Accuracy question, scraping vs CNC - Page 3
Largest Manufacturing Technology
Community On The Web
Close
Login to Your Account

Page 3 of 5 FirstFirst 12345 LastLast
Results 41 to 60 of 92
  1. #41
    Join Date
    Jun 2004
    Location
    Melbourne Australia
    Posts
    4,704
    Post Thanks / Like
    Likes (Given)
    548
    Likes (Received)
    1617

    Default

    Quote Originally Posted by Tyrone Shoelaces View Post
    While we're on the subject - anybody know how many angels can dance on the head of a pin?
    How many angels can dance on the head of a pin? - Wikipedia

    Wiki tells me that's a Scholasticism. Now, I like a big juicy Latin just as much as the next bloke, But that's taking the piss. Couldn't they just say "for wasting time debating topics of no practical value, or questions whose answers hold no intellectual consequence, while more urgent concerns pile up"

    Who'd have thunk it? Me, Pete F & Scholasticism I've never meet Rosie. And I still have my Watch.

    Hence Scraping / lapping vs CNC vs Gauge / Gage / Jo Blocks. Its cactus going on OP @ post #1

  2. Likes Monarchist, Garwood liked this post
  3. #42
    Join Date
    Apr 2006
    Location
    Manchester, England
    Posts
    5,971
    Post Thanks / Like
    Likes (Given)
    689
    Likes (Received)
    2924

    Default

    You need to be careful. You Aussies will be getting a reputation as intellectuals.

    Regards Tyrone.

  4. Likes Monarchist, Garwood liked this post
  5. #43
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    6,049
    Post Thanks / Like
    Likes (Given)
    2390
    Likes (Received)
    2859

    Default

    Quote Originally Posted by Tyrone Shoelaces View Post
    You need to be careful. You Aussies will be getting a reputation as intellectuals.

    Regards Tyrone.
    Unless there has been a dramatic change - as there could well be - it shall have to take third-place as a national sport, even so.

    Back around 1968-69 when I was in discussion with the Embassy staff here in Washington about emigrating, up pops the statistic that Australia was the most heavily unionized of all developed countries. And had the highest statistical incidence of homosexuality. ISTR it was MacKay as wrote that.

    Second part I figured wasn't really any different from anywhere else, just that the Oz didn't generally give a rat's ass who knew what they got into and had answered survey questions more honestly than average. Or were actually straight and chose to take the piss. As a girl-p***y, not man p***y "addict', it didn't matter to ME anyway.

    First part, however, was a total show-stopper. Family have been Union, both sides ever since there have BEEN Unions. USWA myself. I just don't like the sort of rigidity and "we-they" polarization that can create and sustain in a work environment, didn't choose to go back to it, ever again.

    Turns out from many visits since, the far bigger worry was those damned heavily-armed air-superiority BUSH FLIES.

    Not YOU, Phil.. the ones with wings...


  6. #44
    Join Date
    Dec 2008
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    6,836
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1837

    Default

    Quote Originally Posted by pcm81 View Post
    This is purely theoretical question; just trying to learn new things / skills...

    How does accuracy achieved by hand methods like scraping compare to hand Lapping techniques and how do those two compare to accuracy achieved by modern CNC machining?


    I am considering to learn hand scraping and just trying to gauge the usefulness of this skill in terms of "metrology level accuracy".

    The project I am considering is making a gauge block; i will be testing it with 50 millionth gauge (i realize its not a real "metrology grade" block comparator, but that is what i have at the moment) vs grade B gauge block set. What is the practical difference that I should expect to see between lapping and scraping for the final finish? I have a variety of lapping compounds, used to be into knife sharpening big time...

    Other than in a low budget situation when there is no modern machining available for lapping / CNC machining; is there still a place in modern high dollar industry for hand scraping? I am thinking in terms of accuracy.

    Thanks ahead. Throw all info you have at me, trying to learn from "the old folks" what modern day text books won't teach ya...
    .
    .
    i cnc mill surfaces to .0005" per 40" tolerances everyday. usually lap or variations between passes is less than .0003". although modern insert mills can leave a shiny surface that looks like grinding it often has waves +/-.0001" every .010" to 0.100"
    ..... sometimes milled surface is scraped later. to lower even the .0005" error and to add a oil retention pattern. if you want better than .0002" than you have to use something other than milling. even holding a .0003" tolerance is difficult.
    .
    by the way sanding or lapping down even .0002" waves takes considerable time with larger surfaces. you might think it would take only a few seconds but you can be there 30 minutes and still not get all the waves out

  7. #45
    Join Date
    Apr 2014
    Location
    USA FL
    Posts
    100
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    15

    Default

    Quote Originally Posted by DMF_TomB View Post
    .
    .
    i cnc mill surfaces to .0005" per 40" tolerances everyday. usually lap or variations between passes is less than .0003". although modern insert mills can leave a shiny surface that looks like grinding it often has waves +/-.0001" every .010" to 0.100"
    ..... sometimes milled surface is scraped later. to lower even the .0005" error and to add a oil retention pattern. if you want better than .0002" than you have to use something other than milling. even holding a .0003" tolerance is difficult.
    .
    by the way sanding or lapping down even .0002" waves takes considerable time with larger surfaces. you might think it would take only a few seconds but you can be there 30 minutes and still not get all the waves out
    Thank you for the info. It seems to agree with information posted earlier that scraping can get you into couple/10,000 - couple/100,000 range. Ultimately, combined with proper checking vs a calibrated flat the smoothness and flatness that can be achieved by method X will depend on how shallow of a cut such method can produce. It sounds to me like scraping can be in 100,000th range. Although I’d guess it will also depend on how hard the material is.

  8. #46
    Join Date
    Dec 2008
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    6,836
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1837

    Default

    Quote Originally Posted by pcm81 View Post
    Thank you for the info. It seems to agree with information posted earlier that scraping can get you into couple/10,000 - couple/100,000 range. Ultimately, combined with proper checking vs a calibrated flat the smoothness and flatness that can be achieved by method X will depend on how shallow of a cut such method can produce. It sounds to me like scraping can be in 100,000th range. Although I’d guess it will also depend on how hard the material is.
    .
    .
    i routinely mill a extra .0005" pass if surface or edge is curved. often it is a deflection or temperature difference issue. none the less if .0006 off and tolerance is .0005 i often take .0010" pass or 2 passes of .0005"
    .
    boring a hole if diameter .001 small thats .0005 on radius being cut. with a bored hole it, cast iron can be stubborn and not respond 1 to 1 with cuts of less that .001" dia difference. if i am .0002" under size on a bored hole i am more likely to try to sand or lap it
    .
    on cnc mill if servos are oscillating back and forth sometimes this shows up as waves too. this i always have to watch out for
    .
    cnc have backlash and backlash compensation. small amounts of even .0005" has caused problems before too.
    Attached Thumbnails Attached Thumbnails servooscillation.jpg  

  9. #47
    Join Date
    Sep 2010
    Location
    Oklahoma City, OK
    Posts
    3,014
    Post Thanks / Like
    Likes (Given)
    491
    Likes (Received)
    1291

    Default

    Quote Originally Posted by Pete F View Post
    What are you actually asking here? Your question seems so vague and stupid I suggest you head over to Home Sop Machinist and run it by the experts there.

    Your thread title is "scraping vs CNC". WTF? That makes no sense.

    Hopefully The Boss will delete this type of crap.
    Weren't you just recently getting your ass kicked trying to turn 4140ht?

    Insert life turning 4140 HT

  10. Likes Mcgyver, Garwood, ManualEd, henrya, Demon73 liked this post
  11. #48
    Join Date
    Aug 2005
    Location
    Toronto
    Posts
    4,360
    Post Thanks / Like
    Likes (Given)
    2273
    Likes (Received)
    2445

    Default

    Quote Originally Posted by Cole2534 View Post
    Weren't you just recently getting your ass kicked trying to turn 4140ht?

    Insert life turning 4140 HT
    well, as he tries to learn machining, you got to hand it to him....he's sure got the mean nasty grumpy part nailed.

  12. Likes Monarchist, Cole2534, ManualEd, Demon73 liked this post
  13. #49
    Join Date
    Jul 2008
    Location
    Sydney, Australia
    Posts
    3,753
    Post Thanks / Like
    Likes (Given)
    2054
    Likes (Received)
    1277

    Default

    Quote Originally Posted by Cole2534 View Post
    Weren't you just recently getting your ass kicked trying to turn 4140ht?

    Insert life turning 4140 HT
    No, my inserts were getting their ass-kicked, I was fine thanks for asking. I normally machine 1020 dry with cermets, turned out the 4140 HT needs coolant, which I did, problem solved. In the meanwhile I was told by "I'm a professional" that my inserts were the problem, I was feeding too fast, feeding too slow, had too much speed, not enough speed, and virtually every combination in between. So WTF does that thread have to do with this smart-arse?

  14. #50
    Join Date
    Sep 2010
    Location
    Oklahoma City, OK
    Posts
    3,014
    Post Thanks / Like
    Likes (Given)
    491
    Likes (Received)
    1291

    Default

    Quote Originally Posted by Pete F View Post
    No, my inserts were getting their ass-kicked, I was fine thanks for asking. I normally machine 1020 dry with cermets, turned out the 4140 HT needs coolant, which I did, problem solved. In the meanwhile I was told by "I'm a professional" that my inserts were the problem, I was feeding too fast, feeding too slow, had too much speed, not enough speed, and virtually every combination in between. So WTF does that thread have to do with this smart-arse?
    Just a reminder that you've been objectively excluded from the Knowitall club. Maybe go a little easier on the new guys.

    Sent from my SM-G930R4 using Tapatalk

  15. Likes James H Clark, henrya, Demon73 liked this post
  16. #51
    Join Date
    Jul 2008
    Location
    Sydney, Australia
    Posts
    3,753
    Post Thanks / Like
    Likes (Given)
    2054
    Likes (Received)
    1277

    Default

    Quote Originally Posted by Cole2534 View Post
    Weren't you just recently getting your ass kicked trying to turn 4140ht?
    BTW, these were what I was making at the time. Not a great photo as all the flats just reflect the workshop lights. Doesn't look much like I was "getting my ass kicked" to me. Idiot!

    t-bolt.jpg

  17. #52
    Join Date
    Sep 2010
    Location
    Oklahoma City, OK
    Posts
    3,014
    Post Thanks / Like
    Likes (Given)
    491
    Likes (Received)
    1291

    Default

    Neat.

    Sent from my SM-G930R4 using Tapatalk

  18. #53
    Join Date
    Jul 2014
    Location
    Ontario, Canada
    Posts
    746
    Post Thanks / Like
    Likes (Given)
    570
    Likes (Received)
    495

    Default

    Quote Originally Posted by Pete F View Post
    In the meanwhile I was told by "I'm a professional" that my inserts were the problem,
    Inserts were part of the problem. You don't need cermet inserts to turn 4140HT.

  19. #54
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    6,049
    Post Thanks / Like
    Likes (Given)
    2390
    Likes (Received)
    2859

    Default

    Quote Originally Posted by Pete F View Post
    BTW, these were what I was making at the time. Not a great photo as all the flats just reflect the workshop lights. Doesn't look much like I was "getting my ass kicked" to me. Idiot!

    t-bolt.jpg
    If that is what 'the finish is good' looks like I'd guess you've scant need of a TP grinder... it IS 'good'.

    OTOH.. if you'd have admitted that stock drill-rod threaded and Loctite attached to the Tee head was good enough, you'd not have had to use-up any inserts at all to make mostly chips out of a solid clear up at the Tee end's diameter.. or better.

    Someone else can do the math, but IIRC, that sort of one-piece rod to body transition, even if not sharp-cornered, actually fails before two-piece with decent depth threads will do.

  20. #55
    Join Date
    Jul 2008
    Location
    Sydney, Australia
    Posts
    3,753
    Post Thanks / Like
    Likes (Given)
    2054
    Likes (Received)
    1277

    Default

    Quote Originally Posted by RC Mech View Post
    Inserts were part of the problem. You don't need cermet inserts to turn 4140HT.
    I didn't use cermet inserts, if Einstein above hadn't interjected with some pointless attempt to discredit me, that actually blew up in his face and made him look like a tosser, we wouldn't be having this discussion. You'd need to read through the thread as to how it played out. I should take some photos of the finish with my proper camera with close-ups to try and capture it properly, but have coffee to drink, and it has nothing to do with this thread. But I think it's sufficient to say the finish and tolerances at +0.00/-0.05 on all dimensions were "adequate". So much for that "ass kicking".

    Just a reminder that you've been objectively excluded from the Knowitall club. Maybe go a little easier on the new guys.
    Just a reminder you'd been made to look like a dickhead for trying to stir the pot unnecessarily. Better luck next time.

    OTOH.. if you'd have admitted that stock drill-rod threaded and Loctite attached to the Tee head was good enough, you'd not have had to use-up any inserts at all to make mostly chips out of a solid clear up at the Tee end's diameter.. or better.

    Someone else can do the math, but IIRC, that sort of one-piece rod to body transition, even if not sharp-cornered, actually fails before two-piece with decent depth threads will do.
    I have no doubt you are absolutely right, and I think that is the smarter way to go. It was a copy of existing T-bolts, and that's how they did them. Turning from solid was a waste of material too. Not really a problem if I was "CNCing" them , but on a manual noodle somebody has to stand there and flick the levers! I want to do some the same way in 1020 just out of curiosity to compare, and will do that when I get a few minutes, they don't take long to turn out. I was pleased to go through the exercise with the 4140 HT however as I don't normally work with it much.

    The whole point, and what Einstein seemed to have missed, was that there is absolutely nothing wrong with asking something like how does the surface finish of lapping, scraping, and milling on a typical modern VMC with typical tooling compare? People can provide actual numbers for that. Similarly, what sort of tolerances should you expect from a lapped, scraped, or milled part, each being run by typical modern methods? A bit of a fuzzy question, but still reasonable. Asking how to make a "gauge block" either by "CNCing" or "Polishing" is not a reasonable question for this forum. Nothing to do with whether the person if a professional, amateur, new, or been here since the forum started. It's just plain and simple a stupid and meaningless question.

  21. Likes Monarchist liked this post
  22. #56
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    15,704
    Post Thanks / Like
    Likes (Given)
    1114
    Likes (Received)
    2102

    Default

    Quote Originally Posted by machtool View Post
    Your the Engineer. Research is a part of that yes?. Have a look at the post count. 15,262. That bloke would come to the opening of an envelope. Regards scraping, he is a fraud that parrots, what he has read here. Research requires sorting through the chaff from the hay?

    When I first joined decade & half ago, he wanted to know which Loctite, would expand his Conrad style bearings to take out the play and put them into pre-load, on a Logan lathe. He's Harry Home Shop at best, and bother's electrons, as a side line.
    Ah, fuck off.

    You don;t know shit about what I do. You THINK you do, but you are as wrong as a dick on your neck..

    No clue what the heck the loctite thing is, but it's about as helpful as anything else I see you write, so I am not even gonna worry about it.

  23. Likes Monarchist liked this post
  24. #57
    Join Date
    Jul 2014
    Location
    Ontario, Canada
    Posts
    746
    Post Thanks / Like
    Likes (Given)
    570
    Likes (Received)
    495

    Default

    Agreed. Scraping and hand lapping have nothing to do with accuracy and see & see doesn't qualify anything, much less surface finish. A more apt comparison is milling vs. scraping vs. lapping, but even then you can only compare root-mean-squared 'flatness' or surface finish. Even flatness and surface finish are not synonymous. And then consider that scraping has the inherent purpose of allowing oil voids.

    Conclusion: rubbish question, OP hasn't a clue. But I suspect this horse has been beaten already at this point.

  25. Likes Monarchist, Pete F liked this post
  26. #58
    Join Date
    Oct 2013
    Location
    Allentown, PA
    Posts
    511
    Post Thanks / Like
    Likes (Given)
    43
    Likes (Received)
    78

    Default

    I wish I knew how to "feed a model into a cnc." man that would make my life working CAM a lot easier.

  27. Likes RC Mech, Pete F liked this post
  28. #59
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    6,049
    Post Thanks / Like
    Likes (Given)
    2390
    Likes (Received)
    2859

    Default

    Quote Originally Posted by RC Mech View Post
    Conclusion: rubbish question, OP hasn't a clue. But I suspect this horse has been beaten already at this point.
    ROFL! Poor horse has been "CNC'ed", Blanchard ground, hand-scraped, lapped, vapour-de-greased, ion-implanted, and narrowly escaped being sexually dominated by an Australian 'top' with a seminally-fouled keyboard by this point in time!

    Nothing ^^^ left ^^^ to see here. Bar's now open. Pickled eggs are for the single guys. Don't need no more fart-kilt wives.

  29. #60
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    6,049
    Post Thanks / Like
    Likes (Given)
    2390
    Likes (Received)
    2859

    Default

    Quote Originally Posted by xplodee View Post
    I wish I knew how to "feed a model into a cnc." man that would make my life working CAM a lot easier.
    Not if you had to pay the Agency hourly rate for Victoria's Secret class models it would not!


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •