TRAMMING A HARBOR FREIGHT MILL

epanzella

epanzella

Active User
Registered
Joined
Apr 8, 2013
Messages
2,021
I just got thru tramming my HF 33686 round column mill. I thought I had it pretty good. I took a skim coat with a flycutter (first time I ever used one) on both sides of a 5 x 5 x 1/2 inch piece of aluminum. I put it in a vice on parrallels and tapped ti down. The result was .0015 inch difference from the smallest to the largest measurement. Is this what I should be expecting from a low priced Chinese mill or should I be doing some fine tuning on the tramming?
 
The answer to both of your questions is yes. But there can be a lot of other things in play as well. Chip on the parallel keeping the work from seating fully. Jaw lift on the vise. Nod on the head. You checked that right by measuring runout on the y axis? In which axis is the measurement changing? x, y, xy?

And then there is the question of what tolerances you need to hold. 0.0015 in 5 inches is a little over .003 per foot. For most hobby work that is probably good enough. But if you need better you can keep hacking away at it until you get it measuring closer. It can be a real hair pulling process though.

And keep the children away when you make a cut and measure the result, no need to teach them any bad words and they may fly thickly for a while.
 
MontanaLon said:
The answer to both of your questions is yes. But there can be a lot of other things in play as well. Chip on the parallel keeping the work from seating fully. Jaw lift on the vise. Nod on the head. You checked that right by measuring runout on the y axis? In which axis is the measurement changing? x, y, xy?

And then there is the question of what tolerances you need to hold. 0.0015 in 5 inches is a little over .003 per foot. For most hobby work that is probably good enough. But if you need better you can keep hacking away at it until you get it measuring closer. It can be a real hair pulling process though.

And keep the children away when you make a cut and measure the result, no need to teach them any bad words and they may fly thickly for a while.
Click to expand...
There were no chips on the vice or parallels. I trammed the mill using shims under the column and a .001 DI in one of those gizmos that clamps on the quill. I first tried a cheap Chinese .0005 DTI but the POS wouldn't make the same reading 2 times in a row so It's found a new home in the junk drawer. I got the X perfect but I was having a bit of trouble reading the Y with the DI stem jumping over the T slots. It seemed to be within about .001 or so. I thought this would be OK because I didn't really have high expectations for a $960.oo machine. This was a while back and since I have found that what I thought were shortcomings in the machine were actually operator inexperience. I'm raising my sights a bit higher now. Like a dope I didn't number the corners on the 5x5 test piece so I don't know what axis was deficient. I'll have to do another skim pass with my head out in the sunlight.
 
Buffalo21 comment above on Bridgeports seems to imply that they also can be inaccurate. If I misunderstood, my apologies. But what I like, to defend Bridgeports is that my mill is as accurate as I want to spend the time on. The alignment functions are designed in, and easy to use...loosen, adjust, tighten, verify.

Re-machining for squareness, parallel, etc. is what I do to cheap vises, spin indexers, and other accessories.
 
Winegrower said:
Buffalo21 comment above on Bridgeports seems to imply that they also can be inaccurate. If I misunderstood, my apologies. But what I like, to defend Bridgeports is that my mill is as accurate as I want to spend the time on. The alignment functions are designed in, and easy to use...loosen, adjust, tighten, verify.

Re-machining for squareness, parallel, etc. is what I do to cheap vises, spin indexers, and other accessories.
Click to expand...

my comment was meant, that without proper set up anything can be inaccurate, implying its inaccurate only because its an import is wrong.
 
epanzella said:
.... Like a dope I didn't number the corners on the 5x5 test piece so I don't know what axis was deficient. I'll have to do another skim pass with my head out in the sunlight.
Click to expand...
Also make a note whether you flip the work piece front to back or end to end to cut the the backside. You might even try both ways to see if it matters.

Tom
 
There's a good reason engineering drawings have tolerances called out. If you need to work to tenths it might be wise to revisit your machine (and measuring tool) budget. Otherwise it sounds like you are pretty darn close and will have more fun making stuff than worrying about a level of accuracy likely not needed for the work at hand.

Lots of folks do excellent work with the round column mills, even the ones from Harbor Freight. As you gain experience you will get better at making stuff more accurate and the people on this forum can definitely help. Only you can determine what level of accuracy you need for the projects you're working on, assuming you're not making stuff for paying clients.

Most machine shops have a separate room, often climate controlled for metrology, this and verifying parts are within spec is a whole science in itself. Keep at it if you need better than what you have, but do establish a baseline that's acceptable for the work you are doing or it's likely to drive you nuts.

John
 
When I was just getting started with my mill, I trammed the back face of the vise, but could not tram the top of the vise until I took the whole vise apart and scrupulously cleaned all the parts (WD-40) and then reassembled.

Also note: working to 0.001 tolerance requires exacting procedures::
a) deburr machined edges before reinstalling in vise
b) vise faces and parallels must remain scrupulously clean
c) part must be tapped down onto vise
c.1) sometimes both before and after tightening

d) the head must be trammed to essential perfection
e) the nod must be trammed to essential perfection (i.e., shims)

Any unsquareness in the setup can cause errors of the size you mention.
 
You must log in or register to reply here.
Back
Top