Accuracy question, scraping vs CNC - Page 5
Largest Manufacturing Technology
Community On The Web
Close
Login to Your Account

Page 5 of 5 FirstFirst ... 345
Results 81 to 92 of 92
  1. #81
    Join Date
    Apr 2006
    Location
    Manchester, England
    Posts
    5,980
    Post Thanks / Like
    Likes (Given)
    690
    Likes (Received)
    2935

    Default

    Maybe I'm more stupid than I thought but it was the diffuse nature of the question that led me pass up on an attempt to answer it. I don't normally behave as a " great I am " because I'm not that old that I can't remember the stupid things I did when I was learning the trade.

    Regards Tyrone.

  2. Likes Pete F liked this post
  3. #82
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    6,079
    Post Thanks / Like
    Likes (Given)
    2410
    Likes (Received)
    2872

    Default

    Quote Originally Posted by Tyrone Shoelaces View Post
    Maybe I'm more stupid than I thought but it was the diffuse nature of the question that led me pass up on an attempt to answer it. I don't normally behave as a " great I am " because I'm not that old that I can't remember the stupid things I did when I was learning the trade.

    Regards Tyrone.
    We - usually - get issued four fingers, one thumb, each hand.

    Just ONE of my OEM ration of digits has no to-the-bone scar to remind me just how repeatably and consistently stupid I was between 17 and 21.

    Probably because it is the one with the several-times busted extensor tendon that never stayed healed for long.

    "Reminder" bookmarks didn't stop with the hands, either... some of us were fast at book-learning, just fast enough in the flesh and bone to be marked-up here and there, expensively divorced, rather than dead young and long buried.

    Can't complain. At least got a far better wife out of it all..


  4. #83
    Join Date
    Jan 2009
    Location
    Israel
    Posts
    124
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    6

    Default

    And I was told that we (the Israelis) are hot-tempered...

    I work with mechanical designers on a daily basis. Regardless of how talented they are, most of them have a very limited understanding of the machining process that will be used to create the part they design and set tolerances for. I don’t mean general capabilities, limitations and etc. but the details that you know only if you have experienced operating the machines (like in other professional areas). The few who does have practical knowledge are mostly old-timers who learn practical machining in a professional high school. Although it doesn’t make a lot of sense, I can clearly understand where the OP question came from.

    I would like to spot another aspect from my point of view:
    As I wrote earlier I work with many mechanical designers (engineers) and I have few friends who are mechanical engineers- some are over 60 yo. I must say that I never came across any mechanical engineer in Israel who is familiar with the process of scraping. I learn, read and watch a lot on scraping. As more as I learn about it, I wonder to myself how the hell no one here knows scraping (or all of them died before I was born?). Trying to answer this I'm mapping to myself what machining processes can be used for common jobs.

    The more I learn, the more I'm getting the understanding that scraping, as a production process, is pretty limited to heavy machine tools manufacturing (for instance lighter machines often use premade linear rails/ways). Other than this, it is well suited for the small shop machinist who is limited with the variety of high-end big and highly accurate machine tools, and still, can achieve amazing results with hand/power scraping. Of course also very useful where the part cannot be taken to reconditioning, so scraping brings the reconditioning to the part.

    As no one is Israel (at least to the best of my knowledge) design/produce heavy metal-cutting machines (excluding some rare cases), most of the machining here is aimed at parts manufacturing , for which other processes are utilized (perhaps more automatic). These can yield high accuracy or fine surface finish when required. Mostly grinding, lapping and diamond turning. The abilities here are limited to the limitations of the modern machines available. I presume that this is the reason why scraping knowledge virtually doesn’t exist here.

    One final anecdote:
    A few years ago a veteran mechanical engineer who is a friend of mine told me that a former university lecturer of him patented a laser ablation patterning method for bearing surfaces, to increase oil retention. I haven’t paid too much attention to his story but some time later I picked up an unknown top-slide from a scrap yard. It was a well-made top-slide, and I noticed its ways are ground and patterned by something which definitely looks like laser ablation. Back then I thought to myself:
    Grinding + laser patterning = modern(hand scraping + flaking).

    Just sharing my thoughts.

  5. #84
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    5,764
    Post Thanks / Like
    Likes (Given)
    1359
    Likes (Received)
    2144

    Default

    Another factor not mentioned yet is distortion from the force needed to hold the parts in some processes. A shop I frequent makes 2 ft long bars for some sort of alignment fixture. They are T slotted and otherwise machined, then ground. They grind one side close but not all the way to size, then the other side. Before mounting the bar for the second grind, they measure the inevitable bow and put shims under the bar so the magnetic chuck will not pull it flat. The part goes through this process twice, first on a grinder with a face stone driven by a 60 hp motor for fast stock removal, then on a conventional surface grinder. Each time they reduce the amount of stock removal and resultant bow with each pass until they have an almost perfectly straight piece. I believe they hold a couple of tenths over the length, probably about as good as the process will produce. To go much farther probably requires something like lapping, not for its intrinsic accuracy but because it doesn't need a strong workholding method.

    They do cylindrical grinding on a Cincinnati that probably weighs 7 or 8 thousand pounds. Nevertheless, when the stone comes off the end of a shaft, losing contact area, it springs in slightly, putting a taper on the last inch. The operator has learned how much to expect and applies hand pressure to compensate.

    These are things you may read in a book, but you will never be skilled at it until you actually do it. Tune out all the crap and learn by doing.

    Bill

  6. #85
    Join Date
    Jan 2009
    Location
    Israel
    Posts
    124
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    6

    Default

    Bill,

    Thanks for mentioning the distortion related to the holding (and to the process itself).

    Let me advise the OP watch Dan Gelbart's video titled "High accuracy":
    Building Prototypes Dan Gelbart part 17 of 18 High Accuracy - YouTube

    IMHO it is a very shorthand explanation, yet very useful.

  7. #86
    Join Date
    Apr 2014
    Location
    USA FL
    Posts
    100
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    15

    Default

    Given how this thread has progresses and how it diverged into what seems to be 2 separate topics, I feel it is important for me to summarize what has been said plus reflect on it based on my own experience (however limited it may be).

    Communication:
    When I started working in aerospace industry I was interfacing with many senior technicians most of which had 30+ years of experience under their belt. The problem was that these technicians were very used to doing things in one specific (correct) way, AND they were very used to training new technicians by explaining to the new guys "how" to do things. Here comes a young engineer with 2 college degrees who is "in theory" supposed to tell these old guys what to do. Naturally when this young engineer is asking the old techs "why" they do things in a particular manner, the techs feel like they are being quizzed or questioned and so they get defensive. Meanwhile the young engineer is simply trying to understand the reasoning for doing things in a particular way. Too many of these young engineers end-up becoming afraid to interface with the technicians, so they move on to managerial roles, without ever having learned what actually happens on the shop floor.


    Scraping:
    What I gathered from this thread is that scraping is a useful skill to have, which is capable of producing decent accuracy surface, however from the stand-point of sheer accuracy scraping is not a unique finishing method, hence it is often ignored in production environment having been replaced by mechanical (automatic) polishing machines. Hand scraping is a cheaper process but it does not offer uniquely superior accuracy to other finishing methods.
    One place where scraping IS unique would be in the type of finish it produces on the surfaces requiring oil retention characteristics. Polishing (and I have studied abrasive patters under a microscope) produces scratches the width of which is on the same scale as the depth of them. As the result oils have difficulty of penetrating these scratches to be retained on the part. In case of scraping the depth of the cut is many times less than the width of the cut, hence making it easier for the oil to "pool" inside of these shallow cuts, producing a retention layer for the oil film on the surface of the part.

  8. #87
    Join Date
    Apr 2014
    Location
    USA FL
    Posts
    100
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    15

    Default

    Quote Originally Posted by avivz View Post
    Bill,

    Thanks for mentioning the distortion related to the holding (and to the process itself).

    Let me advise the OP watch Dan Gelbart's video titled "High accuracy":
    Building Prototypes Dan Gelbart part 17 of 18 High Accuracy - YouTube

    IMHO it is a very shorthand explanation, yet very useful.
    Already saw the video. In regards to Dan I have to say: "It is a pleasure to hear an expert in the field to present his knowledge in a precise, technically accurate and academically rich way as Dan does it".

  9. #88
    Join Date
    Aug 2005
    Location
    Toronto
    Posts
    4,372
    Post Thanks / Like
    Likes (Given)
    2281
    Likes (Received)
    2455

    Default

    Quote Originally Posted by pcm81 View Post

    Scraping:
    What I gathered from this thread is that scraping is a useful skill to have, which is capable of producing decent accuracy surface, however from the stand-point of sheer accuracy scraping is not a unique finishing method, hence it is often ignored in production environment having been replaced by mechanical (automatic) polishing machines. Hand scraping is a cheaper process but it does not offer uniquely superior accuracy to other finishing methods.
    yeah, maybe. What you say is true at one level, but also "it depend" is real world accurate. It will depend on the nature of the part being made and the available resources. Standard Modern for example up until a couple of years ago while still in production hand scraping the parts to the bed. The average customer wouldn't see it because the bed was ground, but you need some very fancy grinding equipment do get all mating surface with their different geometries in perfect contact while aligning and axis. For all I know that is still standard practice at high end makers.

    Such grinders likely exist, but they weren't accessible to that lathe manufacture of decent repute. I've also take apart china made slideways only to find the bearing surfaces are mill, left that way . Then there are the cheap lathes that you can run back and forth over the work and they keep cutting....because the pancake stack is so poorly fit it flexes and wobbles like crazy. Point being, in the low end market scraping has not been replaced by anything and quality suffers.

    Then there are all the examples of part and assemblies made in shops that just don't have a massive slideway grinder. Or reconditioning work.

    Another area are parts that are thin or spindly and really difficult to old without distorting them. rough scrape to a 4-5 points per inch and they're as flat after the engaging of the mag chuck as before.

    Finally, scraping affords the opportunity to create original accuracy. Not that anyone wants to go down a masochist Maudslay experiment of truth, but you can with scraping create original flatness and squareness. That's not something possible by a lot of other techniques (see Moores Foundations of Mechanical accuracy). These technique let someone with limited resources achieve incredible levels of accuracy, so is at least worthy of mention.

    I've done a fair bit of scraping, and just never has seen it as competing with anything. There are times its indespensible and times its just best way to do it. The more techniques you have available the better, and we're supposed to draw upon the most effective one for the task at hand.

  10. Likes Monarchist liked this post
  11. #89
    Join Date
    Jan 2002
    Location
    West Coast, USA
    Posts
    6,910
    Post Thanks / Like
    Likes (Given)
    284
    Likes (Received)
    3587

    Default

    There are at least two kinds of "accuracy" involved here.

    One related to having surfaces exactly flat. Surface finish is an approximation of this and will vary with the tools and cutting parameters regardless of if it is machined, ground, lapped, or scraped. This surface finish may be variously represented by Ra, RMS, points per square inch etc. -- you could look all those up. As noted earlier, scraped surfaces can be very flat yet still contain those tiny divots from scraping. Good for holding oil and supporting machine slides. Not so good if, say, you want a mirror surface as on actual mirrors or gage blocks.

    The second kind of accuracy involves get plane surfaces at parallel, at right angles, at specified angles etc. Machine tools have those capabilities built in, to some level of tolerance. Scraping requires some external measure or fixture to establish the relationship of planes. It has no intrinsic accuracy; depending entirely upon what combination of surface plates, indicators, right angle plates, etc. are used to spot the surfaces. Someone with lots of time on their hands and top quality reference tools (those plates, angles, indicators) can probably hit something like a .0001" in 6" geometry spec more surely than if they're dealing with a clapped out machine -- or maybe just a new cheap import -- that struggles to do .001" in 6". That accuracy, however, resides in the reference surfaces and the skill of the scraping hand vs. the machine slide accuracy.

    What can be said of scraping is $1000 of good reference surfaces can buy the same sort of geometric accuracy you'd expect in the slides of a $100,000 machine tool (whether cutting, grinding, lapping). However, it also comes with about 1% or less of the productivity and work rate of the proper machine tool.

  12. Likes Monarchist, JST liked this post
  13. #90
    Join Date
    Apr 2014
    Location
    USA FL
    Posts
    100
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    15

    Default

    Who else is salivating after watching this?
    High precision air bearing CNC lathe and grinder - YouTube

  14. #91
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    15,728
    Post Thanks / Like
    Likes (Given)
    1120
    Likes (Received)
    2115

    Default

    Quote Originally Posted by PeteM View Post
    ......

    What can be said of scraping is $1000 of good reference surfaces can buy the same sort of geometric accuracy you'd expect in the slides of a $100,000 machine tool (whether cutting, grinding, lapping). However, it also comes with about 1% or less of the productivity and work rate of the proper machine tool.
    And, it is pretty much of a total loss as a production method in any volume on account of that. But for a one-off, it shines.

    It's like ANY manufacturing... You can spend very little up front (comparatively) and use time and manpower to get the results, as with scraping. OR you can spend quite a bit on machinery and then get parts popping off the line on a regular basis all alike. As with way grinding etc.

    For a one-off, nobody will spend the big money in most cases. For volume production, you have to.

    Imagine what a coffeepot would cost, if a blacksmith had to hammer it to shape and size. That's pretty much the difference between scraping and production grinding.

    Choose your cheapness. Cheap to make per unit, or cheap to tool up to make.

  15. Likes Monarchist liked this post
  16. #92
    Join Date
    Dec 2008
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    6,851
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1838

    Default

    i cnc mill parts to .0005" per 40" tolerance every day. company has machines been doing it for 23 years. they are buying a new $2,000,000 cnc mill to replace a older 23 year old one that is having trouble holding tolerances.
    .
    my company still has some stuff scraped but more for oil retention pattern. if a part is .0005" off perpendicularity on a ball screw face it comes back marked up for recutting.
    .
    part distortion from clamping, warpage from material removed, warpage from temperature changes, warpage from a parts weight and how held, cutting tool forces. many things effect accuracy and tolerances
    .
    picture shows part warpage just from material removed and .0005" passes are used to take magic marker lines off. some parts are flat at certain temperatures and when temperature changes they are no longer flat. obviously part design effects tolerances. part shown when unclamped after initial machining it usually springs to a curved shape and needs recuts. other picture is column that weighs over a ton with turcite reapplied getting ready to remachine. perpendicularity to other surfaces and edges is important there. jack is used to put shims in. i often use .0010, .0012, .0015" shims to get .0002, .0003, .0005 changes for aligning to the .0003 or less tolerance
    .
    for example 10 ton machine frame whose rails are milled flat to .0005" per 40" and when over 1/2 ton of chips and coolant are removed the middle of part bows up .0005" obviously order of operations is to remove chips and coolant first then take final cuts.
    .
    i also get parts hand scraped and machine running and worn from heavy usage of over a decade i machine worn areas often .002 to .004" deep until wear marks clean up. turcite on matching parts is removed reapplied and remachined. i believe they take old machines and for 1/2 the price of new machine take it apart and remachine and reassemble to it is back to new factory spec. i see dozens of machines rebuilt like that every year
    .
    just saying you can hand scrape a machine to less than .0002" per 40" accuracy but it will wear and not stay that accurate. how long it takes depends on load. i often see heavy parts wear a hand scrape surface in a small area (maybe 6" wide out of over 40") as that where the part balance or weight load is heaviest. next time i will take a picture of where you see spots where hand scraping is totally worn off but majority of surface you can still see the hand scraping
    Attached Thumbnails Attached Thumbnails warpageremoving.jpg   400gh_column_79000021_turcite_3rdop_b.jpg  

  17. Likes Monarchist, ferretlegger liked this post

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
  •