Dial Test Indicator Pointers

Gunner:

If your talking about the dial pointer length in my opinion length shouldn't really make a difference. If you draw a strtaight line up the center of the pointer to the scale on the dial it should give you the same reading no matter how long or short the pointer is.

I've only had one beer so I might change my mind later

Jimbo

Any length works just fine! (Been here) The dial doesn't read in inches, it reads in some undefined units that fit into some limits. Any angle on the point, other than 0 or 90 deg will not read in linear units but some progression depending on travel depending on secant height. So, unless you're using a plunger indicator, you can call these units: "Miligunners" or how about "Gunnerdoos" or, my favorite "Gunnerinos"

Sorry Jimbo..I communicate poorly sometimes. The lever pointer..the little ball nosed gizmo that actually contacts the work piece

Gunner

"To be civilized is to restrain the ability to commit mayhem. To be incapable of committing mayhem is not the mark of the civilized, merely the domesticated." - Trefor Thomas

OOOH! I like that...Gunnerinos!

Chuckle..I need to use a DTI for measuring TIR of rotating collet closer bodies. How do I actually measure .0005 runout with a DTI?

Actual tenths..not Gunnerinos. This is why I asked the question about pointer/lever length

Gunner

"To be civilized is to restrain the ability to commit mayhem. To be incapable of committing mayhem is not the mark of the civilized, merely the domesticated." - Trefor Thomas

If this is a plunger type, where the point travels in a straight line and actuates a rack inside the instrument, then the length of the pointer (which screws into the plunger, using an unusual fine thread like 4-48 or something) then the length does not count.

If it is a type similar to a last word, where the pointer swings in an arc, then the dial reading will indeed depend on the length of the pointer from pivot to tip. The longer the pointer, the more sensitive it will be, ie, for a given tip deflection, you will see a bigger dial reading.

If this is the type, and you want to check that the one you have is more or less correct, then simply put the indicator in a lathe toolpost and use your crossfeed dial to wind in a given amount of travel. The only important issue is that the angle between the infeed and the length of the pointer's axis should be nearly 90 degrees.

In those kinds of pivot pointer indicators the error between the actual travel and the indicated travel goes pretty much like the cosine of the error angle you need to get it really misaligned by a lot before you begin to see much discrepancy.

But as mentioned, these devices work best as a null indicator, where one is trying to zero out some error.

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Calibrate to a known with the point at 0 or 90 deg. The secant error will be minimal especially with small movements. At some point, it just doesn't matter when the gravity pull from Uranus will skew your readings.

Yes, it matters if you have to measure in specific units. If all you want to check is straightness or concentricity, any needle of the right thread size should work, but if it has to be an accurate number, it needs to be the correct length stem for the indicator.

Gunner,

It makes a big difference what the length of the lever arm was if you want "true" accuracy. "True" accuracy meaning where the deflection (in tenths) of the lever agrees with the indication of the dial. The pivot on the indicator actually measures angular displacement of the lever, so the lever arm length counts. Suppose you had an indicator with a 1 inch lever, and you measured a 0.010 inch deflection. Put on a 1/2 inch lever and you would measure 0.020.

For concentricity purposes, such as centering a piece of work in a 4 jaw chuck, lever length doesn't matter much, as you are trying to eliminate TIR of the part.

My 2 cents.

Mike Eberle> >

Jim,

I don't agree. The longer the lever, the less of a reading you will see for the same deflection. The indicator is actually calibrated for angular displacement. A shorter lever will give a larger angular diplacement for the same tip deflection.

Mike Eberle> >

Gunner the contact point length is important.. the higher the resolution of the indicator the more important the contact point length and the angle at which the contact point touching the surface to be measured is..

Other then Lufkin docs.. take your contact point.. make sure the contact point is does not have flats.. should be a perfect sphere...just test the movement on a height check.. a good test is plus of minus .0005.. in .0001increments.. with the contract point horizontal to the height check surface....and see how close it is.. should be perfect at .0005 and start to differ around .008 or so.. repeatability is always much better on test indicators then accuracy in travel , depending on the range quality, etc..

Lufkins are a little stiff so don't be discouraged if you can't see a .0001 movement very well... and it could be dirty, etec..

later, Mike

In article , jim snipped-for-privacy@newsguy.com says... The longer the

I don't think so. Just the opposite. Jim

My interapid came with 2 contact points. One is .75 long & the other is 1.5 long. With the 1.5 long point in, my dial reading is cut in half.

Indeed you are correct. My mind was a bit fogged after spending saturday at work. The simple rules of levers says that the shorter pointer makes the dial move more.

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

An additional oops here, to add to the other one I posted. Of course the rule of simple levers says you are correct. My only possible defence is that I was operating with a scrambled head from spending saturday at work....

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

The pointer should be long enough to reach your target. It's similar as your leg length when you walk. If your legs are too long you stumble, if too short you float. Fortunately most persons have the right length legs. HTH.

In article , jim rozen wrote: : :If it is a type similar to a last word, where the pointer :swings in an arc, then the dial reading will indeed depend :on the length of the pointer from pivot to tip. The longer the :pointer, the more sensitive it will be, ie, for a given tip :deflection, you will see a bigger dial reading.

You've got that bass-ackward.

Most are 11/16" long, unless using an extended tip indicator. Don't believe the stories about milli units or whatever. With the pointer as close to parallel to the work surface as possible,(orientation of the body does not matter) the graduations on the indicator should be very accurate to the actual deflection when using the correct length tip.

Yes, the scale will be calibrated properly only for the correct length of probe/finger. (I think of the "pointer" being part of the dial, not what touches the work).

If you can't find what the correct length is for this particular DTI, you can make one of any length. Then figure out how much deflection you get on the dial when you move the point 0.001". You can do the latter by clamping the DTI so it touches the spindle of a mike held in a vise, or use a micrometer head if you have one. Something like a Height Master would be even better.

Basically, you must have *some* way to move the measuring finger reliably a known distance. Then see how much the pointer on the dial moves. The ratio gives you a "calibration constant" for your DTI - now you'll know that it reads 1.2 times the actual displacement, or whatever the value actually is.

If you're really fussy, once you know this value, make a new probe whose length is equal to the length of the first one multiplied by the calibration constant above. (Measure from pivot bearing to contact point on the ball). Now you'll have one that's just the right length.

If you're less fussy, just live with the first one you made.

Dave

I do indeed have to measure in specific units, hence the original post. How do I determine the correct stem lenght?

Gunner

"To be civilized is to restrain the ability to commit mayhem. To be incapable of committing mayhem is not the mark of the civilized, merely the domesticated." - Trefor Thomas