Deckel NC Upgrade Cons & Pros - Page 3
Close
Login to Your Account
Page 3 of 12 FirstFirst 12345 ... LastLast
Results 41 to 60 of 239
  1. #41
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    7,994
    Post Thanks / Like
    Likes (Given)
    1778
    Likes (Received)
    2351

    Default

    Quote Originally Posted by wrench View Post
    Servo amplifiers are an area where there have been enormous gains in the last 10 years. Heck, and amplifier really isn't more than the requisite MOSFETs, a CPU, and glue logic now. Microcontrollers are so powerful that they can run the required DSP and probably still play a video game or two (OK, I'm being flippant now). Seriously, though, the servo amplifiers now are a different beast and are 1/10 the size. It is one area that I would not keep the original.
    FWIW.....I have had success integrating modern servo amps to a Dialog control.
    Have had an FP2NC mocked up running using AMC digital servo amps in place of the Bosch units and they worked fine. The units i used do all the logic switching to allow use of the machine as powered or hand operation
    and fault detection. Original servo's were used with no trouble. Amps preformed all the functions of the full Bosch rack at a smaller size. Only part retained form the original system was the transformer to supply
    the 165v 3 phase to the drives......Nice thing about this setup is that with some additional work the problematic servo tach could be replaced with encoders , solving an additional service problem.....
    Nice thing about these amps is that they can be cycled and their acceleration /deceleratioin curves sampled via PC (tuned) with software supplied by AMC.....Also they are a US maker and have real support based
    right here in Ca.....Down side..cost is about $1000 per axis.
    Cheers Ross

  2. #42
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,283
    Post Thanks / Like
    Likes (Given)
    2785
    Likes (Received)
    5589

    Default

    Quote Originally Posted by Miguels244 View Post
    There's a FP3NC carcass on the 'bay minus the controls.
    Pretty tempting to try building the controls up and having a damn nice little NC mill.
    That statement right there tells me you are clueless in this matter. Then again so is the owner of that FP3 parts machine.

  3. #43
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    Quote Originally Posted by wrench View Post
    I think you misunderstood me. I'm not saying those things are not benefits; I'm saying they are complications. Though they are not insurmountable ones, they do add complexity.

    No, LinuxCNC is definitely not a strong front end control. That really is where Mach came into the picture and made a dent. However, Mach really isn't where I personally wanted to go either. And their macro approach is definitely not what I wanted (I don't think I would classify, even with my limited knowledge, the Dialog as being conversational either). It may have power, but people have tried to make macros that they then claim makes the control "conversation." I don't really agree. I have much more experience with the Hurco which I believe is a true implementation of the conversational approach. In Mach's land, you can create a block (ala the Hurco) but I am almost certain that you cannot execute it in that form. Essentially, the block is a one way translator into G-code. A true block or conversational approach should exist through single stepping too, IMHO.

    Servo amplifiers are an area where there have been enormous gains in the last 10 years. Heck, and amplifier really isn't more than the requisite MOSFETs, a CPU, and glue logic now. Microcontrollers are so powerful that they can run the required DSP and probably still play a video game or two (OK, I'm being flippant now). Seriously, though, the servo amplifiers now are a different beast and are 1/10 the size. It is one area that I would not keep the original.

    I would keep the scales but I'm curious how you are interfacing the resolvers. I should check the Mesa site... I vaguely recall they have a board that will work with resolvers. I must admit, though, that I'm a little skeptical of them. I would keep them primarily because they are difficult to replace. But I've also heard of issues with the Deckels wherein the light bulbs burned out, etc, etc. It just seems like binary counters are more "absolute" than some sine waves... But that is just my digital bias. Clearly resolvers are accurate and I can appreciate that. It is reading them that concerns me though. Buying expensive Hiedenhain (sp?) interface modules seems bulky and expensive.
    Ahhhh, see I come from a time from before digital drives, so I don't really have a bias...except when it comes to tuning.
    Bosch drives are resolver compatible standard.

    Again, for my part, I want a strong Man Machine interface...to ME that's the hard part.
    Partly because I'm more interested in the machine. Especially if it's for me, I'm unlikely to fix things I know how to work around.

  4. #44
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Default

    Quote Originally Posted by AlfaGTA View Post
    FWIW.....I have had success integrating modern servo amps to a Dialog control.
    Have had an FP2NC mocked up running using AMC digital servo amps in place of the Bosch units and they worked fine. The units i used do all the logic switching to allow use of the machine as powered or hand operation
    and fault detection. Original servo's were used with no trouble. Amps preformed all the functions of the full Bosch rack at a smaller size. Only part retained form the original system was the transformer to supply
    the 165v 3 phase to the drives......Nice thing about this setup is that with some additional work the problematic servo tach could be replaced with encoders , solving an additional service problem.....
    Nice thing about these amps is that they can be cycled and their acceleration /deceleratioin curves sampled via PC (tuned) with software supplied by AMC.....Also they are a US maker and have real support based
    right here in Ca.....Down side..cost is about $1000 per axis.
    Cheers Ross
    Hi Ross!

    Would you mind...for prosperity... including the exact model of amp's you have working?

    Was there a thread on this endeavor, or is it not to that stage yet?

    Good to hear from ya.
    Sean

  5. #45
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    Quote Originally Posted by Milacron View Post
    That statement right there tells me you are clueless in this matter. Then again so is the owner of that FP3 parts machine.
    As far as the carcass goes, this is certainly true.
    One of the problems is that starting with a mediocre machine and putting the best controls there are on it is still polishing a turd.
    Starting with a great base, it's easier to justify putting the best on it.

  6. #46
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    Quote Originally Posted by AlfaGTA View Post
    FWIW.....I have had success integrating modern servo amps to a Dialog control.
    Have had an FP2NC mocked up running using AMC digital servo amps in place of the Bosch units and they worked fine. The units i used do all the logic switching to allow use of the machine as powered or hand operation
    and fault detection. Original servo's were used with no trouble. Amps preformed all the functions of the full Bosch rack at a smaller size. Only part retained form the original system was the transformer to supply
    the 165v 3 phase to the drives......Nice thing about this setup is that with some additional work the problematic servo tach could be replaced with encoders , solving an additional service problem.....
    Nice thing about these amps is that they can be cycled and their acceleration /deceleratioin curves sampled via PC (tuned) with software supplied by AMC.....Also they are a US maker and have real support based
    right here in Ca.....Down side..cost is about $1000 per axis.
    Cheers Ross
    AMC has always built good stuff, and they never quit building new stuff either.
    Also, 1000 per axis is not unreasonable for a solid solution.

  7. #47
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    7,994
    Post Thanks / Like
    Likes (Given)
    1778
    Likes (Received)
    2351

    Default

    Quote Originally Posted by Miguels244 View Post
    Bosch drives are resolver compatible standard.
    Can you explain this for those of us that aren't ME's......Thought the only thing the Bosch drive saw was
    a direction and value signal from the control coupled with a tachometer input from the servos......

    Cheers Ross

  8. #48
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    7,994
    Post Thanks / Like
    Likes (Given)
    1778
    Likes (Received)
    2351

    Default

    Quote Originally Posted by Sean S View Post
    Hi Ross!

    Would you mind...for prosperity... including the exact model of amp's you have working?

    Was there a thread on this endeavor, or is it not to that stage yet?

    Good to hear from ya.
    Sean
    Sean:
    Saddened to read the accounts of your loss....But encouraged to read your posts...that is the good from all
    this. Sorry we could not have just jumped over the middle to get here..

    The story on the servo setup goes like this:
    Bought the drives new several years ago now, so perhaps there are better solutions out there now.
    Went with the digital drives as they are more programmable , and have logic inputs and outputs that can handle the changes of state between
    on and off to allow the hand wheel operation if you wish.
    Also the digital drives have an RS232 port that you hook to a PC to program..and dynamically cycle to set and tune the accel. and decel. ramps....
    The AMC software has a built in O-scope that monitors the servo action on your PC screen.....No additional hardware is
    really needed to set these up.

    Only bug in the setup is the fact that the Tach output from the servos can run to 25-30 VDC i believe and that is too high for the
    analog feedback input of the drive...They are really expecting you to use an encoder for servo velocity.....
    What i did to overcome this issue was to scale the tachometer input to the drive....

    I setup my test "Mule" on the FP2NC that i bought from you...and it worked fine as far as i could tell. I would love to be able
    to test the real results by running a ball bar test but at the time that was not an option for my test setup.

    That machine has since been sold and returned to the stock Bosch drives. (Rimcanyon is the current owner)
    Long range plan is to fir the final version of the AMC drives to the rebuilt FP3NC that i have been screwing with for way too long.....

    As to the actual part/model number , i will supply this tomorrow..the drives and my info are all at home...could guess, but i would
    rather be accurate...
    Cheers Ross

  9. #49
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    Quote Originally Posted by AlfaGTA View Post
    Can you explain this for those of us that aren't ME's......Thought the only thing the Bosch drive saw was
    a direction and value signal from the control coupled with a tachometer input from the servos......

    Cheers Ross
    Largely it's a matter of selecting a motor drive combination.
    All the Bosch drives can be had with resolver feedback (AFAIK)
    They can also be had with a bunch of other interfaces, Gray scale, quadrature, various busses...
    Without knowing the model number and options it's hard to know what drives you are talking about.

    Most Rexroth motors come with a resolver if you ask for it.
    I like resolvers because they are TOUGH and absolute...and as fast as you can make them.

    Most of those drives, if not all, can accept both a resolver/motor encoder input for commutation and system tuning, AND an external scale for positional control.
    In fact you can 'blend' these two systems...never had to do it, but I assume it's there for 'springy' systems.

    As far as positional commands.
    Commonly a step/direction signal is used for stepper systems...it's a legacy command system, but solid, common, and easy to understand...used a LOT in CNC retrofits.
    Most OTHER systems use a communications bus to control the drive, SERCOS, EtherCAT, or a few others.
    I'll ignore the old analog controls because, blessedly, they are leaving us.

    So, depending on the axis drive motor, you choose the encoder interface to match, then, depending on the controls you choose the command interface.
    If I knew more about what the existing system was I could probably tell you more about what's happening in that case...

  10. #50
    Join Date
    Feb 2004
    Location
    Napa, CA
    Posts
    2,589
    Post Thanks / Like
    Likes (Given)
    43
    Likes (Received)
    690

    Default

    The problem of adapting the Deckel shifting to a new control is overblown in my opinion. If this is what stops you, you should never have begun.

    I currently have my FP3NC running on an AMC drive for the X axis, in the process of wiring up the AMC drives for the others. I used their analog drive B30A40AC. The digital one would be nice in some ways but way overkill for this job. This analog drive is current product for AMC, I bought 3 of them on eBay out of Singapore for about $800 total if I recall. While they are analog, they are far more advanced than the original Bosch. The reason I picked these is they can deal with the relatively high supply voltage (165VAC), and the high tach voltage (which I have seen up to about 40v). They could be set up with encoders for velocity feedback, but that would be kind of a big job to retrofit the encoders. They operate at a higher frequency and don't "sing" like the Bosch, are generally better behaved and more tunable. The axis sounds somewhat different when moving compared to the Bosch. The one downside of them is the control voltage for Inhibit and Fault needed to emulate the Bosch functionality is 5V, so level shifters are required to get to/from 24V. You need 6 of them, they are about $1 each.

    On my machine they are being driven by the retrofit Heidenhain 410 control, but it would work equally well with the Dialog. I mounted them to a piece of G10, replacing the Bosch chassis including its power supply (each AMC drive has its own).

  11. #51
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    7,994
    Post Thanks / Like
    Likes (Given)
    1778
    Likes (Received)
    2351

    Default

    What we have been discussing here (at me) is the specific hardware form the 8o's Deckel CNC machines....

    Servos are Seimens 250v DC with tachometer output, internal brake ..no resolvers or encoders.
    Series numbers for servos 1 HU3073.......

    Most folke here i would wager do not wish to change the servos as the factory install is neat and reliable.
    Further the change of the servo would cause issues fitting the hand wheel for the "Z" as the servo frame is used to support the right angle drive
    used for hand operation...not the deal breaker, but retention of the existing servos would simplify some of the mechanical mods needed.

    Not sure how better to define the Bosch servo drives but by part number for the cards.....My guess is that this is a Deckel number:

    Entire unit .....6005 04 512001 02
    The chassis consists of 10 slots to provide for having a full 4 axis of servo control (8 slots) with power supply (2 slots)..
    there is an amplifier and regulator card for each axis...along with two slots that cover the power supply for the servos.
    The power supply converts the 165VAC three phase to 240VDC for the servo use.

    Cheers Ross

  12. #52
    Join Date
    Mar 2011
    Country
    UNITED STATES
    State/Province
    Montana
    Posts
    704
    Post Thanks / Like
    Likes (Given)
    27
    Likes (Received)
    153

    Default

    Anyone know what the servo configuration is for the FP5NC? Same Seimens with tachometer out?

  13. #53
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Default

    Quote Originally Posted by AlfaGTA View Post

    Most folke here i would wager do not wish to change the servos as the factory install is neat and reliable.
    Further the change of the servo would cause issues fitting the hand wheel for the "Z" as the servo frame is used to support the right angle drive
    used for hand operation...not the deal breaker, but retention of the existing servos would simplify some of the mechanical mods needed.

    Cheers Ross
    I can't speak for most folks, but you are correctly speaking for me with that statement.

  14. #54
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    Quote Originally Posted by AlfaGTA View Post
    What we have been discussing here (at me) is the specific hardware form the 8o's Deckel CNC machines....

    Servos are Seimens 250v DC with tachometer output, internal brake ..no resolvers or encoders.
    Series numbers for servos 1 HU3073.......

    Most folke here i would wager do not wish to change the servos as the factory install is neat and reliable.
    Further the change of the servo would cause issues fitting the hand wheel for the "Z" as the servo frame is used to support the right angle drive
    used for hand operation...not the deal breaker, but retention of the existing servos would simplify some of the mechanical mods needed.

    Not sure how better to define the Bosch servo drives but by part number for the cards.....My guess is that this is a Deckel number:

    Entire unit .....6005 04 512001 02
    The chassis consists of 10 slots to provide for having a full 4 axis of servo control (8 slots) with power supply (2 slots)..
    there is an amplifier and regulator card for each axis...along with two slots that cover the power supply for the servos.
    The power supply converts the 165VAC three phase to 240VDC for the servo use.

    Cheers Ross
    So, it looks like an OLD Rexroth drive, DC (brushed?) motors, tacho feedback.
    That's not a big deal to work around either.
    Bosch no longer makes drives for that...but may very well still support them if you push up the chain a bit.
    As mentioned AMC builds good ones, Anaheim and Moog both do as well.

    So the 'head' control handles position and the drive handles the velocity feedback loop.

    Bosch's NYCE 4000 might be a good package to look at...no shortage of ways to skin the cat.

  15. #55
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Default

    Let’s take a break in this thread and talk about me for a minute ;-)

    Here’s is who I am….


    I love Deckels.
    I have owned both manual and NC Deckels for over 15 years now.
    I’m a so-so machinist.
    I have a pretty good mechanical skill set, and a basic electrical one. I can solder and crimp.
    I can read a VOM, and I can read an oscilloscope but not be able to tell you what it means.
    In high school, I learned to program in various languages (basic, C..etc)
    I can do a little CAD/CAM but I’m no expert. I can make parts, but sometimes it takes a few tries.
    I usually seek out the best combinations of things one can own or create….and I OCD on it (to my detriment).


    I am telling you this because…


    I am the Steve Jobs of Deckel retrofitting...


    Understanding the difference between AC servos and DC servos is not in my skill set.
    I don’t know how a resolver works.
    I only basically understand how a closed loop works
    I don’t know the difference between an analog drive and a digital drive
    I wouldn’t know how to tune a drive, but I can follow instructions.
    I am dazzled by shiny lights, big displays, and features that help me overcome my deficiencies.
    There isn’t enough time in my life to learn everything needed to retrofit a machine from scratch.
    My understanding of how a machine control works is that data is converted to signals that go to boxes, which send electricity to motors and switches.
    etc etc.


    Still, for 15 years since my first FP2NC, I’ve had this dream….


    The dream is owning a *DECKEL* CNC which could be brought into and maintained in the modern age.


    This machine would not have *any* shortcomings vs the original machine. It would do everything the original did, but include a ton of the fancy abilities that year 2000-2015 controls offer.


    It would be able to be upgraded, adapted, and customized. plus it would be easier to service.


    Combining this concept with what I consider is the best physical machine system ever devised follows right down my OCD pursuit of “The best”.


    Now… I fully understand and have often discussed the virtues of the Dialog. When Ross talks about the control, I fully get what he means when we’ve talked about those big ol’ round buttons where one can “braille” their way around getting the machine set up, and the rotary dial where you don’t have to look to know that you’re going from mode 1 to mode 3 and how that speeds up your setup. The Dialog almost more physically an extension of your hand…like operating a typewriter or something.
    “Click Click Push Push”… you do this for awhile and you really begin to understand that you would miss moving the machine around in this way….especially on a machine which is used “semi-manually” at times.
    Yep… click push click click push and hit “start”. The Dialog was designed by those who really understood what makes Deckels unique in a job shop / prototyping / repairs environment.


    I have never used a control that I liked better for that (although I haven’t tried any new model “interact” style controls).


    The retrofit implementation I’ve always fantasized about is one where the lower half of the Dialog (4) pendant is retained, while the upper half is replaced by a touchscreen which could “identically* emulate the top half of D4 that many of us are used to.
    The machine… on this default screen would function *exactly* like the original if desired. Yes, the buttons would be touchscreen up there, but unlike the lower half, you are usually looking at these versus “feeling” for them.
    The rotary “mode” switch would continua to work…etc, but the upgraded part would be that each mode could have additional information or even alternate windows (via onscreen tabs).


    For instance, the “G-code” mode which originally mostly just showed G-code might have a tab for 3D toolpaths…etc.
    The additional windows would have enhanced features over the original.
    Scrolling through G-code could be done via on screen scroll bars, and modifications cut/pasted, or typed with a real (under pendant) keyboard…etc.


    The manual mode of the machine might have enhancements such as additional, updated, and customizable canned cycles vs the original.


    Because of the advancements in CPU’s, this machine would run faster, smoother, and with less restrictions.


    Now that you are envisioning this hardware/touchscreen hybrid of a Dialog emulator, understand that this whole thing could be dumped and something else loaded… like a HH’s 415. Any controls lacking on the lower pendent for that control could be implemented via touchscreen but the idea is to be able to load “virtual controls” as long as someone was willing to take the time to write one.
    With a little graphics work, you could even make them look just like the originals (onscreen). There would be a bias towards the Dialog because not only is it the original control format, but you would also be retaining the lower hardware panel. I don’t see anything wrong with that bias, but other virtual controls could be made useable with some of the same buttons, or onscreen.


    Many many years ago when I started daydreaming about this, Ross, a person who I’ve always thought used both the Deckel and it’s Dialog to the maximum was resistant….and I think he is right!
    For that, my dream conversion was one where Ross could walk up to it and mount that unobtainium automotive transmission project on it, and do the job just as normal with little to no learning curve…because it is a true virtual Dialog.


    So how does one do this?


    As Steve Deckel Jobs, I have determined that the only way possible is to use a PC front end and a touch screen.
    My only experience at the time (and it hasn’t improved much) is that I had retrofitted a BP using Gecko drives and such (remember those?).


    Mach was coming online about that time…. and it did something that I didn’t know of any other software that could do… design your own screens.


    I set out right away designing screens…. I didn’t even have a mill I could use them on. I just started making to the top half of the D4. It didn’t “do* anything… it was just pretty.


    It doesn’t take long before you start to want making these buttons do something….to start examining the flow chart of how the original Dialog works and trying to figure out if you could write macros, or whatever it is to emulate the function.


    “Oh, you can’t get a Deckel to do *this* and the “kernel” that…” Ugh. Ok, so I’m talking to CNC router guys who’s tables are made of *wood*, then running back to the Deckel board who are on a whole different mindset.


    Years of this and a couple retrofits are coming on youtube where they hacked some hobby servo on the FP and I’m starting to freak out going wondering why modern technology can’t interface with something that seems so electronically historic to me that it might as well have *tubes* in it.


    A million lines of text on the subject, and here and there I’d see “hints” on youtube that it had already been done…. by a guy in Paraguay…who can’t be reached, in broken english, with no parts list or mention of if it’s the original servos, or if the spindle motor had been replaced by a Black and Decker cordless drill motor…UGH!


    Jump forward a couple more years and generic controls are being retrofitted more regularly….no real feedback because that machine is making blender parts in a blender factory somewhere and they just needed it to start cycle drilling again. Thank you for uploading that video tho’.


    Skip forward… ooooh… the Germans are getting involved. If anyone on earth is going to “get it”, it will be the Germans… nope…
    Same damned LinuxCNC DRO looking generic “run my home-made router” looking control, and… to figure if this can be adapted to Mach, or if there’s any way to write a Dialog GUI/layers for it, it’s all in GERMAN and I haven’t gotten anywhere in 10 years in ENGLISH!


    The most frustrating part is my own ignorance… an ignorance that either you all just assume I don’t have, or one that I’ve hidden pretty well.
    15 years of "resolver based, DC amplifier, 10v, 5v, pulse generator..." Chinese.


    All I want is…


    A machine being fronted by a PC that is “proper”. If that involved buying parts…I’ll save up. I’m not trying to cheat it with hobby parts and MacGyvering. Industrial parts.


    This parts will take either signals, raw data, or commands… I really don’t care. The idea is that if I write a piece of code that says “Move X”, the axis moves. If I write “Speed - 500” then the spindle goes 500rpm. If that has to be Relay 1=on, Relay 2 = off, Spindle relay = on…. that ok.


    I am over simplifying this for the sake of concept, and that concept is that a PC is hooked to whatever converter, translator, or industrial control that the machine is correct with and can be commanded from s PC.


    The next step is that the software allows the insertion of subroutines so that one might write code to emulate the functions of the Dialog or whichever other alternate control, and finally that a GUI can be customized to activate those routines.


    Honestly guys, if this had been available 15 years ago, I would have spent the last decade learning how to code it…until it was right, but instead I still can’t even figure out the parts I’d need to get the machine to work *at all*.


    This cannot be this hard.


    Ok then…. I think it’s time to stop talking about me now.

    Edit: Dialog 11 emulator might be fine too.... never had one.





  16. #56
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    I am a huge fan of a tactile interface.
    Even scrolling from tab to tab should be by knobs that fall to hand, not eraser tips.
    Jogs should be handles, not clicks.
    Buttons should be big enough that putting your hand on the controls feels like something.

    What you want, a machine that can be moved by a PC is not hard.
    Nor even particularly complicated.
    Doing it in a serviceable, robust, way with manuals is time consuming and expensive, although in many ways far easier than trying to make do.

    What weaknesses of the original were there?
    Are you just hunting a solid motion control to replace the older system you can't fix any more?

    Either way, the hard part from MY perspective is finding a worthy donor.
    As I said, the nice thing about working on a beautiful piece of machinery is that putting money into it to get it right seems right.
    Putting an $8,000 set of controls on a Chicom BP makes no sense.

  17. #57
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Default

    Quote Originally Posted by Miguels244 View Post
    I am a huge fan of a tactile interface.
    Even scrolling from tab to tab should be by knobs that fall to hand, not eraser tips.
    Jogs should be handles, not clicks.
    Buttons should be big enough that putting your hand on the controls feels like something.

    What you want, a machine that can be moved by a PC is not hard.
    Nor even particularly complicated.
    Doing it in a serviceable, robust, way with manuals is time consuming and expensive, although in many ways far easier than trying to make do.

    What weaknesses of the original were there?
    Are you just hunting a solid motion control to replace the older system you can't fix any more?

    Either way, the hard part from MY perspective is finding a worthy donor.
    As I said, the nice thing about working on a beautiful piece of machinery is that putting money into it to get it right seems right.
    Putting an $8,000 set of controls on a Chicom BP makes no sense.
    I agree about the tactile-ness being key for this type of control and machine. That is my issue with the generic control offerings...even good ones.

    That is the reason that my fantasy retrofit, even if it ends up being a lifetime challenge, will include the lower half of the Dialog's "clamshell". The upper half is no more than screen and a keypad, which I feel could be tolerated as a touchscreen in "classic" mode, and also lends itself to "enhanced mode" features that don't currently exist on the Dialog.
    Put another way, I would like to see the original functionality retained and add "virtual" capabilities to that.

    To answer your question, the weakness of the original was everything it could do was embedded in 1980's technology and electronics, and lacked modern refinements and features.
    Sure, it's small things. Take the little green/black CRT. When I use the Dialog, sometimes I think the 0 is an 8, or the 5 is a 6, because it's...well...a small CRT and I have to squint.
    Program lines were limited, the posts were proprietary and lacking, the machine didn't have enough look ahead and could be jerky on some contours, and it lacks some more modern cutting strategies...etc. The electronics are now 35 or so years old, and sometimes hard to replace...if not expensive...etc.
    Standard stuff on machines of this era really except for a couple things...
    It's the best *iron* ever and the most thought through and extensively adaptable machine line ever made. It's the 1980's man... money was everywhere, and they went nuts on the FP line of machines and accessories expanding and improving on a system that already seen a 40 year true-to-form evolution.. I personally consider the degree of evolution of this line to be historic and monumental.
    That way of thinking didn't survive into the "manufacture overseas", disposable equipment, nobody cares about the difference post 2000 world. The manufacturing ideology became "I can buy 3-4 of these Chinese mills and just keep replacing them for far less than a Deckel cost" People weren't fixing things any more, they were just tossing them and buying new ones, so the ultra-Swiss army knife aspect of the FP line for repair became less attractive. CNC was also becoming the forefront of the way things were prototyped so the manual prototyping aspect of these expensive machines became less attractive.
    The NC versions of the the Deckel FP series were basically designed as "NC assisted manual mills" right through the 80's. Simply, they added it as a feature to what I'm sure in their minds (and few would argue), was a milling system that had already been expanded and improved in every way they could possibly imagine.
    The Dialog was very obviously designed around this concept... that it's a feature on a manual machine, and it works well like this.
    The fact that many, if not most of these NC Deckels still had physical hand wheels is pretty clear evidence that the NC aspect was just adding some useful automation to the manual line. If that isn't enough... even heads from the manual machines would fit and work on the NC capable machines. They didn't drop the hand wheels until the physical size of machine had become so large that human power wasn't enough to move it (Note: Deckels came in different sizes).

    So yeah... more or less the market was going a different direction, and the way people made things overall was following that direction too.
    The problem is that it left some people out. That includes people who *can't* just go buy a new part for whatever they work on (exotic car fixers and such), people who have to work on something different constantly (prototypers, inventors...etc), old school craftsman types and model engine builders...etc.
    The money was in the VMC's, screw the handwheels and one-offs. Here's your VMC and if you want to drill a hole, go buy a Chinese Bridgeport to put next to it. If that's a problem, design around it.

    The hyper quality, hyper expandable, hyper adaptable, and hyper expensive milling system no longer had a large enough market to sustain it's production.

    The people who appreciated these machines didn't cease to exist though and neither did the uses for the machines. It just wasn't enough though to continue the line.

    That has left a lot of people working with 80's controls since no replacement to these machines exists, and that control in my opinion should compliment what is special about the machines and how people work with them vs more generic (or should I say more modern) solutions.

  18. #58
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    Quote Originally Posted by Sean S View Post
    I agree about the tactile-ness being key for this type of control and machine. That is my issue with the generic control offerings...even good ones.

    That is the reason that my fantasy retrofit, even if it ends up being a lifetime challenge, will include the lower half of the Dialog's "clamshell". The upper half is no more than screen and a keypad, which I feel could be tolerated as a touchscreen in "classic" mode, and also lends itself to "enhanced mode" features that don't currently exist on the Dialog.
    Put another way, I would like to see the original functionality retained and add "virtual" capabilities to that.

    To answer your question, the weakness of the original was everything it could do was embedded in 1980's technology and electronics, and lacked modern refinements and features.
    Sure, it's small things. Take the little green/black CRT. When I use the Dialog, sometimes I think the 0 is an 8, or the 5 is a 6, because it's...well...a small CRT and I have to squint.
    Program lines were limited, the posts were proprietary and lacking, the machine didn't have enough look ahead and could be jerky on some contours, and it lacks some more modern cutting strategies...etc. The electronics are now 35 or so years old, and sometimes hard to replace...if not expensive...etc.
    Standard stuff on machines of this era really except for a couple things...
    It's the best *iron* ever and the most thought through and extensively adaptable machine line ever made. It's the 1980's man... money was everywhere, and they went nuts on the FP line of machines and accessories expanding and improving on a system that already seen a 40 year true-to-form evolution.. I personally consider the degree of evolution of this line to be historic and monumental.
    That way of thinking didn't survive into the "manufacture overseas", disposable equipment, nobody cares about the difference post 2000 world. The manufacturing ideology became "I can buy 3-4 of these Chinese mills and just keep replacing them for far less than a Deckel cost" People weren't fixing things any more, they were just tossing them and buying new ones, so the ultra-Swiss army knife aspect of the FP line for repair became less attractive. CNC was also becoming the forefront of the way things were prototyped so the manual prototyping aspect of these expensive machines became less attractive.
    The NC versions of the the Deckel FP series were basically designed as "NC assisted manual mills" right through the 80's. Simply, they added it as a feature to what I'm sure in their minds (and few would argue), was a milling system that had already been expanded and improved in every way they could possibly imagine.
    The Dialog was very obviously designed around this concept... that it's a feature on a manual machine, and it works well like this.
    The fact that many, if not most of these NC Deckels still had physical hand wheels is pretty clear evidence that the NC aspect was just adding some useful automation to the manual line. If that isn't enough... even heads from the manual machines would fit and work on the NC capable machines. They didn't drop the hand wheels until the physical size of machine had become so large that human power wasn't enough to move it (Note: Deckels came in different sizes).

    So yeah... more or less the market was going a different direction, and the way people made things overall was following that direction too.
    The problem is that it left some people out. That includes people who *can't* just go buy a new part for whatever they work on (exotic car fixers and such), people who have to work on something different constantly (prototypers, inventors...etc), old school craftsman types and model engine builders...etc.
    The money was in the VMC's, screw the handwheels and one-offs. Here's your VMC and if you want to drill a hole, go buy a Chinese Bridgeport to put next to it. If that's a problem, design around it.

    The hyper quality, hyper expandable, hyper adaptable, and hyper expensive milling system no longer had a large enough market to sustain it's production.

    The people who appreciated these machines didn't cease to exist though and neither did the uses for the machines. It just wasn't enough though to continue the line.

    That has left a lot of people working with 80's controls since no replacement to these machines exists, and that control in my opinion should compliment what is special about the machines and how people work with them vs more generic (or should I say more modern) solutions.
    Well, get to work.
    Sketch out what you want for an interface, the balance of machine operation is a preskinned cat.
    Just get a good cat skinner to do it if you don't want to or can't handle the hardware end.

    There's no shortage of really good motion control platforms that will gladly accept input, even in G-Code, from a PC.

    FWIW integrating code wheels and buttons is not a particularly big deal so don't worry about how to do it, just select the buttons you like.

    Here's the question.
    Are there enough people out there to make a kit worth while?
    Or are you building a few for you and some friends.

    The other half of the equation is...What would you pay for a kit?

    If it's a labor of love then that question doesn't really apply, find someone that shares your interests and has the rest of the skill set and partner up.

  19. #59
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    7,994
    Post Thanks / Like
    Likes (Given)
    1778
    Likes (Received)
    2351

    Default

    So here is the information on the digital servo drive i used:

    Classed by AMC as "Digiflex Degital Servo Drive"

    type number is: DPRANIE- 030A400A

    This is a drive that accepts from 100-240 VAC three phase input.
    Fully self contained , has its own power supply.
    Output voltage at 127-373 VDC out....
    Has a peak current output of 30 amps with 15 amp output cont.
    Think this is large enough for all but the "Z" servos of an FP4NC....

    Jon in a posting above gave his input about using the analog units made by AMC and his points are
    very valid and of course his cost was significantly lower....

    I went with the digital unit simply because i believed that they would be easier for me to set up and get the best performance given my limited electronics knowledge.

    Digital drive is programmable and has an input where you enter the physical and electrical make up of the servos themselves.....(taken from a Siemens tech servo data sheet)
    Further the software for tuning with its purpose built "O" scope and the ability to "cycle" the servo (accel/decel) using a PC
    helped me get things to run well and make accurate contours.....the point after all.

    Here are some photos of the unit ...Pretty sure the analog item is similar is size....

    If you want more info here is a link to the site:
    DPR series, DigiFlex Performance RS485/232 servo drives













    Cheers Ross

  20. #60
    Join Date
    Mar 2011
    Location
    Denver, CO USA
    Posts
    10,215
    Post Thanks / Like
    Likes (Given)
    58
    Likes (Received)
    5695

    Default

    AMC builds good stuff.
    I have never used those particular drives but they look pretty nice.


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
  •