Jump to content

Welcome to eMastercam

Register now to participate in the forums, access the download area, buy Mastercam training materials, post processors and more. This message will be removed once you have signed in.

Use your display name or email address to sign in:

M_CODE1

Verified Members
  • Posts

    149
  • Joined

  • Last visited

Posts posted by M_CODE1

  1. You guys have always been helpful in the past so I thought I'd come here for some advice on a part we are having issues with. We have a part that we are having trouble putting the woodruff key seats in. The part is 4140 and heat treated to 42-47Rc. We decided to heat treat the stock first then have 2 lathe operations, then a mill operation. We did not think we would be able to hold the keyseat tolerance and position if we did them soft and had a grind operation after heat treat. We are using a Fullerton .875" X .1875" Keyseat Cutter (#607 ). We made a fixture that locates the part on the 1.7722 diameter on both end and is clamped cross the top. The Fullerton book recommends 60-140 SFM for hardened steel for speed of 257-600, then a feed of around .004 IPR, so 1.0 - 2.4 IMP. We have has no luck in that range. The cutters have broke at different times, some at the start of cut and others at the bottom. I tried to program it and roll the cutter into the part to have a thick to thin chip, but that did not seem to help either.

     

    Any advice on a different too to use? Or any other way to better cut this part? Does anyone have an experience cutting key seats in 4140 at 42-47Rc?

     

    A few of the things we have tried

    -Part was in V block vice jaws, made fixture to make it more rigid.

    -Made sure tool was as short as possible

    -Tried rolling cutter in

    -With and without coolant (part was cool after a couple cuts, but then cutter broke)

    -Tried to peel mill .005" step over with a 3/4" keyseat tool to rough out (S7500, F30), Broke cutter after 3 keyseats.

     

    Thanks

     

    Mike

    post-15905-0-72521300-1391623564_thumb.gif

  2. Sweet and my Hotel is only .2 miles away that night.

     

    While we’re in Chicago for IMTS 2012, we will be hosting our first-ever official “Mastercam Meet up.” This is a great opportunity for us to chat with our customers and for you to grab the ears of some key staffers. We would love to hear about what you do and what you think about the software. We have no agenda and no real plans, but if you want to come and just hang out with the Mastercam crew, we would love to spend some time with you off the trade show floor.

     

    We’ll be at the Sweetwater Tavern and Grill on Thursday, September 13th from 7-9 PM. We’ll all be wearing our Mastercam shirts and the name badges. Please join us for a drink or two and share some great conversation. Who knows…you might even score some Mastercam swag!

     

     

    http://events.r20.constantcontact.com/register/event?oeidk=a07e6ckdx4de8853cb4&llr=kftf58bab

  3. Below is a time study I did to give you comparison. The main things the software factors in is type of material, type of contour, material thickness, and what you Jet PSI is when coming up with the time. Flow uses a % that controls your speed and edge quality. They are most likely running in the 10-20% range, when you see a real nice smooth edge in thicker material like this.

     

    12" X 12" Square Cutting 3.5" Thick A36, This is 49" of travel with the Lead In/ Lead Out

     

    10%, 255 minutes, 255lbs Garnet .2 IPM

     

    20%, 132 minutes , 160lbs Garnet .37 IPM

     

    40% 72 minutes , 85lbs Garnet .68 IMP

     

    60% 53 minutes, 64lbs Garnet .92 IPM

     

    80% 43 minutes (most likely will not cut all the way thru material in spots), 51lbs Garnet 1.14 IPM

     

    Garnet is around $.25 a pound for reference.

     

     

    They do have a different kind of Garnet that is better for cutting thicker stuff this this. We just got some for trial but have never used it yet. At 80% you would probably get the parts to cut thru, but the cost is 50% more for it. We do mostly thinner stuff that does not justify the cost.

     

     

    Anything to do out there when I go there got training on the laser?

     

    Thanks

  4. My guess (it's a guess) for 3.5" thick material is only going to be 20 IPM.

     

    I know your only guessing, but with our 86K pump it would be be closer 1.5IPM for a decent cut that thick. Its not a exact IPM feed rate. In the corners it has to slow down before it gets there and it does not speed up right away. It slows down for the stream lag. At 3.5" thick it does not move to fast. I could also run it thru our softare to get a closer cycle time.

     

    On another note, our new Trumpf 1030 laser with auto pallet changer just show up last week and is getting set up. Waterjets are nice but this laser will be able to pump out the under 5/8 steel parts a heck of a lot faster.

     

  5. We recently purchased an EDM at an auction and need some pointers on it. It is a Mitsubishi model DWC90SB. Most of us here do not have any experience with EDM's and the book did not seem to helpful. The part we are cutting is .50" 4140 and has about 4.1" of cutting length. The first part we cut took about 3 hours. That would come out to about .022 IPM. We do not have a tight surface finish to hold, just a profile tolerance of .0015 in certain sections of the cut. Below is the program that we used to cut the profile. What exactly controls the feed during the program? The E101? What kind of IPM can you run of .50 steel? Does anyone have any charts that might help me? Thanks for any help that you might be able to give.

     

     

    %

    L1000

    (06/07/12)

    (NO STOCK ON TOOTH)

    G90

    G92 X2.135 Y.0625

    M20 (WIRE INSERT)

    M78 (RAPID FLUID FILL)

    M80 (DIELECTRIC FLUID ON)

    M82 (WIRE FEED ON)

    M84 (MACHINING ON POWER ON)

    E101

    M90 (STARTS ADAPTIVE CONTROL)

    G41 H1=.006 G1 X2.085 H1

    Y-.035

    X2.4576

    X3.024 Y-.36201

    G2 X3.0315 Y-.375 I-.0075 J-.01299

    X3.0165 Y-.39 I-.015 J0.

    G1 X2.5412

    G3 X2.54286 Y-.3748 I-.0687 J.0152

    X2.4725 Y-.30444 I-.07036 J0.

    X2.40214 Y-.3748 I0. J-.07036

    X2.40998 Y-.40708 I.07036 J0.

    G1 X2.41 Y-.40732

    Y-.525

    G2 X2.395 Y-.54 I-.015 J0.

    G1 X2.12378

    G2 X2.10884 Y-.52631 I0. J.015

    G1 X2.0979 Y-.40161

    G3 X2.1032 Y-.3748 I-.0651 J.02681

    X2.0328 Y-.3044 I-.0704 J0.

    X1.9624 Y-.3748 I0. J-.0704

    X1.96588 Y-.39668 I.0704 J0.

    X1.76799 Y-.44886 I.29412 J-1.51672

    X1.73445 Y-.4754 I.02001 J-.05974

    G2 X1.27469 Y-.73931 I-.47445 J.2941

    G3 X1.23909 Y-.75145 I.00161 J-.06299

    X.94342 Y-1.03881 I.91941 J-1.24175

    G40 G1 X.98281 Y-1.06974

     

    M91 (ADAPTIVE FEED OFF)

    M21 (WIRE CUT)

    M58 (DIELECRIC FLUID TO LOWER LEVEL)

    M02 (PROGRAM END)

    %

  6. Does anybody know what a seat of solidworks goes for these days?

     

    Dont quote me, but this should be fairly close.

    4k for SW Core + 1.2K for maintenance per year

    5.5k for SW Pro + 1.5K for maintenance per year

    Dont know price of next highest package, I think its called Premium, but you should't need it anyways.

    Should be able to work a 10-20% off base price,

    Where are you located? I'm sure someone on here can reccomend a reseller. I know the support I get from mine is great.

     

    We have one seat of Pro and one of the Core. About the only thing I use in the Pro that Core does not offer is the Toolbox.

     

    What version to most other use? Is there anything that Pro offeres that you could't live without?

  7. IMO 800 RPM is slow for a 10" Chuck. Should not have anything to worry about at that speed. Our new 18" chuck is rated for 2000 RPM.

    Our 10" on our 10" Kitagawa is rated up to 3200 RPM.

     

    Get the make and model and look it up. That would be your best bet.

  8. The machine is a Tsugami SS-32 swiss lathe. My part is in the sub spindle and I am going to put a 1/4 ball endmill in the live tool pockets. So I will be using the Y axis, but I will have my c axis turned on for proper orientation. Its just like a vertical mill programing in the Top view, except the X value's need to be in diameter.

     

    What is your email?

  9. Also in the pfxout and pxout post blocks instead of outputting xabs you want to out put xabs_dia

     

    Not sure where to change to the xabs_dia? I changed the two in bold and it did change the x's to double. But now the Z is outputing as a Y, and its not outputting a Z? Example below?

     

     

    pfxout #Force X axis output

    xabs_dia = xabs * 2

    if absinc$ = zero, *xabs, !xinc

    else, *xinc, !xabs

     

     

    pxout #X output

    xabs_dia = xabs * 2

    if absinc$ = zero, xabs, !xinc

    else, xinc, !xabs

     

     

     

    (I added this line)

     

    fmt X 2 xabs #X position output

    fmt X 2 xabs_dia #X position output

     

     

    (THIS IS ORIGINAL THAT I STARTED TO CHANGE THE X's by Hand)

    T252 M6

    G0 G90 G54 X0. Y-.1901 S3500 M3

    G43 H252 Z.25

    Z.0876

    G1 Z-.0124 F5.

    X.0512 Y-.1892 Z-.0136 F40.

    X.1016 Y-.1866 Z-.0172

    X.1506 Y-.1822 Z-.0231

    X.1976 Y-.1762 Z-.0312

    X.2422 Y-.1689 Z-.0411

    X.2842 Y-.1602 Z-.0529

    X.3236 Y-.1504 Z-.0661

    X.3604 Y-.1395 Z-.0808

    X.4264 Y-.115 Z-.1139

    X.482 Y-.0874 Z-.1513

    X.5106 Y-.0695 Z-.1755

    X.5358 Y-.0508 Z-.2009

    X.576 Y-.0112 Z-.2548

    X.6026 Y.0307 Z-.3118

    X.6152 Y.0739 Z-.3707(I STOPPED CHANGING X's BY HAND)

    X.3071 Y.1176 Z-.4304

    X.3033 Y.1448 Z-.4675

    X.2969 Y.1715 Z-.5041

    X.2879 Y.1976 Z-.5398

    X.2765 Y.2228 Z-.5744

    X.2627 Y.2469 Z-.6075

     

     

    (THIS IS THE NEW OUTPUT)

    N120 T252 M6

    N130 G0 G90 G54 X0. Y.25 S3500 M3

    N140 G43 H252 Z0.

    N150 Y.1

    N160 G1 Y-.0124 F5.

    N170 X.0512 Y-.0136 F40.

    N180 X.1015 Y-.0172

    N190 X.1505 Y-.0231

    N200 X.1975 Y-.0312

    N210 X.2421 Y-.0411

    N220 X.2842 Y-.0529

    N230 X.3237 Y-.0661

    N240 X.3604 Y-.0808

    N250 X.4262 Y-.1139

    N260 X.4819 Y-.1513

    N270 X.5106 Y-.1755

    N280 X.5359 Y-.2009

    N290 X.5761 Y-.2548

    N300 X.6026 Y-.3118

    N310 X.6152 Y-.3707

    N320 X.6142 Y-.4304

    N330 X.6066 Y-.4675

    N340 X.5937 Y-.5041

    N350 X.5758 Y-.5398

    N360 X.553 Y-.5744

    N370 X.5254 Y-.6075

  10. Jimmy

     

    I modified the post in 3 spots. But now I get an error when I try to post the program. Any ideas what might be wrong?

     

     

     

    fmt X 2 xabs_dia #X position output (changed "xabs" to "xabs_dia")

     

     

    pfxout #Force X axis output

    xabs_dia = xabs * 2 (ADDED THIS LINE)

    if absinc$ = zero, *xabs, !xinc

    else, *xinc, !xabs

     

     

    pxout #X output

    xabs_dia = xabs * 2 (ADDED THIS LINE)

    if absinc$ = zero, xabs, !xinc

    else, xinc, !xabs

  11. I'll looking to do some surface milling on our swiss lathes to help debur a part. For that machine my X number needs to be in diameter. So I basically need to double all my X's. Anyone know of an easy way to do this. I don't want to have to change them all by hand.

     

    The post I am going to modify is the Generic Fanuc 3x.

     

     

    Thanks

  12.  

     

    Looking to make a meat screw out of a solid piece of cast iron stock (65-45-12). This part is usually a casting, but new tooling will not be ready in time to meet customers demands. This will be machined on a Haas VF4, with a rotary table. The end with the hex will be in the 3 jaw chuck then a tailstock on the other end.

    I am starting to program it with a Rotary 4 axis toolpath for the rough and finish. Is this best toolpath to use, or is there something better? I plan on using a 5/8 ball SGS Z-Carb Endmill to rough and finish. Any other tool that you might think would work great? Any other suggestions would be appreciated.

     

    For Roughing

     

    .015, stock to leave

    .01 Filter Tolerance

    .005 Cut Tolerance

    Rotary Cut Type

    Climb Milling,

    1.0, Axis dampening length

    15, lead/lag angle

    .2, max step

    .1 Tool Vector length (what does this actually do)

     

     

    For Finishing

    0, stock to leave

    .006 Filter Tolerance

    .003 Cut Tolerance

    Rotary Cut Type

    Climb Milling,

    1.0, Axis dampening length

    15, lead/lag angle

    .025, max step

    .1 Tool Vector length (what does this actually do)

  13. FYI

    I asked about the #3001=1 line, and below is what he told me.

     

     

    Since I don’t know how you were going to write the machining code, I wanted to put in a safety so the program couldn’t run all three marks in a row.

     

    #3001=1 () generates an alarm and stops the machine, whatever text you put in between the parentheses will be the message displayed on the control panel.

     

    A good use for this is tool life management as shown here - http://www.todaysmachiningworld.com/shop-doc-custom-macro-programming/

     

     

    Or let’s say you start work at 7:00 AM and like to have fresh coffee waiting when you get in, you could put a macro at the end of the program that checks the time and generates an alarm like this:

    IF[#3012 LT 064500] GOTO 9999

    IF[#3012 GT 065500] GOTO 9999

    IF [#3012 GT 064500] GOTO5000;

     

    N5000 #3000=1 (I’M ON MY WAY IN MAKE A POT OF COFFEE THX – MIKE)

     

     

    N9999;

    M30;

     

    If nothing else it will baffle the xxxx out of the operator and if he’s running the machine between 6:45 and 6:55 AM you just might get coffee.

  14. I cant take full cridit for the macro below but I was in the process of developing something similar when this was sent to me from our Rem Sales guy. Anybody know Dan Murphy, he know his stuff. As usually the idea was driven by bad parts being sent to the customer.

     

     

    Assuming

     

    N1000 = 1st shift mark program (7AM to 5 PM)

    N2000 = 2nd shift mark (5 PM to 9 PM)

    N3000 = 3rd shift mark (9 PM to 7 AM)

     

    N500 (TIME STAMP);

     

    IF[#3012 LT 070001] GOTO 3000;

    IF[#3012 LT 170001] GOTO 1000;

    IF[#3012 LT 210001] GOTO 2000;

    IF[#3012 LE 240000] GOTO 3000;

    #3000=1 (TIME STAMP MACRO ERROR);

    ;

    N1000 (1ST SHFT MARK)

     

    (MACHINING HERE)

     

    GOTO4000;

     

    N2000 (2ND SHFT MARK)

    (MACHINING HERE)

     

    GOTO4000

    N3000 (3RD SHFT MARK)

     

    (MACHINING HERE)

     

     

    N4000;

     

    (REST OF PROGRAM GOES HERE)

  15. I want to be able to mark a part with an end mill plunging in .005" during 3 time period in a day. Each section of time, the mark will be in a different spot. This will tell us what shift made the part. The picture show what I am talking about. The mark will be on the left from 7am-5pm, then in the middle for the 5pm-9pm time, then on the right for the 9pm-7am time. I am not sure how to go about writing a macro for each time frame. This is on a Tsugami SS-32 with a Fanuc 18i control. Anyone have any ideas?

     

    I first thought about using a "IF[#510EQ1]GOTO490" type statement for each mark and have the operator change the macro to (1, 2 or 3) for there shift, but I think they might forget at times, or use the wrong number on purpose

     

     

    N40(MILL SLOTS MILL SLOT)

    T3333(1/8 Z CARB_BACK )

    G98M105(SPINDLE STOP)

    M123S5000

    M50(C-AXIS ON)

    M8

    M276

    G0X-.590Y-.05Z.05

    C45.

    G17

    G1Z0F10.0(F8.0)

    Y.05Z-.04

    Y-.05Z-.08

    Y.05

    G41X-.605F14.0(F12.0)

    Y-.05

    G3X-.590Y-.0575R.0075

    G3X-.575Y-.05R.0075

    G1Y.05

    G3X-.590Y.0575R.0075

    G3X-.605Y.05R.0075

    G1Y-.05

    G40G1X-.590

     

    (JUMP TO PROPER TIME FRAME)

     

    (GO HERE BETWEEN 7am - 5pm)

    Y.05

    Z-.085

    Z-.08

    GOTO6000

     

    (GO HERE BETWEEN 5pm - 9pm)

    Y0

    Z-.085

    Z-.08

    GOTO6000

     

    (GO HERE BETWEEN 9pm - 7am)

    Y0

    Z-.085

    Z-.08

     

     

    N6000(JUMP BACK HERE TO FINISH SLOTS)

    G0Z.05

     

     

  16. Trying to machine the profile in the attached picture. Our machine is a 3 axis VMC, with a rotary table. When I use the contour tool path, using the axis substitution, the corner of the tool cuts into the part.

     

    Is there a way to cut this with my current machine in Mastercam?

     

    I'm thinking that I need to move in the Y axis while following the cam section, but dont know how to do that.

     

    Any Advice?

     

     

     

     

  17. quote:

    Also, I think running in single block will scrap the part. Don't know about overrides. Again, be careful.

    All controls are diferent, but our Tsugmai Swiss Lathes will stop after the [G0 X.7368 (retract)] when in single block.

    I guess to check just raise you X offset up about a half inch or so and try it.

  18. We do somthing like this except the threads are straight first, then it tapers

    The first operation below works on an Okuma lathe. On our Fanuc controlled swiss lathes our PartMaker software will post it out long hand for each pass simalar to what CNCCHUCK has.

     

     

    This is cutting a M16 thread with a taper at the end,

    NAT6

    ( TOOL - 6 OFFSET - 6 )

    ( LFINISH GD THREADER )

    G97 S850 M08

    G00 X0.7915 Z0.15 T060606

    G88 NAT40 H0.1037 D0.005 U0.001 B60 M23 M22 M73 M32

    NAT40 G81

    G00 X.5855

    G34 X.5855 Z-.4921 F.0787

    X.5915 Z-.4921-.3346

    X.7915

    G80

    M09

    G00 X37 Z7 T0600

    M01

     

    This is used on the Fanuc control.

    Its cutting a M12 thread with taper in the back

     

    N4(FRONT THREAD1ST)

    G97S450M13

    T1313(QUADTHREADER )

    G99G0Z-0.1

    X0.5783

    X0.4651

    G32Z0.1379F0.0689

    X.4725Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4535

    G32Z0.1379

    X.4609Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4435

    G32Z0.138

    X.451Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4353

    G32Z0.138

    X.4427Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4287

    G32Z0.1381

    X.4361Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4229

    G32Z0.1381

    X.4304Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4173

    G32Z0.1381

    X.4247Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4118

    G32Z0.1381

    X.4192Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4063

    G32Z0.1382

    X.4137Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.4009

    G32Z0.1382

    X.4084Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.3957

    G32Z0.1382

    X.4031Z0.3543

    X0.5868

    G0X0.5783Z-0.1

    X0.3957

    G32Z0.1382

    X.4031Z0.3543

    X0.5868

    G0X2.28T0

    M1

     

     

    Hope this helps

    • Like 1
  19. quote:

    Just because there is a cracked version of Mcam doesn't mean that it's OK to use it.

     

    The same goes for keeping the Beta guidelines in check.


    I like how you compare something that is illegal to something that is easily searched for on the internet.

  20. LOL. I was told not to share the link/PDF, I downloaded. And I did't, because I respect the guy that asked me not to.

     

    But I went to google and was able to search for the very same link, and it showed up on the first page.

     

    The entire world is allowed to view it, but we can't share it with fellow Mastercam users?

Join us!

eMastercam - your online source for all things Mastercam.

Together, we are the strongest Mastercam community on the web with over 56,000 members, and our online store offers a wide selection of training materials for all applications and skill levels.

Follow us

×
×
  • Create New...