LFP against VRLA at a remote site: the arithmetic that actually decides it

A battery string in a remote cabinet has reached the end of its service life. Someone has to decide whether to put the same chemistry back in or change technology. The lead-acid quote is lower. That is usually where the analysis stops, and that is the mistake.

The decision is not a price comparison. It is a comparison of what the site costs over the period you intend to keep it, and the dominant term in that expression is frequently not the battery.

Write down the actual cost of a site, not the cost of a battery

For a site you plan to hold for ten years, the honest expression is:

Total = (battery cost x number of installations) + (site visit cost x number of visits) + the cost of the outages you take while a degraded string is in service

That middle term is the one that decides most remote-site arguments. A truck roll to an accessible urban cabinet is a scheduled task. A truck roll to a mountaintop site with a locked access road, a weather window, a two-person requirement, and a four-hour drive is a different category of expense entirely, and it recurs every time the string is replaced.

So the question becomes: how many times will you be on that site?

Service life is not a number on a datasheet, it is a function of temperature

VRLA in float service does not fail on a schedule. It fails as a function of temperature, and the relationship is steep. The widely used engineering rule is that VRLA service life roughly halves for every 10 degrees C of sustained temperature above 25 degrees C. That is a rule of thumb, not a law, but it is directionally reliable and it is why two identical strings in two identical cabinets can differ by years.

Which means the specification that matters is not the battery. It is the cabinet temperature profile over a year, and most operators do not have it. If you have thermal logging on the site, that data is worth more to this decision than any vendor comparison table. If you do not have it, that is the first thing to fix, because without it you are estimating the single variable the answer is most sensitive to.

LFP is not immune to temperature, but its calendar and cycle degradation curves are considerably flatter across the range a telecom cabinet actually sees. The practical consequence is fewer replacement events, which is the term that was dominating the expression.

Usable capacity and the number you are actually buying

Two strings with the same nameplate amp-hours are not the same amount of backup.

• A lead-acid string in cycling service is conventionally planned around a limited depth of discharge to protect its life. You do not get to spend all of the nameplate.
• Lead-acid capacity falls off sharply as the discharge rate rises. A capacity figure quoted at an 8 or 10 hour rate substantially overstates what is available during a short, high-current backup event. This is the Peukert effect and it catches people who compare headline numbers.
• LFP holds a much flatter voltage and delivers a far larger fraction of nameplate at higher rates, and tolerates deeper discharge.

So before comparing prices, convert both options to usable amp-hours at your actual discharge rate, at your actual cabinet temperature, at end of life. Comparing nameplate to nameplate systematically favours lead-acid by a margin that does not exist in the field.

The four cases where the like-for-like swap is still the right answer

This is not an argument that lithium always wins. It does not. The swap is correct when:

1. The site has a known end date. If the site is scheduled for decommissioning or a full technology refresh inside the next few years, you will never reach the replacement event that justifies the lithium premium. Buy the cheap string.
2. The plant cannot charge lithium correctly. A rectifier or charge controller configured for a lead-acid float regime is not automatically able to deliver the constant-current, constant-voltage profile LFP needs. Many plants can be reconfigured. Some cannot, and the cost of replacing the power system belongs in the lithium column.
3. The site gets cold and there is no thermal provision. LFP must not be charged below freezing without heating, or you cause lithium plating and permanent damage. A northern site with no cabinet heater and no low-temperature charge management is a real constraint, not a detail to resolve later.
4. There is no way to use the BMS. A lithium battery reports state and faults over a communications bus. If nothing at that site can read it and nothing will alarm on it, you have given up a large part of what you paid for and taken on a failure mode you cannot see.

If none of those four apply, and the site is expensive to reach, the arithmetic usually favours the technology change — and it usually favours it by a margin that is not close.

What to do before you ask anyone for a quote

Gather four things. They determine the answer far more than any supplier comparison will:

• The load, in amps at the system voltage, including everything in the cabinet rather than just the radio equipment.
• The required autonomy, and where that requirement comes from — a generator start time, an SLA, or the realistic worst-case time to get a person on site.
• The annual temperature profile inside the enclosure, not the ambient outside it.
• The fully loaded cost of one site visit, including labour, travel, access, and any escort or permit requirement.

With those four numbers the comparison takes about twenty minutes and produces a defensible answer. Without them, any recommendation from any vendor — including us — is a guess wearing a suit.

If you have the four numbers and want a second pair of eyes on the arithmetic, that is a short phone call, and it does not require you to be considering a purchase.