Do keep in mind that you'd want the fast discharge LiFePO4 variety for UPS service as compared to the slow discharge variety.

Yes, you'd want a new charger that is LiFePO4 savvy. Also remember that you can total LiFePO4 batteries by over-discharging them once whereas VLRA batteries hate being discharged too much too, but short of loosing some capacity will still function. You need to make sure the UPS cuts-off with LiFePO4 at somewhere above 2.5V/cell.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[FreddyNurk]

For various reasons, all of the batteries that I'd seen in DC Telecoms (i.e. -48V) backup applications in recent years had been Lead Acid AGM batteries. I wouldn't have a problem in using AGMs in similar backup applications. I believe that a fair amount of the substation backup batteries were also AGM, flooded cells were not that common.

There is obviously the caveat (although it applies to flooded as well) that up to a certain size, typical makes of UPS units are not nice to batteries. All the DC Telecoms stuff I saw generally included a low voltage cut out, but were also sized to allow an appropriate amount of backup prior to cutting the batteries off. I don't know if the equivalent UPS gear has similar provisions.

 

[Instrument1352]

Thanks Scotty and Freddy.
@itsmoked Keith,thanks heaps for your concise and to the point reply. I really liked it.
Cost is not much of a problem (Odd,I know!), and space is limited, so Watt per cubic inch matters.
Site preference is to have single string of lower voltage, say 120vdc rather than something like 400vdc.
The discharge would be at rare scenario of platform blackout and we need the well-head controls powered for 6 to 10 hrs.
My heart is still with good old lead acid batteries, yet kind of open to newer technologies such as LiFePO4 batteries, should they not add to the risk.
I was looking into deep cycle AGM batteries in particular.
Any comments anyone?

 

[ScottyUK]

Hi Keith,

My experience with good quality VRLA's (Yuasa brand) on a UPS application saw about three years life before we started seeing weak cells in the string. Plante cells usually survive 10+ years, and flooded Nicad's should see 25+ years. If you maintain them then flooded cells will outlast VRLA, but there's a maintenance cost and labour overhead to look after them.

 

[FreddyNurk]

I am aware of at least one maritime installation where LiFePO4 were chosen over the previous traction lead acid batteries due to weight and capacity issues. As far as I'm aware they haven't had any significant issues with them, although that was a 24V system, rather than 400V.

Having said that, that same installation used to regularly kill AGM batteries due to overcharging, which is how they ended up with traction batteries.

In your application I'd be checking with both the battery manufacturer and the UPS manufacturer for reassurances on compatibility before considering Lithium.

 

[Comcokid]

LiFePO4 batteries are appearing in aircraft, especially in helicopters where weight is a premium. These batteries have built-in all the charge and discharge protection circuits needed for the cells, including short-circuit protections. These batteries are currently 28V (same as what is called 24V for trucks and marine). In addition to weight and size savings, LiFePO4 have very low internal impedance which means piston engines and turbine engines start faster from these batteries as compared to lead-acid.

The next step for LiFePO4 in aircraft will probably be a 270VDC battery system to follow what Boeing did for the electrical distribution in the Dreamliner. The Dreamliner still uses 28V batteries to start the engines or APU, but the chemistry of this battery is lithium-cobalt.

 

[itsmoked]

Howdy Scotty! I totally agree with that.

However I'm talking bigger than Yuasa's I ever see.

VRLA AGMs
Like this.


And, as Freddy points out no battery will last if charged by uncaring doofuses. The charging must be prefect or you're wasting boatloads of money.

Instrument1352; If you really are looking to reduce space than you probably should consider LiFePO4.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottyUK]

These are the biggest VRLA's I've seen - over 3000AH at the C10 rate.

https://www.hoppecke.com/fileadmin/...ucts/Downloads/grid_power_vrl_brochure_en.pdf

I'd still pick flooded cells over VRLA, but I agree that incorrect charging and poor maintenance will ruin even the best cells.

 

[itsmoked]

Those are pretty dang large Scotty!

I have just the chargers for them too! Here's a shot of five of the nine 1,200amp chargers.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[FreddyNurk]

The fault current from the battery banks in those installations can get somewhat concerning.

EDMS Australia

 

[ScottyUK]

Yeah, big batteries are scary - they can throw an awful lot of energy into a fault with nothing able to interrupt the current.


Keith - that's a lot of rectifier capacity!

 

[itsmoked]

Place had six battery banks in a basketball size room. They were about 15feet by 8 feet banks. I noticed yesterday that they'd been cabled up to two busbars that are 4" x 1/2".

Random ground cable hanging from the ceiling:

Was going to connect to this bus bar before job went still-born.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Instrument1352]

@itsmoked @Scotty
Thanks for your insight. We selected VRLA batteries. Now there is a second question.

Is there a concern if we split the batteries into two parallel battery strings, each with it's own circuit breaker / isolator?
Both will be connected with a common cable to the UPS.

I'm thinking of the advsbtages, such as easier testing, easier maintenance, probably a short circuit in a single string or when one battery in one string becomes faulty and open circuit will not make the whole system fail. That is more availability, reliability and easier maintenance.

I'm concerned if there is a draw back, e.g. if the charging is not balanced between the two strings and the UPS may not even know about it. Or probably any other drawback.

The whole battery bank will have a capacity of about 250KVAh.

Appreciate if you could share your thoughts on this.

 

[itsmoked]

Ah.. the bean counters got wind of LiFePO4 pricing I see.

Questions:
What are the string voltages going to be?

Can you use two chargers?

What would your scheme be to utilize both banks?

Need to think about this.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[Instrument1352]

Partially the price and the other is the UPS supplier didn't support this and also lacking other approvals we need in place for offshore use.
The battery string voltage is 120v, single charger and we need both battery strings to have the desired autonomy time. I'm thinking if we lose one string only the autonomy tone will reduce.

The other concern is if the voltage of two strings differ, even slightly, one will take all the charging current and the other one will not be charged and then when we need them both, will not have enough charge.

 

[ScottyUK]

Dual strings are fairly common so one bad cell doesn't take down the entire battery. Just don't mix cell types between the strings.

 

[itsmoked]

Scotty how do they solve the riddle of one drained string and trying to switch to the remaining string. They can't ever be connected as the charged one will uncontrollably discharge into the drained one. You can't switch between strings without interruption.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Instrument1352]

Great point Keith,

Other than this, say on the day one, when we want to connect the two strings in parallel; if the two string have a slight voltage difference (which is very likely), there will be a considerable current from one to the other(uncontrolled as you mentioned). That would be also a concern when one string has been replaced for a maintenance and wants to be parallel again.
I am not sure, but do you think a dummy load resistor could help? First the two will be connected via a load resistor and after a while, when the voltages are closer and the current through the load resistor has dropped below a certain valve, then it can be shunt via the tie-in/coupler circuit breaker.