Another bridgeport retrofit, need advise


Page 1 of 7 1234 ... LastLast
Results 1 to 20 of 138

Thread: Another bridgeport retrofit, need advise

  1. #1
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Another bridgeport retrofit, need advise

    Ive been researching for what seems like months before taking the plunge and buying stuff. Everytime I think im all set I find some other quirk or that something isnt going to work the way I expected. Orig i was kinda set on the DMM servos, now after crunching the numbers on cost and speed/resolution, i am kinda leaning tword steppers. I think rapids over 200ipm would probably destroy the dovetail ways if they are adjusted correctly? The 5k DMM servos gave me something like 350ipm with the biggest gear reduction that would fit under the top surface of the table. Onto my issues:

    Since this is still a manual machine and I do a lot of manual work on it, I obviously want to retain that capability. The cost of a quality DRO and power feeds is what led me to the conversion instead. So with the stepper drives on "disable" in Mach, my research shows that the DRO will not keep track of its position? Would be the same if I went with servos correct? Or is there a way to keep power to the encoders still with the drive disabled?

    I initially planned on getting the ether-mach motion controller, but am really leaning tword the HICON integra now. I could hook up glass scales and of coarse paying the xtra fees, have the DRO read the actual table position all the time. With the added benefits of error correction and a few other things.

    Next area of confusion is if I keep the plan of stepper motors (nema 34 around 1000oz/in) it seems most are using 1:1 on the drive pulleys and using microstepping for better resolution. Too much misinformation on the internet for me to make a certain decision on whether its better to do that, or use a reduction and deal with the slower rapid speed. I am not a production shop, just a hobbyist. So bragging rights on super fast rapids mean absolutely nothing to me. Is 100ipm really that slow? This is getting lengthy so I will cut it here

    Similar Threads:


  2. #2
    Member
    Join Date
    Aug 2016
    Location
    United States
    Posts
    14
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    As far as controllers go, I believe Centroid Acorn will still display DRO whether you have servos or steppers. I plan to retrofit my BP Series 1 using DMM servos as well as the Centroid controller. First I need to purchase and install Hiwinn ball screws then go from there. Do a google search for "Centroid Acorn". Centroid has a good page explaining Acorn's capabilities.


    Sent from my iPhone using Tapatalk



  3. #3
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by sohcpwr View Post
    Ive been researching for what seems like months before taking the plunge and buying stuff. Everytime I think im all set I find some other quirk or that something isnt going to work the way I expected. Orig i was kinda set on the DMM servos, now after crunching the numbers on cost and speed/resolution, i am kinda leaning tword steppers. I think rapids over 200ipm would probably destroy the dovetail ways if they are adjusted correctly? The 5k DMM servos gave me something like 350ipm with the biggest gear reduction that would fit under the top surface of the table. Onto my issues:
    Brushed or Brushless DC motors would be my first choice, but steppers can be very fast and very accurate also. It just depends on how you configure the system. The brushed DC motors are my preference for manual/CNC operation, no cogging or turning resistance when disabled.

    Since this is still a manual machine and I do a lot of manual work on it, I obviously want to retain that capability. The cost of a quality DRO and power feeds is what led me to the conversion instead. So with the stepper drives on "disable" in Mach, my research shows that the DRO will not keep track of its position? Would be the same if I went with servos correct? Or is there a way to keep power to the encoders still with the drive disabled?
    I highly recommend keeping manual capability. It is also possible and practical to have a quick change manual/CNC quill. With the work I do, I am back & forth between manual and 2/3/4 axis CNC all the time, with a changeover done by a flip of a lever and a button push.

    Mach3 does not read stepper position, the DRO is where Mach3 ''thinks'' the table should be. It has no idea of where the table is actually at. This can be somewhat corrected with a motion controlled that actually uses encoder feedback, and then is attached to Mach3. In that case, the Mach3 DRO will output the table position, but internally in the software it still has no idea of where things actually are.

    I initially planned on getting the ether-mach motion controller, but am really leaning tword the HICON integra now. I could hook up glass scales and of coarse paying the xtra fees, have the DRO read the actual table position all the time. With the added benefits of error correction and a few other things.
    Also take a look at Galil Motion Control products, many models are compatible with Mach3, a bit pricy to buy new, but there are normally a lot of them on Ebay at prices from cheap to crazy. They will run open/closed loop steppers or servos very well. And later will allow you to upgrade from Mach3, the Galil products are compatible with a number of CNC programs.

    I highly recommend putting scales on the load (table, quill, and knee). Also look at magnetic scales, about $120/axis for 1 micron resolution, hooks right into your motion controller. I've installed them on 4 of my machines. As you say, the advantage is on-the-fly error correction, some backlash compensation (this is a whole subject in itself), and leadscrew error correction.

    Next area of confusion is if I keep the plan of stepper motors (nema 34 around 1000oz/in) it seems most are using 1:1 on the drive pulleys and using microstepping for better resolution. Too much misinformation on the internet for me to make a certain decision on whether its better to do that, or use a reduction and deal with the slower rapid speed. I am not a production shop, just a hobbyist. So bragging rights on super fast rapids mean absolutely nothing to me. Is 100ipm really that slow? This is getting lengthy so I will cut it here
    100 IPM is all I run on my mill, more than fast enough for me,even on a 10x50 table. On my 4x8 router, I'm running 600 IPM rapids at about 760 RPM with 1200 oz/in, NEMA 34 steppers on all axes, and that includes a 700 lb moving table. On a BP size mill I suspect that 1200 oz/in steppers have enough torque to run 1:1. But I can not confirm that because I have never done it on the X/Y, but I do have a 1200 oz/in on my Z, and I reduced the power by 50% because it had too much torque.

    I also recommend that you buy double nut ball screws, they can be adjusted to zero backlash.



  4. #4
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    The hicon has provisions for an xtra encoder board where glass or magnetic scales can be hooked up for true closed loop operation. It is a pricy option but is the main reason I'm strongly considering that board/controller. Most likely as a future upgrade as it is $500 for the software and $300 for the board.

    I'm still kinda fuzzy about the Mach DRO. I do understand that it's only reading the position of the motor, which won't show any inaccuracies in screw pitch, backlash, or lost steps. What I'm curious about is if I want to use the machine manually, I will have to click "disable" on the controller so the steppers or possible servos stop holding their position. After they are disabled, will the DRO still read the position based on the motor? Basically similar to a proto-trak if you have ever used one... the display will read servos off but the DRO will read when turning the handles manually.



  5. #5
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    At $800 for Hicon software and a daughter board, plus the cost of the initial card at $600, you could buy a used top line 4 axis Galil board and the cables, and probably have money left over. And you would have a real high speed industrial motion controller.

    The only experience I have with Mach3 and encoder feedback is brief time that I had it connected to a Galil controller. The Mach3 DRO was reading the encoder position from the Galil controller, motors enabled or not. But from what I understand, even with that feedback, Mach3 itself is not internally aware of the encoder position only the DRO function is looking at the encoder position. In other words, it is only passing that information through to the operator.

    I do not know if any other motion controller passes the encoder position back to the DRO screen, no experience with any other controller.

    Without encoder feedback, the Mach3 DRO would not work at all, it will not read stepper position. In normal stepper mode, Mach3 just displays where it is telling the stepper to go without regard to where it actually is. You could cut power to your drives, and Mach3 would still happily run your part and display the position on the DRO even though nothing is actually moving.

    I hope this clears things up a bit.



  6. #6
    Registered SoCalPlaneDoc's Avatar
    Join Date
    Mar 2012
    Location
    USA!
    Posts
    117
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by Jim Dawson View Post
    Also take a look at Galil Motion Control products, many models are compatible with Mach3, .
    Galil is awesome BUT, the plugin for Mach3 is buggy and devoid of many features.. I'd advise against using it with Mach3 - I'm sorry I used it for my CNC router.
    The machine will stop without any idea as to why and only a reboot will get you out of the failure mode. and the auto-Z doesnt work at all.
    A high-volume poster here who was supposedly an expert told me it was a driver problem and couldn't offer an alternative... quantity over quality I guess

    My new router is running PathPilot perfectly.. Never again will I use Mach.

    Much of my Galil stuff is destined for eBay.

    Automated Machinery Designer - PCNC1100 Series III upgrade, Graziano Sag 12 Lathe
    Solidworks 2016 (SW Certified), HSMWorks


  7. #7
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by SoCalPlaneDoc View Post

    Much of my Galil stuff is destined for eBay.
    Let me know when you're ready to sell, I might be interested. BTW, Galil products work perfectly with my CNC software (DC_CNC), it was written from the ground up specifically to use Galil controllers.



  8. #8
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    I'm pretty set on using Mach 4. Don't recall reading anything negative about the plugin for the hicon. The ether Mach (smoothstepper) is also still an option at this point.

    Here's another crazy idea I had: my next project will be a cnc lathe. Is it possible to use 1 control cabinet to run both machines? I'm thinking if I bought the same axis motors I could unplug from the enclosure when swapping from machine to machine. Use the same drives, pwr supply, and board. The ports and pin settings in the 2 different softwares should allow me to leave them configured and only have to plug and unplug motors and limits? Maybe some xtra difficulty with the spindle but are there reasons other than slight inconvenience that would make this impossible?

    Last edited by sohcpwr; 05-21-2017 at 09:23 PM.


  9. #9
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    I should not have assumed you were talking about Mach3 when you said ''mach'', I know nothing about Mach4.

    Two machines to the same control? I would say it's possible.



  10. #10
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Without encoder feedback, the Mach3 DRO would not work at all, it will not read stepper position. In normal stepper mode, Mach3 just displays where it is telling the stepper to go without regard to where it actually is. You could cut power to your drives, and Mach3 would still happily run your part and display the position on the DRO even though nothing is actually moving.
    Thanks very much for your reply's! Well its a damn good idea I posted here before getting all giddy and buying everything. Retaining a working DRO when not under cnc control is an absolute must. So it sounds like there are 2 ways of accomplishing this. One previously mentioned, the expensive addition of scales on the axis and a controller that will transfer their signals into mach. The second from what I understand, is using either a servo or a dual shaft stepper with an encoder. I'm still confused though on how to set this up for mach to: read the encoder as a DRO. Jim, it sounds like your saying the motion controller also needs to receive the signals from the encoders. They obviously still have to go to the drive too though right? Im guessing something like this would be necessary: https://cnc4pc.com/motion-control/mo...-encoders.html

    I remembered seeing that before and wondering what it was for. I think its starting to make sense now, your encoder input goes into a board like that and it looks like it has an output for the signal to go back to the drive. That board allows the software to use the encoder without the drive's assistance. I wonder if one could connect a linear encoder to that board (glass scales) and use steppers or servos with their encoder run normally.



  11. #11
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Again, I don't know anything about how Mach4 works. Mach3 works as I described above.

    Having said that, I'll try to clarify the different types of systems.

    Lets first define a servo. It is any device that receives feedback from another device and acts on it. For instance, your home heating system consists of a heater of some kind (the controlled device) and a thermostat (the controller). This is a simple servo system. You car is another servo controlled device, where you are the controller and you take the input from what is going on around you, process that data, and then feed the appropriate signals to the car (gas, brake, steering) So you become part of the servo system when driving. A servo by definition is a closed loop system

    Servo does not refer to a particular type of motor, although, the waters have been muddied by the marketing idiots, and anyone without experience specing systems would be confused by the hype.

    So here are the system types

    Stepper motor, implies an open loop system. The controller outputs a stream of pulses to a stepper drive which in turn makes the stepper motor turn. It is fire and forget, the controller has no idea what the motor is doing.

    Closed loop steppers, Implies a stepper that includes an encoder of some type. So the encoder is telling the controller what the stepper position is, and hopefully the controller will correct position errors.

    Hybrid steppers, most times same as Closed loop steppers, but the drive has an encoder input and that is compared to the pulse input and the drive makes the correction rather than the controller.

    Hybrid Servos, many times same as above, but may be a 3 phase BLDC motor rather than a stepper. This is where you really have to read the specs. Sometimes with built in drives, or could have external drives. Normally take a step & direction input.

    There are a few other derivatives of the above, but I covered the main ones until you get into the higher end industrial servos, then things change a bit

    What we normally refer to as a Servo would include an AC or DC servo motor (purpose designed to be a servo motor), a drive, a feedback device (encoder), and a motion controller. Notice this is the first time I used the term Motion Controller.

    A motion controller does the trajectory planning for the moves, if it doesn't do that then it's not a motion controller but rather a breakout board.

    Mach3 is a motion controller, it does it's own trajectory planning and sends appropriate signals to the drives. But Mach3 is only an Open Loop controller, even when running a Galil motion controller. However it possible to close the loop with the Galil controller to at least make the Mach3/Galil system work like a hybrid stepper. You would have to have encoder feedback directly to the Galil controller for this to work. Again, I don't know about Mach4.

    If you were to use a motor with a drive that accepted/required encoder input, then you would have to parallel the encoder to the motion controller to be able to have the DRO when the drive was disabled. BUT....I don't know if this would be possible in actual practice, because the drive would be supplying the power to the encoder. With some wiring magic, it might be possible to get around this limitation.

    My idea of the ideal system is to have the encoder (linear scale) on the load (table), a servo motor and drive (amplifier), and a motion controller. The encoder would feed the position data to the motion controller, and the motion controller would take care of making sure the motor arrives at where it is supposed to go, and at the proper speeds and accelerations. The motion controller would receive it's position data from the CNC program. So the CNC program says to the motion controller move X,Y,Z to some location and here is the speed, Go. Then the motion controller figures out the best way to accomplish that task. This is the way all of my machines work.

    I don't know if I answered your questions or not, I'll get off of my soapbox now



  12. #12
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    First of all, thank you very much for taking the time to go into such detail. When I referred to "closed loop" operation, I was describing the system similar to what you mention in the last paragraph with rotary encoders on the motors, and linear ones on the machine axis. I dont know a ton about how different controllers use the tandem signals but I believe the servos encoder is used for the "meat and potatoes" and the scales are for periodic checks of accuracy. Somewhere I think I read that the scales can not transmit position data fast enough to be used for motion control. The controller will fault if actual position read from the scale does not match where the software thinks it should be from the servo. This is what you are paying for with the high dollar option on the hicon board. Except instead of the controller faulting, it corrects the position for the next move.

    I just read the manual for the board in the link I posted. Looks like its supposed to receive the encoder signal first, then there is an output for it to go back to the drive. The board also requires 5v power, so I think it powers the encoder when the drive is disabled. Here is a crude pic of the schematic from the documentation:

    Another bridgeport retrofit, need advise-encoder-board-jpg

    That board is for the smoothstepper. So it looks like I have to choose between the hicon or some other high end motion controller that will accept encoder input or the smoothstepper with this daughter board. Then decide if steppers and scales is for me, servos with the encoder signal also to the motion controller, or servos with scales. I guess there is also the hybrid stepper motors you mentioned as well. Geez this thing is never gonna get built... I stopped working on my scale model engine to do this retrofit hoping to make some of the complex setups easier. Kinda feels like I'm further behind now! At least I am learning a lot, don't know why but I am fascinated by all the ins and outs of motion control.



  13. #13
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by sohcpwr View Post
    First of all, thank you very much for taking the time to go into such detail. When I referred to "closed loop" operation, I was describing the system similar to what you mention in the last paragraph with rotary encoders on the motors, and linear ones on the machine axis. I dont know a ton about how different controllers use the tandem signals but I believe the servos encoder is used for the "meat and potatoes" and the scales are for periodic checks of accuracy.
    What you are refering to is called a Dual Loop system. Where you have encoders on both the motor and the load. Then depending on how the controller is programmed, you can compare and correct on the fly, or do a compare and correct at the end. I prefer to correct both on the fly and at the end of a move. You do not need a dual loop system to do this.

    Let's take my router, it is a closed, single loop stepper system. I have it programmed to compare the the commanded position (in steps) to the linear scale position (in pulses) and make corrections on the fly, every millisecond, then at the end of the move I do a final compare and correction if needed. I have the final position error limit set at +/- 10 microns (0.0004 inch). Close enough for a router.

    On my mill, I do the same thing, but have brushed DC servos running in a single loop on the X & Y, and a single loop stepper on the Z, and I have that final error limit set at +/- 2 microns (0.00008 inch) On both machines it hits the target consistently. There are no encoders on the motors on either machine.

    Somewhere I think I read that the scales can not transmit position data fast enough to be used for motion control.
    With some older scales that is true, but not in most cases with modern scales. My scales are rated at 3M/sec (about 9 feet per second) at 1 micron resolution or to put it another way 3 MHz output. I think you will find that most modern scales will come pretty close to that or better. At 9 feet/sec your mill table would fly off of the machine and through the wall I heard that actually happened when a large (several KW) servo motor went wild on a large planer mill.

    The controller will fault if actual position read from the scale does not match where the software thinks it should be from the servo. This is what you are paying for with the high dollar option on the hicon board. Except instead of the controller faulting, it corrects the position for the next move.
    If the position error (called following error) gets larger than the set value in the software, then the system should fault out and shut down. This is normally an indication of a mechanical jamb or an encoder failure. In either case you want the system to shut down. The following error limit should be set high enough that you would never reach that value in any normal operation. I have the limit on my machines set for about 1/2 inch worth of pulses.

    I just read the manual for the board in the link I posted. Looks like its supposed to receive the encoder signal first, then there is an output for it to go back to the drive. The board also requires 5v power, so I think it powers the encoder when the drive is disabled.
    I think you are correct.

    Kinda feels like I'm further behind now! At least I am learning a lot, don't know why but I am fascinated by all the ins and outs of motion control.
    We're just getting started Wait until we get into PID loops, motion profiles, and other fun stuff. As long as you are willing to learn, I'm willing to try to help out. I've been doing motion control systems for at least the last 20 years and I am still learning stuff every day.



  14. #14
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Can't thank you enough for taking the time to write such detailed replies. As you've pointed out and I have seen first hand, there is a ton of misinformation out on the web. It is refreshing to stumble across someone with real world experience to get real answers. Your explanation of how your machines are setup opens a whole new set of questions of mine:

    The signals from the linear encoder are going to the motion controller which is then adjusting the signals its sending to the drives for positional error? I would guess that would work for a stepper setup but for some reason Im thinking a servo drive has to receive a signal from some kind of encoder.

    Im also wondering, since the setup you describe and have running is obviously possible... why would any machine have motor encoders at all if linear scales on the axis can perform dual duty? I cant think of a scenario where it wouldnt be advantageous to read direct from the axis. Even the best ballscrews have a little bit of non-linearity, best bearings have some play, etc.. Speaking of, If one went this route, c7 rolled screws would be more than adequate if your controller is correcting based on the table mounted linear scales?

    More questions to come, lunch break almost over. Looking like stepper motors with glass scales is the way to go for me if I can get a controller that will correct for errors instead of just passing the signal through to the DRO.



  15. #15
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by sohcpwr View Post
    Can't thank you enough for taking the time to write such detailed replies. As you've pointed out and I have seen first hand, there is a ton of misinformation out on the web. It is refreshing to stumble across someone with real world experience to get real answers. Your explanation of how your machines are setup opens a whole new set of questions of mine:

    The signals from the linear encoder are going to the motion controller which is then adjusting the signals its sending to the drives for positional error? I would guess that would work for a stepper setup but for some reason Im thinking a servo drive has to receive a signal from some kind of encoder. Im also wondering, since the setup you describe and have running is obviously possible... why would any machine have motor encoders at all if linear scales on the axis can perform dual duty? I cant think of a scenario where it wouldnt be advantageous to read direct from the axis. Even the best ballscrews have a little bit of non-linearity, best bearings have some play, etc.
    Most servo drives require some kind of feedback from the motor for speed control. This feedback could be a tach signal, an encoder, or a resolver (another type of ''encoder''), or in some cases Hall sensors internal in the motor. This motor feedback does not have to be fed back to the motion controller for position data, that can come from some type of encoder attached to the load.

    . Speaking of, If one went this route, c7 rolled screws would be more than adequate if your controller is correcting based on the table mounted linear scales?
    C7 screws would work, but...You want double ball nuts to be able to adjust for zero backlash with some preload. Backlash comp in software is a fallacy. While an encoder on the load will automatically correct (kinda) for some backlash, getting the mechanics right to start with is the correct way to do it.

    More questions to come, lunch break almost over. Looking like stepper motors with glass scales is the way to go for me if I can get a controller that will correct for errors instead of just passing the signal through to the DRO.
    Again, look at magnetic scales. Glass scales have some maintenance requirements (cleaning) and are bulky, where magnetic scales are bulletproof, very compact, and are about the same price as glass scales. I Installed a mag scale on the inside of the cross slide for my lathe DRO, no way that would be possible with a glass scale. The reader head is about an inch square and the mag tape is 10mm wide and about 2mm thick. It is also a non-contact system, the reader head sits about 0.5 mm off of the tape.

    As long as you have a controller that will close the loop you should be fine in this regard. I really can't offer much advice in this area because I haven't worked with a lot of different controllers. If you settle on a controller, I'll be happy to review the specs to at least try to translate what they are saying.



  16. #16
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    So you said DC servo system on your machining center without encoders on the motors, I take it they have the internal hall sensors?

    Do you have a link to any of the magnetic scales you speak of? Ive never seen anything 2mm thick... unless you only mean the tape and not the ground steel bar that it sits in?

    I actually spoke with one of the sellers on ebay about a set of taiwanise made TBI motion ballscrews. He said the single nut was still practically zero backlash, but the double is preferred because as the assembly wears you can adjust it out. Whats your take on that? Elsewhere I have read that c7 lead error is too high for oversize ball type preload which is why a single nut in this grade would have a certain amount of backlash. Again, quite contradictory

    Back to narrowing down the setup... I will email the guy from CNC4PC and find out what the deal with the encoder board is. Its really not making sense, I am almost positive the smoothstepper cannot use encoders for feedback purposes. Encoders can apparently be connected directly to port 3 of the smoothstepper without the board. So all its doing is acting as a breakout board of sorts and possibly powering them as well? As stated the c53 manual leaves alot to the imagination. At that point 99% of what you need for an awesome control system is there, but there is no plugin/macro/software (not sure what to call it) for arguably the most popular motion controller for Mach to DO anything with the position data. Except send it to the DRO screen that is

    It is the motion controller and not the controller software that I am looking for closed loop ability correct? If thats the case I am left with the Hicon, Galil you mentioned, anyone know if there are others in a somewhat reasonable price range? I didnt look into the Galil yet, earlier you said a used 4 axis model would be cheaper than the Hicon. My intentions are to power the quill for the Z axis, but also power the knee for an offset when the 5" of quill travel isn't enough. A axis also a must so I can stop cranking the handle on the indexer like a maniac cutting high tooth count gears. I dont know what Im gonna do when I make the first one under CNC power.



  17. #17
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by sohcpwr View Post
    So you said DC servo system on your machining center without encoders on the motors, I take it they have the internal hall sensors?
    Nope, tachometer feedback. This is common on the older brushed DC servo motors. Hall sensors would be found on 3 phase BLDC motors.

    Do you have a link to any of the magnetic scales you speak of? Ive never seen anything 2mm thick... unless you only mean the tape and not the ground steel bar that it sits in?
    Ground steel bar? The ground steel bar is my mill table. Been on there about 3 years with no problems.

    Here is a picture of my X axis. This particular read head is a Renishaw LM10, 1 micron. But I am currently buying Ditron units, have found them to work just as well. The Ditron units are a new product and there is no internet link as far as I can find. I have attached a PDF brochure.

    Another bridgeport retrofit, need advise-img_1300-jpg


    I actually spoke with one of the sellers on ebay about a set of taiwanise made TBI motion ballscrews. He said the single nut was still practically zero backlash, but the double is preferred because as the assembly wears you can adjust it out. Whats your take on that? Elsewhere I have read that c7 lead error is too high for oversize ball type preload which is why a single nut in this grade would have a certain amount of backlash. Again, quite contradictory
    The double nut is mandatory for a mill in my opinion. For a router you can get away with a single nut.

    Back to narrowing down the setup... I will email the guy from CNC4PC and find out what the deal with the encoder board is. Its really not making sense, I am almost positive the smoothstepper cannot use encoders for feedback purposes. Encoders can apparently be connected directly to port 3 of the smoothstepper without the board. So all its doing is acting as a breakout board of sorts and possibly powering them as well? As stated the c53 manual leaves alot to the imagination. At that point 99% of what you need for an awesome control system is there, but there is no plugin/macro/software (not sure what to call it) for arguably the most popular motion controller for Mach to DO anything with the position data. Except send it to the DRO screen that is
    I'll have to review the manuals to try to figure that out, I'll have to get back to you on that. Maybe some other member can jump in on this.

    It is the motion controller and not the controller software that I am looking for closed loop ability correct? If thats the case I am left with the Hicon, Galil you mentioned, anyone know if there are others in a somewhat reasonable price range? I didnt look into the Galil yet, earlier you said a used 4 axis model would be cheaper than the Hicon.
    The motion controller should be what closes the loop if you plan on having a DRO with the motors off.

    My intentions are to power the quill for the Z axis, but also power the knee for an offset when the 5" of quill travel isn't enough.
    Here is a picture of my Z axis drive case, switches between manual and CNC with a flip of the lever.

    Another bridgeport retrofit, need advise-img_1209-jpg

    I just have a Servo brand power feed on my knee because I hate cranking. It's pretty easy to build your own power feed for the knee if you don't want to spend the money for commercial one. My (K)nee axis has a mag scale on it that reads out on the DRO. I have it connected to one of the auxiliary encoder inputs on my Galil card. Galil has dual encoder inputs for each axis.

    Here is a screenshot of my mill computer screen. R is the rotary axis, what most people call the A axis

    Another bridgeport retrofit, need advise-screenshop-jpg


    A axis also a must so I can stop cranking the handle on the indexer like a maniac cutting high tooth count gears. I dont know what Im gonna do when I make the first one under CNC power.
    Here is my A (R)otary axis. It started out life as a Vertex Super Spacer. Powered by a NEMA 23 stepper, open loop. 3.6 million steps per revolution and near zero backlash <0.6 arc sec. Can be mounted horizontal or vertical by rotating the motor housing.

    Another bridgeport retrofit, need advise-img_0712-jpg

    Attached Thumbnails Attached Thumbnails Another bridgeport retrofit, need advise-dm-pdf  


  18. #18
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    How have I never seen that mag scale tape before?? Thats awesome, normal DRO installation would require "tramming" the scale to the axis to check for parallel and perpendicularity. How is this achieved with the strip? One is obviously covered on x axis by mounting to the flat table face.. but Y, my machine is like a 1958 caddy with smooooooth curves. Do you remember the price per axis for these? Obviously different depending on how long but ballpark is fine.

    For the knee, I think I'd still prefer it to be under CNC control, hence the comment about needing a 5 axis controller. A triple stack nema 34 and large reduction combined with the 2:1 in the bevel gear drive should at least move it up and down. Again its only planned for a work offset if/when I run out of quill travel on a part, so accell, decell, and rapid speed are irrelevant to me. Shouldn't really cost that much more than a quality powerfeed. I'll be making all the brackets and hardware for mounting the motors instead of buying one of the ridiculously priced kits. Building these kinds of projects is the real fun for me. Playing with it afterward is just a bonus

    Is the screenshot of your controller Galil software? I checked out their site awhile ago and only found software for servo tuning and some other stuff non-CNC controller.

    Next up I have some questions about how to integrate my VFD setup. Lemme find some pics as it might be easier to explain



  19. #19
    Community Moderator Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    5717
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Quote Originally Posted by sohcpwr View Post
    How have I never seen that mag scale tape before?? Thats awesome, normal DRO installation would require "tramming" the scale to the axis to check for parallel and perpendicularity. How is this achieved with the strip? One is obviously covered on x axis by mounting to the flat table face.. but Y, my machine is like a 1958 caddy with smooooooth curves.
    You have a lot of leeway in the surface. On my mill it is machined below the Y box ways, so mounting was the same as the X axis, I just stuck the tape to the side of the knee below the saddle. On the Shizuoka we mounted a piece of 1x1 aluminum to the side of the knee, and added some tramming screws. On my router, I used a piece of 1x2 steel tubing and put a screw hole about every foot, then just shimmed to get it pretty good. As I said, you have a lot of leeway in the mounting, the goal is to get the read head as close as possible, but not touching the tape.

    When you get to that point, I'll help you out with detailed pictures. Ditron also sells aluminum mounting rails at about $20/meter as I recall

    Do you remember the price per axis for these? Obviously different depending on how long but ballpark is fine.
    $90/ read head and about $28/meter for the tape. So including shipping, about $120/axis.

    For the knee, I think I'd still prefer it to be under CNC control, hence the comment about needing a 5 axis controller. A triple stack nema 34 and large reduction combined with the 2:1 in the bevel gear drive should at least move it up and down. Again its only planned for a work offset if/when I run out of quill travel on a part, so accell, decell, and rapid speed are irrelevant to me. Shouldn't really cost that much more than a quality powerfeed. I'll be making all the brackets and hardware for mounting the motors instead of buying one of the ridiculously priced kits. Building these kinds of projects is the real fun for me. Playing with it afterward is just a bonus
    Here is an 8 axis PCI controller DMC-1880 GALIL Motion Controller

    And here is an 8 axis stand alone Ethernet controller with either stepper or servo drives built in. This one is the latest top of the line controller. Galil Motion Control DMC-4080 8-Axis Ethernet RS232 Motion Controller | eBay

    You can build a knee power feed out of a stepper for a whole lot less than buying a good one. And as you say, it's fun to do.

    Is the screenshot of your controller Galil software? I checked out their site awhile ago and only found software for servo tuning and some other stuff non-CNC controller.
    Galil doesn't sell CNC software, that's my software. I wrote it to work exclusively with Galil products after I couldn't get Mach3 to work well on a high speed router that I was working on for a customer. I have kind of a short fuse when things don't work the way I want them to and I do crazy things like writing my own software to get it to work the way I want.

    Next up I have some questions about how to integrate my VFD setup. Lemme find some pics as it might be easier to explain
    I've installed a few VFDs.



  20. #20
    Registered
    Join Date
    Nov 2013
    Posts
    71
    Downloads
    0
    Uploads
    0

    Default Re: Another bridgeport retrofit, need advise

    Most of my questions answered finding your website... Maybe you were hesitant to direct traffic from here because of forum rules or something? Good read, I like the quill drive. I think I'm still leaning twords the less attractive ballscrew sticking out the front routine but def not set in stone either way.

    Back to the VFD, I made a panel for my mill that I dont think I could live without anymore. It has speed control, digital RPM display, e-stop, and FWD and REV momentary buttons preset at 100rpm for power tapping. Using the spindle on relay in the VFD, I also have a switch for spray mister where the air line is hooked to a 24v solenoid. Switch on only activates coolant when the spindle is on. Off is off all the time. I was pretty proud of that one. Have to run out to the garage to take a few pics of the completed setup. Problem is, all this is using the same set of inputs that the CNC controller is going to want to use. My idea is that hopefully they make a mechanical switch that will allow me to connect 2 different inputs to a common output. Something like a 5 pole, 2 position, and would you call it a 3 way switch?? To select between manual panel control, or automated. If I knew more about rs485 programming, I could use the modbus I think its called to control it through the computer and not have to bother with that. Limited research tuned up that it might be where I drawl the line on my skill set.

    Another bridgeport retrofit, need advise-bridgeport-vfd-panel2-jpg

    Another bridgeport retrofit, need advise-bridgeport-vfd-panel-jpg

    Another bridgeport retrofit, need advise-bridgeport-vfd-proc-switch-jpg



Page 1 of 7 1234 ... LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Another bridgeport retrofit, need advise

Another bridgeport retrofit, need advise