centering square stock in 4-jaw

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Rayanth

Well-Known Member
Joined
Jun 7, 2011
Messages
336
Reaction score
7
I got my 4 jaw cleaned up (so i thought - until i turned on the lathe with it mounted. THEN it was clean.... nothing else in the rotation plane was, though) this weekend... and decided to give it a whirl. After a quick google on centering parts in a 4-jaw, I got pretty good results with my brass round stock that i've been using to try things out on. (also got a 4-sided diamond hone block with 200/300/400/600 grits to hone my lathe bits and got better results on the brass)

Anyway, after the reasonably success of the brass round I figured I'd use the lathe to face off the rather skittywampus crooked ends of the aluminum square stock I was to use for the EZ engine. This would both give me a nice clean surface, and make it square to the sides - I hoped. It would also let me get used to trying to see what the heck is going on with interrupted cuts and other good fun.

I used a similar method as mounting round stock, to mount my square stock. since each jaw is holding a flat side of the square pretty much at the center, I mounted the Dial Indicator to my cross-slide, and lined up the point with the rear jaw. moved the cross slide to give me sufficient depth to know it was touching in a full rotation, and made adjustments so that at the center of each flat face, where each jaw was holding, ALL of the readings came to within a thou of 0.021" on the DTI. Since they were all equal, I figured I was on center.

I then proceeded to machine the face, and got a very nice finish... still have problems with just a hint of nipple in the very center of the face, but that's just a matter of shimming the bit a little more precisely (i think it's within half a thou), however after shutting it down and pulling it out, I found that I was not on center after all.

I do not have a photo of the part, but mocked up a quick CAD representation of what I do have. The red marks indicate where the jaws were mounted - all readings were at .021 (+/- .001) on the DTI at these locations. the purple crosshair is the center of the part, and the circle is where the center of my facing operation is, with shown measurements.

4jaw.JPG


even the +/-.001 on my readings should magnify to +/-.004 at the most on the part, and I have a difference of about .050 off center in two directions.

What did I do wrong?

- Ryan
 
Ryan,

You did nothing wrong at all, if you are just facing off the end of a bar, you don't need to have the centre in the centre if you know what I mean.

Where your problem lies is when you want to say drill a hole exactly down the centre. You need to get your square bar running exactly true.

I get around it by using 4 jaw self centring chucks, they do it all for you, but in your case, it can be rather difficult.

You would need to mark a centre pop exactly in the centre of the end of the bar, and clock onto that using one of the many methods available. I use a normal solid tailstock centre, tip located into the pop mark, and a rotating centre in the tailstock the tip located into the end of the solid centre (which normally has a centre drilling in the end), and set your clock gauge to read off that so you can then use the chuck jaws to get it truly central. Sketch at bottom.

Now Marv came up with an ingenious little device, for fitting to your dial indicator, that takes care of the problem with square bar. I can't find the article, but I am sure that if Marv reads this post, he will point you to it.


John

img066.jpg
 
Bogs, since I am tackling the EZ with nothing but a lathe, I will be having to drill the holes both dead center and off-center in the square bar, using the 4jaw. I understand it's not so criical when facing, but if I can't figure out how to do it right, now, then I will have some real trouble later.

Drilling a pip mark dead center might be an issue, when doing so would mostly be by eye. Guess I'll add a spring center punch to my list.

I do now have a drill press, but no vice for it until tuesday, and I haven't actually checked to see if it works...

- Ryan
 
Ryan,

You don't have to drill a pip mark, just a centre pop mark will do, just something to locate the point of your sensor bar (the solid centre).

There are a great many methods of finding centre on a bar, be it round, square or hex. A rule, scriber and a centre punch (60 degree for doing this sort of work), and maybe a set of spring dividers or a compass should be your basic layout tools. Later on, surface plates and height gauges will be used for more complicated layouts.

Just scribing a line from corner to corner across the middle will easily find centre of a square bar to within a couple of thou. You can even buy centre squares for a few bucks, and they work on both round and square bar.

The centre pop method is the way most four jaw setups are done, just as I have shown.

Mark your spot, mount into the four jaw, and use your indicator to get the pop mark running true.

There are other more complicated methods, but this way is the easiest for a beginner to go with.



John
 
For irregular shapes it can be very tricky to get the centre. A very crude method which seems to work well for me is to use a flat bar with a bolt/hook at one end, which has one face touching the work and another hooked on something around the same height so that the bar is roughly level with the work surface but can pivot up and down. As the work is rotated by hand the height of the bar changes and the level can be read by a dti. For a square bar it would rely on the corners being accurate/consistent. This method is very useful for hex bar or threaded rod where the surface is irregular but the height of the threads/angles is consistent.

I can only think that in your first try to centre the block using a dti that the needle had rotated 1 full revolution such that the needle pointed to the same place on the dial.

Good luck with your next attempt.

Best Regards

picclock
 
Ha! that would explain .050 off... revolution being .100 ...

oops. Better watch closer next time.

- Ryan
 
Hey Ryan
Limited pc access today, laptop charger is dead
Bogs has nailed it.......and your on to to it....keep going!

;D
Dave
 
There is another way, that is centuries old, first true up a piece of round stock of the same diameter and then replace with your square, remembering to loosen and tighten the same two jaws. It can be quite accurate if you tighten the jaws to the same torque, to achieve this I once made a torque wrench chuck key, nice idea but had its flaws.
Ned
 
What about using a long-travel dial indicator, and carefully dropping the tip onto each face as you rotate the chuck by hand?
 
I have indicated square parts that were 8 feet in length on all sides to cut bearing fits in the
center of them.

I'd find the absolute center using a cutting tool, then mark that position roughly on all four sides.

Find the low spot on the first side by jogging the machine table back and forth.
Set the digital on ZERO and rapid it away for the table.

Go to the opposite side rapid back into the exact digital reading, then find the low spot on that
side using that same method.

Then go the a side 90 degrees away and do the same process.
After you have that side dead nuts, the original indications have moved so you go back and do them again.

Sometimes it would take an hour, sometimes it would take your whole 8 hours to get it perfect.

When I was working in the big stuff it wasn't uncommon for a critical set up to take more than your 8 hour
work shift.

It's a lot easier at home.
Find the low spot rocking the chuck back and forth by hand, go 180 degrees and do it again.
Get it dead nuts that way, then work 90 degrees from there.
Get that zeroed in then go back to re-dial in the first side.

It isn't that hard to do after you've practiced it a few times.
Sure it's frustrating at first so play with it some day before you NEED to do it.

Chuck up a short piece of square stock and indicate it in dead nuts then touch it with a center drill.
Then chuck up a piece of rectangular stock. Indicate it in dead nuts then touch it with a center drill.

Check the corners to your center drill mark and see what the differences are.
It's all about the practice.

Everything is hard to do until you do it a few times! ;)

Rick


 
Doc,

Thanks so much for linking your video, It confirmed that i was effectively doing it correctly (I had been rocking to find lowest point, etc) my only difference was not using tailstock but using magnetic base, mounted to the cross slide, and reading from the 'back' of the part.

This indicates it probably was a simple matter of not noticing an extra revolution, making my centering a full .1 off in two directions.

live and learn!

This brings to question though, if I want to mount square/rectangle stock OFF center....hmm....

- Ryan
 
For "on centre" I normally eyeball it to the tool, turn a small diameter short of the square (yeah this wastes material) then clock to the lowest point. You can clock around the square corners but I find that sometimes throws the clock out and I don't get repeatable results - hence the removal of the worst of the corners. After getting it on centre I then turn the reference diameter further for a later TIR check - if I am going to push it off-centre.

For "off centre" I first get it on centre as per the above and then push it over against a dial gauge to the required offset. In two planes if that is required. As a backup check you can measure the TIR of the turned diameter portion - for a single offset this check is more or less spot on for a two plane offset you need to consult our good friend Pythagarus to work out the TIR - in this case you could still be wrong but it makes a fairly good check that you haven't made a complete hash of it.

I have also done this by taking fore & aft readings with a vernier against a bar in the toolpost (before I bought a long throw DTI) - fiddly but it can be done.

Bog's method is better - but if you don't have a mill to accurately centre drill it might be a bit better than the mark out and centrepuch method (depends on how good you are at marking out and centrepunching - I don't trust my eyesight in that department any more).

Ken
 
Quite apart from the actual indicating, I've found the greatest contributor to easily centring the 4-jaw is to use two (opposing) chuck keys at the same time - it's amazing how much faster it is than just using one. It's worthwhile to make another one if you don't have it.

Gary
 
Rayanth said:
Doc,

Thanks so much for linking your video, It confirmed that i was effectively doing it correctly (I had been rocking to find lowest point, etc) my only difference was not using tailstock but using magnetic base, mounted to the cross slide, and reading from the 'back' of the part.

This indicates it probably was a simple matter of not noticing an extra revolution, making my centering a full .1 off in two directions.

live and learn!

This brings to question though, if I want to mount square/rectangle stock OFF center....hmm....

- Ryan

What I do for putting stock in chuck off a predetermined distance is to mill up a little spacer block. Clamp the stock and the spacer block and then center off the spacer block. If its a small distance I do exactly what you did just use the indicator. There are many ways to do it some a little easier than others.

Gary
Using two chuck keys I agree is faster but with my back the way it is I like to keep the indicator on top so that kind of limits me to one wrench. It sucks getting old with arthritis my back some times flares and I have a tough time bending over at all.
 
doc1955 said:
Using two chuck keys I agree is faster but with my back the way it is I like to keep the indicator on top so that kind of limits me to one wrench. It sucks getting old with arthritis my back some times flares and I have a tough time bending over at all.
Yes, that sucks. Maybe you could make a very short key that could be used under the chuck - it must be very tedious to lower the work-piece when you can only adjust the top jaw.
 
So many great ideas.. but so many that are simply not doable when you only have a lathe and no milling attachment for it.

- Ryan
 
Rayanth said:
So many great ideas.. but so many that are simply not doable when you only have a lathe and no milling attachment for it.

- Ryan
I'm a bit confused, you do not need a milling attachment to center a block either square or rectangular in a four jaw chuck on a lathe.
Watch the video I posted you see no milling attachment just a dial indicator stuck in the tail stock. Or did I miss something.(wouldn't be the first time) :big:
 
Back
Top