quote:

Originally posted by Doug W:

quote:

Originally posted by kcstott:
Home made jigs and fixture. I'm a toolmaker so I don't buy things that I can make unless it's just inefficient to do so.

Thank you!!!

Now the follow up questions.
1. How straight do you know/think/or believe that bar is?

2. When you set it up, how much runout measured with one or two 0.0001" indicators do you consider acceptable say at a 4 inch spacing?

thanks again, Doug

In the one setup I've made I was able to get the bar to with in .0005" And it repeated with in .0007" to .0008" so I had some variance there.

As for max run out. I was on the .0005" band wagon i.e. +/- .00025" I have since rethought this and have now come to the conclusion that with in .001" from receiver face to the rear bearing area in the rear bridge i.e. approx. 4" to be with in .001" or +/- .0005" to be good enough for what i do. And I have an old south bend lathe and it's tough to get anything closer then +/-.0005" I have to stack everything in my favor to compensate if I need to get closer then that.

quote:

Originally posted by Doug W:

quote:

Originally posted by ted thorn:
At break today a couple other toolmakers here at the shop read over this thread. Lathes and tenths don't go together was the opinions of most....

Do that mean that they never bother using indicators graduated in tenths of an inch on the lathes?

What it means is using a lathe, any lathe to hold tenths is a head ache. We toolmakers (and Ted can attest to this) like to grind in tenths not turn. It can be done but that requires talking to a swiss screw machinist but a 4" part isn't going to happen Swiss screw guy operate at one inch or less from the chuck/collet

Because no matter what your indicator says before you make your cut you need to verify your runout after you make your cut And I'm willing to bet the\at with tool load and everything else you are fighting a very uphill battle with tenths and a lathe.

Doug
DME & PCS don't list straightness. But their stuff is as accurate as you can get.
Ever hear of using a "broken indicator" That is where the part you are measuring is so damn close your indicator doesn't move. That is what the ejector pins read and very consistently too.

I have ever confidence in the world that and ejector pin fresh from the box is as straight as my indicators can read. If not better.

...

Damn, that was easy!

quote:

Originally posted by Doug W:

quote:

Originally posted by ted thorn:
At break today a couple other toolmakers here at the shop read over this thread. Lathes and tenths don't go together was the opinions of most....

Do that mean that they never bother using indicators graduated in tenths of an inch on the lathes?

Yes it does.....I never use my tenth indicator on a lathe,mill,or EDM. I will occasionally use it on a grinder set up but primarily it is a surface plate tool.

...

...

For the record I just spent some shop money to check a couple 18" long EX33 .4995 DME ejector pins between centers and on both pins the TIR was .0025

Absolutely brilliant accuracy!!

I know a half inch pin will not slide into a bushing with a half inch hole.What clearance does the pin have to have to slip into the bushings? This is assuming the bore in the action is within how much?
Assuming Doug puts together the properly matched bushings and pin, what kind of tolerance stack are we seeing in the setup?

Butch. Exactly correct. On anything that I make I like .0005" clearance on a reamed hole. That is being optimistic though as even a reamed hole may not be exactly straight or perfectly round. When you work this close things you would never otherwise consider start to show up. Outside influences need to be controlled. Hystarisis.... it all adds up.
But in theory all you need is to be just larger then the out of round variation of the pin.

Even machinists who are reasonably confident of being able to repeat a set-up will prefer to perform all machining in one set-up. This is largely because they are aware that absolute perfection is not really verifiable.
The difficulty which will arise when trying to verify straightness, for instance, is almost insurmountable. In fact, it might be considered impossible to verify straightness with absolute certainty in most shop environments.
In the end, we all do the best we can and learn to work around the limitations imposed by our tooling and equipment.
I use mandrels, jigs, and fixtures which I make for specific tasks. If they allow me to repeat reliably, I consider them good. If not, I try again.
In general, a .0001 dial, when used for lathe set-ups, might tell you when a set-up is not perfect but it can't tell you when it is. At some level, that is taken on faith! Regards, Bill

quote:

Originally posted by Bill Leeper:
Even machinists who are reasonably confident of being able to repeat a set-up will prefer to perform all machining in one set-up. This is largely because they are aware that absolute perfection is not really verifiable.
The difficulty which will arise when trying to verify straightness, for instance, is almost insurmountable. In fact, it might be considered impossible to verify straightness with absolute certainty in most shop environments.
In the end, we all do the best we can and learn to work around the limitations imposed by our tooling and equipment.
I use mandrels, jigs, and fixtures which I make for specific tasks. If they allow me to repeat reliably, I consider them good. If not, I try again.
In general, a .0001 dial, when used for lathe set-ups, might tell you when a set-up is not perfect but it can't tell you when it is. At some level, that is taken on faith! Regards, Bill

Very well stated.
Thank you Bill.

Heck I thought we could start a side-debate about my pictures....we are only on page two

...

You roll the pin under the indicator at v block #1 and zero....

You roll the pin under the indicator at v block #2 to get a referance to pre-set zero from v block #1

Move to center to get TIR at mid point....

BTW the old black Starrett surface plate probably isn't very flat.

The two 1978 era Charmilles edm's in the background have DRO's that read in tenths but only burn in thousanths...however the grinders can hold tenths even though they are getting very old.

Probably would help if they had a good moldmaker to run them.

Guys I've been reading it with interest since all of it is greek to me. Could somebody help me out though, if I remember math tenths are not as fine as thousandths right? but reading this it sounds reverse. Is this possible to explain to a non-machinist or I wouldn't get it?

thanks.

Red

Tenths are assumed to be .0001 when measuring with indicators and micrometers.

Such a blue printing tool does not have to be straight. Only the datum areas that contact the receiver need to be in line with each other the rest of the shaft can be turned to provide clearance. Each end of the tool is center drilled. Then it is turned between centers. Each datum journal area should be turned to the exact diamter that is reqd. The function of the lathe running between centers keeps it straight.

Relieve the areas between the journals. Who cares what diameter they are.

Truing a receiver within +or- a half tenth is silly even if you can prove it.

quote:

Originally posted by Dago Red:
Guys I've been reading it with interest since all of it is greek to me. Could somebody help me out though, if I remember math tenths are not as fine as thousandths right? but reading this it sounds reverse. Is this possible to explain to a non-machinist or I wouldn't get it?

thanks.

Red

Dago...."tenths" is just shop lingo for .0001 but in reality it is really .1 but not used in such fasion in a tooling respect

...

...

Whatever.......back to work $$$$$$

Doug
In reference to you post about not turning mandrels anymore. I would advise against that.
No fixture no mater how well it is made will return to position exactly every time. This is the whole point of having a machinable surface. It eliminates the need for dialing in oh so delicately the fixture. it also eliminates any user induced run out as once you make a cut it is a perfect as you machine can make it.
Which was in a previous post. The fixture is only as accurate as your lathe can make a cut. If you have a dip in the ways the cuter is dropping down and the part gets larger in diameter if turning the od. That was where I was saying this all depends on your machine. We can split hairs all day long on how accurate your set up is. But your lathe will not cut it to what you've got it dialed into. Tool load. material hardness, tool sharpness, machine lack of precision all add up.

This is a big reason why we like to set a part up in the lathe and take a cut to true it up. not indicate it to true it up at least not as these levels of precisions.

I use off the rack S-7 with good results. Just as good would probably be 0-1, all this stuff is straighter than it should be for the price and my choice of S-7 is because I buy it once or twice a year with a minimum of 400 pounds.
Typically I hit the mean to the thou on turning as most tol are .003 or .005 so accuracy is subjective. When turning journals where I have +.0000 to -.0005 I hand hone HSS with 8% cobalt and try to do this operation first so if I screw up I won’t have hours into the job. Tighter than that I will grind myself or spin +.002 and send out if bigger than my grinder can handle. Most of this work is done on a precision Tool and Gauge lathe, not the typical Gunsmith type lathe.
Most guys when typing wishes tend to believe whatever they see posted. Not too long ago it was written about holding zero to the tenth on disposable Chinese lathes with dishwasher grade bearings and beds made out of scrap Chevy’s right off the barge. Lathes typically cut to three places and grinders cut to four. Even the big Cincinnati Milacron grinders I‘ve set up before only went to the tenth and that was just the dials. Earl.


You don't use dial test indicators to measure tenths.