Wire-EDM for part of my production process... Feedback, machine suggestions please!
Close
Login to Your Account
Page 1 of 2 12 LastLast
Results 1 to 20 of 35
  1. #1
    Join Date
    Oct 2014
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,606
    Post Thanks / Like
    Likes (Given)
    1221
    Likes (Received)
    1344

    Default Wire-EDM for part of my production process... Feedback, machine suggestions please!

    Hey All!
    I have been considering wire EDM for part of my production process for a few years, and would like to finally put the idea to bed, one way or another, so I can get on with it or let it go!

    Some background: I make a relatively low volume of high end hunting and kitchen knives. Right now the knives are made from A2 tool steel, but I will likely be moving to a stainless steel early next year. In either case the parts will be hardened to around 63HRC and all machining operations will be done on them in the fully hardened state.

    Currently I machine the knives using a specialized set of fixtures:


    Which work fairly well, but having to make fixtures like this for each knife shape that I want to make really robs me of my flexibility. It is also difficult to control chatter during the machining of the final blade bevel (the angled part behind the edge) because it means that I am clamping against a very thin blade, only 0.016" thick for my hunting knives and likely as little as 0.008" for my upcoming kitchen knives. Tolerance stack-up is also an issue across so many operations, and the fixtures do not allow me to reach all the surfaces that I would ideally like to be able to reach.

    My goals for my process:
    - Quality
    - Consistency
    - Flexibility
    - Stable unattended operation

    I make relatively few knives so speed is not of utmost importance to me. If the EDM operation took 2 hours or less then that would be completely acceptable. Faster is better to some extent because it would make the process more flexible, but speed is not a primary goal. For reference I only make about 30 knives per month.


    The concept:
    Basically I would like to buy steel stock that is oversized in all directions, harden it, then hard-mill the knife blade into the stock, but not actually cut it out. This would give me the flexibility to fully machine both sides of the blade without needing to make specialized fixtures for each model, and would give me more flexibility on the kind of steel stock I'm buying. Right now because I am not able to machine every surface (because fixtures are in the way) I have to buy precision ground stock with tight tolerances and tight surface finishes which drives up price. If I can machine all surfaces myself then I can buy hot-rolled descaled or cold-rolled stock and simply produce all final surfaces myself.

    The output from the hard-milling operations would look something like this:





    The two 3/8" holes at either end of the stock would serve as locating holes for the 2nd machining op and for the wire EDM op... I would then like to skim cut the inside of the pin holes in the handle to get final size and location, then cut out the profile of the knife with a single pass on the EDM. In the following screenshot the red lines indicate where I'd be cutting:



    This would allow me to hold much tighter tolerances between the pin hole locations and the outline of the handle area. This is important because the fit between the metal parts of the handle and the G10 composite handles scales must be fairly tight because that area ends up directly touching the users fingertips, and any mismatch in dimensions or location above about 0.001" is fairly easily felt. Right now tolerance stack-up during the multiple machining operations makes this kind of a tricky target to hit and I would like to make it easier to get a better final result.

    The edges where I'd be cutting with the EDM will generally be bead-blasted or sand-blasted so the matte finish from the EDM is not an issue at all.

    The total length of the cut in the example above is 20.53" with most of the stock being 0.156" thick, though there are some sections that are thinner, namely along the cutting edge where the material is only 0.016" thick.

    The advantages of wire as I see them for this process:
    - More flexibility in design, no need to make specialized fixtures for every design change
    - More flexibility with raw stock dimensions and tolerances
    - Potential for more easily achieving tighter tolerances
    - No burrs, which would reduce hand finishing and re-work from missed burrs

    Disadvantages as I see them:
    - Initial cost of machine if bought new
    - Difficulty with parts availability and cost if EDM bought used
    - Space taken up by another machine
    - Cost of consumables

    I'm curious what you guys think. Do you have any estimates for the time that this cut would take, and what the consumables cost would be? Is it utterly stupid to consider EDM for this process, or is it a process change that would pay off?

    If wire EDM is not an option then honestly my other options for improvement are pretty limited. Making a fixture that would give the same level of access to all the surfaces that need machining is pretty much impossible. I could cut out the parts and leave tabs but then I'd be looking at follow-up machining or hand finishing operations which would make it harder again to hit the tighter tolerances that I want.

    Any and all feedback appreciated!

    Marcus: sorry for revisiting this, I know we spoke about this a while ago but I wanted to take the time to fully flesh out the idea and then finally put it to bed one way or the other!

    -Aaron

  2. #2
    Join Date
    Oct 2014
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,606
    Post Thanks / Like
    Likes (Given)
    1221
    Likes (Received)
    1344

    Default

    Also: if you think the idea for the process is viable please let me know if you have any machine recommendations! I would prefer to buy used to keep the investment reasonable, luckily I am very familiar with working on CNC machines and electronics in general.

    My largest workpiece would be 16x3x0.25" or so, and I have no need for anything other than 90º cutting as far as I have seen yet! The smaller the machine the better honestly as my shop is very small!

    -A

  3. #3
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    6,018
    Post Thanks / Like
    Likes (Given)
    2376
    Likes (Received)
    2988

    Default

    You will need a machine with solid threading capabilities if you want to skin holes and cut profile. Also submersible is the way to go. We do quite well with our sodick machines on unattended machine ing. Wire cut, thread, re-threading, etc

  4. Likes aarongough, greggv liked this post
  5. #4
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    6,018
    Post Thanks / Like
    Likes (Given)
    2376
    Likes (Received)
    2988

    Default

    Also, I would look for a machine that can use .010 or .012 wire add it seems your cuts don't need small radii. Can put more power/speed to thicker wire sizes

  6. Likes aarongough liked this post
  7. #5
    Join Date
    Oct 2014
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,606
    Post Thanks / Like
    Likes (Given)
    1221
    Likes (Received)
    1344

    Default

    Quote Originally Posted by Mike1974 View Post
    You will need a machine with solid threading capabilities if you want to skin holes and cut profile. Also submersible is the way to go. We do quite well with our sodick machines on unattended machine ing. Wire cut, thread, re-threading, etc
    I would be ok with not skimming the pin holes if it would make the process more finicky or needed machine much more expensive. Tolerance stack-up between two operations should still be a lot better than what I'm getting now with 3 operations involved in making the profile...

  8. #6
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    6,018
    Post Thanks / Like
    Likes (Given)
    2376
    Likes (Received)
    2988

    Default

    Another thing to consider, can possibly stack 2-3 plates together in wedm and/or nest the blades to minimize material loss. You don't need to spread them out for endmill clearance, can go maybe .05" apart or so between blades

  9. Likes aarongough liked this post
  10. #7
    Join Date
    May 2011
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    2,715
    Post Thanks / Like
    Likes (Given)
    269
    Likes (Received)
    1475

    Default

    12 wire...one piece, 23" through .156 hardened steel using adaptive...(granted I'm thinking mits) probably average around .100 inches per minute maybe .200 if you tune your epack. Another thing to consider would be using coated wire. I may be a little low on my estimates.

    Things to ponder before getting a wire edm...water resistivity is very important, if the thing sits for the while it's going to take a while for the resin bottle to get the numbers down.

    Tank seals don't like being dry, metal partials can build up on the moving surface and when dried out can shread the seals. This allows water to leak into places your really don't want it (think ways and ball screws).

    Most wire machines take up a lot of space. For for a 13x15 machine you will need about 12 square foot for it maybe more.

    Get spare resin bottles and filters, they always go out when you least expect it typically on weekends.

    Program in a tab or 2 to hold the knife from falling through. Especially if you use a start hole and have retrace with machining on...it will recut everything in it's path and render your part useless. This is because the wire will break due to the piece falling in before it can finish it's cycle. It then goes back and rethreads and will commence cutting back to the start point...unless the part catches the lower head and either moves the entire setup and/or causes the machine to crash. Then you need to restart everything and do a wire alignment blah blah blah. Also the flushing can cause the part to vibrate when there's only a small amount to cut. If critical I toss some neodymium magnets in to hold the part where it should be...if it's magnetic of course.

    That being said, I love using wire edm for all kinds of stuff. The speed sucks but the accuracy that can be obtained after some experience is amazing. I've only used mitsubishi wire edm's so I can't speak for the others but I have heard good things about them. Key thing with any is maintenance.

  11. Likes aarongough liked this post
  12. #8
    Join Date
    Jul 2004
    Location
    Wiggins CO. USA
    Posts
    1,445
    Post Thanks / Like
    Likes (Given)
    11
    Likes (Received)
    74

    Default

    Thinking of this in terms of efficiency and cost (since youre using for your own products, as do I)... and being upfront with you, the newest machine I have any experience with is a 1996 Mits FX10 and Im no pro with it. Youll probably need to figure out your time/cost associated with doing it how you do now - vs the mill time for your new process plan, and then add in the wire edm, wire portion.

    I have no familiarity with thin work pieces, so cant speak to that variable - but am happy to share what I have seen with some of my products that I make using a WEDM that appear on the surface to share some potential similarities.

    On my FX10, a part I make is cut from 1.55" thick hardened A2. It has a .625 bore lengthwise, and the wire cuts in from one of the long sides, into the center where the through bore is, and along it, and through to the opposite side (think a letter J laying on its side, sort of). It calculates out to roughly 2.9" worth of cut length. It takes 40 minutes on the FX10 using .010 wire and the powermaster settings (.072" per min avg). I have a 92 Mits 110SZ that I was able to get it sped up - but the fastest I ever got it to cut was 28 minutes (.103" per min avg). I have not had the time to spend chasing power settings to see if I can get the FX speed up to match the old SZ - but will some day. Wire consumption is roughly 45 parts from a 13.2lb spool - or 0.3lbs per 2.9 long cut.

    So, if you stacked 10 of your plates together yould have 1.56" thick which is very close to the same, compare your thinner areas with the bore in the middle I cut through on my parts and I see a lot of similarity.

    If you got the fastest cut time I have ever had on either machine of mine (.103" per min - I say that figuring you would get a faster machine than I have, but the gaps in the stacked plates would likely slow you down a little), that would roughly equate to 200 minutes to cut a 20.53 path and use 2.13lbs (rough calcs) of .010 wire and leave you with 10 parts cut out. I pay about 78USD per 13.2lb spool, plus shipping and tax - so you would see roughly 12.50USD in wire cost plus the 200 minutes cutting, plus electricity, plus the high maint of a wire edm, plus consumables (DI resin, Filters, guides, cleaning, etc...).

    I may just be rambling useless info for you, I am far from a "real edm machinist". Im just a hack that figures out what I have to do to get my products made as best I can. But hopefully there is something usable in my bumblings that helps you.

    Wade

  13. Likes BugRobotics, aarongough liked this post
  14. #9
    Join Date
    May 2011
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    2,715
    Post Thanks / Like
    Likes (Given)
    269
    Likes (Received)
    1475

    Default

    When I get a chance I'll cut a piece of o1 and harden it at your thickness and do a 1" cut and time it. That should give you a decent idea of cut time. It will be with brass uncoated 10 wire, believe it or not moving up to 12 wire can really make a difference in cut time.

  15. Likes aarongough liked this post
  16. #10
    Join Date
    Oct 2014
    Country
    UNITED STATES
    State/Province
    Maine
    Posts
    1,299
    Post Thanks / Like
    Likes (Given)
    255
    Likes (Received)
    626

    Default

    This might be a stupid idea, but here goes:

    Make a fixture that holds the knife, in its picture-frame at the bottom of a shallow pocket, fasten the picture frame in using the locating holes, pour it full of babbit-esque material, do the pin holes and then the profile. The knife will be locked in place by the babbit, so it shouldn't move as you do the final cutting, especially if you cut around the blade first. Flip it over and melt out the babbit.

    The usual concern with fixturing with low melting point metal is the time it takes, but with your low production rate, that should be less of a concern.

  17. Likes aarongough liked this post
  18. #11
    Join Date
    Aug 2013
    Country
    UNITED STATES
    State/Province
    New Mexico
    Posts
    154
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    50

    Default

    Or a vacuum fixture with a skeleton frame under the knife.

  19. Likes aarongough liked this post
  20. #12
    Join Date
    Nov 2016
    Country
    UNITED STATES
    State/Province
    North Carolina
    Posts
    241
    Post Thanks / Like
    Likes (Given)
    75
    Likes (Received)
    121

    Default

    Aaron, You've probably thought of this and there's a reason it won't work, but why not bolt the knife down through the handle after finishing the 2nd op milling on the blade so you can completely cut the perimeter?

    On a side note, one of our new products might need some WEDM work so I'm interested to see how it goes if you do go that route.

    Edited to add- I must type slow, two new replies with probably better ideas since I started with this post...

  21. Likes aarongough liked this post
  22. #13
    Join Date
    Oct 2014
    Country
    UNITED STATES
    State/Province
    Maine
    Posts
    1,299
    Post Thanks / Like
    Likes (Given)
    255
    Likes (Received)
    626

    Default

    Quote Originally Posted by Duc View Post
    Or a vacuum fixture with a skeleton frame under the knife.
    I think he's trying to avoid needing a custom fixture for each minor variation. With a vacuum, you'd need to match up to the knife pretty exactly.

  23. Likes aarongough liked this post
  24. #14
    Join Date
    May 2017
    Country
    UNITED STATES
    State/Province
    Massachusetts
    Posts
    292
    Post Thanks / Like
    Likes (Given)
    19
    Likes (Received)
    70

    Default

    It does seem like using a standard blank and leaving a thin web to part off would get you down to only a single custom fixture. It would be a good step in the right direction.

    For parting off you might be able to get away with a 3d printed support to match the bevel profile. I am thinking 3d print a top and bottom profile that is backed by a standard metal bottom fixture and thick top plate. Crank down on the clamping bolts to get a little crush in the Z axis for good contact. You only need to watch the heat from the part off to make sure things don't get melty and loose your support.

  25. Likes aarongough liked this post
  26. #15
    Join Date
    Jun 2002
    Country
    CANADA
    State/Province
    British Columbia
    Posts
    2,989
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2693

    Default

    Hi Aaron:
    You wrote:
    "Marcus: sorry for revisiting this, I know we spoke about this a while ago but I wanted to take the time to fully flesh out the idea and then finally put it to bed one way or the other!"

    No need to apologize...obviously this idea still has you intrigued and as you also say, you need to make a final decision one way or another.

    What I wrote to you in that email exchange aligns pretty well with what I'm reading here, so no great surprises there.
    For the benefit of all the other participants of this thread, Aaron and I exchanged emails way back in March of 2019 about this subject, and I gave my two cents' worth back then.

    Aaron, if you wish to, I have no objection to having you copy that email exchange onto this forum if you still have it around.

    Moving on to what to do about putting this idea to bed, I have a proposal for you:
    If you want to run through your method on some plates, I will happily wire cut the shapes for you at cost, so you'll have to pay only for the wire I consume.
    That way you will know for sure, and I will be very pleased to help you to get to a decision this way.
    Make a few so we can see how they cut in a stack, and once we're done we can compile an after battle debriefing and figure out how it worked and what potential it has.

    Between the two of us we will figure this out, and then you can make your final decision once we have fought the actual demons in this plan rather than fighting the imaginary ones we're all afraid of.

    If you can live with getting them whenever I can slide them into the machine rather than needing them on a specific schedule, that would help me a lot.
    I've got a punch die to wirecut right now and that job will be on the machine for the next week.
    After that I have only small jobs to run...they typically occupy the machine for maybe a day worst case.
    That gives us a window of opportunity in December if you can be flexible on the timing.

    So give it some thought and let me know...I have to buy wire again soon, so if you want to try this I will get more than I originally planned.

    Cheers

    Marcus
    Implant Mechanix • Design & Innovation > HOME
    Vancouver Wire EDM -- Wire EDM Machining

  27. #16
    Join Date
    May 2011
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    2,715
    Post Thanks / Like
    Likes (Given)
    269
    Likes (Received)
    1475

    Default

    Quote Originally Posted by implmex View Post
    Hi Aaron:
    You wrote:
    "Marcus: sorry for revisiting this, I know we spoke about this a while ago but I wanted to take the time to fully flesh out the idea and then finally put it to bed one way or the other!"

    No need to apologize...obviously this idea still has you intrigued and as you also say, you need to make a final decision one way or another.

    What I wrote to you in that email exchange aligns pretty well with what I'm reading here, so no great surprises there.
    For the benefit of all the other participants of this thread, Aaron and I exchanged emails way back in March of 2019 about this subject, and I gave my two cents' worth back then.

    Aaron, if you wish to, I have no objection to having you copy that email exchange onto this forum if you still have it around.

    Moving on to what to do about putting this idea to bed, I have a proposal for you:
    If you want to run through your method on some plates, I will happily wire cut the shapes for you at cost, so you'll have to pay only for the wire I consume.
    That way you will know for sure, and I will be very pleased to help you to get to a decision this way.
    Make a few so we can see how they cut in a stack, and once we're done we can compile an after battle debriefing and figure out how it worked and what potential it has.

    Between the two of us we will figure this out, and then you can make your final decision once we have fought the actual demons in this plan rather than fighting the imaginary ones we're all afraid of.

    If you can live with getting them whenever I can slide them into the machine rather than needing them on a specific schedule, that would help me a lot.
    I've got a punch die to wirecut right now and that job will be on the machine for the next week.
    After that I have only small jobs to run...they typically occupy the machine for maybe a day worst case.
    That gives us a window of opportunity in December if you can be flexible on the timing.

    So give it some thought and let me know...I have to buy wire again soon, so if you want to try this I will get more than I originally planned.

    Cheers

    Marcus
    Implant Mechanix • Design & Innovation > HOME
    Vancouver Wire EDM -- Wire EDM Machining
    Can't beat that offer!

  28. Likes aarongough liked this post
  29. #17
    Join Date
    Oct 2014
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,606
    Post Thanks / Like
    Likes (Given)
    1221
    Likes (Received)
    1344

    Default

    Thanks to everyone for all the awesome feedback and replies so far! Really cool to be able to bounce an idea off so many experienced people like this! I will go through and respond as best I can:

    Quote Originally Posted by Mike1974 View Post
    Another thing to consider, can possibly stack 2-3 plates together in wedm and/or nest the blades to minimize material loss. You don't need to spread them out for endmill clearance, can go maybe .05" apart or so between blades
    I had wondered about this in the past as well... Would be pretty cool to just be able to drop a 16x20" plate into the machine and then nest parts into it. My only concern would be chatter toward the middle of the sheet, I think I'd have to mill a bunch of hold-down holes into the sheet as the first op...

    Which brings me to one thing I've thought about quite a bit: magnetic workholding on the mill? Would a large magnetic chuck be stupid if I was going easy with the cuts? I have always figured that the magnet would cause a lot of issues with holding chips in place and killing tool life, but I haven't seen too many people doing it... If I skimmed the top of the sheet to the same Z height over it's whole surface then when I flipped it I could get contact on the bottom of the parts. No idea if that would let me very carefully profile the parts out with light finishing cuts? I assume they would want to move...

    Quote Originally Posted by plastikdreams View Post
    12 wire...one piece, 23" through .156 hardened steel using adaptive...(granted I'm thinking mits) probably average around .100 inches per minute maybe .200 if you tune your epack. Another thing to consider would be using coated wire. I may be a little low on my estimates.

    Things to ponder before getting a wire edm...water resistivity is very important, if the thing sits for the while it's going to take a while for the resin bottle to get the numbers down.

    Tank seals don't like being dry, metal partials can build up on the moving surface and when dried out can shread the seals. This allows water to leak into places your really don't want it (think ways and ball screws).

    Most wire machines take up a lot of space. For for a 13x15 machine you will need about 12 square foot for it maybe more.

    Get spare resin bottles and filters, they always go out when you least expect it typically on weekends.

    Program in a tab or 2 to hold the knife from falling through. Especially if you use a start hole and have retrace with machining on...it will recut everything in it's path and render your part useless. This is because the wire will break due to the piece falling in before it can finish it's cycle. It then goes back and rethreads and will commence cutting back to the start point...unless the part catches the lower head and either moves the entire setup and/or causes the machine to crash. Then you need to restart everything and do a wire alignment blah blah blah. Also the flushing can cause the part to vibrate when there's only a small amount to cut. If critical I toss some neodymium magnets in to hold the part where it should be...if it's magnetic of course.

    That being said, I love using wire edm for all kinds of stuff. The speed sucks but the accuracy that can be obtained after some experience is amazing. I've only used mitsubishi wire edm's so I can't speak for the others but I have heard good things about them. Key thing with any is maintenance.
    Thanks mate! This gives me a basic idea of cutting params which is great. I think I could shoe-horn a 10-12sq ft machine into my shop with only a bit of a squeeze, much larger than that and it would be impractical though unfortunately. A desktop WEDM would be super cool, but I think I have only ever seen one of those...

    Quote Originally Posted by Wade C View Post
    Thinking of this in terms of efficiency and cost (since youre using for your own products, as do I)... and being upfront with you, the newest machine I have any experience with is a 1996 Mits FX10 and Im no pro with it. Youll probably need to figure out your time/cost associated with doing it how you do now - vs the mill time for your new process plan, and then add in the wire edm, wire portion.

    I have no familiarity with thin work pieces, so cant speak to that variable - but am happy to share what I have seen with some of my products that I make using a WEDM that appear on the surface to share some potential similarities.

    On my FX10, a part I make is cut from 1.55" thick hardened A2. It has a .625 bore lengthwise, and the wire cuts in from one of the long sides, into the center where the through bore is, and along it, and through to the opposite side (think a letter J laying on its side, sort of). It calculates out to roughly 2.9" worth of cut length. It takes 40 minutes on the FX10 using .010 wire and the powermaster settings (.072" per min avg). I have a 92 Mits 110SZ that I was able to get it sped up - but the fastest I ever got it to cut was 28 minutes (.103" per min avg). I have not had the time to spend chasing power settings to see if I can get the FX speed up to match the old SZ - but will some day. Wire consumption is roughly 45 parts from a 13.2lb spool - or 0.3lbs per 2.9 long cut.

    So, if you stacked 10 of your plates together yould have 1.56" thick which is very close to the same, compare your thinner areas with the bore in the middle I cut through on my parts and I see a lot of similarity.

    If you got the fastest cut time I have ever had on either machine of mine (.103" per min - I say that figuring you would get a faster machine than I have, but the gaps in the stacked plates would likely slow you down a little), that would roughly equate to 200 minutes to cut a 20.53 path and use 2.13lbs (rough calcs) of .010 wire and leave you with 10 parts cut out. I pay about 78USD per 13.2lb spool, plus shipping and tax - so you would see roughly 12.50USD in wire cost plus the 200 minutes cutting, plus electricity, plus the high maint of a wire edm, plus consumables (DI resin, Filters, guides, cleaning, etc...).

    I may just be rambling useless info for you, I am far from a "real edm machinist". Im just a hack that figures out what I have to do to get my products made as best I can. But hopefully there is something usable in my bumblings that helps you.

    Wade
    Awesome info Wade! Thanks for taking the time to give me some concrete examples of time and cost... The cost of my current process is likely lower, I would say it's currently about $2-3 per knife in tooling plus about 15 minutes cycle time to rough and finish the blade profile, fairly fast as I currently do that part of the process soft. The real 'cost' on my end is mainly in the lack of flexibility and the fact that right now to make it practical I have to machine that part soft. If I hard-mill the profile (which is better because it allows me to heat-treat rough-cut rectangular blanks which reduces bottlenecks) then the profile machining goes up to about 45 minutes per blade and $4-6 each in tooling which is definitely getting closer to the numbers you ballparked for the WEDM...

    Quote Originally Posted by plastikdreams View Post
    When I get a chance I'll cut a piece of o1 and harden it at your thickness and do a 1" cut and time it. That should give you a decent idea of cut time. It will be with brass uncoated 10 wire, believe it or not moving up to 12 wire can really make a difference in cut time.
    That would be awesome mate! Would be great to have some concrete numbers! If you'd like me to mail you some pre-hardened stock just let me know!

    Quote Originally Posted by DanielG View Post
    This might be a stupid idea, but here goes:

    Make a fixture that holds the knife, in its picture-frame at the bottom of a shallow pocket, fasten the picture frame in using the locating holes, pour it full of babbit-esque material, do the pin holes and then the profile. The knife will be locked in place by the babbit, so it shouldn't move as you do the final cutting, especially if you cut around the blade first. Flip it over and melt out the babbit.

    The usual concern with fixturing with low melting point metal is the time it takes, but with your low production rate, that should be less of a concern.
    I have wondered about this in the past too... Will a fixturing alloy hang onto a flat surface though? There's no real areas for the material to 'key' into on the underside of the knife and I've never been sure if it will hang onto a flat surface... I actually bought some special thermoset 'hot glue' (a urethane based one) last year with the idea of trying it for the same purpose. I found it didn't hang onto the blades as well as I would have hoped, but I might need to try again with the blank pre-heated or something.

    Quote Originally Posted by Duc View Post
    Or a vacuum fixture with a skeleton frame under the knife.
    I actually have a vacuum generator here (venturi style) with an eye toward trying something like this. I imagine that minimizing the material removal necessary with the vacuum fixturing would be important? Maybe tabs on the part leaving the outline oversize and then skim with it held down by the vacuum? I have never had the chance to use vacuum fixturing as yet so that would be new territory for me...

    Quote Originally Posted by LOTT View Post
    Aaron, You've probably thought of this and there's a reason it won't work, but why not bolt the knife down through the handle after finishing the 2nd op milling on the blade so you can completely cut the perimeter?

    On a side note, one of our new products might need some WEDM work so I'm interested to see how it goes if you do go that route.

    Edited to add- I must type slow, two new replies with probably better ideas since I started with this post...
    I have actually tried this in the past, and unfortunately I wasn't able to stop the blade from chattering when trying to finish near the tip, far away from the bolt locations at the handle end. This is a problem that would be worse with the kitchen knives I'd like to make as they will be even longer.

    With that said last time I was trying a spring 'finger' to hold down the tip:



    The 'finger' on top was spring-tempered A2 (not hardened in that photo) and Maritool even made a series of 'downcut' endmills for me to try to make this fixture work but I couldn't get rid of the chatter... The tip would move sideways under cutting forces and that would destroy the endmills quickly!

    Maybe a large shaped steel block like 3/4 thick on top would be better? Or vacuum or magnetic fixturing for the blade section where there are no bolts?

    One thing that might make this easier is that the blade edge area does not really need to be finish machined as I sharpen that area away anyway, so perhaps I could mount toggle clamps overhead in that area to hold the blade down. That would potentially be fairly simple and only require a single customized fixture per blade shape... I was mucking around with this concept a little last week:



    The idea being that I would machine the blade in the 'picture frame' similar to what I show in my first post, but I would cut out the outline oversize and leave tabs, then finish the outline in this fixture. The red marker indicates the areas where I would take a finish cut. The biggest issue I see is that the toggle clamps are fairly tall which would get in the way of having nice stubby tooling, but perhaps a different kind of clamp or something could be used.

    Quote Originally Posted by DanielG View Post
    I think he's trying to avoid needing a custom fixture for each minor variation. With a vacuum, you'd need to match up to the knife pretty exactly.
    Yeah ideally I would really LOVE to sidestep the need for custom fixtures at all... That would really give me a lot of flexibility. However if that flexibility simply comes at too high of a cost (either initial or ongoing) then I might have to settle for 'second best' lol

    Quote Originally Posted by newtonsapple View Post
    It does seem like using a standard blank and leaving a thin web to part off would get you down to only a single custom fixture. It would be a good step in the right direction.

    For parting off you might be able to get away with a 3d printed support to match the bevel profile. I am thinking 3d print a top and bottom profile that is backed by a standard metal bottom fixture and thick top plate. Crank down on the clamping bolts to get a little crush in the Z axis for good contact. You only need to watch the heat from the part off to make sure things don't get melty and loose your support.
    Interesting idea on the 3D printed clamping plates! That sounds like a fun option, I could do with an excuse to buy a nicer printer

    Quote Originally Posted by implmex View Post
    Hi Aaron:
    You wrote:
    "Marcus: sorry for revisiting this, I know we spoke about this a while ago but I wanted to take the time to fully flesh out the idea and then finally put it to bed one way or the other!"

    No need to apologize...obviously this idea still has you intrigued and as you also say, you need to make a final decision one way or another.

    What I wrote to you in that email exchange aligns pretty well with what I'm reading here, so no great surprises there.
    For the benefit of all the other participants of this thread, Aaron and I exchanged emails way back in March of 2019 about this subject, and I gave my two cents' worth back then.

    Aaron, if you wish to, I have no objection to having you copy that email exchange onto this forum if you still have it around.

    Moving on to what to do about putting this idea to bed, I have a proposal for you:
    If you want to run through your method on some plates, I will happily wire cut the shapes for you at cost, so you'll have to pay only for the wire I consume.
    That way you will know for sure, and I will be very pleased to help you to get to a decision this way.
    Make a few so we can see how they cut in a stack, and once we're done we can compile an after battle debriefing and figure out how it worked and what potential it has.

    Between the two of us we will figure this out, and then you can make your final decision once we have fought the actual demons in this plan rather than fighting the imaginary ones we're all afraid of.

    If you can live with getting them whenever I can slide them into the machine rather than needing them on a specific schedule, that would help me a lot.
    I've got a punch die to wirecut right now and that job will be on the machine for the next week.
    After that I have only small jobs to run...they typically occupy the machine for maybe a day worst case.
    That gives us a window of opportunity in December if you can be flexible on the timing.

    So give it some thought and let me know...I have to buy wire again soon, so if you want to try this I will get more than I originally planned.

    Cheers

    Marcus
    Implant Mechanix • Design & Innovation > HOME
    Vancouver Wire EDM -- Wire EDM Machining
    Thanks Marcus! Both for your patience and for the very kind offer! I think part of the reason I haven't been able to put this idea to bed in my mind is because my experience with WEDM is purely from watching videos so I have very much a 'grass must be greener' idea of it... If cost/time considerations were nonexistent then I think WEDM would be a great process but all the comments here alongside your original thoughts in the email are definitely making me think that perhaps another option (ANY other option) would be better... The thing that most appeals to me is simply the ability to get away from custom fixturing as that would really give me a freedom that I can't get with my current methods.

    I have looked at some other options to get a similar kind of process working, but these would all be very much DIY alternatives rather than an 'off the shelf' option like WEDM is:

    - A CNC 'abrasive wire saw'. The abrasives wires I've looked at would come from from Diamond wire saws, diamond wires and accessories - Diamond WireTec and they quoted me pricing of about $50 each for an 800mm long diamond grit wire loop with 0.018" diameter wire. This is an interesting idea but I'm not sure that the tolerances I'd be able to hold would be good enough. Machine would require some maintenance but definitely lower than what I'm hearing would be required for WEDM and such a machine would be fairly 'easy' to prototype

    - One pie-in-the-sky option would be "waterjet guided fiber laser" which uses a high pressure water-jet as an optic guide for a fiber laser, really cool tech but fairly new I think:

    https://www.youtube.com/watch?v=L1tZWDKMYdI
    https://www.youtube.com/watch?v=GywQVNA6qoY

    Quote Originally Posted by plastikdreams View Post
    Can't beat that offer!
    Seriously!

  30. #18
    Join Date
    Jun 2002
    Country
    CANADA
    State/Province
    British Columbia
    Posts
    2,989
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2693

    Default

    Hi again Aaron:
    So now you've got me hooked, and I want to try!
    So let's plan to give it a whirl and see what we can come up with.

    If you want to keep this adventure on the public forum I'm totally cool with that...ditto if you want to move it back to a private email exchange to flesh out the details of what to do and how, and keep those details private to help keep competitors at bay.

    I assume you still have my email address...if not you can look it up easily from my website...I hate to publish it directly on a forum because I'm always worried I'll end up with a torrent of SPAM if I do.

    Cheers

    Marcus
    Implant Mechanix • Design & Innovation > HOME
    Vancouver Wire EDM -- Wire EDM Machining

  31. Likes aarongough liked this post
  32. #19
    Join Date
    Oct 2014
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,606
    Post Thanks / Like
    Likes (Given)
    1221
    Likes (Received)
    1344

    Default

    Quote Originally Posted by implmex View Post
    Hi again Aaron:
    So now you've got me hooked, and I want to try!
    So let's plan to give it a whirl and see what we can come up with.

    If you want to keep this adventure on the public forum I'm totally cool with that...ditto if you want to move it back to a private email exchange to flesh out the details of what to do and how, and keep those details private to help keep competitors at bay.

    I assume you still have my email address...if not you can look it up easily from my website...I hate to publish it directly on a forum because I'm always worried I'll end up with a torrent of SPAM if I do.

    Cheers

    Marcus
    Implant Mechanix • Design & Innovation > HOME
    Vancouver Wire EDM -- Wire EDM Machining
    Sounds amazing Marcus, let's try it! Really appreciate your willingness to help out!

    Let's keep the trial public, I think it's good to get a bit more info out there regarding this as I've certainly never seen anyone else try to make knives this way, and maybe it will help someone else in the future!

    Unfortunately I won't be able to machine my end of the parts until the first couple of weeks next year if that's ok with you? Trying to catch up with all my backorders before Christmas which has me working crazy hours!

    I'm really looking forward to trying this out.

    -Aaron

  33. #20
    Join Date
    Jun 2002
    Country
    CANADA
    State/Province
    British Columbia
    Posts
    2,989
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2693

    Default

    No worries Aaron:
    Give me a shout whenever you're ready to move forward.
    This should be fun!

    BTW I just spent a bit of quality time with the Avonisys website and now I WANT one!
    It is so damn cool...I REALLY want one.

    Cheers

    Marcus
    Implant Mechanix • Design & Innovation > HOME
    Vancouver Wire EDM -- Wire EDM Machining


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
  •