In this series of articles, Dr Mike McDonagh looks into the processes involved in the assembly of lead-acid batteries and cells. In this article he starts by examining the manual welding of single-cell plate groups to the bus bar and pillar.
In almost all lead-acid battery designs the plates are first stacked together, with negative-positive-negative configurations. The lugs are arranged with the positive and negatives on opposite sides to be parallel joined by a welded, lead alloy bus bar to form an element. During this welding process the terminals or pillars are also attached. These elements are then placed in an insulating container, followed by the lid being sealed on the top of the container. The lid has holes that allow the terminals to project above it for electrical connections.
In the case of most monobloc designs, the pillars attached to the element bus bars are welded to lead inserts in the battery lid. The cells or groups are then connected in series, by either resistance welding, or by use of gravity castings that are joined over the top of the partition cell walls, to form a battery. There are several processes involved in the assembly of lead-acid batteries and cells:
1.1 Manual welding of lugs with top bar and small parts components (gas welding)
1.2 Stack assembly with separation (stacking and plastic welding)
1.3 Cast-on-strap with top-bar and connectors (metal casting)
1.4 Intercell through-partition welding of automotive groups, (resistance welding)
1.5 Box-to-lid heat sealing – plastic heat welding and adhesive welding
1.6 External post/connector welding to the lid and groups.
In this article we will be examining 1.1 i.e. the welding of single-cell plate groups (elements) to the internal cell bus bar and pillar/terminal Fig 1. This is generally a manual process and mostly applies to the larger two-volt cells for plates between 250 and 650 millimetres in height.
Because of their physical size, automated processes such as cast-on-strap that turn whole battery elements upside down in order to immerse the plate lugs into a molten lead bath in a strap-mould, require expensive equipment. For this reason, many companies (mostly SMEs) opt for a manual solution for the process.
The method described here relies instead on loading the plates and separators into jigs, mostly by hand, then placing machined castellated moulds (burning combs) onto the plate lugs Fig 2. The lugs and pillar are then joined by adding molten lead alloy into the burning combs to form a top-strap.
The most common method for adding the molten lead is to use a gas flame to melt a hand-held lead alloy bar into the recess mould of the burning comb Fig 3. After that the elements are inserted into polypropylene, ABS or SAN cell containers. There are two broad divisions in this welding category: complete-cell group welding, and separate positive and negative plate welding. There are advantages and disadvantages to both methods.
Group welding
In the first category of complete-cell welding in a single jig, the plates and separators are loaded in order of negative plate/separator/positive plate/separator. The positive and negative lugs are placed either side of the jig for subsequent strap welding. This sequence is then repeated for as many plates as are required for the complete cell. It is normal practice to finish with an end negative.
A set of burning combs is then attached to the positive and negative plate lugs on either side of the jig. The battery lead-alloy pillars are then placed in their location slots in the burning combs Fig 4. Once the group or element is assembled, the lugs are welded together by adding molten lead into the recess of the burning tool. This is completed for each burning comb to make two straps for the element, each with their positive or negative pillars attached. Once the strap lead has cooled, the group is removed, generally by dropping it straight from the jig and into the cell container.
Separate welding
The second category differs by welding the positive and negative plates separately, then putting them together by interleaving the positive and negative plates, then sliding the separators between the plates. The welding process is identical; however, the complete cell is manufactured in three stages:
- Positive plates with bus bar and pillars.
- Negative plates with bus bar and pillars.
- Group assembly with separator insertion. The finished element is then dropped into the battery case for lid heat sealing.
There are advantages and disadvantages for each method. For group welding, there is the efficiency and speed of the operation – production rates are faster and fewer people are involved, particularly if rotary jigs are employed. This enables the plate loading on one side whilst the other side is welding the preloaded groups.
The advantage of the separate plate group method is that of quality control. With the group burning method, there are the possibilities of causing hidden short circuits due to molten lead runs between the plates, plus separator heat damage from the gas flame. If the positive and negative groups are kept separate, then any lead runs would be noticeable and removed before the cell is assembled. The separator cannot be damaged by heat since it is placed into the cell after the welding process.
A moulding process
This gas welding is usually executed using either a natural gas or a propane-based flame. The type of gas used and the design of the burners and nozzles is dependent upon the design of the strap and lug system and the physical size of the parts to be made or joined in this process. In real terms, it is also a moulding process as the burning jig is designed to make a top strap of particular dimensions from the added lead which is melted into a recess in the burning combs holding the plate lugs.
The process begins with an experienced operator creating a molten pool of lead alloy in the burning comb recess. Further lead is added from a hand-held lead-alloy stick, whilst partially melting the lugs as molten alloy pool progress along the length of the tool. Fig 5 Is a schematic showing the progress of melting a lead-alloy stick onto a burning comb containing the plate lugs. In this a solidified strap is formed as the molten lead alloy freezes.
The plate lugs are attached by partially melting into the strap, and partially by forming an intermediate bond when coated by a low-melting-point tin solder, a process called tinning. Ideally, the tin solder enhances attachment by creating a low-angle-bond, which is formed at the higher temperature in the molten lead alloy. Failure to treat the lugs will likely result in a high-resistant joint with minimal welded area Fig 6.
Ensuring success
For this process to be successful, it is important to ensure that the lugs are in good condition and that there is sufficient material above the tool face to allow a good contact area. Equally important is the treatment of the lug. Paste residue, or oxidation from long-term storage, curing, or dry-charge formation left on the surface, must be removed in order to ensure that there is a good quality weld. However, further treatment of the lug surface is recommended in order to ensure that the strap properly adheres to the lug surface.
The lead alloy of the lug should form a low contact angle interface with the underside of the strap. The lead alloys used for the grid and the strap do not normally provide a low contact angle. However, this can be achieved by use of flux and a molten tin-lead solder into which the plate lugs are dipped prior to the strap welding process. As just mentioned, the strap, lug and pillar may be made of different lead alloys. There are three principal considerations for the alloys used in the plates, components, and strap lead. These are:
- The differences in their melting points. The lugs should ideally partially melt when the molten lead alloy is dropped into the burning comb assembly. Operator skill is required to achieve this, but the difference between the top-bar lead and the lug alloy is also important. When dropping the molten lead into the welding zone, the lug should be partially but not completely melted. The lug will be warmed by the gas flame and some operators may partially melt it this way before adding the lead. This ensures a fresh, clean surface when the molten lead alloy is dropped onto the lug in the burning comb. Most operators prefer that the added top-bar lead has a lower melting point than the lug and component alloy. This is largely due to the speed of solidification and the ability of the operator to use the heat content of the melted alloy to partially melt the lugs and create an actual intermetallic bond between the components. However, differences in melting point of more than 200C should be avoided.
- Their chemical differences due to alloy composition. It is possible that lead alloys that have different alloying elements can create an electrochemical cell when immersed in sulphuric acid. For this reason (and the above) those companies using lead antimony alloys should ensure that the difference in Sb content is less than 5% by weight. Experience has shown that this alloy difference can be particularly damaging, particularly in VRLA batteries.
- The contact angle formed between the molten strap alloy and the solid lug and component alloys. The wettability of one material onto another, i.e. how well it sticks, is a vital component of any welding or joining process. This can be observed in the strap-to-lug weld by examining the underside of the weld after the group has been removed from the burning jig. If there is a high degree of wettability, then there will be no gap between the lug and the underside of the strap. If the two molten alloys have low wettability and a high contact angle, then there will be a positive meniscus with a gap leading into the strap at the lug/strap interface, Fig 3. It is important to ensure that there is a good area of contact and that the contact point or joint has a low resistance. Lug cleanliness, use of a proprietary flux, before dipping the lugs in a lead-tin solder bath would ensure a better, low resistance contact.
Coating the lugs
A lack of wettability between the plate lugs and the welded plate strap as explained above is remedied by the use of a lead-tin alloy coating on the lug Fig 7a. This process is commonly known as “tinning” and is usually applied well in advance of the lug-to-strap welding process, before storing the plates for the cell assembly processes.
However, this is not the best practice. The coating’s surface will oxidise a little and pick up dust, which can result in a surface layer that will make adhesion to the strap more difficult – by changing the coating’s surface characteristics.
The process of coating is straightforward: the plate lugs are brushed, to expose clean metal, then dipped into a fluxing agent followed by dipping into a molten bath of lead-tin alloy. It is important that this coating reaches the part of the lug that is in contact with the lead-alloy strap.
The temperature and time of dipping are important as it determines the bond between the coating and the lug, as well as the thickness of the coating. If the temperature is too low, the coating will adhere as a lump on the lug rather than a coating. Should the temperature be too high, there is a danger of melting the lug.
Approximate guidelines for dipping times and temperature of the molten bath are given in Table 1. However, the size of the lug and the temperature will also play a role in the process, so some experimentation will be required by the manufacturer to optimise the process.
Alloying considerations
Fig 7a is a schematic showing the alloying pattern of the lug after the tinning process. Some of the tin in the coating-alloy has diffused into the surface of the lug. An alloy is formed at the interface with variable and decreasing tin content with distance from the interface.
The importance of this for the lug to strap bond, and the wettability of the strap alloy on the lug surface is illustrated in Fig 7b. This shows that the lead-tin alloy forms another diffusion bond, with tin migrating into the strap alloy.
The diagram also shows that the residual heat in the added molten-lead-alloy also melts the surface of the lug coating before the solidification of the added lead. We then have a common, tin-rich, molten alloy interface, forming a bond between both lug and strap. It is also beneficial to add a high content of tin to the strap-lead alloy, preferably at least 2%. Limitations here would be cost considerations.
Avoiding the undercut
Fig 8a shows a typical welded group for a traction cell. Points to note are the level of the top-end-strap and the degree of undercutting (component melting) at the point where the pillar is joined to the strap. This is a common problem with this component and strap welding. It is caused by excessive heat input to the area where the base of the component adjoins the top-bar or strap. It requires an experienced operator to control the width of the flame, as well as the flame’s residence time in that area, in order to perfect the method of joining the lugs, component, and top-bar.
Because this is a vital part of the battery production process, it is advisable to take at least two cells or batteries per shift for QA examination. Fig 8b is a photograph of the underside of a lead-alloy-strap that has been bent to show the adhesion characteristics of the strap-to-lug weld. In this case the welded area does not show a gap extending into the contact area between the strap and the lugs. The weld has a low contact angle showing good adhesion. This destructive test is best conducted using bare, unpasted grids to make a group at the beginning of the shift.
Capital or labour investment
In summary, the group burning process is practiced mostly for larger 2-volt cells. This is due to the engineering difficulties and expense of proprietary cast-on-strap equipment available for larger lead-acid batteries and cells. Whilst there is excellent large-scale assembly and cast-on-strap equipment available, there is a threshold turnover required to make the capital investment economically viable. For the smaller industrial battery manufacturers, this manual process will remain a skill-dependent segment of their manufacturing processes for many years.
