The A B C's Of Joining Two Pieces

by Richard C. Roth, Ph.D.
Superintendent, Allegheny Highlands Division

Part 1

This is the first in a series of articles intended to assist the modeler in joining materials together without mechanical fasteners.  The first several will address joining plastics with the first subject being that of welding.  Materials such as metals, paper, plaster, and some others will be handled later.

To most people the thought of assembling a model means they will be joining at least some of the components with "glue".  Well more or less, for the term "glue" is a catch all for several different systems used for attaching.  To get the best from the material you are using there are some important considerations that should be observed.

The first thing that must be considered is the type of joining that is desired.  There are really two different kinds, welded and bonded.  Both involve the application of some sort of substance to the joint but, each works in entirely different ways.  The choice of which to use depends both on the type of materials being joined and the integrity of the joint area desired.

A welding agent is one that is intended to soften or liquefy the plastic at the joint area.  The welding agent temporarily dissolves the surface of the pieces being joined allowing material from each piece to intermingle.  When they reharden the two form a joint that is about as strong as the material that was being joined.  It is not unlike welding two pieces of steel together.  For plastics we use solvent welding agents while for steel heat is used to temporarily cause liquefaction of the materials being joined.  Figure # 1 shows some types of weld joints including a.) butt; b.) lap; and c.) miter butt.  The shaded area represents the area of the joint in which the two materials have intermingled.  The value if this joint is that no additional material is added at the joint and therefore the parts do not have to be adjusted dimensionally to allow for the joint.  There are downsides as well as will be discussed later.

The welded joint can be made stronger by increasing the area of intermingling of the two pieces being joined.  This is best done by keeping the joint wet with repeated applications of the welding agent.  The joint can also be strengthened by pre﷓treating the surfaces to be joined immediately prior to assembly by applying a coating of the welding agent.

The effects of surface imperfections  on the pieces being joined can be minimized by pre﷓treating and by extending the wet period for the joint.  This allows the material to become softer and allow the high spots to be pressed down to increase the joint surface.  Care should be taken doing this though, because if the pieces being joined are pressed together with too much force some of the softened material may be forced from the joint requiring it be removed later.  In some applications this is done intentionally to eliminate any gaps in the joint.  The excess is trimmed and then sanded after it has hardened. 

With a little experience the modeler can even control the amount of squeeze﷓out and the direction to which it goes by the manner in which the forces are applied and by controlling the application of the welding agent.  The modeler can force the squeeze﷓out in a miter joint to the inside by coating the inside area of the joint more heavily than that at the outside edge.  This is shown in Fig. # 2.

When the two pieces are brought together the points are brought together first and the parts rotated slightly to bring the remainder of the joint areas into contact.  Several applications inside the completed joint will also aid in obtaining the best possible joint.  Because the two pieces will already be in position then, squeeze﷓out will be minimized.

Years ago, the most common plastic was styrene and most of us are familiar with the Testors Airplane Glue that was sold in tubes.  This "glue" had some unique properties that allowed it to serve both as a bonding and welding agent.  Because we will discuss bonding the next time let's take a look at the welding properties of it that made it desirable to the modeler.  (I'm not talking about getting high on it either!)  This "glue" was nothing more than some clear styrene dissolved in solvent, usually a blend of toluene or xylene with some hexane added.    It was really  a welding agent with filler.  It was the filler that gave it gap filling properties that were of great value in early models.  The "glue" softened styrene and allowed the welding action to take place.  The filler kept the glue at the joint while the joint hardened.

When we talk about solvents there are various levels of solvents.  Some will dissolve a lot of a material while others will dissolve just a little.  A solvent such as toluene will dissolve a lot of styrene and is called a primary solvent for that material.  Hexane on the other hand dissolve very little and is called secondary solvent.  When blended in the proper proportions the blend of solvent will dissolve as much as the primary solvent and have other properties that are between the two.  The blend used by Testors was chosen because it would provide a relatively thick glue that would fill gaps well and would not dry too quickly.  The slow drying was the accomplished through the use of the Hexane.

There are several kinds of plastic that can be joined using the welding process.  They include styrene and styrene based materials such as ABS, the vinyl family and polycarbonates.  The one difficulty with welding plastic is that one is required to stay within the family of plastics to get a good joint because the welding agents for one family may not work well for the other.  Fortunately, most of the low to moderate priced plastic kits contain materials in the styrene family for which welding agents are readily available.

There are down﷓sides to using welding agents just as there are benefits.  One of the most common is their ability to attack surfaces around the joint area causing discoloration or surface roughness.  Much of this can be avoided by careful application of the welding agent. 

Prepare well the surface for joining before introducing the welding agent.  Make sure the surface is smooth and free of high spots.  Make sure the surfaces fit while dry then add the agent.  When possible roughen very smooth surfaces slightly with very fine sand paper to raise fine particles of the plastic to hold the welding agent.    

Use a small brush to apply unfilled weld agent to the parts to be joined.  Pre﷓treat the joint areas.  The welding agent will flow better on plastic that has already softened by pretreating then areas still dry.  Get rid of excess agent on the brush before the brush is placed in contact with the parts to be joined.  After joining brush additional agent on the inside of the joint.  Also, work one joint at a time and give it time to firm up before moving on to the next.  Keep a soft cloth handy to wipe up any spills on the work bench immediately to prevent transferring it to the model. 

Today there are a variety of welding agents ranging for very fast to very slow and with a viscosity range from very low (thin like water) to the very high (almost like putty).  The lowest viscosity can be applied to very tight joint areas and it will seep in by capillary action just as water will move up a paper towel when just the corner is placed into water.  The highest viscosity can fill gaps of almost unbelievable proportions.  While I mention the heaviest or highest viscosity materials, I don't really recommend them because they just cause us to be less careful in forming joints, and take much longer to harden.  Like most other things, they do fill a need so they are there when needed.

The next time, the discussion will be on bonding plastics. This involves the use of materials that do not soften or attack the plastic in any appreciable manner but give good joint strength and can join dissimilar materials.  See you then.




Part 2

Last time we discussed welding plastics together, basically a chemical softening of the base material and then joining the softened areas to allow mingling together.  Bonding too can be chemical in nature but, it can be mechanical also.  The modeler can use both types to their advantage and here's how.

Most materials such as plastics, glass and metals have a relatively smooth surface, or to the naked eye looks smooth.  When we go down to the molecular level however, the surface would most likely look like one side of a bale of hay.  This is because at that level, the chains of molecules that make up the part are linked together in very specific orders forming chains having not only length but also width and height. 

One analogy would be to have a number of bottle brushes, each with its bristles sticking out at various angles from the center roughly forming a cylinder.  If we take a number of these brushed and line them up side by side but not taking care to have their ends even we begin to form a solid object.  If the brushes are very gently pushed together from their sides, the bristles would start to intermesh and if more pressure were applied the would compact even more.  The bristles sticking out that mesh first with another brush make good chemical bonding sites while the gaps between provide good mechanical bonding sites.

This same analogy can be used to illustrate why certain materials will not bond well or not bond at all.  If the surface is very compact with a very small number of bonding sites extending above the surface, or if the surface is very wide open such as would be the case with a brush with only a few bristles, it becomes very difficult to obtain good strength of bond between the surfaces.  That is why some materials like Teflon® will stick to very few other materials (no bonding sites) and thin super glue will not bond plaster, wood, or paper very well.  These materials are porous, too wide open.

From these examples the reader should be able to see a pattern emerging.  It is necessary to have receptors extending above the surface and the surface cannot be too open (porous) if a good bond is desired.  On the same note this same information can be used to gain temporary adhesion of two parts where it is desired to separate them later.

Before we get into what makes a good surface for bonding and what does not, there is one more important consideration to be discussed and that is the types of bonded joints.  There are two basic kinds, mechanical and chemical bonds.  Mechanical relies on the texture of the surface to provide a place for the bonding agent to get a grip, if you will.  White glue and plaster is a good example of such a bond.  The glue creeps into the pores of the plaster and grabs hold.  A chemical bond relies on those molecules extending above the surface for good strength.  Many sites means a strong joint.

Figure # 1 shows some examples of surfaces that are good and poor candidates for bonding.  Sample "a." would most probably yield a poor bond because the surface is very irregular.  Only the high spots would be in good contact with the surface to which it was being joined.  For a good mechanical bond sample "b." would be much better. 

For chemical bonding, sample "d." offers many more sites for the bonding agent to grasp as compared to sample "c.".  To obtain the best joint, the surface should be relatively smooth but not a polished mirror finish for most materials.  Because both mechanical and chemical bonding require access to the bare surface, cleanliness of the surfaces to be joined is very important.

Bonding agents come in a variety of forms from those that are applied to the surfaces and the joint made while wet to those that require the surfaces to be dry to the touch before being joined.  An example of the first would be "ACC" or "super glue" while the latter would be contact cement.  Each has its good side and its bad. "ACC" can be applied after the joint is formed.  In the thinnest forms it will creep into the joint to form a bond.  The joint need be held in position for only a short time before it can be moved.   Most "ACC" allows a few seconds for adjustment before the bond forms.  Contact cement on the other hand requires almost perfect alignment when the joint is made because once made there is not the capability of adjustment.

From the above, it would look like "ACC" has it all over contact cement, but it doesn't.  "ACC" is a relatively hard material once it hardens and thus does not stand up well to flexing.  Contact cement or some of the other bonding agents will remain semi﷓flexible after hardening to tolerate some movement of the joint area. 

Many materials used for bonding are what is known as curing systems, that is the chemicals that make up the bonding agent go through a chemical reaction to form the finished form.  This is true for most of those that harden to form the joint.  There are two common  forms of these chemical systems; two part such as the epoxies, polyesters, acrylics, and some urethanes and the catalyzed systems such as "ACC", RTV and some other urethanes.

In the two part systems the resin is mixed with a curing agent.  After a given period of time the material hardens.  This hardening is the result of "curing" or a chemical reaction taking place.  Most of these systems will give off heat.  Care must be taken because in some instances considerable heat can be produced if large volumes of the material are mixed and poured at a time.  Two part bonding systems can be obtained that offer from a few minutes to hours of work time before they begin to cure and can no longer be moved safely.  Some of these are used to form such features as rivers, lakes and streams.  Their color ranges from water clear to dark amber.  Most two part systems can be pigmented using small amounts of oil based colors to blend in or add color to their application.

Catalyzed systems contain unstable chemical combinations that cure in the presence of some "catalyst" or trigger material.  For most that the modeler would use, the catalyst is water.  That is the reason that "ACC" reacts so fast when you get a bit on your fingers.  The moisture from the skin is sufficient to cause the chemical reaction.  For that same reason, it is important to keep such materials securely closed between uses.  Because the air also contains water in the form of humidity, it is a good idea to buy these materials in small containers.  In that way, if some moisture enters by way of the air drawn back into the bottle, only a small portion is lost.  Most modelers have had the experience of forgetting to put the lid securely on a bottle at one time or another only to find it rock﷓hard in the bottle later. 

If the cost is such that larger containers are preferred, on a dry day, transfer the material to smaller new and clean containers.  Do not use containers that were used previously as even a very small amount of cured material will be enough to cause fresh material to cure as well.  After sealing the small containers well, seal each in plastic bags such as zip﷓top freezer bags.

Some precautions must be observed when working the with two part systems as well.  Always use new and clean cups for measuring and mixing.  Do not place any item into the bulk containers that may have material that has already been mixed or cured material.  Clean off the container top with a fresh clean paper towel immediately after pouring, replace the lid, then throw the towel away.  Don't keep it around to be a possible source of  contamination later.  Trash it now.

The next consideration that must be given is the type of joint to be formed.  What must be considered is the strain, if any, that will be on the joint.  By far the weakest joint is the butt joint.  This is formed by merely bring edges of two parts together.  The strongest is the lap joint.  There are variations of both that can offer the modeler alternatives to fit the application.  Figure # 2 shows the basic joint types, butt and lap.  The butt joint is reasonably strong when the forces are  vertical, however it may not stand up well to lateral forces.

Figure # 3 shows the directions of loading on the two basic joints and their strengths.  You will notice that the lap joint has little weakness relative to direction of forces.  When the two are combined in the reinforced butt joint, a very strong joint is attained.  There are a number of methods of reinforcing that will be taken up next time.  The main idea is to achieve as much surface area as possible in the joint area to obtain a strong bond.  Thin sections such as walls are difficult to join using butt joints.  When a reinforced butt joint is made such as the one shown in Fig. # 3,  the strength increases appreciably.  If a wall section of 0.100" is desired, start with the center section measuring 0.050" and the reinforcing layers each 0.025".  The thickness would be the same as for one that was made up of pieces of 0.100" thick but with much more strength.  Yes it does require more time to make, but the end result is usually much more desirable.

Next time we will continue the discussion of bonding agents for plastic.   We will talk a bit more about reinforcing joints and also about some of the methods of preparing the areas to be joined.  With those topics covered we will begin the discussion about the bonding agents themselves by discussing their properties that make them the choice or not the choice and some methods of best using them. 



Part 3

In resuming the discussions on reinforcing joints, a good point to remember is that the key to making a strong joint is to gain the maximum amount of surface contact at the joint area.  If the joint contact surface area is relatively small, it is advisable to reinforce it in some manner.

The most common means of reinforcement is by lamination, the process that was touched on near the end of Part II.  We saw that lamination can be used to extend the bond and thus spread the area of the joint many fold.  Lamination works well where there are edges or ends that are being joined edge to edge or end to end. 

If a piece is to be joined to another in the middle, a new problem is posed.  Lamination is not possible, but reinforcement is.  Diagram # 1 shows two joints of this type which are sometimes called "tee joints".   Fig. a. shows the joint without modification.  Slight up or downward

pressure would cause the joint to break with little effort.  In Fig. b., the joint has been reinforced with sections of angle.  The angle increases the contact area of the joint by 3 or 4 fold for each side that it is added.  This method of strengthening is very efficient for joints that are hidden. 

Exposed joints reinforced in this manner may be considered an eyesore because of the added material at the joint area.  There is a reasonable alternative.  This is called a joint filet.  Diagram # 2 shows such a joint.  To make a filet joint requires two or more steps to complete.  First, the two pieces are joined using the appropriate bonding agent.  After the bonding agent has cured,  applications of a product such as epoxy are  make to build up the filet.  A filet can be made on one or both sides of the part being joined.  It is best to build up the filet in layers since the material being used will be somewhat free﷓flowing and may run if too much is placed at the joint at one time.  The filet can be built up a bit faster if pieces of tape are places across the ends and the two parts rotated to form a channel in which the material can be formed. 

At times, it is desirable to be able to remove one part from the other.  An example of this might be the floor between the first and second floors of a structure.  Instead of attaching the floor with a bonding agent pieces of angle could be placed  around all the walls to form a support for the floor above.  The angles then form a supporting ring for the floor.  One trick that can be used to form the support for a floor and provide a duct for wires, is to use a section of hollow tube slit lengthwise.  This is attached to the walls using a bonding agent as shown in Diagram # 3.  Because of the curvature of the section being used, a hollow is formed behind it where wires may be placed for such things as ceiling lights and other uses.

Useful hint: It can be desirable to electrify a structure without having wires that can be seen.  Several of the methods discussed will solve that problem.   Use thin gage brass angle for reinforcement at joints.  Before attaching in place, solder a fine wire to the ends in an area that cannot be easily seen.  If forming a lamination for a wall or floor, use thin brass sheet or strips in place of some of the plastic or other material.  Attach wires in areas that are not visible.  Multiple circuits can be formed using this method as is shown in Diagram # 4.  The strips need not be all aligned in one direction.  If desired, a tab can be made by making the strip of brass slightly longer and then folding it over at the top and/or the bottom to provide a means of circuit continuation.


When using bonding agents rather than solvents for lamination thin materials can be joined together without the problem of the solvent attacking the thin sections.  Some of the solvent adhesives can dissolve to a depth greater than the thickness of the material being joined.  This is caused by solvent being trapped between the layers without a ready escape to the outside air.  The only challenge to the modeler then, becomes keeping it flat while the bonding agent cures.  

When considering reinforcement of joints, keep in mind that the reinforcement may be placed inside or outside.  The outside of a wooden structure may have a vertical batten that covers the corners of the siding.  This could be formed using plastic or brass channel bonded to the siding.  It adds to the strength to the corner and still it aesthetically pleasing.  The insides of corners are the best places for reinforcement.  Consider also plastic or brass angles attached to the base for a structure.  If they are positioned properly they can serve as locators to hold the structure in place without its being mounted permanently.  This allows the later addition of features, such as interior detail, and also maintenance such as replacing lamp bulbs.


Bonding Agents

Bonding agents come in a variety of viscosities (thicknesses).  Each being  suited for certain applications.  So, too, there are a wide variety of curing rates for the agents from a few seconds to days depending on their thickness.  One of the things that is almost universal, is the ability of the bonding agent to cling either through chemical bonding as was discussed earlier, or mechanically to form the joint between parts. 

There is strength at the bond, but the bonding agent itself in many situations is much weaker.  Therefore, a thin film is preferred as might be indicated on the instructions for its use.  Diagram # 5 shows this.  Since the strongest joint is one with only a thin film of the bonding agent, it is best to have the pieces to be joined as smooth as possible, or have the surfaces to be joined conform to one another.  Some "ACC" comes in a "gap fill" variety that is very thick.  This will form a bond, but the question of the strength of the bond is pertinent.  It is much better to work the surfaces a bit using sandpaper or other tools to the point that only a very thin gap is present. 

Of the common bonding agents used by modelers, the one most widely used is "ACC".  This material, while available in a variety of forms, is not the best for many applications.  When cured, this material is very brittle.  Slight flexing can cause total failure within the bond area.  The material is also prone to failure if either of the joined members are subject to mechanical shock such as cars or locomotives coming together when coupling.  Many a coupler pocket needed to be replaced after a hard couple.  Because of some of the components of "ACC", it can also cause problems with paint in the area.  As it cures, it can cause a white film to develop on adjacent surfaces to the joint.  The "ACC" can out﷓gas, that is, give off gases, and discolor paint for some time after curing.  It  offers a quick bond, but not necessarily a good one.

A better choice for bonding is epoxy.  This material, while it to has its shortcomings, is better suited for modeling use.  Most epoxies will withstand a small amount of flex and shock without failure.  Because epoxy itself is a plastic, it is much stronger away from the bond area then many other bonding agents.  None of the epoxies have the cure time of "ACC" but there are some that offer cure rates less than an hour. 

Epoxies, like polyesters and acrylics, are usually available as two part systems.  Some come in dispenser packs that will dispense the proper mixture at the push of a finger.  Others require measuring.  When mixing the two part systems it is best to mix gently so as not to whip air into the blend that can weaken the material.  Since most of the two part systems are water clear, it is usually easy to watch the blend as it is mixed.  As mixing is started, streaks will appear.  The material should be mixed until all of the streaks are gone.  Make sure to scrape down the sides several times and also the bottom.  

Many modelers find polyesters and acrylics objectionable because of their strong odors.  They also lack some bond strength.  Most urethanes and silicones are catalyzed and require only the moisture in the air to cure.  They offer good adhesive properties, are strong even when slightly thicker, make good cushioning materials, and give adequate work time for positioning.  They are most often used for bonding materials other than plastics.

In the next installment we will discuss some techniques to get strong joints by forming the joint areas and other preparatory methods.  See you then.




Part 4

A bit of preparation before applying the "glue" to a joint can make the joint very strong.  There are two primary types of preparation that can be done to improve joint strength, joint design, and surface modification.  The first, joint design, is most often the primary consideration when working with plastic.  Surface modification is much more important when working on other types of porous materials such as paper, cardboard, wood, and plaster. 

A good joint is one that will take the forces exerted on it and remain stable.  As was discussed previously, some joints, by design (butt), are fragile while others (lap) are stronger.  Still others can be designed to increase the strength even more.  Consider the two coupler pockets in Diagram # 1 for a few minutes.  One has all indications of being able to withstand severe abuse while the other is questionable.

Fig. (a.) shows the method most often used to attach coupler pockets to the underside of car bodies.  The pocket is attached directly to the bottom with a drop of ACC.  Because the ACC is so brittle, any hard shock to the coupler will be transmitted back to the pocket and very possibly cause it to break free.

Fig. (b.) shows a better way of attachment.  Use a welding agent to attach the coupler pocket and a small piece of rectangular styrene stock to the floor behind the pocket.  This will act as a reinforcement to the joint and also serve as a stop.  Before adding the coupler and the cover, drill a small hole in the front section and insert a small round piece of stock through the top of the coupler pocket and into the floor.  This, too, will serve as a stop and will absorb some force if exerted from the coupler end or either side.  Another method would be to select a piece of round stock that just fits into the hole in the coupler pivot post.  Drill a hole the diameter of the stock through the post and into the floor.  Insert the stock using either weld or bond agent to permanently attach it.  This, too, will absorb much of the forces that may be exerted on the coupler and prevent them from causing damage.

Many structure kits are being introduced in cast materials such as urethane or polyester.  Many modelers are also using materials such as epoxy and other two part systems to cast their own components.  Many of the materials used offer some challenges to good adhesion.   Wall, floor, and other sections can be made to fit together and bond well if a step is placed at the mating edges.

A step is cut into each section to be joined.  A bonding agent, such as epoxy or even ACC, is applied and the sections joined.  The step serves as a stop very similar to what was done in the discussion above on coupler pockets.  Regardless of whether the force is applied to the outside of the horizontal or vertical pieces, it will press one part into the other.  This serves as a stop that will prevent breaking the bond.  Some kits come with mitered angles on the edges.  These, too, can be modified to be similar to the stepped corner.

Diagram # 3 shows a typical mitered corner.  Two pieces of plastic stock have been added, one rectangular and the other square.  Both could be rectangular if desired.  The only reason the

square one was chosen, was to reduce the amount of materials used and to aid clarity.  Both pieces are bonded to each other and the adjacent panels.  This serves much the same purpose of a fillet.

Structures usually do not present the challenges to the modeler that rolling stock does, because the stresses exerted on the structures are minimal compared to the rolling stock.  Because of this, it is usually not necessary to go to the extent of reinforcing corners near as much with the structure.  However, there are some exceptions.  One is small details extending off of flat surfaces such as chimneys, posts, and other details.  One is often prone to just glue them to the surface rather than integrate it into the surface onto which it is attached.  Diagram # 4 shows such an installation.  Notice in Fig. a. that the stove pipe was attached to the outside of the building.  There is considerable stress at the point of attachment, not unlike a butt joint.  In Fig. b. the lower section of the pipe was inserted through a hole in the wall and a weld or bond agent used to secure it in place.  In this Figure, the strength of the material forming the pipe caries the load instead of the joint.  In some instances model kits contain parts that have small pins molded off the end to serve as a locator and attachment means into a hole.  The small diameter of the pin makes it very fragile and prone to breakage.  The solution is to open the hole up to accept the full diameter of the part being attached as shown in Diagram # 5.

Do not be too critical on the size of the hole matching very closely the size of the piece going into it.  It is better to have just a very small bit of clearance so that the bonding or welding agent can do its job.  Too tight of a fit can prevent good attachment.  When joining walls the modeler relies mostly on finger pressure or some small clamping devices to hold the joint together as it forms.  When a piece such as a smoke jack is placed into a hole, much more force can be generated that will cause a squeezing out of the bonding or welding agent.

The last topic dealing with plastic that deserves some discussion is how to prevent migration of the welding or bonding agent to areas where it is not wanted.  The first consideration is to follow the admonition that says use only what is needed and no more.  Too much is going to squeeze out or run.  Most people, at least, start out by using too much. 

The next consideration is to use some form of applicator other than the container to apply the agent.  This makes it very difficult to control either the amount dispensed or where precisely it is applied.  Many have found toothpicks to be a useful tool.  Two sewing needles that have been aligned and wrapped together using some thread can also serve well.  To prevent the liquid from reaching the fingers because of capillary action, make a handle of a piece of runner from a plastic kit or a thin piece of wood dowel.  Cut a section off and  drill a hole into which the two needles will snuggly fit.  Glue them into place using ACC.

Make up applicators from several different sizes of needles.  Work with them on some scrap material to develop a feel for using them and a knowledge of which work best for a given size and style of joints and different types of bonding and welding agents.  No one is going to be the best for all applications, but, with a bit of trial and error on scraps, the proper ones will be identified.

Many bonding or welding agents, because they are very thin, will migrate because of surface tension to areas on which they are not wanted.  Some of this is because too much was applied, but there is still the surface tension to consider.  A small chamfer, such as shown in Diagram # 6, can prevent migration from continuing beyond the joint.  The chamfer should be on all possible sides at the joint to best control the liquid. 

This pretty well brings us to the end of the story on plastics.  We have covered a number of topics that I hope have been of some value to you, the readers.  I know there are going to be some who are already aware of what was presented and more.  For those, the articles will have been of little use.  I say this to them.  Use the articles and your knowledge and talents to aid others.

The next time may be the last in this series.  It will address joining materials other than plastics.  These can include such things as wood, plaster, metals, fabric, paper, and other flexible materials.  I say it may be the last, because at the end I usually try to answer questions that have been posed.  If anyone has questions, I encourage you to write, phone or e﷓mail me so that I can address them.  Until next time, continue your trip down modeling lane.



Part 5

This time we will discuss joining such things as ground covers, ballast, and loose granular  materials and plaster.  These have been chosen for coverage together because the ground covers most often are glued to some form of plaster that was used to form the terrain of the layout.  The glue can be applied to the plaster or applied in it as will be seen in the discussions.

 White glue is usually a bit cheaper and has a bit wider application for modeling than does yellow glue.  Both are widely available and are water suspensions of various resins and tackifiers. They remain fluid as long as the level of the water remains at a minimum level.  Once it drops below, the glue begins to set up and will cause plugging in dispensing nozzles and tubes. 

White glue may be thinned with water to improve viscosity almost to the point of being able to be sprayed. This can be of benefit when gluing "loose stuff" such as ballast or ground covers.  When diluting white glues, it is best to add at least one additive to the glue solution that is a surfactant or wetting agent.  The purpose of the surfactant, a couple drops of liquid dish soap will do, is to reduce the surface tension of the solution so it covers the areas into which it comes in contact and gains a better bond. 

Diagram # 1 shows the effects of the surfactant.   With it the drop very quickly spreads out and not only covers more area, but also will travel into small pores in the materials to gain better adhesion.  Without it, the drop of glue will, for the most part, just sit on the surface.  While this may be OK for some projects, it is usually not the desired result.  Small objects that have only slight contact with the glue may be easily displaced if bumped later.

When thinned glue with the surfactant comes into contact with a pile of loose material such as ballast, it will literally climb up between particles while wet to make a better bond. 

When the white glue is incorporated into some systems, it can be coaxed to the surface while still wet to bring about adhesion to materials applied over the top.  Dia. # 2 shows this action.  A layer of material, such as PermaScene™, containing some white glue, is laid down over layers of foam board.  The PermaScene / glue system is troweled over to form the contours.  While the PermaScene is still wet, sifted soil or ground foam is dusted over the surface and sprayed with water to which has been added a couple drops of soap (surfactant).  The glue will be drawn up from the PermaScene and will hold the foam in place. 

The advantage of drawing the glue up from below is that it will not coat the top as it would if sprayed or drizzled over the top.  The ground cover will have a bit better texture this way.  When the glue is drawn up from the bottom, the lower areas of the foam particles will be better coated also assuring better adhesion.

The yellow glues are designed to withstand a bit more humidity than white glues while still being water based.  Because both are water based, they offer relatively easy cleanup if it is done soon after working.  If left to cure, it will take considerable work to displace.  It, like the white glue, can be thinned by the addition of water.  Because it can withstand humidity better than white glue, it is the glue of choice when attaching structural members, but it does have some drawbacks for hobby purposes.  One is that it does not dry as clear as white glue.

Plaster to plaster attachment can be a problem if the two surfaces are uneven or the plaster is somewhat soft even after it has cured.  To have the best possible chance of getting a good strong bond, prime the surfaces to be joined with a slightly diluted white glue.  This does two things.  First, some of the diluted glue will seep into the pores of the plaster and strengthen it near the bond area.  Second, it will provide a smoother surface that will yield more contact area with the mating surface.

Many modelers use molds and plaster to form rock castings to be used on the faces of mountain sides and even cuts along the railroad rights﷓of﷓way.  Some wait until the plaster is just firm but not yet set up and then remove it from the mold.  They then immediately press the still wet castings to the surfaces to which they intend them.  The author prefers to allow the plaster to harden before attaching.  After removal, the castings can be glued into place using white glue as noted above.

By waiting for the plaster to harden, there is less probability of cracking the casting.  After the plaster is set, it can be sculpted using various blades and gouges while lying flat on a firm surface such as a table top.  This also allows multiple sections to be fitted together with lines running across several sections instead of trying to work on vertical surfaces that may or may not be real convenient to reach.

One of the many methods used to decorate plaster can dictate what method is used to glue the pieces of plaster to vertical faces.  Some people like to use diluted water based paints or stains to provide a base color from which to build when using plaster for their mountains and hills.  The stain will not penetrate any area where the white glue has been applied because the surface has been sealed.  This, too, is a good reason for attaching cured plaster rather than still soft sections.  The cured plaster can be pre stained and then glued into place.

Attaching items such as plastic, metal, and glass to plaster can be frustrating.  Most of the usual welding and bonding agents work poorly on  plaster.  One good example of this would be when using plastic retaining walls and tunnel portals.  Most modelers can get a fairly good job initially, but over time the plaster moves because of the movement of the wood supporting it and the two, plaster and plastic will separate.

There are, however, alternatives that allow firm attachment without using mechanical fasteners such as screws or nails.  Try painting a coat of white or yellow glue on the surface of the plaster to which the metal or plastic is to be attached.  Use a first coat of 50:50 white glue and water.  Allow this to harden a day or so and then apply a second coat of full strength white glue.  To attach the plastic, metal or glass to the plaster, use a medium viscosity ACC.  Try to use a minimum amount to obtain the strongest joint.  The white glue primer provides a smooth surface to which the ACC can bond.

Another method that can be used very satisfactorily when bonding porous materials, such as plaster or even wood, is to use an adhesive material that will remain a bit flexible.  Adhesives, such as panel and dry﷓wall adhesive, contain synthetic rubber based materials to provide some give when there is movement between the parts.  These materials also offer a much higher degree of resistance to mechanical shock.  That is why GOO™ is so much sought after for attaching weights inside covered hoppers and box cars.  The adhesive will withstand shock and minor flexing that occurs through handling and running.  Unlike a more brittle material, such as ACC.

Plaster, after staining, can be attached using thin applications of the panel adhesive.  Another advantage of the panel and dry﷓wall adhesives is the availability of colors such as browns, tans, and grays that will allow using the adhesive for more than just an adhesive.  It can also be used to fill in joint areas between casings while providing more natural colors.  When the adhesive has set, most can also be painted.

One modeler I know built some really fine looking mountains with strips of foam board.  He joined the strips together using panel and dry﷓wall adhesive.  To the top and sides he attached  pre﷓stained plaster castings made in rubber molds.  These castings were all attached using the same adhesive used to join the foam. 

He only used  3 or 4 molds to make about 40 of the castings.  Because the castings were rotated, contoured, or even broken into sections, the overall project went quickly.  At the same time, it yielded a result that looked like it was built from scratch using many different textures on a hard shell.  Between some of the castings, the panel adhesive is forming bulges or overhangs that are natural looking.  With some additional color washes, it looks very realistic.

One word of caution should be presented here, DO NOT USE PETROLEUM SOLVENT BASED ADHESIVES ON FOAM.  Some will literally dissolve the foam before your very eyes.  Read the label to be sure that the adhesive can be used on foam before taking the plunge and having to start over.  It may also be advisable to try a bit of the adhesive on a piece of scrap to be sure.

The next installment will deal with wood and some metals.  As we near the end of this series, I hope that modelers have been able to use as least a few of the things presented.  Please address any questions you may have to me through your editor that I may cover them in a question and answer forum at the end.  Until next time, keep on modeling.

"This article may be copied for personal use by the modeler without prior permission.  It may be reprinted in other NMRA Region or Division publications without prior consent of the author if: a.) it is published unedited in its entirety and b.) 5 complete copies of the publication are mailed postage paid on the publication date to the author at: 6075 Spires Drive, Erie, PA  16509-3459.  Any other use requires prior permission."

Richard C. Roth, Ph.D.
6075 Spires Drive
Erie, PA   16509-3459
Phone: (814) 868-5147
E-mail #1:
E-mail #2: