how bad is my mill?

i have facemilled a part 650 mm long on a deckel type machine. its 0.02 mm high in the middle. how does that compare?

That's 0.00078'' out over 25.6''? .................I'd say that mill was bloody good if not chuffin amazing!

It's a milling machine - not a surface grinder

FWIW and to give you a rough idea - from memory!! ;- as a rule of thumb, Payen - the gasket people used say (and might still do) that a cylinder head or black face should be flat with 0.003''

P.S. I wasn't far out http://www.fme-cat.com/Docs/1423.pdf

Depending on your inserts its easy to get enough compression in the milled surface to cause a perfect bit to bow a lot more than that much! Im with sami, time to start worrying about something else, Martian invasion? Over bearing goverments? Terrorist attack? Tin foil hat will probably help there.

so i would change the title to "how good is my mill" if i could. that was before unclamping, btw.

dian said:

so i would change the title to "how good is my mill" if i could. that was before unclamping, btw.

Click to expand...

How did you measure?

dian said:

so i would change the title to "how good is my mill" if i could. that was before unclamping, btw.

Click to expand...

Well... culturally.. the Swiss can exhibit a dry, droll and understated way of "backwards" bragging...

And it AIN'T "bragging" since you actually DID it, so.. Guess you have the right to call it either way.

Until someone else figures out how to steal that mill without being shot, caught, or jailed, anyway..



Bill

Screwmachine said:

How did you measure?

Click to expand...

Hopefully not by putting an indicator in the spindle and running it down the length of the part.

dian said:

i have facemilled a part 650 mm long on a deckel type machine. its 0.02 mm high in the middle. how does that compare?

Click to expand...

.
.
common to take a light finish passes of less than 0.5 and then 0.05 mm after rechucking part. that is loosen part in fixture or chuck and lightly retighten. this machines part warpage and chucking distortion out

Tedger said:

Hopefully not by putting an indicator in the spindle and running it down the length of the part.

Click to expand...

That's what I'm guessing!

0.02m/m high in the middle thats a sign of a bad m/c-suppose I could take it off you for rough work

onecut said:

0.02m/m high in the middle thats a sign of a bad m/c-suppose I could take it off you for rough work

Click to expand...

You're too late mate - the best peas went to Farrows - the deal's done.

Screwmachine said:

How did you measure?

Click to expand...

straightedge and shims.

(then it was blued up and sanded flat on a granite plate.)

Prepare for crickets as some lose their shit-talking points. Someone will prefer you use some thermal stabilized argon laser chamber with quantum field generating anti-gravity containment, using a ceramic sphere accurate to .0000001 per degree with automatic sampling software on a cray supercomputer... it'll happen - this is PM.

i was expecting a few people comming forward with what accuracy their mills produce to have a comparison, as i dont have the chance to use the full traver on mine very often.

All mills run correct with the ways by definition. If the ways are stuffed you can’t make anything flat (from the machine point of view).

It’s like Sami said (WTF that’s really good)… On the bench you check part section point to point many – many times (reference over a datum, part free), then you check bow (however you wish).

If all that stuff checks out right @ your numbers, “you don’t have to take any shit from anybody” & that’s some really stable stuff you machined there…. Let the world know where you got it!!!

Nice job,
Matt

To get that level of flatness and tolerance is indeed an excellent result.You wish to know how it compares to other peoples machines,I cant speak for others but from my past experience of of various machines I would say it cant get much better the next level of improvement would be the grinding machine.I have one word of caution I would wait untill I had done a few more batches before I was certain of my results you might put the next repeat order on and find things arnt the same and the tolerance is varying believe me it happens

Did you mill an arc or did the material distort from having material takin off of one side?

SIP6A said:

Did you mill an arc or did the material distort from having material takin off of one side?

Click to expand...

.
.
flatness, straightness and perpendicularity on my machine is
.0003" per 40" or .0075mm per meter
.
when machines were new it could do .0002" per 40"
.
i routinely machine surface so they will not need scraping or if hand scraping is done it is just for oil retention. cnc machines been doing it over 2 decades on thousands of parts
.
obviously part needs to be on Airy points or it will sag from its own weight. and final machining cuts are after rechuck at low bolt torque often only 30 in/lbs and is often .001" or .0005" passes. i check machine by indicating granite square with .00005" indicator. then i indicate part. after parts are checked again with granite squares in inspection. older machines you have to be careful about reversing direction. even .0005" backlash can create problems at times. also servo oscillation often needs programmed delay time to wait for machine to stop bouncing
.
2nd picture shows small part about 0.8 meter long having .0005" cuts to remove part warpage after light rechuck. note the marker lines being removed as passes done.
.
Modern facemills usually can create surfaces as good or better than rough grinding. usually waves are under +/-.0001" although some parts i have seen +/-.0002" waves. Feeds and speeds and insert type obviously are important.
We got some newer facemills in and the surface they leave are shiny not quite mirror finish but almost. you usually see reflection off surface fuzzy a bit and not mirror sharp image but not bad for a facemill. they run higher sfpm and part vibration is often the real limit not the carbide insert sfpm limit
.
also part temperature control is critical. obviously part will change size as it warms up

Hardened said:

Prepare for crickets as some lose their shit-talking points. Someone will prefer you use some thermal stabilized argon laser chamber with quantum field generating anti-gravity containment, using a ceramic sphere accurate to .0000001 per degree with automatic sampling software on a cray supercomputer... it'll happen - this is PM.

Click to expand...

Your wrong on your math, the ceramic sphere accuracy is off by two places, not enough 0's in there, common man, this aint trade school.

many a part can be machined on a perfect machine and indicate perfectly flat but once unchucked it can spring or warp considerably.
.
i have seen big parts warp when unchucked easily 0.1 to over 1mm quite often. it requires careful work procedure and programming to machine flat surfaces. not something you can do carelessly and with no effort.
.
i have often modified programs to take 3 extra .0005" passes to meet a .0002" spec
.
still sometimes parts come back from inspection and need recut cause over .0005" out of perpendicularity spec, why, not sure maybe part warps more after sitting for days and weeks