The purpose of this article is to set down certain observed facts regarding the loss of water and electrolyte from the individual cells of submarine storage batteries. It is not intended to advance any new theories or hypotheses with regard to these losses. Excessive water losses from submarine installations have existed for years, but the causes of these losses have remained unchanged although certain fundamental facts have become established.
The fact that any wet secondary cell loses water is a matter of common knowledge. To submarine personnel it is a matter of constant annoyance. In the past the need of watering batteries periodically has been accepted, by submarine crews, as one of the necessary hardships peculiar to that engaging pastime known as submarine life. The task of watering a submarine’s batteries is not one that may be arbitrarily delayed or undertaken at one’s leisure. It requires from three to five hours. On all submarines except the larger V’s, it requires lifting about 60 per cent of the free deck space in the boat interior. This obstructs traffic, interferes with the proper conduct of other ship’s work and more or less upsets the daily routine. In addition it has a most unfavorable effect on the tempers of such gentry as the captain, the chief-of- the-boat, and the ship’s cook. It is a thoroughly unpleasant task. In peace time it remains as something which has to be done and is accepted as such. In war time it assumes a much more serious significance.
Consider an, S boat conducting a war-time patrol. To begin with, the boat must carry aboard sufficient battery water to provide for at least two complete waterings. On the S-class this means about two tons dead weight. The boat will, in all probability, be submerged all day and may have to conduct several submerged attacks during this time. This means charging batteries every night. Under these conditions ten days is considered a liberal period to expect between waterings. On such occasions three to five hours must be subtracted from the surface period at night during which time the batteries may not be charged and any other ship’s work is seriously hampered. The watering might be accomplished during the daytime, submerged, it is true. However, the resulting confusion and difficulty in controlling the boat submerged would be sufficient to cause a prudent commanding officer to avoid this except under exceptional circumstances. This would be particularly true if there were any possibility of making contact with the enemy. Couple all this with the weight of the evaporating plant, which is installed primarily to provide battery water, and you have a formidable problem.
From the above it may be deduced that the necessity for watering batteries imposes the following handicaps on a submarine: (a) reduces the military effectiveness by say one torpedo; (b) restricts its mobility; (c) imposes a definite task on the crew which, in war time, might become a factor of some importance.
The above discussion would seem to provide a substantial foundation for conducting an investigation into these water losses and their causes; although another cause actually brought the matter under observation. The Bureau of Engineering has steadily applied itself and its resources to eliminate the danger of battery explosions aboard submarines. Under its supervision various tests, investigations, and experiments were undertaken, for the most part, at the Naval Research Laboratory at Bellevue, D.C. As the work on explosion hazard progressed the importance of the question of water loss became more and more apparent. Investigations were, accordingly, undertaken. Shipboard tests were conducted on two submarines sent to Bellevue for that special purpose. Additional work has been done at the Submarine Base, New London, Connecticut. As a result of this activity, a clearer understanding of water losses and their causes has been arrived at.
Theoretically, water and electrolyte leave the cell due to the following causes:
(a) Loss due to gassing.—Gas is formed due to the excess charging current causing an electrolysis of the electrolyte. The gas so formed is carried out of the cell by the ventilating system.
(b) Loss due to evaporation.—There is neither the time nor space to quarrel over the correctness of the designation of this cause. Simply it is intended to mean that the relatively dry air of the ventilating system, passing over the electrolyte surface picks up moisture and carries it out of the cell.
(c) Loss due to mechanical sweeping out. —This loss is caused by the high velocity of the ventilating air. It is a process difficult to describe but is at once apparent when it can be seen. Tests conducted on the experimental cell at Bellevue have shown that gas bubbles, forming on the cell plates, detach themselves and rise rapidly to the surface of the electrolyte. On reaching the surface they have acquired considerable velocity under the action of which they forcibly break through the surface and are projected upward into the air space above. By air space is meant the space between the cell cover and the surface of the electrolyte. It has been determined that in breaking through the electrolyte surface each gas bubble becomes enclosed in a coating of electrolyte. The bubble, with its coating of moisture, is projected upward into the rapidly moving air stream and is mechanically swept out of the cell.
These then are the three fundamental causes of water loss. Let us inspect them again with more attention to detail. A little study will show at once that all causes are in their most aggravated form when the cell is charging at the finishing rate. This may be briefly summarized as follows:
Gas.—The greatest evolution of gas in the cell occurs during the finishing rate of a charge. Hence, the greatest loss due to gassing exists at this time.
Evaporation.—During the finishing rate all the ventilating blowers are being run at full speed. In other words a much greater volume of dry air is being swept over the electrolyte than at any other time. More dry air; more evaporation. In addition the surface of the electrolyte is considerably agitated, at this time, due to the high evolution of gas. Further the electrolyte is warmer due to the heating effect of the charge. Both of these factors assist evaporation.
Mechanical loss.—As the greatest number of gas bubbles are emitted during the finishing rate then the greatest number of electrolyte coated droplets are being projected into the air space during this period. In addition the ventilating air is moving at its highest linear speed due to the running of the blowers at full speed. Hence the “sweeping out” effect is more pronounced during this time than at any other.
From the above, then, it seems reasonable to expect that the greatest loss of water will take place during the finishing rate of a charge. To reduce it a little further, water losses in service should be at a maximum under the following conditions: (a) repeated charging; (b) when “floating” a fully charged or nearly fully charged battery on the line; (c) during overcharges and equalizing charges.
It is considered that these findings are fully substantiated by service experience. It is a well known fact, among submarine people, that an idle battery loses little water. Lying alongside a dock or tender, putting in a routine charge now and then, takes little water from the cells. But under operating conditions requiring a charge every night, “floating” when under way to keep the wheels from racing, then the water disappears truly like “water.” There is nothing new or novel about all this; it is something which has been a matter of service knowledge for a good many years.
From the above it may be said, with reasonable accuracy, that it is now rather definitely understood what is causing the loss of water from the cells. Having arrived at this understanding there remains the question as to what can be done to either eliminate or control these losses.
Most submarine personnel are, by this time, familiar with the glass plate which has been authorized for installation in submarine cells. For those who are not it will suffice to say that it consists of an oblong sheet of ordinary window glass about one- eighth of an inch thick. This plate rests on top of the cell plates, in a horizontal position and covers the whole area between the negative and positive terminals. In this position it is submerged about two inches below the normal electrolyte level. It thus covers almost completely the central area through which the gas bubbles find their way to the surface of the electrolyte. The presence of this plate alone should cause a considerable reduction in water loss. Its action with regard to the three fundamental causes of water losses may be described as follows:
Gas.—Gas bubbles form, as before, and proceed upward until they strike the under surface of the glass plate. Here their progress is arrested and their velocity reduced to zero. Several small bubbles then form into one large bubble which, finding its way to one of the edges of the glass plate, slips around the edge and proceeds upward to the surface of the electrolyte. During the period of this final ascent, however, the bubble acquires little velocity. Hence it is not projected as violently through the surface nor as far into the air space above. In addition instead of a myriad of small bubbles we have many larger bubbles, all coming to the surface at the sides of the cell rather than over the entire electrolyte surface. This description has been entered into here merely to picture the course of the gas bubbles in the cell interior. The glass plate provides no reduction in water loss through gassing. The same amount of gas is being evolved, as previously, and finding its way into the air space is carried out of the cell as before.
Evaporation.—As described above, the bubbles break through the electrolyte surface around the edges of the plate and at the sides of the cell jar. The surface of the electrolyte over the entire central area remains placid and unbroken. This unbroken surface is directly under the path of the ventilating air stream and is, hence, where most of the previous evaporation took place. But whereas before this surface was agitated and so with considerable moisture laden air above it, now it is smooth and unbroken. This fact alone seems to have materially reduced the loss from evaporation. There is still some evaporation taking place, without question; but not in the quantities previously experienced. So, by the use of the glass plate, we have caused some reduction in the water loss due to evaporation.
Mechanical loss.—The velocity of the gas bubbles breaking through the electrolyte surface is materially reduced below that experienced without the glass plate. Secondly, there are not nearly so many of them. Finally they are not projected directly into the path of the rapidly moving air stream of the ventilating system. These several features seem to have caused an almost complete elimination of water loss due to coated droplets being mechanically swept out of the cell.
In summary, then, the glass plate has the following effect on the three basic causes of water losses: gas, none; evaporation, partial reduction; mechanical loss, almost complete elimination.
Just how pronounced the over-all reduction in water losses will be due to the installation of glass plates remains for service test to demonstrate. However, it is hoped that their adoption will prove to be the first step on the pathway of improvement, which leads to the broad highway of progress. A great deal remains to be done on the complete question of water losses. There is the matter of how far water loss from a cell may be reduced without running into difficulty due to high cell temperatures. The baffle discs now installed in the cell goosenecks have come under a severe fire of criticism. One suggested substitute employs the principle of a water separator in a steam line. In other words by reversing the direction of flow of the ventilating air, one or more times, any entrained moisture is trapped out and may be drained back into the cell. This device not only recovers water from the air stream but reduces the resistance to air flow experienced with a baffle installation. If it brought no other benefit the adoption of such a device would be acceptable on the grounds that it would close forever the question of how many baffles should be placed in each individual gooseneck. It will, in all probability, meet with opposition from the battery people both ashore and afloat. They would have nothing left to fight about. The sun may still rise on the day, however, when the electrical gang can stow their gear, after watering, with the knowledge that the job is done for at least three or four months.
Here, it is hoped, is a message of cheer for those who still dwell in darkness. Although in the dawning of this coming day of relief one seems to hear the voices of the oppressed chanting their age old battle cry— “Hell, it ain’t no job to water batteries. We used to do it in an hour on the B boats.”