First off, let's not have another dispute over whether batteries are recoverable by desulfation or not. I intend to do some science regarding the matter and that's what this thread is about.
I have devised an approach which I'd like to discuss. Here it is:
Assumptions
1) The equalizing charge (@15V+) serves the purpose to minimize creeping sulfation due to under charging which can happen to individual cells as they age and chemistry alters cell to cell. Thus occasional higher voltage charging has benefits.
2) Analyzing each cell (6 in a auto batt) voltage under constant current charge will reveal differences in cell internal resistance. A higher relative cell Pd => sulfation or grid corrosion etc. An unusually low cell Pd => a developing short etc.
3) After a charge, discharge, equalize cycle, data from step 2 can reveal weak or faulty cells and serve to justify attempts at service or designate for recycling.
4) Swept frequency pulse charging modulation on top of the regular charging current won't harm the battery and 'may' enable enhanced ionic activity in a sulfated cell.
5) The enhanced ionic activity is derived from sharp rise time pulses treating a sulfated cell as a capacitor. Such a cell would otherwise have reduced ionic activity due to IR and Pd increase. Such a cell would induce other series cells to be undercharged due to a faulty end of charge battery voltage being detected. This in turn induces 'creeping' sulfation in other cells.
6) Perhaps 40 to 50% of battery 'failure' is due to sulfation from poor charging regimens.
The science I intend to perform involves dismantling an old battery with bad cells as determined by spec. gravity after proper charging. Reassemble said battery in a transparent polycarbonate case with access to all cells. Examine samples of battery plates under a microscope and take imagery of the samples. Log all this data and assess which plates are sulfated.
Assuming that several cells have sulfated plates (quite likely from a battery put down for a year or 2) the following is carried out:
a) Internal resistance of each cell is measured by both ac and dc methods.
b) Sulfated cells are discharged and then pulse charged with varying frequencies for a fixed time with the same pulse charger.
c) New measurements of internal resistance are taken. Plate samples are taken for inspection under microscope.
d) Results are compared and an assessment is made which may justify further pulse charging cycles.
Am I missing anything useful?
First off , for those who don't believe in the restorative effects of certain pulse charging ( not all), no explanation possible, for the rest, no explanation necessary.
Secondly, any scientific evaluation; must include a Test Plan, Test Methods, and Planned Analytical methods, such as critical specs, Metrics, Figures of Merit (FoM), a parametric model and schematic of the equivalent circuit, definition of terms and relevant Industry Standards with modifications and rationale to save time if there is good correlation.
For Example a Schematic is shown at the Battery University ( courtesy of Cadex) as follows;
We know this analog is good for estimate the differences in ESR for fast and slow charge or delta V/delta I and C is often in Farads per Vol or Amps per Farad or just Farads , where Ic=C dV/dt.
But we know this isn't entirely accurate because if you applied a constant current for charge and discharge and 2 different distinct levels ( e.g. C/20, C and 4C or different distinct time intervals or constant power drain vs constant current , that the model changes and is no longer accurate.
So analyzing the effectiveness of a method can be predicted by the model and compared with the results.
e.g. ESR1, SG Ah, ESR2, Rp or leakage, remaining charge cycle life or Total Life Cycle watt-hrs [Wh], charge transfer efficiency ( Watts out/in), Thermal coefficients for each variable
These will all become important when we get OFF OIL and we all use EV's
The best report format I have used as a Test Engineer for 20yrs uses 1 page per Test plan with Diagram , overview with acceptance criteria , graphical/table results, brief conclusions and pass /fail. Statistics give weighting factor for accuracy of predictions like R^2.
Then the exec summary is just like a Table of Contents with results. pass/fail. , statistics , sample size.
hmmm.. Also there are other effects to used batteries such as dendrite growth, ionization, oxidation, piezo-mechanical effects and passivation, which also affect the same metrics as sulphation.
For ionization, rise time is inverse to gap size, and contaminants in other purified electrolytes such as transformer oil have ionization rise times in the sub-nano-second range. For battery electrolyte ( sulphuric acid) , it can vary from microsecond range to tens of nanosecond range depending on size of contaminating particulate mattter.
Thus Pulse rise time peak current or I^2T of pulse duration can be more significant than a slow big pulse to sulphation dissipative effects and damage from cell mismatch loading effects. Equalization timing is also critical to prevent over charging the weakest cell.
What Else?
Mass Spectroscopy profile? Such as Lead oxide , Suplhates, Calcium Oxide and Antimony contents, etc.
Vaporized solids laser particle size counter ( histogram) .... Size vs Pulse rise time effectiveness.
Metrics of importance; are ESR, CCA, Capacity [Ah],
Maximum sustained current worst case ( to emulate cranking for say 10 seconds) similar to (0'F ) CCA test at a designated lower threshold ( e.g. ?not sure? 9V (JAE),8.5V (IEC), 7.5V SAE))
Sometimes crank current drops rapidly after 1 second, 10 seconds 1 minute or 10 minutes depending on battery condition and as we know the only purpose is to crank the engine, otherwise all we need is the Alternator, so tests critical to these requirements at local temperature ranges are most relevant.
For other applications like Marine deep cycle, different metrics would be more relevant.
This is just off the top of my head.