[lumberjack_jeff]

Lots of questions. As a noob, I have questions which I haven't even formulated yet.
1) You opted to replace the stepper motors with servos. Elsewhere, I read that servos are considered by some to be a second-choice option, particularly when integrating with Mach3. Why did you choose servos?
2) How do you power the cnc controls? Are they run on a separate 110v circuit, or are they run off the three phase power coming from the vfd?
3) In your situation, is there a reason you chose to run a vfd for the Boss as opposed to a RPC? Have you integrated the CNC controls with the vfd for programmatic speed control?

I would like to do exactly as you have done, except I would have a bias toward keeping the stock servos (absent a compelling reason to dump them e.g. incompatibility with controller or software), and I'd like to have one RPC run both the mill and the planned lathe. At this point, I don't see the clear necessity of a vfd in my situation.

[damae]

Hi Lumberjack Jeff. Good questions; see answers below:

1) You opted to replace the stepper motors with servos. Elsewhere, I read that servos are considered by some to be a second-choice option, particularly when integrating with Mach3. Why did you choose servos?

Servos are actually much better performers, especially in speed, and control. For instance, most steppers systems are open-loop, which means that if you have a momentary load that is greater than the power the stepper is putting out, the control has no way of knowing that position is not what it is supposed to be. Aside from losing steps, steppers typically lack speed, whereas servos commonly are rated for 3000rpm.

Of course, there are ways to add an encoder to a stepper motor to prevent missed steps. And modern micro-stepping controllers are much better than the original BOSS control. So even though the original BOSS control would have 1 step = .001", a new GeckoDrive could easily get much finer resolution.

On the other hand, Servo controllers usually have a lot of extra features packed in, like programmable acceleration profiles. The Emerson units I'm using allow me to select (in software) what resoultion I want the motor to operate at. I can tell it to be a 5000 count/rev motor or a 100 count/rev motor.

But the biggest reason I switched to servos was because I could replace older steppers with much larger servo motors and get better resolution, speed, and power. At the time, it looked like the servos would be bolt-on replacements. Of course, reading my project log here, you see that I ended up needing custom motor mounting brackets to make it work.

2) How do you power the cnc controls? Are they run on a separate 110v circuit, or are they run off the three phase power coming from the vfd?

That's another advantage of the Emerson servo controllers. They have their own power supply built-in. All I do is supply 240v single phase and it rectifies it and generates its own 3-phase motor power. Pretty simple. I didn't have to build any power supplies, as I would have if I kept the original steppers.

I was too far down the servo path before I realized how simple it is to convert a BOSS to run off of Mach. That would have been a true shortcut. It might have been a better first step, because I could use that setup to machine all the parts I needed to convert to servo control.

3) In your situation, is there a reason you chose to run a vfd for the Boss as opposed to a RPC? Have you integrated the CNC controls with the vfd for programmatic speed control?

Yes, but in the end, some of the reasons are flaky:

1. I didn't want to buy a RPC, too much money. Funny thing is I just bought one to run my lathe and the other mill (Kearney & Treker). So had I just committe to a RPC to start with, I could have reused the original power supply and stepper drivers and just plugged it all into Mach, using the MaxNC mode.

2. VFD allows you to run the motor even faster (or slower) than 60hz. The motor can probably handle 2x the speed safely, but the spindle bearings ultimately limit the max RPM.

3. Spindle speed control from Mach was the ultimate goal. I don't have that hooked up yet. (it doesn't take step/direction signals)

4. The VFD allows me to run the whole machine off of single phase 240v power, and I got to ditch the original motor starting controls (another metal box hanging off the mill)

I would like to do exactly as you have done, except I would have a bias toward keeping the stock servos (absent a compelling reason to dump them e.g. incompatibility with controller or software), and I'd like to have one RPC run both the mill and the planned lathe. At this point, I don't see the clear necessity of a vfd in my situation.

Were I to do it all over again, I would have just bought a RPC and a set of Geckos. Later I would have redesigned the machine and have it build its own parts. Leaving the stock motors on has a huge advantage, because you don't have to mess with belt tensioning, motor clearances, motor wiring, etc. And after all, those motors were sized for that machine... back in 1975! =) By all accounts (see elsewhere on the 'zone) this method works beautifully!

[damae]

Damae,
Keep the pics coming. You seem to be attracted to these excellent deals. Will the garage floor support that weight.
We will be watching as your efforts unfold.
Jason

Hey Jason, thanks for showing interest in my thread! =)

I do have about 15,000lbs of machine tools in there. But a coworker just put it in perspective for me yesterday: his Suburban weighs 10,000 lbs, but that's sitting on only 4 contact patches!

Yesterday, I was in a molding factory in China, where they had about 25 injection molding machines in a row, on the second floor of the building. These machines easily weigh 5tons each. When one of the employees accidentally dropped a 500lb mold onto the ground (from about 1 foot) the whole floor shook and it really made me think about how amazing it is to have so much weight up on the second floor! Then again, no more amazing than a parking garage really. =P

Anyway, I think my garage floor will be OK. No threat of putting any more machines in there -- they just won't physically fit!

[lumberjack_jeff]

I was too far down the servo path before I realized how simple it is to convert a BOSS to run off of Mach. That would have been a true shortcut. It might have been a better first step, because I could use that setup to machine all the parts I needed to convert to servo control.
(snip)
Were I to do it all over again, I would have just bought a RPC and a set of Geckos. Later I would have redesigned the machine and have it build its own parts. Leaving the stock motors on has a huge advantage, because you don't have to mess with belt tensioning, motor clearances, motor wiring, etc. And after all, those motors were sized for that machine... back in 1975! =) By all accounts (see elsewhere on the 'zone) this method works beautifully!

Thanks so much for your thorough reply, this thread will be an excellent resource for those contemplating similar conversions.

You mention, and I've read elsewhere, anecdotes about others who have performed conversions using the existing steppers. As I understand the ways to use a 80's vintage BP with a Boss controllers, in order of complexity;
1) use as is (and with its limitations) and repair electronics with scavenged but expensive parts
2) hook a pc up to the existing boss controls via rs232 to store and organize programs
3) replace the boss electronics with new stuff. I've read that the Geckos (201's?) won't handle large stepper voltages, apparently the steppers that came with the BP are under this threshold?
4) replace the boss electronics and the original steppers.

In your research, have you seen a discussion from someone who took door #3? If so, could you provide a link? (I've googled my fingers to the bone trying to find one )

Also, I'm not that far from you. Where did you find your machinery? The Seattle area seems to be (except for boeing) devoid of any machine tool selling activity. I suspect it's because of the sales tax. Any recommendations for someone trying to get started?

[damae]

...
3) replace the boss electronics with new stuff. I've read that the Geckos (201's?) won't handle large stepper voltages, apparently the steppers that came with the BP are under this threshold?

In your research, have you seen a discussion from someone who took door #3? If so, could you provide a link? (I've googled my fingers to the bone trying to find one )

Also, I'm not that far from you. Where did you find your machinery? The Seattle area seems to be (except for boeing) devoid of any machine tool selling activity. I suspect it's because of the sales tax. Any recommendations for someone trying to get started?

Hi Jeff,

I did a quick search and found a few others doing Gecko conversions with existing steppers (Door #3):

http://www.cnczone.com/forums/showthread.php?t=23079
http://www.cnczone.com/forums/showthread.php?t=23605
http://www.cnczone.com/forums/showthread.php?t=19637

There are lots of others who have posted diagrams and instructions on how to wire a gecko to the OEM Bridgeport steppers. This may have been a smarter route to go, but I'm happy so far with Door #4. =)

It's funny you mention a lack of machines in the Seattle area, because I bought my last two mills from Arlington, just north of Seattle. Watch Craigslist and ebay (sort by distance). I find that weeks may go by with no listings. Then all at once, there will be a flurry of unrelated posts offering mills. Patience pays off big time! That and being the first guy to come out and take a look at it.

[damae]

Faced with the prospect of scratching out my missing aluminum motor mount on a very small CNC machine, my progress on the Bridgeport had stalled. Having a very busy life made this little excuse enough to delay any effort.

Then I had a breakthrough -- a few simple realizations that were enough to get me motivated again:
1. This part really doesn't need to be aluminum, or even acrylic. It's under very little stress.
2. I can cut the part from MDF and have it done in an hour or so!
3. The design of the bracket is already done. All I need to do is make it.

Cutting the part from aluminum would take a long time: 6+ hours on the small machine available! I also could not find a suitable piece of aluminum stock to cut the part from. That was enough of a mental barrier to keep me from attempting the build. I had "builder's block!" You know, like writer's block. And so the project sat.

The idea that I could have the part in my hands in an hour changed my electronic design into reality. It was very satisfying to make chips again, even if the chips were really just MDF dust. The part bolted up perfectly, toleranced just right for a snug, but not permanent fit with the locating circle on the NEMA 42 servo motor. The bracket can also accept a NEMA 34 motor, if I need to change my mind again.

Another nice discovery -- the bracket fits! Someone asked me for the CAD models of this design a few months back, but just before I sent the models, I realized that the "star" points of my design would probably interfere with the vari-speed housing on the spindle! So I didn't send the models, thinking the design needed more adjustment. If you're following this thread and still interested in the IGES or Pro/E models, please PM me.

Of course, there is a new stopper: My helical coupling fits a 1/2" motor shaft, which is what the NEMA 34 motor used. But the NEMA 42 motor needs a 5/8" coupling. Fortunately, McMaster Carr has good prices and I was able to order one online. Hopefully it will be here on Tuesday. If you want to take a look, go to www.mcmaster.com and type in "6208K14" which is the P/N.

[damae]

WooHoo! The new Z axis works!!

The motor is overkill, but it works. By default, this motor was set to 20,000 steps per inch (4000 steps/rev). Fortunately, I can change that to anything I want -- that's a nice advantage of using a resolver instead of an encoder. With so many steps/inch I decided to set the Z speed to 50 IPM, which it handled easily.

One potential flaw in the design is a bit of springyness when I force the spindle to reverse directions very fast (2 or 3 times per second). It has no problem doing a hard stop, but it seems to have more settling time than the X and Y axes. Perhaps the helical coupling that I used is not as torsionally stiff as I expected? I know the Z axis timing belt is not tensioned as well as it needs to be. Perhaps that could be another source.

In any case, I'm extremely pleased that all 3 axes are powered and working. I can start cutting steel! The first job will be to machine a replacement for that temporary MDF bracket.

Also on my list:
-Replace the Z axis limit switches (two of the 3 seem to not be working).
-Wire-in some E-Stop switches
-Mount LCD monitor on swing-arm
-Rewire control box so that computer is on a seperate circuit
(so computer can be turned on independant of drives)
-Replumb the Z axis to the Biijur lube pump
-Order a new ER32 collet holder, modify on lathe to fit my spindle nut
-Build custom electronics enclosure that mounts above the ram!

The giant electronics cabinet I have takes up far too much room. Real estate is at a premium in my garage! I really like the idea of splitting the enclosure into two sides: servo drives + computer + BOB on one side; VFD + power distribution + fuses on the other side. Both sides would be about eye-level with access through a hinging door. This configuration will also greatly reduce the cable clutter.

My quickie sketch is shown in the attached images. The object in the middle of the sketch is a crude aproximation of the Bridgeport's ram.

Any suggestions about what which components to seperate or which should be in the same compartment? Do you have successes or failures related to electrical interference you can share? I'll even take general tips about the best way to arrange components for servicability, aesthetics, cleaning, etc.

[Kevin Taylor]

damae Looked at the pic.s of your Kerney being mooved in Your neibourghs must be understanding !!! I'm about to put power to my boss3 I had to get a transformer to drop 230v to 208 diden't see that it was a 208 machine when I got it . Did you have a tracermill in your colection? If so would it be for sale?
I'm working on gathering stuff for a lathe conversion will have servomoter's next week and I think I have the ball screw's nailed down If you have extra control part's left over from your Boss let me know I may need help Carry on Kevin

[Kevin Taylor]

damae Just looked at the pic.s of your Boss3 when you got it and I was woundering what the box on the left side of the power encloser was maybe a frecdrive or a transformer mine dosen't have that box? let me know Thank's Kevin

[damae]

damae Looked at the pic.s of your Kerney being mooved in Your neibourghs must be understanding !!!

Yes, they are! I just recently sold the K&T. Having 4 mills and a lathe in one garage made it so difficult to do anything that it just sucked the fun out of doing my CNC conversion. With the K&T gone, my lathe is usable and I even have a small area for welding with my new TIG welder! It's starting to be fun a again!

... Did you have a tracermill in your colection? If so would it be for sale?...

Yes, it's just sitting at the moment, waiting to have its retrofit finished. I intended to finish the retrofit myself, but if you're really interested, PM me and we can work something out. I wouldn't mind reclaiming even more of my garage space! =)

...If you have extra control part's left over from your Boss let me know I may need help...

I junked the control except for the BOSS computer. It's yours if you want it, but it's been sitting outside (out of the rain) for a while. I'd be glad to help if there's something I'm able to help with. Do you have a project log thread for your mill?

damae Just looked at the pic.s of your Boss3 when you got it and I was woundering what the box on the left side of the power encloser was maybe a frecdrive or a transformer mine dosen't have that box? let me know Thank's Kevin

I don't have that box anymore, but I think it was the spindle motor control.

[Willbird]

You have far more done than I do yet. But my thinking on control cabs was to seperate AC and DC componanats for one thing. The PC if you use a micro ATX board can fit in the space of a small pizza box :-).

Bill

[damae]

You have far more done than I do yet. But my thinking on control cabs was to seperate AC and DC componanats for one thing. The PC if you use a micro ATX board can fit in the space of a small pizza box :-).

Bill

I just read your project log -- your build is progressing nicely!

This cabinet will be custom, of course. So I can make it just the right size and subdivide it if needed. Your suggestion to keep the PC seperate is good. So now I'm thinking to do this:

Left side cabinet
-Servo Drives
-Incoming 230v and 115v power blocks
-Star ground point
-Main pPower switch

Right Side Cabinet (main compartment)
-Computer
-Breakout board
-Relay board

Right side cabinet (secondary compartment)
-VFD all by itself. This compartment forms a shilded chamber to keep some of the hi freq interference at bay.

[Willbird]

Where is your 5 volt/ 12 volt power supply and your 24-72 volt power supply going to go ??

http://www.wikihow.com/Convert-a-Com...b-Power-Supply

You probably already have that covered.

I need to read up on Star ground as it is greek to me and I have seen it mentioned a few times.

With the micro atx PC mainboards really a case is pretty simple.

I'm not sure if my toroid will kick off enough magnetic field to disrupt an ATX power supply in the same 12 x 12 box or not ?? The wall plug power supplies from walmart are $15 each and few of them put the rating my cnc4pc breakout board wants....so I figuired an ATX supply is tons better than the wall plugs...and cheaper than the $30 for a 5 volt(didnt even find one) and a 12 volt.

Bill

[Kevin Taylor]

Damae did you ever power up your machine as it was when you got it I thought I read that the guy demoed it for you in MDI I woundered If I was missing somthing about the start up noticed that the lub res was not full and on my BTC it will shutdoun if there is no lub and it dosen't give you an error message . Maybe the mail will bring my manuals today and I can mack some progress Kevin

[damae]

Where is your 5 volt/ 12 volt power supply and your 24-72 volt power supply going to go ??

Actually, I don't need a power supply, since my Emerson servo drives take AC line voltage input. The drives have their own internal power supply. That's why they are so large.

...I need to read up on Star ground as it is greek to me and I have seen it mentioned a few times.

It might be a complex theory to understand why it works. Fortunately, it is actually very simple to do: just ground your devices at one point only. Most people use a 1/4"-20 bolt mounted to the cabinet and attach all the ground wires with ring terminals.

so I figuired an ATX supply is tons better than the wall plugs...and cheaper than the $30 for a 5 volt(didnt even find one) and a 12 volt.

I agree completely. You can also find lots of wall adapter power supplies of all voltages at your local Goodwill. A slicker solution is to buy real power supplies. I get mine from MPJA.com -- very good prices and nice power supplies.

-D

[damae]

Damae did you ever power up your machine as it was...

It was working at the machine shop, but I didn't power it up after trucking it to my garage. The Biijur lube pump on my machine has no level sensor, just an AC input.

The BTC is a more recent and higher end Bridgeport, so it makes sense that it would have a level sensor.

[damae]

Here's a novel idea -- use an engine stand to mount your milling machine head. I had this idea a while back, but finally got to try it out today. It was nice to get the Shizuoka head up off of the floor (on a dolly) and on wheels. When your shop is as packed as mine, you begin to value having everything on wheels.

It is also very convenient for rebuilding, cleaning, and powering-up to test out the power drawbar, speed changer, etc.

To celebrate the new working Z axis, I bought a very nice Kurt vise (model 3600V) to replace the "import" vise I sold with the K&T milling machine. I bought it from a machine shop who is replacing all conventional vises with high production vises (tombstone-style). The vise is immaculate, but didn't include the handle. Can you believe Kurt handles are $50? Just the handle! With the handle, I spent $400 for the vise. New these are $7-800.

I included a picture of the shop to show how things were rearranged with one milling machine gone. Much nicer to use when there is a little space to move around!

[Barry_ward]

Hey D,

Great to see you working on your shiz. I'm using the heck out of mine but it's far from done. Tonight I ran a 2" corncob shell mill to trim some extrusions.

The next biggie for me is a uniform method of chip containment...I made a billion chips. at least 1% will wind up in the living room carpet......

[jb2590]

mxtras said "Those covers on the head are where the quill feed mechanism would go if you had a manual machine head (from post #6)." it may be a step backwards but, i wonder if it is possible to install the manual quil feed equipment as well as run it via servo. that way if there was a simple operation that dident reguire cnc, you could just use it as a manual machine and use dual shaft servos so that you can operate x an y axis also manualy ?

[damae]

Hey D,

Great to see you working on your shiz. I'm using the heck out of mine but it's far from done. Tonight I ran a 2" corncob shell mill to trim some extrusions.

The next biggie for me is a uniform method of chip containment...I made a billion chips. at least 1% will wind up in the living room carpet......

My wife is on my case if she spots any chips in the house (and there have been a few!) Not good for the dogs paws or for barefoot humans either! So I leave the shoes in the garage when I come inside. But somehow they still migrate indoors! I think it's a lesser known corollary to the "socks dissapearing from the dryer" mystery.

Barry, your Shiz is looking great! (check out www.barryfish.com if you haven't seen his Shizuoka retrofit) Your machine really came out sparkling!

....i wonder if it is possible to install the manual quil feed equipment as well as run it via servo...

There is a lot of discussion here on the 'zone about making your CNC with manual handles. Generally, I discovered two main reasons people don't usually do it:

1. If you use ballscrews, they are so efficient that the reaction forces from the cutter could backdrive whichever axes you don't have your hand on. Manual machines have low efficiency ACME screws and so there is lots of friction to prevent backdriving.
2. CNC machines (especially routers) would have the handles in very awkward locations. That wouldn't be a problem on this machine.

The Z axis is the only one that has any appeal to me as a manual axis. Since I don't have a drill press in my garage, I use the Bridgeport, or else I just drill by hand. There are times where it might be nice to use the BP in jog mode to position the part in X and Y, but do the downfeeding by hand to 'feel' my way through the drilling operation.

To do this I would have to somehow disengage the Z axis ballscrew, or else force it to backdrive the Z axis motor. I think it will be too much work for the return. (I'll just get a drill press). =)