Hot Rods Surfacing Cylinder Heads With Sandpaper and Granite Surface Block...

I'm building a mild (9:1, maybe 325 hp?) AMC 360 on a budget and would like to surface the cylinder heads in my shop with granite surface plates. The heads are slightly warped - .0025" on one and .004" on the other (checked with a machinist straight edge and feeler gauge). They don't need to be perfect, but I'd like to get down to .001". The general spec is no greater than .001" per cylinder, so .004" is the max for a V-8.

I bought two granite surface plates from Amazon - one Grade A 18" x 12" x 3" (about 80 lbs) and a Grade B 12" x 9" x 2". I wanted something bigger, but the 18 x 12 x 3 is the largest I could find with free shipping.

I will not be using MLS head gaskets so I'm shooting for a 60-80 Roughness Average (RA). If I'm not mistaken, 80 grit should result in about 71 RA.

The plan is to use some 3M spray adhesive to attach a sheet of 8.5 x 11 sandpaper to the 12 x 9 block and use it as a sanding block.

Anyone have any experience doing this with automotive heads? There are a few YT videos that I need to watch.

FYI - I realize that the machine shop is the way to get the job done right, but I don't want to wait for several weeks for them to get to it and I'd like to have the tools and know how to do this in the future if needed.

 

This guy did

I’ve seen flatheads done similar
Friend did stuff like this with brigs engines on go cars.

 

They used to hand-s****e the bed ways on machinery way back when; those old time machinists were really good! I've done it myself, definitely a process of patience and persistence.

 

Three thousands is a lot of cast iron to remove on a flat surface with sandpaper. Sure, it can be done if you've got the arms and patience.

Personally, I'd go for the machine shop.....Roger

 

Vote for Emory cloth

 

I have a large flat piece of heavy material, 2' x 5' x 1" thick wood with vinyl outer layers, with three sheets of heavy sand paper discs stuck to it, https://www.fastenal.com/product/details/99332155. I cleaned up a couple sets of Pontiac heads and they are still in good shape for more. It doesn't take all that long with the length of sanding material I have. With discs, there is room between each one for the grit to go so you are not moving the head through it all the time. You still need to keep sweeping the paper to keep the grit at a minimum. You also need to learn how to move the head, you push from the ends and let the weight do the sanding, pushing down and forward will not give a level surface. Go slow so the edges don't dig in to the paper. I went forwards and backwards along the length of head so all the grooves ran the same direction. The heads were installed with Felpro blue head gaskets with no issues. I have a commercial floor mat / carpet where I do the sanding, so I added sand paper to the back side of the flat wood to help hold it from sliding around. Flat home center sand paper will not last long, you need heavy emery type or self stick type paper made for commercial use.

I gave a couple smaller sheet of paper disc to friend that stuck them to a heavy gl*** sheet and leveled out his beater car aluminum head. It worked perfect and looked as good as any machine shop cut head.

 

Having the pair shaved at a local machine shop prolly costs less than the plates.:\

 

Back when I was young/dumb/cheap I used to use a whetstone.

 

I wouldn’t do it with regular sized sheets of sandpaper because the edges of the head will always wander off the edge of the paper. At the very least get a belt from a belt sander.

You want the entirety of the surface being worked to be within the abrasive perimeter if that makes sense. Every time the head runs off the edge you’re automatically taking more material from the middle.

 

Getting the head flat is one thing but getting them square is another. If you have to take more off one end than the other I’m not sure how you could accurately do that without a fixture like a resurfacer has.

 

Sanding is fine if you are just resurfacing or removing dings and rust from storage, it will not work to correct poor machining or casting flaws. My sanding fixture is longer than a head, so like mentioned above, the head never leaves the sand paper discs ( it stays on a flat plane ).

 

I have done this with Ford 6.7 Powerstink valve covers. The valve covers have to be very flat because of the gasket design and the fact that the intake charge flows through the valve cover. If they aren't flat, they can leak charge air into the cylinder head rocker box, and then you end up with very high crank case pressure and a bunch of oil leaks and seal failures.

I used an 3' x 8' Sturm cast iron surface table and some 120 grit 37" x 75" sanding belts from Grizzly. Sliced the belt, and taped it down to the surface table, and got to pushing those stupid valve covers around. It worked well as a proof of concept and confirmed that the bake oven was warping the valve covers enough during cleaning to cause air leaks.

We weren't worried about perpendicularity like RmK57 mentioned above, and that would be important on a cylinder head. You don't want the chamber on cylinder one to end up at 69 cc and #4 be 74 cc.

 

Have any woodworking friends? With one of those giant belt sander machines? Those belts are wide and long enough to cover the entire surface plate, with material to spare.

 

I've got this one at the shop, but not quite wide enough. Shouldn't be difficult to source the right size.

Thanks

 

Poor man’s surface grinder, a couple thousands isn’t hard to achieve, no more than .004 or 0.005. Like was stated keeping everything plumb and square is very hard. I’ve used the sandpaper on a flat surface to true up exhaust manifolds and a header I built for my OT Sprite with hood results. My dad used to draw file stuff to well within 0.001. The trick to draw filing is using a very good file and letting the file do the work. Draw filing takes time and patience.
Filing is as traditional as it gets.
Dan
PS 80 grit is probably the best in my opinion, I’d use copper coat also unless using mls gasket.

 

I'll copper coat as well. Not using MLS gaskets.

Thanks

 

My arms got tired and sore just thinking about all that. But I'm another who tries to do it almost all himself, so good luck.

 

I disagree with the guys who say don't run your parts off the edge of the sandpaper. I'm a retired machinist, and I used to occasionally flatten parts this way, and I would always run the part off the edge a little, because when you're sanding, you are wearing the paper down too. if you don't go off the edge, you'll be sanding in a shallow dip in the paper, and the edges of your part will be sanded a little deeper when they are pushed over the less worn area of the paper.

 

Just for information and the hope that it will help you with your search.
This process is called "Lapping" and there is lots of information on the web.
Here is just one example.
The Complete Guide to Lapping in Machining - Kemet USA

 

Using a figure 8?pattern while lapping is the best , did many industrial compressor valves this way however it would hard to find a sheet of sand paper large enough for doing a head would probably need 6 to 10 sheets.
Dan

 

BoyScout way. Rub 2 heads together with valve grinding compound. Twice as fast too. 8^)

 

Just did a search on Facebook Marketplace and there's a bunch of granite slabs available for cheap or free. I think I will go check some out with a machinist's straightedge and a feeler gauge and if I can get a flat (within .001"?) piece that's about 24" x 36", I'll buy that and just toss it on top of one of my toolboxes. Once I'm done, it can stay there.

EDIT: did another search and it looks like the acceptable variation is 1/16" over a 4' section. That won't work.

 

Thanks for the link. That was informative.

Scott

 

The method historically involves 3 surfaces to be lapped together to achieve flatness...

 

I agree with your point however in the original post he was asking about using a standard 8x11 sheet of paper which in my opinion would cause problems.

 

You folks have most likely all been on a playground teeter-totter at one time or another. Think about what is happening when the head slides off of the edge of the underlying sandpaper support.

Do you really think it's a good idea to NOT have a greater than head length sandpaper bed so that this is not an issue?

The evidence is in the red ellipses in this screenshot from the Youtube video below.

 

@Ebbspeed, that looks like a terrible setup to try to get something flat within thousandths. First off, it looks like the sandpaper is taped to Plexiglas.( I see no green tint of gl*** at the edges.) From more than 45 years of working with it, I know that the thickness of a Plexiglas sheet varies a lot! Second, it looks like it's clamped to a plastic work table, which is also not going to be anywhere near flat. Then he's got it clamped only in the middle, so it's probably bowed in, it's no surprise there's not as much sanding dust in the middle of the sandpaper. When I say run the part off the edges of the paper, I don't mean so far that the part will tilt, just so you are using the whole sheet, and there will be no thick spots at the edges.

 

I have never done an automotive head that way. I have done quite a few single cylinder heads that way though.

 

I tend not to over think this stuff these days. As I see it offering my opinion, the weight of the head will hold it flat on the sandpaper surface.


Push pull repeat.




This is how I did a resurface to my aluminum heads. 4 sheets 80 grit., back taped with that thin red vapor barrier tape, flipped and now taped with masking tape along the edge of a gl*** pane with a cushioning matt surface under that gl*** on a piece of 3/4" plywood.
Want to make the job go faster, improve on the paper and the back and forth action. But it's not hard nor does it take that long. Mentioned in many a how to in the car magazines, and these days video's, it works. Cast Iron would take more effort with the push and pull but a wet paper some oil would have also been a better paper to use.