Checking hydrometer accuracy with charged battery voltage

I adjust 1.280 SG down to 1.265 for average temps near 100. I have never heard of OEM using 1.300 except for polar applications.

At 25C 1.300 SG will eat a battery 20% faster than 1.280 acid.

Maybe I read this wrong.

mena661 wrote:

HiTech wrote:
12.78v predicted. Pretty close given likely error in hydrometer and volt meter. Even more so if it had settled slightly at 24 hours.

I know you didn't ask but I use a Victron battery monitor for voltage (does other stuff), it uses a shunt (like the Trimetric), and a Freas hydrometer which I understand to be pretty accurate.

I'm using three shunt meters as well - a watts up and two turnigies. They are not quite as good because there is 3 inches of wire between the battery and shunt vs the cool on post shunt a lot of you use, but both seem pretty accurate. I desoldered the shunt on one of my watts up with too much transient current. I am thinking of getting a larger one and remote mounting it at the battery post. They have directions on how to do this on their web site.

Jim

Boundary conditions were of course a great test. At a SG of 1.000 volts are 0 so of course this is way off as you take it out of normal operating ranges. Good approach.

I wonder if there is a better formula that actually correlates specific gravity to voltage. I assume it must be a curve rather than a linear relationship.

The paper shows charge *pretty* linear down to 2 molar acid, then it avalanches.

Ok so it's a log function.

Voltage= voltage at 1 molar acid concentration (say 11.7v) + 26mV*ln(molarity of acid)

At normal temp and for a battery where it is 11.7v being discharged down to a molarity of 1.

Jim

Ok more reading lol. I only got are far as organic chemistry. I considered it just a specialty of physics which pissed of chemists no end.

Yes I believe Freas to be the best commonly available.

Jim

On edit: ktmrfs add "battcon" to google searches for deep papers. I have found it helpful.

HiTech wrote:
12.78v predicted. Pretty close given likely error in hydrometer and volt meter. Even more so if it had settled slightly at 24 hours.

I know you didn't ask but I use a Victron battery monitor for voltage (does other stuff), it uses a shunt (like the Trimetric), and a Freas hydrometer which I understand to be pretty accurate.

HiTech wrote:
Is it true that:

Specific Gravity = Cell Open Circuit Voltage - 0.845?

Could one average the specific gravity readings across cells in a battery, and use this formula to check the accuracy of the voltmeter and/or hydrometer (based on which one is calibrated better)?

Jim

turning the equation around:

open circuit voltage=SG +0.845

Now apply this to two boundary conditions, fully charged, fully discharged.

Fully discharged SG=1.1ish
voltage = 1.945*6=11.67
even using 1.0 for discharged SG yields a voltage of 11.1V

fully charged. for trojan T-125's the "nominal" fully charged SG = 1.275.
NOTE: Various batteries have noticeable differences in initial SG. Trojan T-125 GC are 1.275, while the "renewable energy T-125" version is 1.26 or so IIRC. Some batteries are in the 1.28+ range some well below 1.275. So one needs to find out the initial SG from the mfg for your battery rather than "assuming" an initial SG.

voltage = (1.275+.845)*6= 12.7

so for the boundary conditions, the voltages are IMHO "in the ballpark" for batteries at "room temperature". graphs I've seen show fully discharged batteries near 10.8V and fully charged 12.6-12.7. However the link below gives 11.7V for the fully discharged case for the battery electrolyte concentration they give.

so, I suspect the formula is a "rule of thumb" and if you know all the specific conditions it may apply pretty accurately. I suspect at a bare minimum this "rule of thumb" applies at a specific temperature, and there is a temperature coefficient that would need to be applied.

Otherwise, it appears while the voltages calculated are certainly close, probably not close enough to "calibrate" a voltmeter or hydrometer.

so some searching and I found this paper on battery chemistry which expands on the above equation.

lead acid battery chemistry

but then, it's written at a college level chemistry class so see what you remember!

Anyway, I'll trust my Freis (sp?) hydrometer and Tek TX3 true RMS DMM which is calibrated with NIST traceability rather than the formula.

It's not clear to me if you temp correct for voltage. No mention. I think SG would have to be. I imagine the formula works in some band of conditions. No mention of how wide. It's interesting in the implications for the rest voltage for some batteries. If you knew the batteries very well and the voltage came up on an accurate voltmeter to exactly what you expect, odds are good SGs in every cell are close together and close to nominal.

Once the voltage is a bit low, now the mystery gets geometrically deeper faster without a hydrometer.

Jim

My example may not take into account temp correction on SG and voltage so ignore.

12.78v predicted. Pretty close given likely error in hydrometer and volt meter. Even more so if it had settled slightly at 24 hours.

BFL what kind of voltmeter and hydrometer do you use?

Jim

I've had a 12.83V resting voltage (12 hrs) with a 1.285 SG (97F/36C).