Thread: Hole locations on Prototrak mills ???

Assuming a retrofit Prototrak mill is in reasonably good shape, what kind of hole location tolerance do you think a machine should be able to hold if you are center drilling, drilling and reaming the holes ???


Frank

We can hold'm to .0005" most any time.

Originally Posted by Zahnrad Kopf

We can hold'm to .0005" most any time.

is that what the machine says or actual measured parts on something repeatable in that accuracy?

True position of .0005? I would like to see proof of that!

I would guess you should be able to hold .010 true position, maybe .004 if the pre-reamed (post drilled) holes were bored first, by boring head or u/s end mill.

http://blog.pptvision.com/Portals/39.../TP_Image1.jpg

Originally Posted by rockfish

Assuming a retrofit Prototrak mill is in reasonably good shape, what kind of hole location tolerance do you think a machine should be able to hold if you are center drilling, drilling and reaming the holes ???
Frank

I had a old Prototrak Plus I would check DRO every year. It was a simple test. Clamp a 6" parallel in Y axis direction in a Mill vise opened up 6". Then clamp a digital caliper to it. Extend caliper all the way, zero it and zero DRO. Use mill crank dials to push caliper closed 5.000" on DRO and see if caliper says 5.000 as picture shows.
...... Their is a procedure to correct DRO reading which I have seen 0.001 to 0.004" off per 5"
.
Also all turret head machines like a Series 1 Bridgeport need their head trammed. Basically if mill has it's drill quill at a angle then as quill is extended it's drill bit tip is drifting in the X and Y axises. I check the alignment of the head every day on critical jobs. I have seen heavy milling move it out of alignment 0.5 degrees , often small enough not to be noticed for some time.
.
Machine rigidity of tool in collet check with a dial indicator, usually it does not take much to move it 0.001" or more. Basically drilling requires spotting or using a short center drill, then a pilot drill bit, then a bigger bit. If drilling a big holes with no spotting or pilot hole can drift easily 0.005" or more. With care I have seen holes on location within 0.002" on average most of the time. With no checking and little effort holes can easily be off 0.005"

I would not guarantee anything better than +/- a couple of tou on a very good machine (short pitch jobs).
For longer jobs, tolerance rises.

check it with indicator and size blocks , couple thou from what ive seen. as good as the scale that on the machine i guess, readout can be calibrated to blocks etc after cleaning but its still only going to repeat w/in xxxx

On a knee mill you have too many places for a stack up of "lost motion" to ever think you can reasonably hold locational tolerances closer than +/- .002 and as has been noted +/- .005 is likely. On the Proptotrak bedmills its a different animal so that you could probably hold locational tolerances of half of that which is possible on a knee mill.

Frank - if you have a job that is calling for closer hole positioning than that and you don't have the capabilities in house, then I usually drill and ream the holes undersize and then let the edm'er work his magic and put the holes right on the money and he can hold size within a couple tenths all day long. If they're blind holes it's a differerent issue.

I think +/- .002 is pretty reasonable. Use a spot drill, not a center drill. I use a 90 degree and spot to about 75% of the drilled hole diameter. Then drill and ream. You can use the spot drill again to chamfer after the hole is drilled.

http://www.guhring.com/Documents/Cat...SpotDrills.pdf

I can hold +/- .002 location, and +/- .0002 diameter on reamed holes using a rigid ram CNC knee mill.


BTW, that caliper method is pretty shady. That method would only be as accurate as the calipers, which is no going to be better than +/- .002.

A better method would be to ring two pins on the end of a gauge block stack, or use tool maker's buttons. Sweep in the first pin and zero DRO. Sweep in second pin and note the DRO reading. It should be the length of the gauge blocks plus the diameter of one pin (assuming they are the same diameter).

My apologies - it was late and I have very fat thumbs... One too many zeros in there. It was _supposed_ to have some humor invoked... I suck.

Originally Posted by Joe Miranda

On a knee mill you have too many places for a stack up of "lost motion" to ever think you can reasonably hold locational tolerances closer than +/- .002 and as has been noted +/- .005 is likely. On the Proptotrak bedmills its a different animal so that you could probably hold locational tolerances of half of that which is possible on a knee mill.

Frank - if you have a job that is calling for closer hole positioning than that and you don't have the capabilities in house, then I usually drill and ream the holes undersize and then let the edm'er work his magic and put the holes right on the money and he can hold size within a couple tenths all day long. If they're blind holes it's a differerent issue.

Yes, I'm trying to hold stuff within .002 of location. It's not happening. No, EDM isn't an option......there is absolutely no way this customer would pay that kind of money. They want perfection, but they aren't willing to pay for it.


Frank

Originally Posted by rockfish

They want perfection, but they aren't willing to pay for it.


Frank

Frank,

That sounds like 99% of my customers

We just finished a job...The machine was just tightened and tuned (annually done) and a .047 drill even with drill bushing guide found hole true position to vary from best .0009 to .0039 as measured on a 200k new CMM.

Can it be done better...maybe, we think we can improve our process to eliminate the .0039's and keep the holes around .002 TP or better with about 2% fallout.

With the machine (AcuMill, Kent) tuned and tight as said above... repeat on position is realistic at +/_ .0005 that what about .0013 TP? Anyway, drill drift, and material inconsistencies are the problem.

Steve

A buddy of mine's got a pair of Prototraks and just went through this same discussion. The machines were new to him, so he asked the service tech what should be acceptable and was told, "about 5 thou."

Frank, these are bridgport style turret mills or "bed mills"? makes a difference in this discussion. The bed mills are generally better on repeatability as thats ow basically all vmc's are.

I have a similar Milltronics bedmill with ballscrews that have 0.003" slop. I often make parts that are located to each other with 3/16 dowel pins up to 8" apart. The dowel pin holes are on the same plane. If I program the machine to approach the holes from the same direction I can get parts that fit together.

I worked on an old Prototrak Plus converted Series 1 Bridgeport turret knee mill.
.
Backlash compensation procedure measures and remember backlash but it uses the DRO reading and not the ball screw turning. I had an old Trav A Dial encoder wheel which is a fine tooth gear wheel in contact with side of machine that confirms movement. If a chip was stuck in it it would give full power to feed until it confirmed movement. Usually a couple of inches which breaks end mills very fast. I learned to replace chip scraper / wiper often. The gibs or slide tightness needs to be moderately tight. If not you tend to have parts easily move 0.002" or more. Also make sure DRO either Trav A Dial or glass scale encoder is tight. If screws are loose you will have hard time getting repeatability.
....... Tightening the gibs, adjusting backlash compensation and calibrating DRO is needed at least every year. A problem you might get is on slides the gib feels looser in center of travel compared to ends of travel. This is wear and why old timers would deliberately have vise off center to the right for months then to left of table for months to try to get wear more even. Making sure gibs or slides are moderately tight is .... well basic beginner stuff. By the way sometimes a Gib has a screw on each end, one pushes tighter but the other end needs loosening to allow gib to move. Basically gib is pushed on each end. I once tightened screw so tight I busted head off screw as I could feel and dial indicator measure slop / movement. Once I got my head to see other end of gib had a screw and backed that one off, the replaced forward gib screw tightened slide up with no problem.

I put together a small fixture on a prototrak bedmill. Main thing was the plate and 2 nesting blocks to capture the part. It wasn't any really tight tolerance, just like maybe +/- .002". They wanted me to put the screw holes in the plate, mount up and have a guy set it tight on the CMM, then go back and transfer dowel holes. I went ahead and put everything in as to be finished. The part to go on was only about 14-16 inches, the dowel holes were about 2 inches apart in each block. I put the dowel holes in the blocks and plate all indiviually on the prototrak. Reamed .2500" in the blocks for press fit and .2505" in the plate for slip fit. They fit well. It went in to the CMM to find the length between the blocks was out about .0006" or maybe .0008", and this is measured between pockets in each block to add more "potential" error to the location. Of course, the foreman person grawled about it when he had seen I already pinned it, but it worked out well.

I know there is a lot more potential error on a knee mill, but if you guys cant figure out how to do better than +/- .005", maybe you should get into woodworking. One should easily be able to do that on a machine that's well worn out.

As for calibrating...... a caliper?... come on...... use a damn guage block, preferably a 6".

Yes, the pick-ups on those old prototraks can get crap under them and can make you angry. Proper maintenance can go a long way

Anyway.... just saying.... I spect I'll make everybody mad. Tee hee. But really, come on ................. thought some you all was machinists???

calibrating with a caliper works very, very well, yes you can use a gage block too. i check caliper to gage block to see it it agrees to within 0.001" per 6". never had one not do that.
.
using digital caliper gets DRO to agree with my caliper and has quickly detected loose screws on glass scales or Trav A Dial not under enough tension
.
i have used quick digital caliper calibration on ALL lathes and mills in the shop quickly picking up problems and correcting them. I check all equipment if necessary before EVERY critical job. My repeatability rate went up at least 10x. Use what ever you want to calibrate a DRO. Loose screws on a DRO pickup can be not easy to detect. I have easily detected them, tightened screws and fixed the problems.
.
By the way as a apprentice I measured with my caliper a gage block 0.001" under size. No one believed me. I insisted and check with 2 other calipers and then 0.0001" micrometers. All had agreed. I insisted and went to teacher who pulled out gage block set inspection paper showing that the one gage block was 0.0012" undersize. All the apprentices but me failed the test. It was a test to NEVER trust even a gage block.

Originally Posted by aerodark

I would guess you should be able to hold .010 true position,

.010 thou? geez, guys could do that with a scribed line and center punch on the drill press.