Measuring Tolerances

When measuring tight tolerances with an indicator, how hard should it be pushed against the measured surface to register a correct tolerance?

Are you referring to a DTI?

The plunger is spring loaded, and the spring decides

I use as little as possible but you must ensure the nib stays in contact with the piece being tested , it also depends on the range or travel of your indicator and what variation in tolerance you are expecting as there is no use trying to measure parts that may very by more than 2mm with and indicator that only has 1mm travel.

If i was using a dial indicator with 1 inch travel ( 100 thou per rev of needle ) i would pre load about 50 thou or half a turn .

For a dti 5 -10 preload though i find works for me

If as i suspect you may be talking about dial or digital callipers ? - very little pressure and it is important that the pressure ir fairlt consistant each time .

Micrometers - use the ratchet until you develop a feel for them - some people never do though .

I use a torque wrench on my micrometers.

So do I, my index finger and thumb.

All contact measuring tools are flxible at some level of force. Consistency is the important factor. Think here of measuring internal and external dimensions for a close fit - ie comparative readings.

Outside measurementsis apparently easy and could be measured (probably inaccurate in the absolute sense) at any pressure. However measuring the inner dimension would not be satisfactory as one would be using a secondary transfer tool and likely deformed differently. Common sense needs to be applied - a light but adequate pressure and, above all, consistency.

Remember, too, that sensitive measuring tools do not take kindly to heavy handling. Micrometers, for instance use a fine screw thread which would easily be damaged if overtightened and would definitely wear out much faster than they should if used in that manner.

So you are the one that bent all the micrometers i found in a cupboard at a place i used to work at ! 😄

A few were swiss made too 😭

One doesn't measure tolerances, one makes a measurement to a certain precision, and that measurement will have a tolerance that determines the accuracy of the measurement. The tolerance can include inherent inaccuracies in the measuring instrument as well as the technique.

A DTI is not a precision measuring tool. The clue is in the name "indicator", it indicates variations but isn't intended to accurately measure them. By a DTI I mean one with very limited travel and a movable stylus - normally used for setting work or vice alignment. Those with a fixed axis plunger are better but still normally intended for comparative measurements.

It's important to understand the difference between precision and accuracy.

Andrew

I use a stilson on mine !

With a micrometer, lock the spindle at the required reading and with a nylon hammer, not metal as this damages the paint work on the frame, tap the mic until it fits on the piece being measured. The number and strength of blows indicate how much interference there is.

When I was at school our teacher told us that within a few years we could use all the imperial mics as G clamps, as we would be using metric from then on, that was 1963, I now have one metric mic, 25 to 75 mm ex Russian fishing boat( the sailors flogged off anything that wasn't nailed down, and a few things that were), the rest are still imperial.

Precision, hit it hard with a hammer, and if at first you don't succeed, get a bigger hammer.

Ian S C

The originator of this thread signs himself absolute beginner, if he is then this levity may put him off a bit. I will say that he is getting a bit mixed up anyway. First there is always a tolerance, on the part being measured and on the measurement itself. For home modelmaking it is not important , make stuff to fit. Eg make a bore as near to drawing as you can, then make the bit that goes in the bore to suit, by trial and error. May need a good sliding fit, eg a piston. Or may need an easy fit for a swivel pin. Actually using ,say , a micrometer takes a bit of practice, the " feel" is different for different people, one reason we have tolerances.

Of course a tolerance is important. You are introducing a tolerance when you say "a good sliding fit" . It is no different than saying ± 0.002" from a given dimension. By stating a tolerance you do away with making something by "trial and error"

Edited By Eric Cox on 03/11/2017 11:26:54

It would be helpful if you could elaborate on what you mean by "indicator". Dial indicator? Dial test indicator? Or something completely different, like say a digital caliper or a micrometer. Otherwise, all answers so far are just guesses.

Also, what do you mean by "measuring tight tolerances"? One usually measures a dimension. Tolerances are the assigned allowable deviation from a nominal dimension and can vary widely. Are you asking how to measure a piece of metal with the above mystery indicator device with accuracy? Or something else?

Not getting on your case, just trying to clarify what you are asking to increase the chances of getting a useful reply.

Edited By Hopper on 03/11/2017 11:18:07

For clarity, error bar would be a better way of describing uncertainties in a measurement (my definition being the range of uncertainty each side of the actual measurement).

Tolerance would be the acceptable size of the error bar.

They are connected in the sense that the chosen meathod of measurement should have an error bar somewhat smaller than the tolerance require of the part.

regards Martin

I'll try to explain myself a little better :- Tolerances are mostly to ensure inter-changeability. If I needed a cover on a casting and the holes are 2" centers, tolerance +/- .002" centers, then I would mark off the cover at 2" as best I could. Then drill the cover and use that as a jig to drill the casting. It's only when a replacement is needed that the tolerance the 2" dim. is important. By trial and error I mean turning the part and trying in the hole for fit. When possible turn a short length until the fit is correct and then do full length at that setting. Probably not much help to the OP, but my usual method of working.

AB, how can your question ever be answered?

You can't measure a tolerance. A tolerance is the variation allowed from a nominal dimension and is stated on a drawing or spec., not measured on a workpiece.

You must already know that different instruments are manipulated differently to obtain a measured value.

And even if somebody did come back with a force in pounds or pascals or whatever, however would you measure that force in practical use? How would you know which dimensions carried what tolerances on the force applied to which type of measuring instrument?

Ahem, let's assume that someone signing themselves 'Absolute Beginner' is unlikely to be familiar with meanings of all the terms we use and help them learn.

There are no daft questions if you don't know the answers.

Neil

well said

jim