Print

Email

Heads Up - Rebuilding Aluminum Cylinder Heads

By Dick Schaffner

As you’ll remember, in our first of four articles (see Automotive Rebuilder, March 1999, page 32), we started to review how the staff at Aluminum Head Service (AHS) performs the necessary steps in remanufacturing aluminum heads. Items covered included: teardown, selecting a cleaning process, and component part inspection.

Teardown was decided to be a very important step in the process. During this operation, potential defects which could render the head irreparable are to be looked for. We saw that selecting a cleaning process depended upon the severity of build-up and the particular head being repaired.

The processes used at AHS include aqueous jet washer with special cleaning powders, an oven system, glass beading, and a parts tumbler with clear mineral spirits. During the component part inspection process following cleaning, we addressed identification of defective areas and additional techniques used by AHS to bring about a meaningful inspection. In this and the upcoming two articles we will cover the following areas: pressure testing castings; checking tolerances; making a repair estimate; and the actual repair of the cylinder head.

Pressure testing is an important procedure required to rebuild aluminum cylinder heads properly. Without a pressure tester, a cylinder head cannot be completely inspected for cracks. There are other ways to check for cracks, but in our opinion none are as accurate as the pressure tester.

Some shops, (including ours) use a spray-on or brush-on dye penetrant that will enter and stay within the crack when excess dye is wiped off the casting. A developer is then sprayed on to reveal any cracks. This is not an alternative to pressure testing, but an additional tool to help find the beginning and end of a crack which may be in a difficult place to locate.

The following are the methods used in pressure testing cylinder heads. We inject 30-40 psi air pressure into the coolant passages, blocking off the ports to the manifold and block. This is more than adequate pressure as most cooling systems run between 15-18 psi. We have fabricated slates with rubber backing to facilitate pressure testing heads which we frequently remanufacture. These plates have the combustion chamber exposed in order to easily inspect this area. We use an aqueous soap solution (usually dish soap and water) to spray the head. Some pressure testers have a tank in which to submerge the head. Cracks that leak air pressure will usually cause bubbles to appear.

There are several companies which make different types of pressure testing equipment and fixtures (see Automotive Rebuilder, March 1999, Pressure Testing Equipment Profile, page 18). We have used three types of pressure testers over the years, and have learned which ones work and which ones do not. The first machine we tried had a plate with head bolt holes in it and used long bolts and nuts through the cylinder head bolt holes. This required many plate configurations and 15-20 minutes per head to test.

The latest pressure testing fixture we use is a four-post design with lever action pressure (no bolting). This latest unit requires 2-5 minutes of testing time per head. We have used several designs and have decided that a quicker set-up time will increase productivity exponentially as most heads get pressure tested several times throughout the rebuilding process.

The most common area where cylinder heads crack is in the combustion chamber, usually between the valve seats. Of course, this varies depending on the type of head. Other heads crack on the top side (valve spring side) out of the casting plugs, guide boss or cam journals. Less common areas to crack such as spark plug holes on the outside of the casting near a freeze plug, or head bolt holes, should always be thoroughly checked.

Bolt holes for manifolds, housings or brackets sometimes are drilled through to coolant passages. Make sure these studs or bolts are properly sealed with an appropriate sealer such as pipe compound or a pliable sealant. Cracks on the face of the cylinder head (head gasket surface) are sometimes hard to detect because the pressure test plate covers them, so be sure to inspect the head for surface cracks. All cracks should be clearly marked with a permanent marker or wax crayon for future identification.

At this point we know what it is going to take to repair the casting, and we make a determination to fix the head or not. The decision is based upon how much work is required, availability of a replacement casting or core, and time required for rebuilding.

A repair estimate can be made if the head is a customer’s, rather than an inventory item. Areas that should be considered are cracks in the casting, warpage, camshaft journal alignment, valve guide work, valve seat condition, and condition of the parts, such as valve springs, keepers, rocker arms and shafts, and camshaft. We recommend bidding the job a bit higher than actual cost to allow for any unforeseen factors. This also will make the customer happy if the job comes in under, rather than over bid.

In order to prepare the head for repair it will be necessary to determine if peening or cutting out the crack will be utilized. If peening is in order, a peening tool will be utilized to seal the crack by moving material into the crack area and peening that material such that the area is completely sealed. This is done mainly for cosmetics.

Once you have located cracks that you want to fix, you need to decide the method that you want to cut them out with and where to start. If the crack is between the seats, you will need to remove the seats first. You can do this by using a cut-off wheel that goes down through the seats. Then use a small chisel to go down on either side of the cut you have made at the bottom of the seat to pop the seat up and out.

A TIG welder may also be used to remove seats. Set the TIG to weld steel. Weld half way around the seat; if you weld all the way around this will not work. You may have to go back and fourth once or twice. This will heat the seat so that the seat will shrink. When it cools you will be able to work the seat up and out of the hole with a 90-degree pick or similar tool.

The other way to get seats out is to use a seat cutter. We use an undersized seat cutter to perform this removal process. A thin piece of a seat will be left around the hole. Use a little punch or chisel to break out the thin piece that’s left. It won’t take much; the piece of the seat is only about .030" to .050" thick. You could also use an adjustable flat cutter to cut the seat out. The key here is to not remove any aluminum. You don’t want to open up the seat pocket yet.

Now you should be ready to cut out the crack. We use an Enco vertical mill with a ball nose end mill. The ball nose end mill is good for this because it leaves a rounded cut at the bottom of the hole. You can use a flat nose end mill, however this causes more work when you start welding because it’s harder to get that 90-degree angle at the bottom of the hole to flow.

Once you think all of the crack is cut out you can check this with the TIG welder. You don’t have to preheat the head for this step. If you run the torch across the area that has been cut out this will tell you if the whole crack is gone. If the crack is not gone it will split apart slightly and blacken a little. You can also do this step before you start cutting to see the crack better.

If you have a crack on the top of the head you can use a drill and a drill bit. The size of the drill should be around one-quarter of an inch. Determine the start and ending points of the crack, then drill a hole at those points. Follow the entire crack, drilling holes right next to each other. Next move the drill bit back and forth along the crack to open and connect the holes. You can check to see if the crack is gone by checking it with the TIG welder as we discussed before.

After the crack has been prepared for welding, glass bead around the area, so it’s clean and free of any oil and antifreeze deposits. Areas in the seat pocket may have to have guides driven down or out of the head prior to welding. Check to make sure there are no surface dings and/or water etching.

If there are any surface dings, most of the time you don’t have to grind them before welding, as long as there is no debris in the ding. In the case of water etching, grinding will be necessary in these areas to get the antifreeze deposits out of the aluminum. If this step is not performed, welding will be difficult because the deposits will float up in the weld and make it tough to weld.

The head should be preheated to 450°F. A hot plate may also be used to heat the head. The oven works well because it heats the head evenly, instead of just heating one surface, as with a hot plate. The 450°F temperature is one that we have found works best for welding. It makes the aluminum flow much better. It also relieves some of the welding stress so that the head doesn’t crack more.

First, take the torch and flow the whole area that you have cut out. This will make sure that the surface is clean and there are no more cracks. After flowing the area you can start adding aluminum. Add a little, then flow it so all the edges are smoothed. When this is done, we have found that it makes for a much better weld. There are less chances of leaks and more control of the aluminum and where it goes.

Planning out the weld before starting will save time in future steps, such as knowing where and how much aluminum will be added. If there are water jacket holes on the surface you can leave the holes open so you don’t have to go back later and re-open them.

The previous way that we straightened heads was to bolt them down to a plate made of a thick piece of steel that had been surfaced flat. There were two holes in the center of the plate so that the head could be bolted through the center head bolt holes. On the ends of the head between the head and the plate, we used .010" shims to space it up. We then tightened the head down with bolts in the center. The head was baked on the plate in the oven at 450°-500°F for six hours. After it cooled down, the head had minimal warpage. Most of the time align boring the head was still required to make the camshaft fit correctly. This process obviously took much time.

The current way we straighten heads is with the TIG welder. This works on most heads. A head which needs straightening will have warpage and cam rock. Flow the aluminum on the flat spots under the valve cover rail first, both on the inside and outside of the head. Don’t add any aluminum, just flow what is there. The amount of flowing needed depends on how much warp the head has endured.

You may have to flow more spots on one side than the other. The spots on each side may not be the same size, but you’ll want to do it on both sides of the head. You’ll have to experiment with this process at first. After welding a little on the head, let it rest for a few minutes so the heat will dissipate throughout the head. Then try the cam in the head. If it still has rock, weld some more, then try the cam again.

After doing a few heads, you will get a feel for how much welding and where you need to weld a head to get it straight. If you happen to over straighten a head, this is easily fixed. The areas that have been welded on the head can be peened using an air hammer or a peening tool. This will push the metal out and the center back up.

When a head is welded to straighten it, the aluminum is super heated so when it cools, the aluminum shrinks and pulls the ends of the head back up to eliminate cam rock. The head usually doesn’t have to be align bored because you’re straightening the head to the cam journals.

In closing, let’s recap what we have covered. The best pressure testing fixture we have found to date is a four-post set-up with lever action. It’s very important to identify cracks during pressure testing in order to facilitate crack removal. We saw that planning how the weld is to be completed will make the process more successful. And, finally, we addressed the technique used at AHS for straightening aluminum heads.