one horizontal vs. how many verticals? (Brothers at that) - Page 2
Close
Login to Your Account
Page 2 of 22 FirstFirst 123412 ... LastLast
Results 21 to 40 of 430
  1. #21
    Join Date
    Dec 2013
    Country
    UNITED STATES
    State/Province
    California
    Posts
    899
    Post Thanks / Like
    Likes (Given)
    264
    Likes (Received)
    536

    Default

    The multi-million dollar question. Answer: depends! Smallish parts with a good # of tool changes and positioning moves and some tapping? Hard to beat the brother. Two faster spindles versus one slower spindle? Automating, first op with straight forward raw material and work holding, pretty doable. How about a hybrid set-up? What is the cycle time for first op versus second? Let's say your 1st op is a longer cycle time and you have 6 clamping stations on each pallet. Set up 4 stations to run 1st op at night with robot tending. During the day have an operator load the 2nd ops in the other stations. Unless you come up with a quick change jaw set up. Has Allison been doing a good job running the machines? Many of my client's shops utilize women and they do a good job. My .02

  2. Likes wheelieking71, empwoer liked this post
  3. #22
    Join Date
    Jun 2019
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    512
    Post Thanks / Like
    Likes (Given)
    185
    Likes (Received)
    535

    Default

    Quote Originally Posted by Orange Vise View Post
    8 vise stations per tombstone is very "last decade". We just delivered four tombstones to a customer who opted for 16 vises per pallet.
    And in other news I have to go pick up my new yacht.

  4. Likes Hardplates, Bobw, Oldwrench liked this post
  5. #23
    Join Date
    Feb 2012
    Location
    California
    Posts
    1,450
    Post Thanks / Like
    Likes (Given)
    909
    Likes (Received)
    1612

    Default

    Quote Originally Posted by CAMasochism View Post
    And in other news I have to go pick up my new yacht.
    These were the lower priced Delta vises.

    Total price came in at less than half of what most of our competitors would have charged for a comparable setup, so no yachts (and no apologies) here.

  6. #24
    Join Date
    Jun 2013
    Location
    Northern Il
    Posts
    1,510
    Post Thanks / Like
    Likes (Given)
    771
    Likes (Received)
    1379

    Default

    This is a rather challenging problem.

    Without knowing a lot about your business, the first thing I would ask you is if you currently have enough square footage to support the 2X production numbers?

    The real challenge in this will be that you will the normal raw parts arrived, parts in process, and finished parts ready to ship but what happens when you suddenly double the size of these? Running lights out really complicates the material handling issues in that now you also need to have enough floor space to accommodate the extra space requirements of the weekend transportation gaps.

    The concept of utilizing robot loaders sounds good but a robotic approach will exacerbate floor space requirements. If you go down this path, you will need to thoroughly analyze all the parts handling from how the raw material comes in to how it goes out the door to the truck.

    I could see a cell consisting of the 4 Brothers machines and being serviced by one robot as being very feasible if you can come up with a method to properly address the parts handling issues. One downside to the 1 robot and 4 Brothers machines is that all production will stop if any one of the machines or the robot need repair work done. A one robot and two machine approach would be more robust but require more floor space and further complicate material handling.

    Have you analyzed how much time it currently takes to load and unload your existing pallets? Are you using air hold down systems or are you still using mechanical hold downs requiring operator time which is really a loader/unloader person?

    Thee answer to this question is an indicator in how close to an automated approach you really are and if a horizontal with tombstones is more appropriate vs the four Brothers machines.

    Business growth is wonderful until it comes time to increase production capabilities in geometric increments. You have a very good opportunity with plenty of land mines to really complicate your life.

  7. #25
    Join Date
    Jan 2013
    Location
    Gilbert, AZ
    Posts
    6,581
    Post Thanks / Like
    Likes (Given)
    8402
    Likes (Received)
    8390

    Default

    Quote Originally Posted by Orange Vise View Post
    Horizontals, especially machines with pallet pools, allow you to load tombstones in a more mentally relaxed manner. Even if it takes the same total number of minutes to load the parts, these can be "junk minutes", e.g. 5 minutes here, 5 minutes there, with interruptions in between. It's a lot less taxing mentally and is a lot more efficient. If you were able to collect all the "junk minutes" over a given week, you might find yourself with an entire day or two's worth of free time.
    This right here ^^^^^^. I think it would be far more pleasant feeding a pallet-pool vs. 4 hungry Brothers.

    Quote Originally Posted by Orange Vise View Post
    8 vise stations per tombstone is very "last decade". We just delivered four tombstones to a customer who opted for 16 vises per pallet.
    My plan would be the same as the plan in my other current thread. Double station vises on 4 sided tombstones.
    OP-1 would be raw stock clamped with Pitbulls on top of the move jaws. OP-2 clamped in the vises normally.
    This would get me 8 OP-1, and 8 OP-2 parts per stone. Effectively 8 finished parts.

    One beautiful aspect of this for me. I already have the vises. All the jaws already exist.
    Just need to buy 4 stones. Bolt it all together. Set origins, and go! (I realize it will not be anywhere near that simple, but ya'll get the point).

  8. Likes mhajicek, Oldwrench, DrHook liked this post
  9. #26
    Join Date
    Jun 2015
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    594
    Post Thanks / Like
    Likes (Given)
    1202
    Likes (Received)
    252

    Default

    Sounds like a challenging decision. Sadly I have nothing to add but will be rooting for you and will closely follow the thread to glean as much info as I can about the process. Good luck man.

  10. #27
    Join Date
    Sep 2006
    Location
    Abingdon, VA
    Posts
    3,544
    Post Thanks / Like
    Likes (Given)
    4778
    Likes (Received)
    3437

    Default

    Horizontal all the way.

    Horizontals do require more up-front tooling, programming, and setup time, but it sounds like by using standard vises and already-machined jaws you could be cutting metal quickly.

    Another thing to keep in mind: even though a horizontal may not have the max RPM’s and/or rapids of a Brother VMC, , the typical Japanese 40-taper HMC is a heavy and rigid bastard, and cutting performance will be light years ahead of a 30-taper VMC.

    (No idea about the rigidity and performance of a Haas HMC...)

    Chip management is on another planet in an HMC versus a VMC as well....

    A pallet pool of some kind sounds almost like a necessity, as two pallets would work you to death — with little “lights out” capability compared to a pallet pool.

    Good luck!

    ToolCat

  11. Likes Hardplates, Ox, empwoer liked this post
  12. #28
    Join Date
    Jan 2013
    Location
    Gilbert, AZ
    Posts
    6,581
    Post Thanks / Like
    Likes (Given)
    8402
    Likes (Received)
    8390

    Default

    Quote Originally Posted by cnctoolcat View Post
    Horizontal all the way.

    Horizontals do require more up-front tooling, programming, and setup time, but it sounds like by using standard vises and already-machined jaws you could be cutting metal quickly.

    Another thing to keep in mind: even though a horizontal may not have the max RPM’s and/or rapids of a Brother VMC, , the typical Japanese 40-taper HMC is a heavy and rigid bastard, and cutting performance will be light years ahead of a 30-taper VMC.

    (No idea about the rigidity and performance of a Haas HMC...)

    Chip management is on another planet in an HMC versus a VMC as well....

    A pallet pool of some kind sounds almost like a necessity, as two pallets would work you to death — with little “lights out” capability compared to a pallet pool.

    Good luck!

    ToolCat
    Looking at a Doosan NHP4000 which actually rapids FASTER! than the Brother. By a lot 2362ipm for the Doosan vs. 1969 for the Brother.
    I don't know the acceleration rates? Or even how to compare them.
    And, I am sure chip-to-chip tool-change time between the two isn't even close (Brother by a land-slide!).
    But, I'll bet the Doosan is no slouch in the speed department. And, 40 taper FTW!

  13. Likes empwoer liked this post
  14. #29
    Join Date
    Feb 2008
    Location
    MI, USA
    Posts
    892
    Post Thanks / Like
    Likes (Given)
    68
    Likes (Received)
    363

    Default

    Quote Originally Posted by wheelieking71 View Post
    Looking at a Doosan NHP4000 which actually rapids FASTER! than the Brother. By a lot 2362ipm for the Doosan vs. 1969 for the Brother.
    I don't know the acceleration rates? Or even how to compare them.
    And, I am sure chip-to-chip tool-change time between the two isn't even close (Brother by a land-slide!).
    But, I'll bet the Doosan is no slouch in the speed department. And, 40 taper FTW!
    How many tools do you have per part? I don't think a landslide in tool change time will matter here.

    Lets say 10 tools. and lets say you get one part per tool change (you'll obviously get more)

    I am not looking up what chip to chip times are, but lets say the brother is 1 second, and the doosan is 5. So you have 10 seconds of tool changes vs 50 seconds. For 1000 pcs, thats an extra 11 hours. But if you pallet pool it up and nest lots of parts, how many more hours of unattended cycle time can you get by letting it run while you leave? 20 hours a week? 40 hours more a week? I bet you'll be getting a ton.

    I am sure you know how to do this math already. Sorry if you have already thought of it.


    Edit: Just looked, a quick google says a Speedio is 1.4 seconds chip to chip, while a brochure for the nhp4000 says 2.3 seconds chip to chip. That is even less of a gap than I figured.

    Funny enough, the Doosan brochure I found calls it "Chip to Chop" time. That thing looks like it can chop material pretty well..

  15. Likes Ox, Garwood liked this post
  16. #30
    Join Date
    Jan 2012
    Location
    indiana, usa
    Posts
    198
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    48

    Default

    On our Horizontal, I made the tombstones. The first one I made 6 sided ones, that was a 400mm machine I think. We then upgraded to Mazak FH-6000 and I made 8 sided tombstones. I think you could probably get 6 sided and get 12 parts per cycle. I am pretty sure you can buy 6 sided.

  17. #31
    Join Date
    Feb 2008
    Location
    MI, USA
    Posts
    892
    Post Thanks / Like
    Likes (Given)
    68
    Likes (Received)
    363

    Default

    Quote Originally Posted by chad883 View Post
    On our Horizontal, I made the tombstones. The first one I made 6 sided ones, that was a 400mm machine I think. We then upgraded to Mazak FH-6000 and I made 8 sided tombstones. I think you could probably get 6 sided and get 12 parts per cycle. I am pretty sure you can buy 6 sided.
    Just make sure your machine can index to where you can load it if you get more than 4 sides.

    We almost bought a 6 sided tombstone until I mocked up how awkward it would be to load on our machine with only 4 pallet "stations".

  18. Likes wheelieking71 liked this post
  19. #32
    Join Date
    Apr 2015
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    25
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    7

    Default

    Quote Originally Posted by wheelieking71 View Post
    Looking at a Doosan NHP4000 which actually rapids FASTER! than the Brother. By a lot 2362ipm for the Doosan vs. 1969 for the Brother.
    I don't know the acceleration rates? Or even how to compare them.
    And, I am sure chip-to-chip tool-change time between the two isn't even close (Brother by a land-slide!).
    But, I'll bet the Doosan is no slouch in the speed department. And, 40 taper FTW!
    Doosan says 1G. Apparently Brothers are over 2G 2.3s chip to chip v. 1.4s.


    Meet the NHP Series: Legendary Horizontal Machining Centers - YouTube

  20. #33
    Join Date
    Feb 2007
    Location
    Boston MA
    Posts
    1,140
    Post Thanks / Like
    Likes (Given)
    512
    Likes (Received)
    1036

    Default

    I have a Speedio S700X1 and the shop I do programming/setup work for has a Speedio R650X2 that I program/setup as well as a bunch of Kitamura horizontals ranging from 250mm to 500mm with 2-pallet, 8-pallet and 10-pallet systems which I also program/setup.

    Here's my .02...

    Horizontals are great in terms of keeping the spindle turning and giving you time to let the operator load more parts at a reasonable pace. But they are not a magical solution to every machining problem. They usually tend to have slower rapids than verticals, and even the fast ones are going to pale in comparison to a Speedio.

    I would never want to load 4 Speedios. Hell, even two R650's would be a royal pain in the ass to keep up with - you'd constantly be running between machines to keep them fed.

    My ideal solution would be robot loading. I know you mentioned chips and load errors, but I have to think you could solve those issues with a combination of a chip fan + air blow on the robot loader (can adjust pressure and angle to make sure the area you're loading your part in is clean).

    Whether you run a horizontal or a Speedio, IMO it is IMPERATIVE that you implement tool breakage detection with your tool setter AND fixture loading checks with your probe. That way, you could detect if a part is misloaded because a chip is stuck under the part and halt the operation before you make a bunch of bad parts - or even better - unclamp the vise, lift up the part with the robot and re-blow off the fixture and re-seat it. All of that is not really very complicated to do, and is key to making consistently good parts.

    Running lights out isn't all that easy and there are a million things that can go wrong. You really need to either start off with an experienced person who can cover most of your bases on what things go wrong and pre-empting them, or you just need to keep chipping away at it and iterate your process so that each time a stoppage occurs, you take steps to stop it happening next time. Probing is key. The best luck I have had is using the probe to check for part misloads, then using the probe to pick up part X/Y/Z zero and measuring tilt/skew and compensating with WCS updates and G69 rotations - once I started doing that, I started succeeding. Recent job I programmed/setup ran 1,086 pieces and there were 3 bad parts (mostly due to dings in the original stock from the supplier). The key was probing to make sure the part was loaded right and probing to compensate for tilt/skil/translation offset in X/Y/Z, combined with tool breakage checking. Once I dialed that all in, it was just a function of keeping the machine (or robot) fed.

    You can add in tool groups, tool-life/expiration and then you really start getting things automated to where you load a bunch of parts in the morning and then just keep tabs on the robots as you do other stuff (or as you're at home sleeping).

    But for as much advantages as horizontals have, they are definitely not faster than a Speedio. Forget quoted numbers for rapids...they are irrelevant. All the 'little things' are what kills you. Tool change time, the time it takes to probe a part, the time it takes to index/load a new pallet. Those are where all your inefficiencies come from. Plus, if you are doing dynamic/HSM toolpaths, the speed at which a hefty box-way machine moves compared to a Speedio - the Speedio will run circles around anything else. I would much rather start with a fast machine and solve the loading problem than start with a machine (pallet pool horizontal) that doesn't have a loading problem but has a speed problem. Much easier to solve the loading issue than scratching for seconds here and there in a machining program to make your deadlines.

    PM me if I can be of any more assistance. I've spent the past couple of years solving exactly the problems you have. Not looking for $$$ or a job or anything, just happy to share my experience. My specialty has been high volume production on parts that run for years, fast efficient programs that make full use of the machine, and making jobs run 24/7 without issues making good parts day after day even with dopey operators that forget to install clamps or who load a part badly. I've had lots of success but it was a long road getting there!

  21. #34
    Join Date
    Jan 2013
    Location
    Gilbert, AZ
    Posts
    6,581
    Post Thanks / Like
    Likes (Given)
    8402
    Likes (Received)
    8390

    Default

    Quote Originally Posted by dandrummerman21 View Post
    Just make sure your machine can index to where you can load it if you get more than 4 sides.

    We almost bought a 6 sided tombstone until I mocked up how awkward it would be to load on our machine with only 4 pallet "stations".
    That never occurred to me! Yea, that could make for a bad day if the stone will only turn in 90' increments at the change out station.

  22. #35
    Join Date
    Jan 2013
    Location
    Gilbert, AZ
    Posts
    6,581
    Post Thanks / Like
    Likes (Given)
    8402
    Likes (Received)
    8390

    Default

    Quote Originally Posted by SRT Mike View Post
    I have a Speedio S700X1 and the shop I do programming/setup work for has a Speedio R650X2 that I program/setup as well as a bunch of Kitamura horizontals ranging from 250mm to 500mm with 2-pallet, 8-pallet and 10-pallet systems which I also program/setup.

    Here's my .02...

    Horizontals are great in terms of keeping the spindle turning and giving you time to let the operator load more parts at a reasonable pace. But they are not a magical solution to every machining problem. They usually tend to have slower rapids than verticals, and even the fast ones are going to pale in comparison to a Speedio.

    I would never want to load 4 Speedios. Hell, even two R650's would be a royal pain in the ass to keep up with - you'd constantly be running between machines to keep them fed.

    My ideal solution would be robot loading. I know you mentioned chips and load errors, but I have to think you could solve those issues with a combination of a chip fan + air blow on the robot loader (can adjust pressure and angle to make sure the area you're loading your part in is clean).

    Whether you run a horizontal or a Speedio, IMO it is IMPERATIVE that you implement tool breakage detection with your tool setter AND fixture loading checks with your probe. That way, you could detect if a part is misloaded because a chip is stuck under the part and halt the operation before you make a bunch of bad parts - or even better - unclamp the vise, lift up the part with the robot and re-blow off the fixture and re-seat it. All of that is not really very complicated to do, and is key to making consistently good parts.

    Running lights out isn't all that easy and there are a million things that can go wrong. You really need to either start off with an experienced person who can cover most of your bases on what things go wrong and pre-empting them, or you just need to keep chipping away at it and iterate your process so that each time a stoppage occurs, you take steps to stop it happening next time. Probing is key. The best luck I have had is using the probe to check for part misloads, then using the probe to pick up part X/Y/Z zero and measuring tilt/skew and compensating with WCS updates and G69 rotations - once I started doing that, I started succeeding. Recent job I programmed/setup ran 1,086 pieces and there were 3 bad parts (mostly due to dings in the original stock from the supplier). The key was probing to make sure the part was loaded right and probing to compensate for tilt/skil/translation offset in X/Y/Z, combined with tool breakage checking. Once I dialed that all in, it was just a function of keeping the machine (or robot) fed.

    You can add in tool groups, tool-life/expiration and then you really start getting things automated to where you load a bunch of parts in the morning and then just keep tabs on the robots as you do other stuff (or as you're at home sleeping).

    But for as much advantages as horizontals have, they are definitely not faster than a Speedio. Forget quoted numbers for rapids...they are irrelevant. All the 'little things' are what kills you. Tool change time, the time it takes to probe a part, the time it takes to index/load a new pallet. Those are where all your inefficiencies come from. Plus, if you are doing dynamic/HSM toolpaths, the speed at which a hefty box-way machine moves compared to a Speedio - the Speedio will run circles around anything else. I would much rather start with a fast machine and solve the loading problem than start with a machine (pallet pool horizontal) that doesn't have a loading problem but has a speed problem. Much easier to solve the loading issue than scratching for seconds here and there in a machining program to make your deadlines.

    PM me if I can be of any more assistance. I've spent the past couple of years solving exactly the problems you have. Not looking for $$$ or a job or anything, just happy to share my experience. My specialty has been high volume production on parts that run for years, fast efficient programs that make full use of the machine, and making jobs run 24/7 without issues making good parts day after day even with dopey operators that forget to install clamps or who load a part badly. I've had lots of success but it was a long road getting there!
    Good post. I appreciate the input. One thing that needs to be stated about this:
    I've been making these parts for years now. I have a very reliable process dialed. Most of the tools that are in the haas, have been in there in measures of years.
    And, I don't pussy-foot around either. That machine is making those parts as fast as possible for that machine. Its just dialed, and reliable.
    Same with the Brother, save the rougher. I swap that every 30 days. Everything else are the original tools I started the job with the first day I ran that machine.
    I did break a saw in the brother once when the power went out. And, I have tried a few different drills looking for more efficiency.
    I was having problems in the beginning with a deep 1/8" hole, breaking drills. But, I have that sorted.
    The process is dialed. Its not like I am starting from scratch with these parts.

    I will gain an hour a day, just not having to deal with chips at the end of the shift. That's pretty efficient.
    The Brother is not as bad as the haas. But, its not great like a horizontal either. It still needs its share of attention (the haas is absurd!)

  23. #36
    Join Date
    Mar 2013
    Country
    UNITED STATES
    State/Province
    Oregon
    Posts
    2,075
    Post Thanks / Like
    Likes (Given)
    840
    Likes (Received)
    2579

    Default

    I think you need to reconsider the Flex 2 cell.

    The problems of robots are well documented and have lots of solutions. On our automation setups, we have fixed air lines on a solenoid valve that blast the workholding pockets out during the robot load. We also have part-presence detection that uses air pressure to make sure the part is seated in the vise. Deploy those two tricks and a little tool-break detection and straightforward 3 axis work is quite very reliable. Yamazen can implement both as part of the package (a few hundred bucks in total).

    As to the accuracy issue? That is a matter of workholding design where we have a bunch of tricks to throw at the problem. In your case, we're talking a pair of Schunk air vises with hardened tops that you can send out for a little WEDM work. Allan or Scott at Yamazen can get you dialed in, and I have a guy I work with who could have this solved for you with a couple of hours of CAD work - he has set up dozens of parts in this sorta configuration. The stuff you are working on is pretty easy to automate - easy to grip, easy to hold.

    Once you go Flex 2? You're going to ditch the R650/Haas setup for the other SKU in short order. You are looking at *hours* of unattended production, with extremely high reliability, way less workholding to manage, and the problems are all ones that have been solved before and you have lots of resources to pull to get it up and running. Oh, and all at about the same price as an optioned horizontal, spitting out 2x the parts, and with your only interaction being "Lay blanks on conveyor, pick up blanks from other conveyor."

  24. #37
    Join Date
    Dec 2013
    Country
    UNITED STATES
    State/Province
    California
    Posts
    899
    Post Thanks / Like
    Likes (Given)
    264
    Likes (Received)
    536

    Default

    Quote Originally Posted by dandrummerman21 View Post
    Edit: Just looked, a quick google says a Speedio is 1.4 seconds chip to chip, while a brochure for the nhp4000 says 2.3 seconds chip to chip. That is even less of a gap than I figured.
    What is the tool change time at 16k rpm? That is where Brother is 1.4 seconds chip to chip. Rapids don't mean squat. Accel Decel is what makes a difference. How much floor space and power do you have available? Brother runs on 30 Amp breaker. Can you make more parts with one slower spindle instead of two faster spindles? Do you need a 30(?)hp spindle to make small aluminum parts? Here is a machine like wheelie's machining at 16k, stopping the spindle, checking the tool for breakage, changing the tool and back in the cut at 16k:

    Brother Speedio R650 / Blum Z-Nano broken tool detection - YouTube

  25. Likes wheelieking71, Jared, empwoer liked this post
  26. #38
    Join Date
    Sep 2010
    Location
    Oklahoma City, OK
    Posts
    5,136
    Post Thanks / Like
    Likes (Given)
    941
    Likes (Received)
    2167

    Default

    Just here to say- that seems like a great problem to have. Good luck!

    Maybe Tonytn will stop by, this seems up his alley.

  27. #39
    Join Date
    Jan 2003
    Location
    Canada
    Posts
    11,440
    Post Thanks / Like
    Likes (Given)
    5977
    Likes (Received)
    3581

    Default

    I'd be concerned about the power required to run a horizontal, unless a bigger building and real 3ph is on the shopping list.
    I'd also be extra worried about how fast that work dried up last year. This is a time to be putting in the hours and piling the cash, 7 days a week.
    Work on your game plan for if it holds, but be very careful. Making chips is the easy part.

  28. Likes Bobw, macds, Garwood, TeachMePlease, dstryr and 3 others liked this post
  29. #40
    Join Date
    Apr 2019
    Country
    UNITED STATES
    State/Province
    Arkansas
    Posts
    1,574
    Post Thanks / Like
    Likes (Given)
    750
    Likes (Received)
    614

    Default

    Can you make it in a live tool lathe? We took a high volume part off the mills and into the live tool lathe and let it run 3 shifts with a bar feeder.

    Sent from my Pixel 3 using Tapatalk

  30. Likes Booze Daily 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
  •