Automotive Bodywork: Repairing Collision Damage in a Decklid

This job is to repair the lower area of the rear decklid from a 1937 Chevrolet sedan. The original panel was punched in by impact from another vehicle hitting it from the rear. Most of the damage was to the right side of the decklid, but some damage also occurred to the center and on the near left side of the lower panel.


This Tech Tip is From the Full Book, AUTOMOTIVE BODYWORK & RUST REPAIR. For a comprehensive guide on this entire subject you can visit this link:


SHARE THIS ARTICLE: Please feel free to share this article on Facebook, in Forums, or with any Clubs you participate in. You can copy and paste this link to share:


The damage amounts to a series of related and unrelated deformations in this decklid’s skin. This combination results because most of the damage was indirect; it was forced into the metal by direct impact against the heavy trim pieces attached to the panel. The impact from the second vehicle was primarily against the trim, which in turn pushed the decklid skin in from several points of the trim’s attachment to the panel. In some areas, one dent created or modified another. Some areas of the damage are quite unrelated to each other.


This is the starting point for the metal straightening, finishing, and filling job. It amounts to correcting several related and unrelated dents in the decklid of a 1937 Chevrolet sedan. My good friend, Herb Statz, is shown working in many of these photos.

This is the starting point for the metal straightening, finishing, and filling job. It amounts to correcting several related and unrelated dents in the decklid of a 1937 Chevrolet sedan. My good friend, Herb Statz, is shown working in many of these photos.


This photo illustrates a subtle but important point: The innocuous-looking scrape between the two white arrows was the only area of direct impact that deformed this panel. The other damage was done by impact against trim, which also deformed the panel. Understanding this is key to repairing the panel.

This photo illustrates a subtle but important point: The innocuous-looking scrape between the two white arrows was the only area of direct impact that deformed this panel. The other damage was done by impact against trim, which also deformed the panel. Understanding this is key to repairing the panel.


As hopeless as it may look, always remove body gaskets and trim carefully, as if you will have to reuse them, because it may just come to that. If you don’t have new items in-hand before you remove the old ones, keep the old ones for patterns, just in case.

As hopeless as it may look, always remove body gaskets and trim carefully, as if you will have to reuse them, because it may just come to that. If you don’t have new items in-hand before you remove the old ones, keep the old ones for patterns, just in case.


There was one small but particularly significant area of directimpact damage. This small area was hit directly by the second vehicle, and has severe direct damage that is locking in a large area of undeformed metal that will begin to return to its proper shape and position when the direct damage is removed.

After removing the trunk lid’s trim and body gaskets, we obtained a better idea of the type and extent of the damage. Damage is revealed by close inspection, and that inspection is aided by removing anything that interferes with viewing and feeling the damage as minutely and directly as possible.

This job divides into three major areas of damage, and a few smaller ones. First the trunk lock and associated trim pushed in the panel metal under them when they were hit. That damage was evident on both sides of where the latching mechanism was mounted.

Second, the damage traveled upward, pushing in the license plate bracket, which may also have sustained some direct impact. That deformed the trunk panel into a second area of damage that extended up to and slightly beyond the top mounting point for the license plate bracket.



These two views of the damage illustrate how different things can look when seen in different light, and from different angles. Only by examining damage from several viewpoints, and manipulating the light on it, will you fully understand it.

These two views of the damage illustrate how different things can look when seen in different light, and from different angles. Only by examining damage from several viewpoints, and manipulating the light on it, will you fully understand it.


This view and lighting shows the nature of the damage particularly well. Three distinct damage areas are easily visible. The direct damage point is obvious, as are the deformations caused by the trim and bracket items. It is easy to see where and why the damage stopped moving through this panel.

This view and lighting shows the nature of the damage particularly well. Three distinct damage areas are easily visible. The direct damage point is obvious, as are the deformations caused by the trim and bracket items. It is easy to see where and why the damage stopped moving through this panel.


The substructure under this decklid panel severely limits access to the panel’s back side, making it much more difficult to work on. Actually, there is great rear access to many areas that do not need attention, and very poor access to most that do.

The substructure under this decklid panel severely limits access to the panel’s back side, making it much more difficult to work on. Actually, there is great rear access to many areas that do not need attention, and very poor access to most that do.


A third, separate, area of major damage was evident on the lower right side of the trunk. This was caused by direct impact in the center of the dent. It produced a deep and severe rolled buckle, upward from the point of impact. A less serious rolled buckle extends to the right of the direct impact point on the panel. Metal was also deformed and displaced downward from the point of impact. The damage in this dent was stopped by the crown of the metal at its top, and by edge substructure at its bottom and to its right.

As is usually the case with collision damage, there is good news and bad news. The good news is that the supporting structure of this decklid held its shape. There is no overall dimensional deformation in this panel, in the sense that it still perfectly fits in and aligns with its jamb. Its diagonal measurements remain exactly symmetrical.

If there was no substructure behind the decklid skin, it would be relatively simple to rough the damage out of it. Some hammer-off-dolly and hammer-on-dolly work would successfully remove the buckles and ridges, and relieve the relatively small areas of damage that are locking most of the out-of-place sheet metal into its presently deformed state.

Now the bad news: The decklid’s substructure prevents direct access to the areas where almost all of the corrective work needs to be done. The exception to this dismal situation is that there is good access to the back of the center dent’s upper area, behind the license plate bracket.


The Approach

There are two possible approaches to repairing this damage. One would be to cut away enough of the deck-lid’s substructure to gain access to the critical areas on the back of the panel, and then approach the job with conventional hammer-on-dolly and hammer-off-dolly work in those areas. With this approach, after the metal in the panel had been returned to its original format, the removed substructure would then be welded back into place. This way of doing this job would probably be faster than any other way of accomplishing it, such as the one that we decided to use. Also, it would yield a panel surface that would require little metal finishing and filling. The panel surface could be brought pretty close to its original format easily and quickly.

The other approach the one that we used is to leave the sub-structure in place, and to work around it with pry tools and other specialized tools and techniques. This is a much more cumbersome and time consuming way of doing this job, but it has two distinct advantages. First, it maintains the shape and alignment of the trunk lid. If substructure members were removed to straighten the panel surface, it would be almost impossible to maintain the shape of the decklid, and a difficult refitting process would be required. The fact that this decklid is still in near-perfect alignment with its jamb is an advantage that should not be ignored.

The other major problem with removing and replacing substructural members for access is that it would require considerable time to re-weld and refinish them. This would probably offset and possibly cancel out any time saved. Finally, the substructure around the lower and right edges of the panel could not be removed without severely disturbing and deforming the panel in those areas.

For all of these reasons, we decided to work around the sub-structure, rather than to remove it. Please note: The decision to work around the decklid’s substructure does not change the basic approaches to removing collision damage, outlined in earlier chapters. The theories and sequences of damage removal remain the same because the elements of cause and effect in working damage out of the sheetmetal do not change. What does change is some of the tools and techniques used to accomplish the job.


The Early Steps

Coming up with an effective strategy for dealing with collision damage requires seeing the damaged area clearly. In this case, applying some paint stripper to the decklid, and removing the loose rust with coarse steel wool, revealed previously invisible detail in the lower-right part of the panel.

It would be difficult to overstate the importance of using all of your relevant senses to understand impact damage. We’ll pass on smelling and tasting it, but feeling damage and looking at it from many different angles are necessary to gain enough information about it to formulate a good plan to repair it.

Analysis of the damage in the right dent suggested that it was caused by direct impact. A key to unlocking the damage in the entire dented area is to gently work out the crease that is evident at the impact point. Following this, apply gentle pressure against the dented area, while lightly hammering on the ridges that surround it. This will be difficult at the right and bottom edges of the panel because there is supporting structure behind the metal in those areas.

The favorable factor in this dent is that most of the metal in the large damaged area is not badly deformed. It will spring back into its approximately original and correct positions, once the relatively small areas of deformed metal that are holding it out of position are removed.

It is critically important that the attempt to pry the direct damage area out, and to release the metal that it has locked out-of-place, be made with the greatest possible accuracy. A near miss would create a serious problem that would have to be corrected. For that reason, we transferred the exact position of the direct damage line from the front to the back of the panel with a large pair of calipers. Then, we made a chalk mark, indicating the exact area under the decklid’s substructure, from where the prying force would have to be applied.

We selected a suitable body-prying tool, and marked its depth behind and into the substructure hole. We then inserted it through the hole to accomplish the prying maneuver. The whole process was performed as carefully as possible, since one false move could worsen the damage. The initial move in this job was to work out a small but strongly deformed knot of metal, as a first step in releasing the damage. This meant applying considerable but very accurate force to a very small spot. The twisting motion of a pry bar was ideal for this purpose because it confined the applied force to a small area.

Step 1:

1After chemically removing the paint and scuffing the loose rust with coarse steel wool, in the lower-left part of the panel, its surface looked very different. It was then much harder to spot the area of direct impact.


Step 2:

2Feeling damage with your fingers often divulges information beyond what your eyes can see. Tactile data can be as important as visual data in planning approaches to removing collision damage from autobody metal.


Step 3:

3While the outside of this panel revealed what needed to be done to repair it, its underside showed how difficult it would be to gain access to several key areas. It is always best to understand the extent of such problems before you begin to work.


Step 4:

4We started with a plan. First, we would correct the direct damage. Then, we would push and hammer out the long, vertical V-channel above it, while hammering down its rim. Similar strategies were planned for the rolled buckles on the right, below, and to the left of the direct damage.


Step 5:

5After the work was completed, most of the displaced metal in this dent was released. Then, with a little underside persuasion with mallets, the metal returned to normal. The key to this result was in understanding the role of the small area of direct damage, and relieving it.


Step 6:

6Usually, feeling the back of a panel reveals where things are. However, in this case, substructure interfered with feeling the knot of direct damage under it. To ensure accuracy in prying it out, we used a large set of calipers to precisely transfer its position to the decklid’s back side.


Step 7:

7Next, a line was drawn to indicate the exact position of the direct damage, under the sub structure. It is along this line that force was applied with a pry bar, against the decklid skin, to work the direct damage out.


Step 8:

8The prying was done with the bent part of a pry bar, inserted through the sub structure hole in the center of this photo. The pry bar shaft was marked for depth of insertion, indexing it to the left edge of the hole.



Step 9:

9Once the pry bar was positioned, the bending operation was accomplished by twisting it. After several attempts, the chalk mark on the pry bar smeared, but we could feel the proper depth of insertion by then, and did not need to see it.


Step 10:

10It was critical to watch, and to evaluate, the effect of each pry bar twist. A second worker sometimes provided another set of hands, stabilizing the panel, while hammering lightly on the rim of the direct damage dent, as Herb applied prying force to it.


Step 11:

11Once the small knot of direct damage had been pried out, we went to a broader prying tool, to force out the area of metal surrounding it. That look visible on Herb’s face is genuine teeth-grittin’ determination.


Step 12:

12With the locking factor of the metal in the direct damage relieved, a soft-rubber mallet was used to push out the un-deformed metal that comprised most of the area of the dent. Access to this area was refreshingly good.


Step 13:

13Some areas of displaced metal resisted movement by our soft-rubber mallet. A dead-blow plastic shot mallet proved ideal for persuading these areas back into place. Both the rubber and dead-blow mallets were used more to remind the un-deformed metal where it belonged than to form it.


Step 14:

14As the straightening work progressed, Herb kept feeling the surface of the panel to confirm its progress toward its final format. The importance of checking work by feeling it cannot be overemphasized.


Step 15:

15It is always useful to mark areas of a panel that require straightening work. As Herb tapped lightly on the panel’s topside, he felt and marked where he was tapping on its reverse side, identifying an area that still needed to be lifted.


Step 16:

1616 Pry bars are useful for raising broad areas of low metal when you cannot access them directly. Here, Herb lifted a shallow dent. He lightly hammered the ridges surrounding it, on the other side, while lifting its central crease with pressure from a pry bar.


The prying was done very slowly, and in many small, incremental steps. It was critical to watch the metal that was being pried move, from the outside of the panel, to determine when it had been moved far enough to accomplish the first stage of restoring the panel to its original format.

The actual amount of movement in the direct damage area was so small that a photo of it would be hard to distinguish from a photo of the area before prying force was applied to it. Nonetheless, that small movement of metal in the panel released the locking force at the point of the direct impact. Next, we applied broader force with a large, curved prying spoon. This worked out the larger area of metal around the direct impact area.

The back of the panel was then lightly worked with a rubber mallet, and with dead-blow mallets. During this work, a dolly was held against the ridges on the outside of the panel at the edges of the dent. In some cases, the buckles at the end of the damage where the panel finally stopped the deforming metal from advancing during the impact were hammered down lightly, as the panel was reshaped with light pries and soft blows on its back side.

Additional areas of panel damage were located visually, and by feeling the panel, and were marked and hammered out with appropriate tools.

This work requires imagination in how best to use tools in ways that will accomplish your goals. You look at a problem and try to find a tool, or tool combination, that will solve it. We used several different prying tools to remove this damage.

Sometimes, the best tool to do part of a job isn’t in your collection. When that happens, you have to improvise imagine what is needed and then create it.

Sometimes a simple maneuver, like inserting a small block of wood under a spoon and hammering it through a hole in substructure, will accomplish what is needed. Small, specific solutions that accomplish tasks that address the right problems are critical to doing autobody damage repair work efficiently and well. In this work, the simplest and most direct approach is often the best approach.

In a relatively short time, we accomplished a dramatic improvement in the panel. After the initial roughing out of the dents, we felt the damaged areas, and located and relieved a few points that were still locking in displaced metal. Then, we applied mild force to move the undamaged metal out, and into its proper positions. Those few operations greatly improved the panel. However, we were still far from finished with this job.

By its nature, autobody panel work is incremental. It takes many well-planned steps to accomplish goals. Each action should improve the result. Each action, when possible, should be small enough to be reversible because some moves may not achieve their desired results and may need to be undone. Many small actions are often far better than a few big ones.

The area of damage in the right side dent has been greatly reduced. There is some overwork in the area of the original impact damage, where it was pried out, but this can be corrected later.

Next, our priority was to remove the remaining big ridge in the right dent area. This was done by hammering down its locking edges, while applying pressure to the back-side of the V-channel through what access points were available. It would be far better to hammer directly on the V-channel, off-dolly, but limited access prevented this approach.

Similar prying work was applied to the rest of the major dents in the panel, until most of the metal in it was in roughly the correct positions.

Mapping damage helps to keep tabs on the amount of work that remains to be done, and may suggest sequences for attacking it.

The major crease in the right dent area was mapped on the reverse side of the panel, so that it could be driven out by hammering against it on a curved body spoon. It was important to back up the area of the crease to limit the deformation by the spoon that was being hammered against it. This was done by placing and backing up the panel on a resilient surface made up of several layers of corrugated cardboard.

Step 17:

17Inserting this little scrap of wood between the back side of the decklid panel and its substructure allowed two other tools to accomplish a job that they could not have done without the wood block. Because the wood was soft, it did not leave its outline in the metal.


Step 18:

18A spoon was inserted over the wood block to apply force to it. However, using the substructure above the spoon would not work for a prying fulcrum. The structure was not strong enough.


Step 19:

19The solution was to use a rod, and to hammer down on the back of the spoon. A few strong blows were enough to do the job. Three simple tools, used together, accomplished what, in this case none of them could do individually. Step 20: 20 At this point, the panel showed marked improvement. There was some overwork on the direct damage crease, just above Herb’s fingers. And one buckle above that still had to be worked out. The mild dent, below and to the left of the buckle, was pried out easily.


Step 21:

21Working through the substructure, behind the panel, required innovative uses of tools. The action replaced simple blows with a body hammer with very heavy blows struck against a body spoon’s shank. This is not a good situation, but it sometimes comes with the territory of panel work.




Step 22:

23In this view of the panel, you see the ridges and buckles that were still locking what was left of the dent. With good access to the back of the panel, it would have been easy to hammer the V-channels out, off-dolly, while supporting the buckle ridges with a dolly.


Step 23:

24Pressure, applied against the main V-channel in the lower-right dent with a big, curved body spoon, combined with light blows against its surrounding ridges on the panel’s surface, allowed much of the metal locked in this dent to move back into place. Body metal has a long memory. If you can identify and correct the deformations holding damaged metal out of place, you can rely on that memory to return much, or most, of the metal in a damaged area to its correct positions.


Step 24:

25The damage map, chalked on the remaining damaged areas of the decklid, indicates where metal still needed to be raised (minus signs) and two areas where it had to be lowered (plus signs). Areas that are particularly obvious, like the area of the direct damage, were not mapped.

Step 25:

26The center of the major remaining crease was transferred to the back side of the panel’s substructure. This made it easier to locate the crease by feeling for it with the body spoon that was used to work it out.


Step 26:

27A curved, double-ended spoon was chosen to work this crease, because it had the right curvature for the job. To avoid damage to one of the spoon’s working surfaces, a bronze hammer was used to beat on it.


Step 27:

28This small dent, in one of the decklid’s edges, is typical of damage that can be dealt with effectively with the traditional hammer-off dolly approach. It offers perfect access from behind to hit it out with a few blows.


Step 28:

29The dolly was held so that hammer blows moved the metal edge out, as pressure and rebound from the dolly pushed it in. Metal was never squashed between the tools, and no stretching occurred. You could reverse this approach, hammering the bulge down, while supporting the back side with the dolly.


At this point, the major metal moving had been completed, and it remained to work out several small areas of damage, the ones indicated in chalk on the panel surface. A small dent in the edge of the decklid jamb area was corrected with a few hammer-off-dolly blows. The rest of the job was mostly small corrections like this one.

Some of the small-damage areas that required repair had inevitably occurred in the process of prying out the major dents in this panel. If the panel had not presented such severe access problems to its back side, it would have been possible to remove the major damage with traditional hammer-on-dolly and hammer-off-dolly techniques. That would have greatly reduced the amount of small, collateral damage repair necessitated by approaches like prying and spoon hammering that were used to work around the panel’s substructure.

This can be tedious and repetitive work, but each blow should bring the panel closer to its correct shape. It is important to avoid stretching the metal. This means hammering it off-dolly, to raise low areas. If a hammer blow is accidentally struck on-dolly, its sound and feel will announce the mistake. One such accidental blow does little harm.

It is critical to use tools that are shaped correctly for the job at hand. The crowns of hammers must roughly match the crowns of the panel areas that are being worked out, and dollies have to contact the metal that you are trying to stabilize.

Soft tools like rawhide mallets, shot mallets, and hand held shot bags can be very useful in limiting unwanted deformation, as you move metal with impact.

A few areas of the decklid had such deep and locked-in small areas of damage, that no amount of clever tool use allowed us to access them from the panel’s back side. In two such cases we resorted to stud welding to pull out these areas. Working damage out with hammers, pries, etc., is preferable to stud welding, but sometimes you run out of conventional bodywork impact-tool options. Then, stud welding may be necessary. One key to getting good results in stud welding is to pull out the damage in controlled steps, and avoid creating over-pulls.

Stud welds are easily clipped off, and if they are done right, clean up easily and leave no adverse effects in the metal. The area of this stud weld was disc sanded almost flat, and left for final filing and finishing when the panel leveling was completed.

The rest of the job consisted of continuing to pick up low spots and push down high spots. Because this panel had considerable crown in every area of its surface, it was possible to accomplish the entire job without having to shrink any metal. This is because high-crown sheetmetal conceals minor stretches that are not so large as to affect its basic symmetry.

Various pick hammers were the choice tools for making finer adjustments to the decklid’s surface. Hammering was done lightly, and interrupted with many inspections. At this point hammering was almost entirely off-dolly. For particularly delicate adjustments, a plastic dolly was used for on-dolly hammering.

Step 29:

30After more work, the panel was remapped, and looked like this. Visible sanding-board marks helped indicate low and high spots. Note the tic-tac-toe item, top left. I need to state that this kind of fooling around has no place in serious work. By the way, I won.


Step 30:

31As the work progressed, the crowns of the tools used to perform it became more critical. This highly crowned, modified Proto 1427 hammer worked well for raising areas of the decklid that it could reach. This combination dolly had several surfaces that were good for backing up off-dolly hammering.


Step 31:

32With proper technique, hammering off-dolly can be used to move metal thousandths of an inch. If this technique is applied properly and with sensitivity, metal cooperates in seeking its pre-collision format.


Step 32:

33Backing devices like this leather bag, filled with lead shot, were helpful in smoothing out the decklid surfaces, in areas where very minor adjustments were necessary. Shot bags limit metal movement, but are very forgiving regarding exact placement.


Step 33:

34This rawhide shot mallet was great for leveling some areas of the decklid in a gentle way. Here, it is being used against a spoon dolly that is wedged between the back of the panel and its substructure.


Step 34:

35A stud welder was used to pull out a small, but deep, dimple dent that was inaccessible from the back of the panel. This type of device resistance welds a copper-plated steel stud to the panel’s surface.


Step 35:

36While stud welders produce fusion welds between studs and panels, there should be no permanent damage to the panels. When pulling out dents with studs, it is critical to locate the stud(s) in exactly the right place(s). Sometimes, multiple studs are necessary to pull out defects like creases.


Step 36:

37A simple slide hammer was attached to the stud to pull it, and the metal welded to it, out. The slide hammer was used incrementally, to avoid over-pulling and stretching the metal.


Step 37:

38Here is the result of the stud welding operation. The deep dimple, where the stud was welded, was now level with the panel. If stud welding had not been used here, excessive amounts of body filler would have been necessary to fill the dimple.


Step 38:

39After welding, the stud was cut flush to the panel. The heat from stud welding anneals panel metal near the stud. As Herb cut off the stud, he pulled on it and rocked it slightly, to finely adjust the level of the metal around it to the panel.


Step 39:

40The last step in the stud welding operation was to grind the stud stub almost level to the panel. A little of the stud was left for final shaving when the panel was metal finished.


Step 40:

41Because the stud metal is very soft steel, a body file can level it to the panel, easily. Note the burned ring around the stud weld area. This is where the outer rim of the stud welder contacted the panel, and supplied current in the circuit to the stud.

Step 41:

42Using a blunt pick hammer and a plastic-clad dolly, Herb went after the low metal that could be accessed from the back side of the panel. Plastic dollies only work for very limited on-dolly situations, but that is exactly what was called for here.


Step 42:

43We discovered depressed areas, where the trunk lid latching mechanism and trim had pushed in the surrounding metal. Although we thought that we had corrected this problem, we found that it needed more work. We had good access to hammer this area from the back, off-dolly.





Metal Finishing

Metal finishing is the final stage in sheetmetal work, before body filler is applied to a panel’s surface to correct any small remaining inaccuracies. In the best situations, no body filler is used. Metal finishing techniques can only be applied following the completion of all of the major movement of metal in a job. After metal finishing, the metal workers in body shops turn their finished work over to painters and think, or say, “Now, buddy, it’s up to you.”

Up to this point, we have done little more to clean the decklid than to chemically remove the paint from it and scuff off the loose rust with coarse steel wool in the repair area. There were two reasons for not cleaning the rust and paint residues to base metal until this point in the restoration. First, it would have been difficult to use abrasives effectively in the depressions and creases in the damaged metal. Those gross defects have now been corrected. Second, leaving the panel surface uncleaned also made it easier to indicate low and high spots by just dragging a body file or sanding board over it.

The panel was made level enough to clean it down to bare metal. This step was accomplished with a 7-inch disc sander and an 8-inch rotary orbital sander. Note that abrasive blasting was not used to do this.

Any blasting method, or media, that would remove the pitted rust from this panel would also tend to stretch and warp it, unless it was slowed down to the point that disc sanding would be faster. Thus far, we have tried very hard to avoid stretching or warping this panel.

At this point, the decklid’s official portrait shows the metal basically in place, but in need of many small, local surface adjustments. These involve using processes similar to the ones that we have been using, with the addition of filing and sanding the panel to achieve finer adjustments to its surface. In the metal finishing stage, all procedures used to move metal are milder and less violent than those employed in the roughing out and bumping stages that preceded it.

Given the extreme access problems to the back of the panel, we had to use some inventive approaches, like a homemade slide-hammer attachment, to move the metal for metal finishing. The rest of the work was mostly very light hammer work, backed by spoons held against the back of the decklid, and dollies against its outside, as pick hammers were used to raise metal. Pick hammers play a major role in most metal finishing operations. Due to the extreme access problems in this job, they played a more minor role. You can’t pick metal up when you don’t have access to swing a pick hammer.

When the metal work was completed, and the panel was ready for filling, it was given a final cleaning in the areas that had not come completely clean in previous sanding operations. Various tools and methods were employed to remove as much rust as possible, particularly from the pits in areas of the metal that had been covered by trim.

Step 1:

1As work progressed, Herb periodically dragged a body file or sanding board over the decklid’s surface, to indicate low and high spots. This is a very effective way to get a visual sense of how a panel is progressing.


Step 2:

2There are faster ways to remove paint and rust from a panel than disc sanding it. Some of them stretch metal and should not be used. Disc sanding doesn’t stretch metal, and has the added advantage of indicating high and low spots, and even correcting them to a limited extent.



Step 3:

3We used an 8-inch orbital sander, after the disc sander, to get the panel clean and smooth. The orbital sander was too slow to do the whole cleaning job, but it was ideal for use after a disc sander to finish it.


Step 4:

4The decklid still needed many small surface adjustments to make it smooth. Some of this will be done with further metal shaping applications, and some will have to be accomplished with filler.


Step 5:

5This homemade slide-hammer attachment was very useful for working through holes in the decklid skin to raise metal. Its contact point can be slid to where it is needed, and stays there during use.


Step 6:

6The tool is inserted through available holes in the panel. Then, the slide-hammer’s weight is slid up its shaft to provide outward impact against the panel. With this end mounted on it, it has a range of about 31⁄2 inches from the edge of its insertion access.


Step 7:

7The rest of the metal finishing phase of this job was accomplished through repeated applications of the techniques used before, but in finer and finer increments as the panel came into shape.


Step 8:

8Next, it was time to give the panel a final cleaning, before applying metal conditioner and body filler to it. A coarse-wire cup brush was used to dig rust out of the surface pits in some areas of the panel. Note the old lead repair below the wire brush.


Step 9:

9An old-fashioned, carbon-steel hand wire brush was worked into the pits in the metal with a circular motion. This is a very effective way to dig rust out of pits.


Then, the panel surface was blown free of all loose debris and wiped down several times with sol-vent. We ran solvent-soaked towels and drying towels over it until they came up unsoiled.

The next and last preparation, before applying polyester filler to a few areas of the decklid that needed filling, was to chemically neutralize the surface of the panel, and particularly to treat any remnants of rust that might have remained trapped in pits in the metal. To that end, we diluted a commercial metal-conditioner preparation per its manufacturer’s instructions, and scrubbed into the panel’s surface with a woven nylon pad and a stainless-steel-bristled cleaning brush. We worked sections of the panel, roughly 2-feet square, this way, and then wiped them dry, again, as per the metal-conditioner manufacturer’s instructions.

The purpose of applying the metal conditioner was to stabilize the surface of the panel, and to prevent the clean metal from flash rusting. A good metal conditioner does this by depositing a very thin phosphate coating on clean metal. It also stabilizes any small amounts of rust that remain on the panel, and that cannot be sanded off effectively. Metal conditioner must be applied to very clean metal; but it is no excuse for painting over rust. It neutralizes and stabilizes very minor amounts of rust, no more than that.

Finally, metal conditioner gives clean metal tooth, that is, a microscopically craggy surface to mechanically interlock and bond with coatings or filler applied over it.

Step 10:

10The panel was blown with compressed air, to remove all loose rust and debris from it. This small step was absolutely necessary to get good results in this job.


Step 11:

11The panel was then vigorously wiped down with solvent-soaked rags, and dried with clean rags. This process was repeated until the rags came up clean. Solvent wipe down is another small but critical step before filler is applied to a panel.


Step 12:

12The last step before applying filler was to use metal conditioner on the bare panel surface. The metal conditioner was mixed to its manufacturers’ specifications, and applied with a nylon scrubbing pad to 2-foot-square panel areas. In pitted areas it was worked in with a stainless-steel wire brush.


Step 13:

13After the metal conditioner sat on the metal for a minute or two, but before it could dry completely, it was wiped dry with clean rags. These rags did not come up clean, because metal conditioner tends to dissolve some rust as it neutralizes it.



Polyester filler was chosen for this job. We mixed it thoroughly in its can before adding hardener to it, then we mixed the combination to a uniform consistency. Preparing poly-ester filler is not difficult, but it is important to keep things very clean when you mix and apply it, and to avoid all contamination to it by things like oil and debris.

When the filler was completely mixed, we applied it to the panel with a squeegee, in slight excess of the build that we were trying to achieve. Note: Never apply the poly-ester filler in thicknesses exceeding 1/8 inch, overall, and it is best to keep individual applications to less than 1/16 inch.

Unfortunately, some people who do autobody metal work do not understand or observe these limits, and greatly exceed them. The results can be disastrous. Several thin coats of filler always trump one thick one. And total thickness of filler that exceeds 1/8 inch is risky.

Polyester fillers come in many types. All of them have slightly different application considerations. The one used here, Bondo, is very well known and works well. Depending on temperature and mixing pro-portions, it is easy to apply after mixing because it begins to set at a workable rate. It can be spread evenly, smoothly, and in a uniform way. It also provides good adhesion to properly prepared base metals.

Depending on conditions, polyester fillers can be roughly shaped with a cheese grater-type file, for several minutes or more after initial setup. After a little more time, again depending on product, temperature, and mixing factors, it hardens to a point where it can be contoured accurately with a body file.

Next a board sander with a semi-firm backing and 80-grit open-coat abrasive paper was used to further shape the hardened polyester. After a single application of filler, the fender surface was not only back to its original contours, but was also approaching a pretty smooth format.

Our decklid had come a long way since we started working on it, but there were still some low spots in the panel. More polyester filler was applied to those low spots, allowed to stiffen up, and grated to rough contours. Then, the filing and sanding steps were repeated.

After the filler hardened and was grated, we used an in-line air sander, followed by a sanding board, to shape and smooth it. Then, we performed a little, light, detailed hand sanding with a semi-hard rubber pad in a few areas to blend the edges of the filler and the metal.

The decklid was now ready to send out for painting. This panel was somewhat protected from rust by the metal conditioner that we used before applying filler to it, but it was still very vulnerable to rusting. It was stored in a laminated plastic bag with some desiccant dryer, until it could be painted.

The finished panel looked good. It would look better when refinished. Diagonal measurements indicated that it was still perfectly symmetrical, and would fit perfectly into the trunk jamb from which it was removed. The polyester filler was no more than 3/32 inch thick at any point on the panel, and much thinner than that in most areas. This panel should look good and provide great durability for many years.

Step 1:

1Polyester filler was mixed in its can, prior to combining it with hardener. This often overlooked step is necessary to get the best results with this type of product.


Step 2:

2Metal or plastic tools can be used to mix filler. Aluminum foil is an excellent mixing surface for filler. Never mix it on newspaper or wood. Avoid surfaces that can absorb filler components, or that can release contamination like waxes and resins.


Step 3:

3Exact mixing proportions for polyester filler, and its hardener, are not critical with most polyester filler products, however thorough and uniform mixing are. Also, it is crucial to mix these components on a clean surface.


Step 4:

4Filler must be applied before it begins to set up, otherwise it becomes hard to spread and porous. If filler is mixed correctly, there should be plenty of time for its proper application.


Step 5:

5This is an example of what can happen if filler is applied incorrectly. This paint was less than six months old when it failed. In this case, the filler under it was more than 1/2 inch thick. Paint failure was inevitable.


Step 6:

6We used a cheese-grater-type file to roughly shape the hardening filler. This is a crude process, and no attempt should be made to achieve final contours with it at this point.


Step 7:

7There are many ways to shape hardened filler. We used traditional body files, which is a slow but sure way of shaping plastic fillers. Some people use power tools in the first stages of shaping, but I find files are fast enough.


Step 8:

8Filing was followed by board sanding the filler. We could have used an in-line air sander at this point, but we were so close to the desired final surface that we saw no need to use this tool at this point.


Step 9:

9After shaping the filler, we still had some low spots, and areas that needed building up to get the correct, final contours. A second application of filler was made to these areas.


Step 10:

1010 In some areas, our second filler application was much too ambitious, resulting in very high spots of filler. We used an in-line air sander to quickly remove excess filler. This was followed by a sanding board, to achieve final contours.


Step 11:

11A foam sanding pad and 120-grit abrasive paper were used sparingly to blend a few edges where filler met metal. This was a delicate step and we exercised care not to overdo it.


12The finished panel looked like this. A few coats of primer, and some wet sanding, will bring this panel to perfect contours that reflect light uniformly and symmetrically. There is no need for the dreaded spot putty to make this surface work.


Written by Matt Joseph and Posted with Permission of CarTechBooks


If you liked this article you will LOVE the full book. Click the button below and we will send you an exclusive deal on this book. Deal-Button