Battery-Charging Voltages

[Rick]

Having spent 30 years working with batteries, designing charging systems and battery monitors AND referring to electrochemistry textbooks and technical papers regarding lead-acid batteries (yes, AGM's of all types and gel-cell are still lead-acid) it is evident that there is NO set battery charged voltage that is "correct" with the exception of float voltage (specified at a specific temperature).

Having first-hand experience working with Xantrex, Heart Interface, Trace Engineering and Cruising Equipment I can tell you that NONE of the "charts" recommending charge voltages came from anyone other than marketing personell, NOT from electrochemists who design batteries. The marketing people are forced to attempt to give uninformed users some relevant form of recommendations in order to sell their products.

The basic electrochemistry for ALL lead-acid batteries is essentially the same as far as users are concerned. To be sure there are different chemicals used in the plate "pastes" as well as the plate lead make-up. The formula describing the reversability of sulpheric acid, lead sulphate and lead oxide is the same for all as far as energy transfer (charge and discharge) is concerned. The fact that what is called half-cell equations differ slightly is of no concern as far as how one might charge the battery (at least for the batteries applicability to deep-discharge and high rates of current charge/discharge is concerned for cruising).

MY past posts describe just how fast one can charge a battery and there are various discussions regarding the so-called "Amp-hour law" taken from electochemistry texts. The three-step charging algorithm is a first appoximation to the Amp-hour charging method and CAN be exceeded.

There are VRLA (valve-regulated lead-acid) batteries used for things light emegency lighting that are not designed with sufficiently large plate interconnects to be used for high rates of charge/discharge and are, therefore, limited to charge rates...not applicable to us here where we use batteries designed with large plate interconnects.

None of the charger or inverter/charger marketing information (that I have read) properly describes just what constitutes a valid equalization cycle. Equilization is done with a constant-current source that terminates with time qualified by observation of the temperature, cell voltage and charge acceptance. NO manufacturers of chargers have that capability built in and, therefore the marketing people must compromise with a half-baked method which has no where near the efficacy of the proper method. Again, any time that you read what is an "equalization voltage" there is no such thing in the rigorous sense of equalization.

Here is a short description of charging: Float voltage is determined by the designed acid concentration and temperature period. There is a simple formula describing the at-rest cell voltage as a function of the specific gravity of the electrolyte. ALL AGM batteries follow this as well. With ANY lead-acid battery of any type use for deep-discharge high current applications You can safely charge the battery without excessive temperature rise or gassing by applying a charge current equal to (or slightly greater than..following my experience) the value of Amp-hours "missing" from the battery. This concept cuts through all of the crap regarding charge voltage because the voltage is always different as the battery comes up to full.

Now when the three-step method was first promulgated by Ample Power and then by Cruising Equipment and then implemented by Heart Interface (with their inverter/chargers) it was with the realization that simple voltage regulation in steps could approximate the Amp-hour law. We all wished early on that it was affordable to create a charger to exactly follow the Amp-hour law yet knew that a battery monitor would have to be married both to the battery as well as the charger to do that.

Further research in the lab and in the field revealed that ALL of the battery types would respond remarkably well using a true Amp-hour law charging regimen (I did this for years) and that many of the AGM batteries rejected by many users could be brought back from the dead by using a true equalization method. One REMARKABLE result of doing this is that if subsequent charge/discharge cycles were done using Amp-hour law charging they never needed equalization again as long as they were not left to long in a state of discharge before charging.

One observation that I have made is that the beginning charge voltage of the Amp-hour law charger would always be higher than 14.4V and usually below 15Volts. Does that give anyone a clue? It makes no difference weather one is using an AGM with a specific gravity requiring a float voltage of 13.2 Volts or one requiring a float voltgage of 13.9 or 14.0 Volts. Gel-cell batteries love this as well and tended to "look" more like the bettery quality AGM's like Fullriver today. Again, do not confuse float voltage with upper limit acceptance voltages that would keep charge current within the limit of the number of amp-hours missing from the battery as stated.

Hope that this helps cut through the crap regarding all the different acceptance or equalization misinformation out there.
Rick

[Rick]

BTW: After reading the previous it makes sense to have a "real" good battery monitor. A good monitor simultaneously reads voltge and current with high accuracy and time stability for making the Amp-hour and kilo-Watt-hour calculations and accumulations.

The monitor will tell you at a glance just how many Amp-hours are missing from the battery. This value is independent of battery type or capacity and is useful for setting up the initial charge voltage in "working" an Amp-hour law chaging regimen. This assumes that one has some control over setting the charger voltages. If the charger is in the bulk mode just set the charge voltage as high as possible when the charger is not current limited and observe your digital monitor. With a few setting changes you should find a value optimum for your setup.

[Sandero]

Speaking of monitors my Link20 doesn't seem to want to show that the house bank is fukll up - blinking light, but it does for the start bank which is only charged via an echo charge.

The boats' been on a tru charge shore charge all winter plus 100 wats of solar and I only use the occassional load of some cabin lights, the Espar and the stereo when I am working there. But for the entire week there are no loads except things like radio memories, and a small bilge pump, some pilot lights like the S2a and the Blue Seas LEDs and the Link20 itself. It would seem to me that it should have reach full charge on Bank1 and started to blink, only to stop the moment I used any load. The voltage reads anywhere from 14+ to 13.75 I suppose depending on what the tru charge and the solar are doing. The Tru's been off as they redo the dock wiring for 3 weeks. So no work and no shore charging, only solar. But still no blinking light.

I can see why the start bank is full up and blinking full as it hasn't been used and gets the "left over" from the echo charge and finally become fully charged - blinking.

Or is that blinking stuff just rubbish?

[GordMay]

George Wood Vinal* makes some interesting comments at:
Optimizing Battery Performance

* Author of the seminal text:
Storage batteries: a general treatise on the physics and chemistry of secondary batteries and their engineering applications

[ess105]

Originally Posted by Rick

One observation that I have made is that the beginning charge voltage of the Amp-hour law charger would always be higher than 14.4V and usually below 15Volts. Does that give anyone a clue?

Are you implying we'd improve charge times / battery lifetime if we raised our bulk charge voltage to somewhere in between?

[Rick]

Yes, ess105, that is correct. It is with the use of a battery monitor and knowledge of the Amp-hour law that allows us to use much higher acceptance voltages than otherwise without a monitor. By monitoring the charge current against the number of Amp-hours "missing" from the battery you can charge much faster than otherwise.

In addition, the internal battery resistance will be lower after a full charge; Charge current acceptance will be lower as well after a full charge indicating that lost capacity has been recovered. The necessity for having to equalize a battery will be lower as well due to less sulphation

[Rick]

Defjef, you gave the answer: After the green LED blinks to indicate full any discharge current resets the full indicator. Your operation appears to be normal.

[GordMay]

Originally Posted by Rick

... knowledge of the Amp-hour law that allows us to use much higher acceptance voltage ... you can charge much faster than otherwise.

Excepting any limitations imposed by maximum input voltages (often ± 15V*) on equipment operating whilst charging.

Ie: Most 12V nominal lamps prefer 15V or less (for longevity)

[Rick]

Gord brings up a good point about operating some equipment. After several years of using high acceptance voltages I have found no unnacceptable halogen or incandescent bulb life AS LONG as they are on before starting the charge cycle. Turning them on with a high voltage is harder on them than is ramping the voltage up like with an alternator.

[GordMay]

Originally Posted by Rick

... I have found no unnacceptable halogen or incandescent bulb life AS LONG as they are on before starting the charge cycle. Turning them on with a high voltage is harder on them than is ramping the voltage up like with an alternator.

Thanks for expanding my good point to an excellent one.
At home, all of our halogen pot lites are on soft-start dimmers. Pre-set a lighting level, turn the lights on, and they ramp up to the selected light level. Upon turning them off, they also ramp down.

[bob kingsland]

Rick, my link 20 never shows full... the charger always tapers down and shuts itself off after a period of time at 2 or 3 amps. This is a big bank, 4 lifeline 8Ds and a prosine 2000 inverter/charger, plugged into shore power. With no loads at all the link 20 draws a couple of amps/day, and I start accumulating negative amp hours. I can get to -255 ah with the voltage still at 12.6. Hit the charger and I see 25-30 amps at 14.4 for not very long, and then tapers thru absorption and float and overnight shuts itself off... still with maybe -150 ah showing, and batteries at 13.2. After removing surface charge they're usually around 12.8 or a little more. Same occurs under long periods of power, with a 140 amp leece neville and balmar 612.
All settings on each of these instruments are by the books for my setup. I do have temp sensing as well. I just shut it off and rezero the ahours on the link, but there must be a better way.

Best, Bob S/V Restless

[CarlF]

Rick,

This is really interesting.

I have both a charger that permits me to set voltage and what I think is a good monitor (both Victron). Temperature sensor, etc.

Could you take me through the higher voltage bulk/acceptance charging strategy step by step. If I have amp hours missing according to the monitor I should charge at ???V. Once I reach full (e.g. no amp hours missing) I should drop to float? Sorry if I seem slow...

Second question. Do you have an opinion of the Odyssey AGM batteries that Nigel Calder is trying? Is there a reason that they should have better cycle life than other AGM's? Does their claim (and evidently Nigel's experience) of faster charge acceptance seem reasonable?

Thanks,

Carl

[Rick]

O.K. What you do is pick a time when the number of Amp-hour missing is slightly less than the maximum charge current capability of your charge source. You jack up the charge voltage to as high as possible. Note the charge current. If the current is say 40A and you have 38Amp-hours missing then leave it there, it doesn't matter what the voltage is. Keep watching the charge current versus the number of missing Amp-hours. If the charge current exceeds the number of missing Amp-hours by say 10% or more then consider lowering the voltage again to make the numbers match.

Note that on a succeeding charge cycle that the voltage might have to be set slightly lower because the battery has recovered lost capacity due to a better acceptance charge. Note as well that as the battery heads towards full that the charge current acceptance of the battery decreases tending to not make you have to change the charge voltage perhaps at all.

If the decreasing charge acceptance current "tracks" the missing Amp-hour number within 10% or so you are in luck. The time to go to float is when you are almost full you will note a complete lack of tracking. Say the current is 5A and you are only missing 2 A-hr. You should also note at this time that no significant temperature rise has occured in the battery. If it has then on the next cycle go to float earlier. Hopefully your system can be forced to float at any time, otherwise you will have to fool it.

The Odyssey AGM batteries do not charge accept better than other good qualityn AGM's such as Fullriver or even good Gel-cell batteries. I'm not convinced that Lifeline AGMs are high quality or that they will repetitively survive high charge/discharge currents compared to others. I could be wrong.

[CarlF]

Thanks. I followed that. I'll give it a try.

So how come in an era of $150 terabyte disk drives no one can make a battery charger that can match current to missing amp hours?

Carl

[Rick]

Before Xantrex bought Cruising Equipment we were poised to do exactly that because the technology existed to do it at a decent cost. Since then my attempts to get anyone in a position to fund the development has fallen on deaf ears.

Back when I worked at Heart Interface there was a similar problem with getting the world to accept the concept and practice of using the 3-step charging regimen. Now everyone (in the deep-discharge world is using it.

Perhaps this forum will spark some of those in the industry (some I have talked to) to do so...Magnum???

The formula for implementing the Amp-hour law is fairly trivial, a curve of Eqsilon with an exponent of -t for time (having a coefficient of one) and the coefficient of epsilon is C, the number of missing Amp-hours.

Not everyone grasps the concept and still keep asking for values of voltage that the equation describes as being always variable until time to float.