Basic Soldering Guide Part 2 – How to solder

Alan Winstanley covers how to actually solder in this second instalment of an article originally reproduced by Everyday Practical Electronics magazine

© Alan Winstanley 1997-2010

How to solder

Turning to the actual techniques of soldering, firstly it’s best to secure the work where possible so that your accuracy isn’t affected should the work happen to be moved accidentally. In the case of a printed circuit board, various holding frames are fairly popular especially when densely populated boards are being soldered: the idea is to insert all the parts on one side (a process often known as “stuffing the board”), hold them in place with a suitable foam rubber pad to prevent them falling out, turn the board over and then snip off the wires with cutters before soldering the joints (see photos).

The frame saves an awful lot of turning the board over and back again, especially with large boards: all the soldering can be performed in one “pass”. Hence not only is the printed circuit board held firmly, the individual components being soldered cannot move either. However only the more serious constructor is likely to go to the expense of purchasing a holding frame, and it is not uncommon for hobbyists to retain parts in place by improvising in a variety of ways – including adhesive tape or blobs of Blu-tac! Other parts could be held firm in a modeller’s small vice, for example.

Solder joints may need to possess some degree of mechanical strength in some cases, especially with wires soldered to, say, potentiometer or switch tags, and this means that the wire should be looped through the tag and bent over before any solder is applied. The down-side of this is that it will be more difficult to de-solder the joint (see later) to remove the wire afterwards, if needed. Otherwise, in the case of an ordinary circuit board, components’ wires can simply be bent to the correct pitch (distance apart) to fit through the board, inserted flush against the board’s surface, splayed outwards a little so that the part grips the board, and then soldered.

In the author’s view – opinions vary – it’s generally better to snip off the surplus wires leads first, to make the joint and any neighbouring joints more accessible and also to avoid applying a mechanical shock to the p.c.b. However, in the case of diodes and transistors the author tends to leave the snipping until after the joint has been made, since the excess wire will help to sink away some of the heat from the sensitive semiconductor junction. Integrated circuits can either be soldered directly into place if you are confident enough, or better, use a dual-in-line socket to prevent heat damage. The chip can then be swapped out at a later date if needed. Parts which become hot in operation (e.g. some resistors), are best raised above the board slightly to allow air to circulate. Some components, especially large electrolytic capacitors, may require a mounting clip to be screwed down to the board first, otherwise the part may eventually break off due to vibration. In the case of these or, say, p.c.b. mounting power transistors, it is a good idea to bolt such components firmly into place before soldering their terminals, in order to avoid placing a strain on the soldered joints or the components when fasteners are tightened. As you will read later, the perfectly soldered joint will be nice and shiny looking, and will prove reliable in service. In general, the key factors affecting the quality of the joint are:

• Cleanliness – dirt or impurities drastically hinder good solder coverage
• Temperature – the right level to enable the solder to flow freely!
• Time – apply heat for just the right amount of time!
• Adequate solder coverage – enough to form a good joint without touching neighbouring areas

A little effort spent now in soldering the perfect joint may save you — or somebody else —a considerable amount of time in troubleshooting a defective joint in the future. Let’s discuss the basic principles outlined above in more depth.

Really Clean

Firstly, and without exception, all parts – including the iron tip itself – must be clean and free from contamination. Solder just will not “take” to dirty parts! Old components or copper board can be notoriously difficult to solder because of the layer of oxidation which builds up on the surface of the leads. This repels the molten solder and this will soon be evident because the solder will “bead” into globules, going everywhere except where you need it! Dirt and contamination are the enemies of a good quality soldered joint! Hence, it is an absolute necessity to ensure that parts are free from grease, oxidation and other contaminants. Note that in any case, some materials or surface finishes just cannot be soldered using ordinary tin/ lead solder, no matter how hard you try – e.g. aluminium parts would require special aluminium solder to be used.

In the case of old resistors or capacitors, for example, where the leads have started to oxidise, use a small hand-held file or perhaps scrape a knife blade or rub a fine emery cloth over them to reveal fresh metal underneath. Stripboard and copper printed circuit board will generally oxidise after a few months, especially where it has been fingerprinted, so the copper strips ought to be cleaned using an abrasive rubber block, like an aggressive eraser, to reveal shiny copper underneath.

Also available is a fibre-glass filament brush, which is used propelling-pencil-like to remove any surface contamination. These tend to produce tiny particles which can be highly irritating to skin, so avoid accidental contact with any debris. Afterwards, a wipe with a rag soaked in cleaning solvent will remove most grease marks and fingerprints. After preparing the surfaces, avoid touching the parts if at all possible.

Another side effect of attempting to solder unclean surfaces is the tendency for the novice to want to apply more heat in an attempt to “force the solder to take”. This will often do more harm than good because it may not be possible to burn off any contaminants anyway, and the component or the printed circuit board may be overheated and damaged in the process. In the case of semiconductors, temperature is quite critical and they may be harmed by applying excessive heat for more than a few seconds. Furthermore, extreme heat applied to printed circuit board tracks may also cause irreparable damage, because the tracks will be lifted away from the substrate underneath especially on a delicate or badly designed board. If you are lucky, you may be able to bypass the damaged part of the board by soldering in some wires.

Before using the iron to make a joint, the hot tip must be “tinned” by applying a few millimetres of solder, then wiped on a damp sponge in order to prepare it for use: you should always do this immediately with a new bit, being used for the first time. Some old hands re-apply a small amount of solder again, mainly to improve the thermal contact between the iron and the joint (the molten solder fills the small void between the parts and the iron tip) so that the solder will be encouraged to flow more quickly and easily.

It’s sometimes better to tin larger parts as well before making the joint itself, but this is not generally necessary with p.c.b. work. (Note that all EPE printed circuit boards from the P.C.B. Service are “roller-tinned” to preserve their quality and to help with soldering.) A worthwhile product is Tip Tinner & Cleaner, a small 15 gram tinlet of paste onto which you dab a hot iron – the product cleans and tins the iron ready for use.

Solder Grades

General purpose electronics grade solder is usually 60% tin and 40% lead (“60/ 40”) and it contains a “flux” which helps the molten solder to flow more easily over the joint. It does this by removing oxides which arise during heating, and will be seen as a pungent brown fluid bubbling away on the joint, accompanied by some fume emission which some will find a slight irritant. Those coming into electronics from other engineering industries should note that the flux is already contained within “cored” solder and on no account should any acidic flux be applied separately before using the soldering iron.

Other solders are available for specialist work, including aluminium and silver-solder. Different diameters are produced, quoted in Standard Wire Guage (Imperial SWG) or American Wire Gauge (AWG) — 20-22 SWG (19-21 AWG) is 0.91- 0.71mm diameter (the higher the gauge, the smaller the diameter), which is fine for most general printed circuit board and interwiring work. Choose 18 SWG (16 AWG) for larger joints requiring more solder.

Unfortunately certain diameters may only be obtainable in large reels designed for professional use, though savings can be made by using 40% tin/ 60% lead (40/ 60) which is approximately 15% cheaper than 60/ 40, but has a higher melting point. (Tin is eight times more expensive than lead, which explains the price difference.) For many constructors the difference in performance will be negligible.

In view of the increasing focus on health and environmental issues these days (namely, the need to reduce human contact with lead, plus the environmental effects which the disposal of old equipment containing lead/ tin solder have on our surroundings), a “lead-free” solder has become available which consists of 99.7% pure tin and 0.3% copper. This lead-free solder can be up to 50% more expensive than an equivalent 60/40 solder (metre for metre) and also demands a higher melting point, but it is probably the direction in which solders will ultimately be forced to steer in future years for environmental reasons.

Another solder variant which finds favour amongst professionals is “Smart” wire which contains a small degree of silver. It produces very clean results and is often associated with SMD (surface mount devices), though some engineers use it for routine printed circuit board work for producing the best possible finish by hand.

Temperature

The next step to successful soldering requires that the temperature of all the parts is raised to roughly the same level, before solder may be applied. Imagine, for instance, trying to solder a resistor into place on a printed circuit board: both the copper p.c.b. and the resistor lead should be heated together so that the solder will flow readily over the joint.

A beginner will often mistakenly just heat one part of the joint (e.g. a component wire protruding through a printed circuit board) and hope that the resultant blob of solder will be sufficient to bond all the components together. This is unfortunately completely wrong, because the remainder of the joint will be quite cold when molten solder is flooded on to it. The joint will be weak, incomplete or unreliable. The secret of success is to touch the tip of the iron onto the workpiece so that it is in contact with all the parts. Within a fraction of a second, heat will conduct from the iron and raise the temperature of the entire joint, after which solder can be applied.

The melting point of most solder is in the region of 188ºC (370ºF) and the iron tip temperature should be set for typically 330-350 ºC (626-662 ºF). Many soldering iron workstations have a control which permits adjustment of the tip temperature. Depending on the alloy used, some solders require higher temperatures, and indeed at least one type of solder is specifically designed for use in equipment which operates at high temperature; it requires a tip temperature of well over 400ºC (752ºF).

Now is the time

The joint should be heated with the bit for just the right amount of time – during which a short length of solder is applied to the joint. The heating period depends on a combination of factors, including the temperature of the iron, the size of the tip and the size of the joint — larger parts need more heat than smaller ones — but some components (semiconductor diodes, transistors and i.c.s), are sensitive to heat and should not be heated for more than a few seconds. An average printed circuit board joint can be made within roughly two seconds or less. A common mistake is to use a soldering iron to carry molten solder over to the joint — don’t do this!

Until they have gained some practice, novices sometimes buy a small clip-on heat-shunt, which resembles a pair of aluminium tweezers. In the example of, say, a transistor, the shunt is attached to one of the leads near to the transistor’s body. Any excess heat then diverts up the heat shunt instead of into the transistor junction, thereby saving the device from any possibility of thermal damage. Beginners find heat shunts reassuring until they’ve gained more experience, but modern semiconductors are more forgiving in this respect anyway.

In due course a novice can judge how much solder should be applied to any particular joint. An excess is just an unnecessary waste and may cause short circuits with adjacent pads, especially on densely-populated boards. Professionally-produced p.c.b.s have a green solder resist coating which helps to ensure that solder does not stray onto adjacent pads. If too little solder is used, the result may be an incomplete joint which may become the source of an intermittent fault later on.

There are a small number of components which can create hazards during a soldering operation.

Coin cells if heated excessively will ultimately explode due to the build-up of internal pressure. It is not uncommon to need to solder wires to such a cell but depending on their size this should be done as quickly as possible. Similar dangers exist with other types of battery, although an increasing number of cells are designed for p.c.b. mounting and have solder tags fitted for this purpose.

Some memory back-up capacitors or electrolytic capacitors may retain an internal charge. Molten solder makes a perfect liquid electrical conductor and in some cases a technician could accidentally short the component’s contacts during the soldering (or desoldering) operation. If the device is suddenly discharged (e.g. a short to the 0V rail, or molten solder shorts it out) then molten solder globules can be spattered outwards, possibly inflicting eyesight damage. Always take care to ensure that such components are electrically inert before they are installed. Capacitors or memory back-up caps should be discharged first. Take care with cells, batteries and battery packs to ensure they are not accidentally shorted for any reason during the soldering process.

After the soldering is complete, many assemblers tidy up the joint by snipping any excess wire or solder away from the joint using a pair of “end cutters”. These have blades specially angled to help snip the joint flush against the circuit board. It is worth taking time out to inspect the work closely, looking for any possibilities of whiskers of solder or swarf shorting out any solder pads, and all such potential problem areas should be dealt with prior to testing the board. Lastly, there is nothing more infuriating than finding that a joint has been missed just as the final interwiring and testing is about to be performed, so double check to see that no joints have been overlooked!

Desoldering methods

A soldered joint which is improperly made is likely to be electrically “noisy”, unreliable and will probably become worse over time. It may even not have made any electrical connection at all, or could work initially and then cause the equipment to fail at a later date! Noisy joints can also introduce intermittent problems which can be maddening to locate and resolve. By following the guidelines given and putting in some practice, there is no reason why you should not obtain perfect results and eliminate most potential problems.

A joint which is poorly formed is often called a “dry joint”. Usually it results from dirt or grease preventing the solder from melting onto the parts properly, and is often noticeable because of the tendency of the solder not to “spread” but to form beads or globules instead. Alternatively, if it seems to take an inordinately long time for the solder to spread, this is another sign of possible dirt and that the joint may potentially be a dry one.

A moderately complex circuit board can easily require 1,000 solder operations or more and it is inevitable that occasionally one or two joints may be made imperfectly. They may appear to be adequate given a cursory inspection but a closer look under a magnifier may reveal that some joints will have to be re-made. Consequently there will undoubtedly come a time when you need to remove the solder from a joint, possibly to replace a faulty component or fix a dry joint. Naturally, there are tools and techniques which help with such tasks.

The usual way of removing solder from a joint is to use a desoldering pump. These work like a small spring-loaded bicycle pump, only in reverse! A spring-loaded plunger is pressed down until it locks into position. It can be released with a thumb-press on a button which sucks air back up through a pointed nozzle and any molten solder is drawn up into the pump. It may take one or two attempts to clean up a joint this way, but a small desoldering pump is an invaluable tool especially for p.c.b. work. More demanding users using CMOS devices might need a pump which is ESD safe.

The pumps themselves have a heatproof P.T.F.E. nozzle which may need replacing occasionally. Every time the button is pressed, this action clears the nozzle but sometimes solder particles and swarf will be ejected in the process, and it’s a good idea to direct the nozzle into a small pot or old aerosol top to catch the debris. Remove the spout and clean out the pump from time to time.

With particularly stubborn joints where the last traces of molten solder cannot be shifted, it can sometimes prove effective to actually add more solder and then desolder the whole lot again with a pump. Care is needed, though, to ensure that the boards and parts are not damaged by excessive heat. It is not very difficult to apply so much heat in the desoldering operation that the adhesive which holds the conductors onto the p.c.b. can eventually fail, causing the copper track to lift away. If this should ever happen, remove the iron immediately and permit the area to cool (a freezer aerosol is valuable at such times). It may well be possible to repair the lifted track using a droplet or two of Super Glue.

An excellent alternative to a pump is to use desoldering braid, including the famous “Soder-Wick” (sic) which is packaged in small dispenser reels. It is a flux-impregnated fine copper braid which is applied to the molten joint, and the solder is then drawn up into the wick by capillary action with remarkable effectiveness. For certain tasks, the action of braid can be more thorough than a desoldering pump and it is recommended that a small reel is bought, especially for larger or difficult joints which would take several attempts with a pump. The correct way to use the product is to press the end of the braid down onto the joint using the tip of an iron, and let the solder melt underneath: the braid will then absorb the solder.

However, be aware that it is possible to damage a printed circuit board accidentally when removing the desolder braid if it is not removed quickly. The solder can soon harden which effectively solders the braid to the printed circuit board! A careless tug may cause copper tracks or pads to be lifted away, stuck to the braid. You can also drag solder “whiskers” onto neighbouring pads unless the braid is removed cleanly. Why not practice with some surplus components?

<<Part 1  Part 3>>

---

The on-line version of the Basic Soldering Guide (with colour photo gallery) can be accessed at http://www.epemag.wimborne.co.uk/solderfaq.htm

Other useful web sites:

• http://www.antex.co.uk – Antex soldering irons
• http://www.multicore.com – This web site has some background to Multicore Solders’ products.
• http:// www.metcal.com – considerable advice on reworking, presented by Metcal Inc.

Text and Photographs © Copyright © Alan Winstanley 1997-2010 The author can be contacted by E-mail:alan@epemag.demon.co.uk