What's new
What's new

Surface plates

Vernon Tuck

Stainless
Joined
Oct 26, 2008
Location
Brenham, Texas
Folks,

I am thinking about acquiring a surface plate. However, as will soon become apparent, I am not a machinist. I will explain my perceived need in hopes that y'all will be kind enough to educate me.

Here's some background:

First, we are motor heads. We own dozens of old VWs and Mercedes Benzes, engines and components. Some of the cars are runners. Others are parts vehicles. These represent my lifetime collection of vehicles.

Second, my wife and our older son, are newly minted apprentice machinists. They just graduated from a one year, rather comprehensive manual machinist course at a Houston, Texas community college (Lone Star).

Third, we now have the makings of a rudimentary little machine shop. This is not to say that we have commercial aspirations but it's also not to say that we don't. At present we have a South Bend 10L tool room lathe and two manual milling machines.

I mention all of this to explain that we have both short term and long term aspirations. In the short term I need to inspect a VW diesel engine head in order to determine if it has excessive warpage. I recently bought an unfinished project from a fellow consisting of an '81 VW rabbit diesel into which he had retrofitted a completely rebuilt engine and transmission. The machine shop rebuilt the engine and head but the previous owner himself installed the head. I bought the car from him before he ever started it.

After we got the car and tied up the loose ends, we got it started. Long story short the car overheated for (at that time) unknown reasons. We NOW know that the fellow put the head gasket in upside down. The engine did not PEG in the red but it got pretty hot during a test run. Those engines (aluminum heads) are very unforgiving of overheating. Hence, I need to remove the head, inspect it, and (knock on wood) re-install it with a new head gasket.

I'm very unclear on at least three points: First, it would be good to know what grade of surface plate I need for this kind of work. Second, I would appreciate knowing what the best and most versatile size would be for a tiny shop such as ours. Not only is price an issue so is available space.

Third, I assume that when inspecting a cylinder head on a surface plate that the plate has to be slightly bigger than the head. Is this correct? If so, I would say that the biggest size I would need would be one that's big enough to inspect the head from a Cummins B, 12 valve, 6 cylinder engine such as used in Dodge trucks until 1998.

Finally, I would appreciate receiving some advice on how to determine a used plate is a good one, and how much I might reasonably expect to pay.

In the short term I'd like to inspect several cylinder heads. In the long term, my wife and son will hopefully evolve into machinists and become competent to do our own machine work - at least surfacing heads.

I will genuinely appreciate your help. I hope the metrology forum is the right place for this question.

Sincere regards and many thanks.

Vernon
 
You need a straitedge and a set of feeler gages to inspect a head for warpage.

Take it to any automotive machine shop and they will inspect it for free.

If it is warped they will expect to get the repair job though.
 
I would say that you would need a Grade B plate that is larger than the head you are trying to inspect. I have no idea how large a Cummins B head is, so I can't comment on size, or how much deviation from flat is acceptable.

You would most likely want to set the head up on machinists jacks, or similar, on the plate so that the surfaced side of the head is facing up. Support the head at three points. Draw a small circle with a Sharpie or similar on the surface of the head above each support point. Use a height gauge and DTI, or a surface gauge and a DTI, and raise or lower the jacks until a DTI measurement in each circle is the same, indicating that the head is parallel to the surface plate at each of the three points. Now sweep the head surface with the DTI and it will indicate the amount of bow or crown in the head.

Basic version...
 
You would most likely want to set the head up on machinists jacks, or similar, on the plate so that the surfaced side of the head is facing up. Support the head at three points. Draw a small circle with a Sharpie or similar on the surface of the head above each support point. Use a height gauge and DTI, or a surface gauge and a DTI, and raise or lower the jacks until a DTI measurement in each circle is the same, indicating that the head is parallel to the surface plate at each of the three points. Now sweep the head surface with the DTI and it will indicate the amount of bow or crown in the head.

Basic version...

I would say this is way too complicated for normal head work. As has been mentioned before, a "good" machinist's straight edge and some feeler gauges should be all you need. Not saying you can't do it with a surface plate but why make things over-complicated. Check with a few automotive machine shops and I doubt you'll find any that use a surface plate for checking heads...
 
...Not only is price an issue so is available space... I assume that when inspecting a cylinder head on a surface plate that the plate has to be slightly bigger than the head. Is this correct?
The least expensive grade of granite surface plate will be flat to better than 0.001", which I believe is as good as you need. I suspect you will ignore any head warpage you find that is this small (I know I would).

Next is size. Ideally you would have a surface plate ~4" longer than the longest head you will put on it, so you can slide the base of your dial indicator all the way around. However, heads are long and narrow, and surface plates are closer to being square, so if you get a very long plate, you automatically get a very wide plate. This adds a lot of dollars and weight (which is a problem for both shipping cost and cost of a mounting table), as well as takes up a lot of space, even though you don't need the extra width. As a workable compromise, no matter what your dial indicator has to be mounted on an arm that extends at least half the width of the head, to reach the center of the head. Let's say the biggest head you will need to measure is, say, 8" wide. That means the indicator has to overhang at least 4". Because of this, you can have a surface plate that is significantly shorter than the length of the head. If the head overhangs 4" at both ends, but if the indicator is at the end of a 6" arm it still will be able to reach the center of the head at the far end (6" x cosine(45-deg) = 4.2"). In this scenario, the surface plate could be 8" shorter than the longest head you need to measure, as long as the width of a plate that size is at least ~6" wider than the widest head (to leave ~3" on each side for the base of the DTO).
 
I've never done any head work yet but a quick google shows likely flatness around .002-.004 flatter for performance or race work. As stated straight edge and feeler gages is likely fine. One thing not mentioned is you state you have 2 manual mills - the tables on those should be flat enough for you needs to check as well. Of course you will need to check table flatness but you need to know that anyway if you are planning on surfacing the heads there.

Another thing mentioned in head flatness with aluminum heads is taking some of the warp out by clamping and heating if appropriate.
Good luck
Paul
 
For checking flatness on cylinder heads I do not think you will find a better solution than a straightedge and a feeler gage. If you use a surface plate, you will only be able to check flatness on the perminter of the head. With a straightedge you can check in a straight line from any two points on the plane. Looking in the McMasterCarr catalog just for a ballpark figure, you can get a 36" precision straightedge for about $145 a 36" surface plate would run around $500 and would weigh aprox 400lbs.
 
... you can get a 36" precision straightedge for about $145 a 36" surface plate would run around $500 and would weigh aprox 400lbs.
Having given my answer to the question the OP asked, I now feel free to digress. I agree that a straightedge does this one job faster/easier than a surface plate, but the OP gave us additional clues to his longer term interests. Before proceeding, given unlimited space and money, more tools are always better to have than fewer, even if some of those tools are used very infrequently. However, given limits on space and money, choices have to be made.

Given he has two mills and a lathe, as well as a son and wife interested in machining, a surface plate will find many more uses than a straightedge. Having said that, if he needs to check several heads per day, serious thought should be given to a straightedge. On the other hand, if he needs to check one head per month, and setup time is less of a limiting factor than money, a surface plate could be the better choice, given its many other uses in a machine shop.

I work on motorcycle engines, whose heads are of smaller dimension than those on Cummins diesels. I don't do this as a business, so setup time to make a measurement is not a huge factor. For measuring warp, I put them on the bed of my mill and use a DTI in the spindle. I do have a surface plate (Laboratory Grade AA), but for the same space reasons the OP mentioned, it's only 12"x12". However, that size always has been completely sufficient for the precision measurement applications I use it for. If I only used it for motorcycle-related measurements, a much cheaper Toolroom Grade surface plate would be fine for the work I do (which is on old motorcycles). Not that machining to better than a ten-thou. is hoped for, but it's possible that some tolerance on a modern motor vehicle would be such that it would be worthwhile moving up to Inspection Grade, at least for a small surface plate. But, for a truck-head-size plate, I believe Toolroom Grade would be completely sufficient.
 
If you look on ebay you may find a great deal. A surface plate is very heavy, and expensive to ship. If you can find one within driving distance chances are you can get a good deal. I found a nice one about 70 miles away and got it cheap. Just remember, the things are very heavy.

D
 
Vernon - By now you've noticed one thing that has been mentioned repeatedly: weight. Here's a link to a website that has a chart giving the weights of various sizes of surface plates. (I have no affiliation with the company, I simply liked the chart!)
www.deterco.com/products/Precision Granite/precgranite.htm

There MAY be some slight differences in weight based on the grade / type of granite used, this should give you a pretty good idea of what you'll be dealing with, should you decide to acquire one.
 
I buy my precision granite products from Precision Granite. :D Great company to deal with.

The weight of granite is about 166 pounds per cubic foot, so a surface plate 24" x 36" x 12" thick would weigh about 1000 pounds.

A 24" x 36" x 6" thick plate would weigh half that, or about 500 pounds.

- Leigh
 
If you expect to achieve rated accuracy (as opposed to just having something "flat"), the surface plate must be supported
at three points, as shown in the following diagram:
Surface Plate Supports.png

The points are 1/4 to 1/5 of the length of each side.

I use a frame with carriage bolts at the support points for small plates. This permits easy leveling.

- Leigh

NOTE: Disregard the thumbnails below. This stupid BBS s/w doesn't give me any way to delete them.
 

Attachments

  • Surface Plate Supports.jpg
    Surface Plate Supports.jpg
    10.4 KB · Views: 681
  • Surface Plate SupportsLg.jpg
    Surface Plate SupportsLg.jpg
    9.8 KB · Views: 469
I have an inexpensive 2' x 3' x 4" that came with a frame. There were no points to set it on so I used carriage bolts, but they are out near the corners and in the middle on the other edge. I'm not doing lab grade work, but I am curious how long before it will bow to the point I could measure it.
Off course after I bought it, I saw much better and thicker plates for sale for near the same price. That happens too often.

Dave
 
Check with a few automotive machine shops and I doubt you'll find any that use a surface plate for checking heads...
Quite a few just run 'em across their belt surfacer and call it good. Praise be for Fel-Pro head gaskets. :D
 
Quite a few just run 'em across their belt surfacer and call it good. Praise be for Fel-Pro head gaskets. :D

Fellow Diesel guy here, just got my 83 rabbit running.


No belt sander, but I've seen some people use a chunk of plexi or normal glass (new, flatish) and sanding disks from floor grinders. I do believe the head's still being used. I'm a member over at vwdiesel.net - They know ALL about those cars. Just like Peachparts ;)

On the gasket and overheating: Theirs two different gaskets for the older diesels, with various thicknesses between them (notches) They very between solid lifter and hydraulic heads. Although, it is rather easy to put the gasket on upside down.


Good luck with the project and the shop!
 
and on OHV/alumn heads like these: if it's WAY out of tolerance (.004+) you may want to look for a new head.

Sure, you could recut it. But the cam journals are integral to the head. They warped with it.
 
and on OHV/alumn heads like these: if it's WAY out of tolerance (.004+) you may want to look for a new head.

Sure, you could recut it. But the cam journals are integral to the head. They warped with it.

I 'spect you meant overhead cam there.

Some years back I re-built a 22R(toyota) for a little plow truck(mine). Someone had milled the head -.020" and of course, when fastened down to the block you needed a wrench to turn the cam... Nowadays the aluminum heads are straightened in an oven fastened down to a heavy CI plate. Anyway...

Solution - Had the motor shop take another .020" off the head & get a .040" spacer (need this to keep the cam timing right with OHC engines). Then fastened the head to the block & set up portable bar using outside bearing saddles. Bored for automotive type bushings (the ARA motorhead shop found them for me).

Worked a treat, but I'd hate to have paid someone to do it for me (I was just hoping to get away with the $800.00 short block treatment). Oh well - Live and learn...

To the OP, for auto type work, get a Straightedge first & think about the rock later. :)

Matt
 








 
Back
Top