TL;DR - The whole test in ten minutes
A specific gravity test reveals which cell inside your battery is failing. You use a $15 hydrometer from an auto parts store. You squeeze the bulb, draw up a little fluid from one cell, read the number on the float, write it down. Repeat for every cell. Healthy cells read 1.265 to 1.285. One dramatically lower than the others means that cell is dead. Replace the battery containing the dead cell. The test only works on flooded lead-acid batteries with caps you can unscrew. Run it once a quarter. The reason this matters: a single dead cell can hide behind the average voltage of the bank for months. The hydrometer finds it in ten minutes.
He'd been checking voltage every Sunday for four years. Six flooded batteries in his shed, 24-volt bank, all reading right where they should. Then one Sunday in February the inverter started cutting out during breakfast. Coffee maker on, microwave on, toaster on, alarm. He pulled the cover and put his meter on each battery. All six read normal. Nothing visibly wrong. He almost ordered a new $1,400 inverter that afternoon. Then his neighbor showed up with a $14 hydrometer from O'Reilly Auto Parts. They tested every cell. Five batteries came back tight, healthy numbers across every cell. Battery six came back with five normal cells and one that read 1.085 - basically dead. The other five cells in that battery were carrying the dead cell's load until the moment a heavy draw demanded more than they could give, at which point the whole bank voltage dropped and the inverter alarmed. He replaced that one battery for $185. The bank ran clean for three more years. The voltmeter had been telling him everything was fine while one dead cell hid behind the average voltage of the other twenty-three cells. The hydrometer cost less than a tank of gas and saved him from spending $1,400 on an inverter he didn't need.
Who this is for
This guide is for the Vermont retiree who's owned the same flooded battery bank for four years and has never tested a cell because nobody told him he was supposed to. The Florida coastal family whose humidity wears out batteries faster than the manual says it should. The Arizona rancher whose summer heat boils water out of the cells and exposes the lead plates if nobody's watching. The Maine cabin owner whose batteries spend winter in an unheated shed and seem to "die" every January. The Texas off-gridder who watched his neighbor throw out a $2,000 battery bank that turned out to have one bad cell and could have been saved for $185. The Colorado homesteader who bought a used system from someone who swore the batteries were "in good shape" with no paperwork to back it up. The Pennsylvania DIYer who inherited an off-grid system from the previous owner and has no idea what shape the batteries are in. The Idaho first-time off-gridder who bought flooded lead-acid because lithium was out of budget and now needs to know how to keep them alive.
Three things are true of every one of them.
The voltmeter is showing them an average, not the truth.
A single dead cell can hide for months before they see it.
A $15 tool from the auto parts store finds problems that thousand-dollar meters miss.
Why your voltmeter isn't telling you the whole story
Open up a 12-volt lead-acid battery and you'll find six smaller cells wired together inside, like six small batteries acting as one. Each cell produces a bit more than 2 volts. Add them up: 12.6 volts at full charge. Your voltmeter measures all six cells at once, not each one separately. That sounds fine until one of those six cells gets weak.
When one cell starts dying, the other five work harder to keep the total voltage near normal. Your voltmeter still reads 12.6 volts. Everything looks fine. But under heavy load the dying cell can't keep up, and the whole bank suddenly drops voltage and your inverter alarms. By the time the voltmeter catches the problem, the cell is dead and the other cells have been overworked for months.
The specific gravity test is the only way to see what's happening inside each individual cell. You're not measuring voltage. You're measuring how strong the acid is inside the cell, which tells you whether that cell is still doing its job.
Strong acid means the cell is healthy. Weak acid means the cell is in trouble. The hydrometer measures acid strength directly. Numbers go up to about 1.285 for a healthy fully-charged cell. Numbers below 1.200 mean the cell is sick. Numbers below 1.100 mean the cell is dead.
According to U.S. Department of Energy research on battery health, this kind of cell-level testing catches problems months before voltage testing can. You don't need a science degree to do it. You need a $15 tool and ten minutes.
One catch: this test only works on flooded lead-acid batteries - the kind with caps on top you can unscrew. Sealed batteries (AGM, gel, lithium) don't let you in to the fluid, so this test doesn't work on them. Different test methods for those.
WATTSON'S HYDROMETER TRUTH: First time the instructor at the US Solar Institute showed me a hydrometer I thought he was joking. Glass bulb with a colored ball that floats. Looked like something from a hundred years ago. Then he set up two batteries on the bench. Both reading 12.6 volts on the meter. Both looked identical. He tested every cell on the first one - all six cells came back tight, healthy numbers. Then the second battery. Five normal cells. One cell at 1.105. That cell was effectively dead, and the voltmeter never saw it. I'd been off-grid six years at that point and nobody had ever told me about this simple test. The glass bulb saw what every digital meter I owned couldn't see. Lithium people skip this whole topic because their batteries don't work this way. Flooded lead-acid people who skip it are the ones I see replacing $2,000 banks when $185 would have fixed it.
Find the dying cell before it kills your whole bank.
Wattson's Hydrometer Reading Card - the simple chart you tape next to your battery shed. Numbers in plain English. What's healthy, what's dying, what to do about it.
GET THE READING CARD ->How to run the specific gravity test
The test is dead simple. Reading the numbers is where it gets useful.
What you'll need
- A bulb-style hydrometer from any auto parts store. $10-15. They usually come in a plastic case with a thermometer built in.
- Safety glasses (the battery fluid is acid - splashes burn skin and eyes)
- Gloves (rubber dish gloves work in a pinch, electrical-rated gloves are better)
- A small open box of baking soda and a spray bottle of water for any spills
- A notebook and pen
Total cost to set up: under $20.
Before you start
Three things have to be true or your readings will be wrong.
-
The batteries need to be fully charged. Run a normal charge cycle and let the bank rest at least 30 minutes after the charger stops. If you test partially charged batteries, every cell reads low and you learn nothing about cell health.
-
The fluid level inside the cells needs to be normal. Open a cap and look in. You should see fluid covering the lead plates by about a quarter to half an inch. If plates are showing or barely wet, top off with distilled water (never tap water, never acid), run a full charge cycle, then test the next day.
-
The batteries need to be at normal room temperature, roughly 60-90 degrees F. Cold batteries give artificially high readings. Hot batteries give artificially low readings. If your battery shed is freezing or 100+, wait for a better day.
The actual test
- Turn off the system. Open the disconnect between the batteries and the inverter. Open the disconnect between the charge controller and the batteries.
- Put on safety glasses and gloves.
- Unscrew or pry off the cap on the first cell. Most flooded batteries have six caps per battery, one for each cell.
- Squeeze the rubber bulb on top of the hydrometer to push the air out.
- Stick the tip into the cell, into the fluid.
- Slowly release the bulb. Fluid gets pulled up into the glass chamber.
- Hold the hydrometer level at eye height. Look at where the colored ball or float is sitting. The number on the side of the float is your reading.
- Write it down. Battery number, cell number, the reading.
- Squeeze the bulb again to put the fluid back into the same cell. Never move fluid between cells.
- Wipe the hydrometer tip with a clean rag. Move to the next cell.
- Repeat until you've tested every cell in every battery.
A six-battery 24-volt bank has 36 cells. The whole test takes 15-20 minutes once you get into the rhythm.
Close every cap when you finish. Wipe up any spills with the baking soda and water.
Reading the numbers
Here's what your numbers mean.
| Reading | What it means | What to do |
|---|---|---|
| 1.275 to 1.295 | Healthy cell, fully charged (top-of-line battery) | Nothing. Re-test in 3 months. |
| 1.265 to 1.275 | Healthy cell, fully charged (standard battery) | Nothing. Re-test in 3 months. |
| 1.225 to 1.265 | Cell isn't fully charged | Charge longer, retest |
| 1.180 to 1.225 | Cell is sick. Could be sulfation. | Try equalization charge, retest in 48 hours |
| 1.120 to 1.180 | Cell is in trouble. | Battery is near end of life. Plan replacement. |
| Below 1.120 | Dead cell | Replace this battery before it damages the others. |
A healthy bank shows readings all clumped together in the 1.265-1.285 range. That's it. That's the whole game.
The single most important pattern in a specific gravity test
The number that matters more than any individual reading is the difference between your highest and lowest cell in the same bank.
If all cells are within 0.030 of each other (say, 1.270 on the lowest and 1.285 on the highest) the bank is healthy. Cells are working together.
If the difference is 0.030 to 0.050 something is starting to go wrong. Probably one cell sulfating slowly. Run an equalization charge (more on that below) and retest in 48 hours.
If the difference is 0.050 or more you have at least one cell failing. Find which cell it is, look at the reading, decide if it's the kind you can save or the kind you can't.
This is the pattern that catches the dying cell early. Voltage testing can't see this. Specific gravity testing can.
What an "equalization charge" is and when to use it
Equalization is a fancy word for "overcharge the batteries on purpose for a few hours to fix two specific problems."
Problem 1: Stratification. The heavy acid settles to the bottom of each cell over time. The water sits on top. The battery starts giving weird readings because the fluid isn't mixed. Equalization mixes it back up by causing gentle bubbling inside the cell.
Problem 2: Mild sulfation. White crystals build up on the lead plates over months and reduce how much surface is available for the chemistry. Equalization at the right voltage can break some of those crystals back down.
To run an equalization charge: most charge controllers have an "equalize" mode you can trigger from the front panel or app. It runs the bank at about 15.5 volts (for a 12-volt system) for 4 to 8 hours. The charge controller manages the time and voltage - you don't have to babysit it.
Important: equalization only works on flooded lead-acid. Never equalize AGM, gel, or lithium batteries. It will damage them or destroy them outright.
After equalization, wait 30 minutes for the cells to rest, then run another specific gravity test. If the readings tightened up, the bank recovered. If readings are still low or variance is still high, the cells are too far gone and the bank is at end of life.
When to run the specific gravity test
- New battery bank: at 30 days, 90 days, and 180 days. Builds your baseline.
- Healthy bank, years 1 to 3: once a quarter, same day each season.
- Older bank, year 3 and up: once a month during fall and winter, when problems tend to show up.
- After any time the bank got drained below 30% charge - test within two days of recharging.
- After any equalization charge - confirms whether the equalization actually helped.
- Before winter if you're leaving the cabin - know the bank is healthy before you walk away.
- Any time the bank acts weird - voltage drops you can't explain, inverter alarming, batteries running warm.
Quarterly is the minimum. Monthly is best for older banks.
When to stop and call a pro
Some things you find during a battery test mean stop now, call somebody who does this for a living.
- You smell rotten eggs near the batteries (hydrogen sulfide - serious fire risk)
- You see active bubbling or boiling in any cell when nothing is charging
- A battery case is bulging, leaking, or shaped wrong
- The fluid is dark brown or black, or has visible chunks in it
- One cell reads below 1.000 (lower than plain water - something serious is wrong)
- You break the hydrometer and get acid on your skin (rinse immediately with water, then a baking soda paste, then more water - get medical attention if your skin keeps burning)
- You feel any kind of tingle when you touch the metal parts of the battery setup
- One battery is much warmer than the others (something is shorted inside it)
Any of those, you're past routine maintenance. Open the main disconnect, ventilate the space, call a licensed solar electrician or a US Solar Institute trained tech.
Frequently asked questions
Can I do this test on AGM or lithium batteries? No. The test only works on flooded lead-acid batteries with caps you can unscrew. AGM, gel, and lithium are sealed. For those, voltage under load testing and capacity testing are the right diagnostics. See battery voltage drop diagnosis for those.
How do I tell if my batteries are flooded or sealed? Look at the top of the battery. If you see round caps you can unscrew or pry off, it's flooded. If the top is completely smooth with just terminals sticking up, it's sealed. Most batteries also have the type printed on the side - look for words like "flooded," "wet cell," "AGM," "gel," or "lithium."
The fluid in my batteries is low. Can I just add water? Only distilled water. Never tap water (minerals damage the cells). Never battery acid (changes the chemistry). Top off until the fluid covers the plates by a quarter to half an inch. Run a full charge cycle before you test specific gravity - otherwise your readings will be wrong because the new water hasn't mixed with the acid yet.
Can I bring a sulfated battery back to life? Sometimes. Mild sulfation responds to one equalization charge. Severe sulfation is permanent. The test is simple: run an equalization, wait 48 hours, retest. If the numbers came back up, you saved the battery. If they didn't, it's done.
Should I run a specific gravity test on a used battery before I buy it? Yes. It's the cheapest pre-purchase check you can do. Bring a hydrometer to the seller. A healthy used battery should test 1.265 or higher on every cell with the cells within 0.030 of each other. Anything else and you're buying someone else's problem. Walk away if any cell reads below 1.225.
What if just one cell is bad in a battery? Do I have to replace the whole battery? Yes. You can't replace individual cells in a sealed battery case. Replace the whole battery. And in a series-wired bank, replace it soon - one bad battery slowly damages the batteries next to it. The dead cell drives up resistance and forces the neighbors to work harder.
Can I mix a new battery with my old ones to make the bank last longer? No. The old batteries pull the new battery down to their level within months. You'll just kill the new battery early. If the bank is dying, replace the whole thing at once. If you need more capacity, add a separate parallel string of all-new matched batteries.
Why does my reading drop in winter? Cold batteries give artificially high readings on the meter and artificially low readings on the hydrometer. The chemistry is just slower in the cold. Quality hydrometers have a temperature correction chart printed on the side. Use it. Or wait for warmer weather to test.
Is there a better tool than a $15 hydrometer? Yes. Optical refractometers (the kind you look through like a tiny telescope) cost $40-80 and are more accurate, easier to read, and don't break as easily. Same principle, fancier tool. For a small bank the cheap hydrometer is fine. For a big bank or a serious off-grid system, the refractometer is worth the upgrade.
Is the specific gravity test dangerous? Moderate risk. The fluid inside the batteries is sulfuric acid. It burns skin, eats clothes, and damages eyes if it splashes. The off-gassing during testing can produce small amounts of hydrogen, which is flammable. Wear gloves and safety glasses, ventilate the space, no open flames nearby, and keep baking soda water handy for spills. Done right it's no more dangerous than checking the oil in your car.
My readings are all over the place but the system seems to work fine. Should I worry? Yes. Erratic readings on a battery that still works mean the cells are going downhill at different rates. The variance is the warning sign before the failure. Run an equalization charge, wait 48 hours, retest. If readings tightened up, you caught stratification in time. If they're still scattered, the bank is past its useful life regardless of what the voltmeter says today.
Conclusion
The specific gravity test is the most undervalued tool in off-grid solar. Fifteen dollars at the auto parts store. Ten minutes per bank. Catches problems your voltmeter can't see for months.
A healthy bank shows readings clumped tight between 1.265 and 1.285 on every cell. One cell sticking out low means that cell is dying. Replace the battery containing it before the others get damaged carrying its load.
Voltage testing is your dashboard light. The specific gravity test is your mechanic looking under the hood. You need both. The voltmeter tells you the average. The hydrometer tells you which specific cell is the problem.
Run the test once a quarter. Write the numbers in a notebook. Compare quarter to quarter. The patterns will show you which batteries are healthy, which are going downhill, and which are about to fail.
The complete Maintenance and Troubleshooting guide ->