There are four components to take into consideration:

• ball diameter
• thread diameter
• contact point length
• ball composition

New indicators come supplied with .080" diameter balls as standard, but for special needs other diameters are often available. Thread diameters vary from manufacturer to manufacturer but are generally metric threads. Manufacturers differ in their method of measuring contact point lengths. It is imperative that the correct length point is used.

MTC contact points are turned from #303 stainless steel and have carbide, ruby, Nylon or Teflon balls.

Alloy 303 is a non-magnetic, austenitic stainless steel that is not hardenable by heat treatment. It is the free machining modification of the basic 18% chromium / 8% nickel stainless steel." [source: pennstainless.com/stainless grades]

We recommend carbide balls over chrome or steel for its durability. For special applications you may need the ruby, sapphire, teflon, nylon or some other exotic.

The contact points listed here are home grown. They're every bit as good as the originals yet they cost less. We give you the manufacturer's ordering number for ease of reference.

These contact points are used for special applications, typically in optics. The ruby point outlasts standard chrome points and does not expand or contract with heat and cold. It does not conduct an electric current so it can be used on your machines without creating interference. It is also less likely to scratch delicate surfaces. It's also the material that will hold up to silicon carbide surfaces. The Teflon point is softest made of a solid Teflon ball. It is very white in appearance. It is least likely to scratch but is not as long lasting as the others. The smallest Teflon diameter available is 1/16" (.062"). Smaller points can not be manufactured in this material. The Nylon point is not as soft as Teflon but will last longer. It can also be used where surface damage is an issue. Nylon appears opaque.

We know that the length of the contact point determines the ratio of leverage. In order to get accurate dial readings the length must be as specified by the manufacturer.

Changing the angle of the contact point changes the ratio just as if you had installed a somewhat shorter point. As a result, the readings on the dial will be higher than they should be.

To compensate for this cosine error it will be necessary to multiply the reading on the dial by the cosine of the angle between the contact point and the measuring surface.

How do you measure the angle? Most of us can intuitively judge an angle of 45° and, perhaps with a little less accuracy, an angle of 30°. Beyond that, we need help. A simple protractor like we used in grade school will suffice. Lay the straight edge on the measuring surface and, by eye, make a judgement of the angle. A few degrees one way or another won't be significant.

Where do you find the cosine? Use a scientific calculator or a printed chart.

Enter the value of the angle: for 20° enter "20" and press the COS (cosine) button. The display will read .9340 (allowing for rounding up).

Now it's simply a matter of multiplying the reading on your test indicator with this value.

• For example: when the reading is .0006" you multiply .0006" x .9340 and the result is .00056"
• This is just as expected. Your reading of .0006" was larger by .00004" than it should have been.

From this example you can see that angles of 20 degrees or less have relatively insignificant effects on the reading. But, when the angles are larger you'll be wise to make cosine error adjustments.

This same cosine error applies to all makes of test indicators regardless of contact point lengths and resolutions. Some Fowler test indicators feature "pear-shaped" contact points which compensate, in theory, for the cosine error. These may be a good choice if you frequently use a variety of measuring angles.

Interapid indicators pose a special problem. They're designed to work without cosine error at 12°. When setting up, use a protractor to make sure you're in the ball park. If you're off, then you'll have to compensate, just like the others. If the angle is 32° then the cosine will have to be of the difference between 32° and 12°, in other words 20°. Make the calculations as above.