Believe it or not, one of the most overlooked problems in qualifying gages is unanticipated deflection of the fixture due to the force of the probe on the part. Who would have guessed? After all, fixtures are used to provide stability.
Most fixtures are made of several component parts and are a variation of the well-known C-frame. If the user is aware of some common problems that can affect the use of the C-frame and other fixture designs, he or she can quite easily detect and eliminate possible error sources such as deflection.
All materials, regardless of their hardness, have some degree of elasticity. That also applies to the frames we use to fixture our parts for gaging. Small as it may be, this elasticity is a real and vital consideration in a precision gaging setup. Even the slightest pressure will cause some deflection of the frame. If the deflection is great enough, it will throw off the calibrated accuracy of the gage.
There are several possible solutions. You can (1) increase the spring rate of the gage frame to the point where deflection is no longer great enough to affect calibration, (2) reduce the spring rate of the indicating system until the deflection of the frame becomes insignificant or (3) compensate for the deflection. While it is possible to compensate for deflection with a reasonable degree of accuracy, your best bet is to take the problem completely out of play with one of the first two choices.
The tendency for an object to deflect is known as the "spring rate." It is the ratio of the load applied to the fixture component (expressed in pounds) to the resulting deflection (expressed in inches). So the higher the spring rate, the less the frame will deflect under a given load. The size of the part to be gaged also figures heavily on whether or not spring rate will be a large or small problem. Large frames designed to accommodate sizeable workpieces are much more susceptible to error caused by frame deflection than are the ones for small pieces.
As a rule of thumb, the spring rate of the fixture should be at least 100 times greater than the spring rate of the indicator. Fortunately, there is an easy way to test this without doing a lot of math. With a workpiece in your gaging system and the indicator on zero, place a known weight (for example, 1 pound) on the arm of the frame at the center line of the indicator's spindle. The deflection of the frame will be shown on the indicator of the gage. Let's say, for example, the deflection is 0.004". By applying the 100:1 rule, we know that the load on our probe to make a 0.004" measurement should not exceed about 1/100 of a pound or approximately 4.5 grams. Otherwise, deflection of the fixture may alter the result unacceptably.
Next, we find out how much load it actually takes to move our gage 0.004" during measurement. If you haven't removed the 1-pound weight, do that. Using a dynamometer, place the lever underneath the contact point of the measuring indicator (dial indicator, electronic probe and so on). Zero out the indicator, and then using the dynamometer apply pressure to the contact until you make the indicator move to 0.004", and note the reading on the dynamometer. If the force is less then 4.5 grams then you're home free.
If the ratio is smaller than 100:1, try making the fixture more rigid or reducing the gaging force of the indicator by going to a lighter force spring. As a last resort, you can introduce a compensation factor to all indicator readings. For example, if the spring rate ratio of the fixture to the gage is as low as 10:1, you may try multiplying the indicator reading by approximately 110 percent to approximate the proper answer.
This added 10 percent will compensate for the fact that one out of every ten units of part dimension away from zero or nominal results in the spring force of the indicator deflecting the frame arrangement rather than moving the indicator mechanism. It should be understood that this formula for an acceptable spring rate ratio applies only to comparative gaging, for example, gaging in which the instrument was initially zeroed with a master of the same size as the part to be checked. Spring rates do change with displacement. Also, this process assumes that there is no part deflection as a result of the measuring forces.
The concept is more difficult to explain than it is to test. Work through the steps for one gaging setup and you will have mastered a valuable skill to use whenever you suspect that fixture deflection may be causing a problem.