Thread: Locating a bore

Originally Posted by Stustoys

Re tram, haven't you flipped the work over since boring?

Flipping the work over won't affect tram - the head has not been touched. In fact, the thing that started this whole thread off was doing that - that is, how do I accurately locate a bore once I do put something up?

Michael

Originally Posted by Michael G

Flipping the work over won't affect tram - the head has not been touched.

Well no, not the tram of the mill. If you clocked the bore after you machined it it should clock round. Once you move the work you may end up with an oval, depending on which way you flip it and which way the tram is out. Having said that, I haven't even tried to do the maths on what sort of error you would be looking at, though its almost certainly not the first 0.004 of the 0.005 you're looking for.

Have you tried the Z move?

Stuart

When locating a bore I use a DTI with a long tip mounted directly in a collet with an 8mm dovetail adaptor. I use the rear dovetail for bores up to around 1.5" and the top dovetail for bores up to 2.5".
I centre the bore roughly with the probe not quite touching the work then bring the tip into contact before preloading it by adjusting the position on the dovetail.

When making the setup in the photos I centred the rotary table to the spindle first by spinning the DTI in the table centre then mounted the 4 jaw chuck and adjusted it to centre the cylinder bore, both within a division on the DTI. Interestingly, when I spun the rotary table I had to tweak the 4 jaw slightly in one axis (about 2 divisions) to make the bore true to the stationary indicator. I'm not 100% sure what's going on with that but it's close enough for what I need to do.
Cheers,
Greg.

I scrounged around and got my chinese indicol clone out and attached it and tested it out...

It does work OK for a cheap chinese tool

These threads are wonderful for getting the grey matter working.

One of the reasons I reversed the DTI when using it to find the minima was that I was concerned that the plane of probe would not be the same as the spindle axis and so make the apparent centre different with respect to the true centre. However, I did not think about that when using a DTI myself!
Seeing all these pictures of DTI's being used has made me think -
DTI.jpg
Ideally if the DTI is set up then getting centre is straight forward but if the probe plane is offset (doesn't need to be much) then if axis of rotation (that is, the spindle) is not the same as the axis of the bore, the offset may make the hole look like it it is lobed, as the readings just off the major axis may appear to be greater than the major axis readings. (Note that if the axis of rotation and the bore axis are co-incident then it will look normal). The other thing that this will mean of course is that even if you adjust the readings so that they are equal say left versus right, if the bore axis is offset from the spindle front to back and you adjust, it will change the left right readings (if the probe plane and rotational axis are coincident, the left right readings will change but equally so).

Yes, yes. I know it's obvious to you guys but I can be a bit slow some times.
End result is that the elliptical result that Simon and I were seeing could be this effect. Looks like I need to tune up both the homemade Zero-it and the Taster probes so that they are co-incident (it never occurred to me with the DTI holder that the offset may cause issues)
Michael

Hi Michael,

While I see what you are getting I think it depends how you clock.
If you are indexing the spindle but using the face as 0, then 90 , 180, 270. Yes you'll see the oval error you are talking about.
But if you sweep the "area around 0, 90, 180, 270" to find the low/high spot, while the probe plane is offset from the spindle, isn't the "measurement" you are taking along a line passing through the center of the spindle and the center of the probe? Therefore, to use your drawing as an e.g. the spindle/DTI will just be rotated a little anticlockwise for each reading.

Stuart

The way I see it is the connection between the probe and the spindle can go around in circles 6 times inside (or even outside) the bore and it still won't affect the true plane which is an imaginary line between the spindle axis and the contact point of the probe. This assumes rigidity and that the probe surface is an accurate sphere for the areas of contact.

My take anyway.

Dean

Originally Posted by Stustoys

But if you sweep the "area around 0, 90, 180, 270" to find the low/high spot, while the probe plane is offset from the spindle, isn't the "measurement" you are taking along a line passing through the center of the spindle and the center of the probe?

Originally Posted by Oldneweng

The way I see it is the connection between the probe and the spindle can go around in circles 6 times inside (or even outside) the bore and it still won't affect the true plane which is an imaginary line between the spindle axis and the contact point of the probe. This assumes rigidity and that the probe surface is an accurate sphere for the areas of contact.

It's like setting up stock in a four jaw chuck on a lathe. Once it is concentric then you can sweep around all you like and the reading should stay the same. Initially however you set up your indicator so it is directly measuring the movement of the stock when acted on by the jaw - that is, the indicator is in the plane of the jaw movement and the spindle axis. On a mill it is similar except that you want the displacement of the probe to be directly related to the leadscrew movement. If you find a minima or maxima a cosine component creeps in. How big that is depends on your DTI, the offset between the spindle axis & the probe plane as well as the bore diameter. Does it matter? It's not critical really except it makes things more difficult if you don't realise that is what is happening and can indicate results giving the impression that the bore is out of round.

As I said a few posts ago, I've never considered this applying to a DTI, having always thought that the probe plane and attachment point axis are co-planar.

Michael

I've been thinking about circles all day.Its doing my head in
It occurs to me (and I don't think its been mentioned) the oval thing. I think the axis that is closest to being centered will measure shorter than the other axis. So that my explain some of the ovals.



Stuart