Best path to CNC a Bridgeport knee mill (no, no no, not what you think)?

I have a Haas VMC, not looking to build a real CNC out of my knee mill or even pretend this would be close to a good idea. I get it. No worries. Hear me out.

So, I am looking at this machine and every so often I think: It would be useful if it could do some basic CNC while keeping manual control.

Limits? Well, I don't have to list them all for you. From making an absolute mess with chips flying around to the same thing with coolant to everything else you guys know about. Again, I get it.

Yet, it would be useful to have the ability to, for example, drill hole patterns or do simple stuff with it while the VF-2 is busy.

And yet, again, we all agree that you have to be very careful about how much you invest into a conversion because it just isn't worth it. I really like the Acu-rite Millpwr G2 control but it would be nothing less than insane to take that path when I could literally buy another VF-2 for the price.

So, cheap, yet reliable and useful without losing valuable manual operation. What are the options?

I'll take this on as a hobby project to do with my kids and teach them about the process, so that adds value to it. I can obviously machine all the brackets and parts on the VF-2, so, if I take that as a learning opportunity for the kids, the cost is basically zero.

Ball-screws? $1,500? Not sure it's worth it, don't care about feed rate and backlash is manageable.

For controllers I see stuff like this on eBay:

4 axis CNC milling controller ,Support absolute, G code control panel ATC PLC , | eBay

And this at the other extreme:

4 axis CNC controller USB Stick G code Spindle Control for Servo Stepper Motor | eBay

I really hesitate to spend $1K on a control. Yes, I would like a decent control panel rather than a PC with a keyboard and mouse, that would be a mess.

That said, I have half a dozen laptops and a few desktops laying around that could be repurposed (I never get rid of old computers).

I also have a bunch of steppers and DC servo motors laying around from other unrelated projects, including a set of Teknic Clearpath SD servomotors with 800 step encoders, power supply and cables.

In other words, I might be able to name this tune with a bunch of parts I machine on the VF-2, a laptop or desktop (as long as I forgo a real control panel), belts and pulleys, software like Mach 3/4 and maybe a ballscrew retrofit if it makes sense.


OK, I think you get the idea. Not worth spending a bunch of money on this. Set me straight, what might be the right approach?

Thanks.


PS: It absolutely hurts me to the core to buy cheap Chinese crap. Sometimes there are no options. I will spend a bit more for either US or European hardware. Definitely.

If I had to get rid of my CNC kneemill or my manual, I would get rid of the manual

with joysticks and handwheels, using it in manual mode is not the most efficient thing, but it can do things the manual cannot

TO me the best CNC knee iron is still the CNC Bridgport Rigid ram

To be clear, I already have the manual Bridgeport. I am not looking to add a machine. Just thinking it could be useful and a good educational opportunity for my kids to convert to CNC for very basic stuff (hole pattern, whatever).


Keep in mind that if you are teaching your kids something useful 100 hours all of a sudden can become time very well spent and the value derived goes well beyond the time or money invested. My 14 and 21 year old boys are very comfortable setting-up and running the VF-2 and manual machining on the Bridgeport. That took hundreds, if not thousands, of hours over the years. Time well spent as far as I am concerned.

My thinking is: Within what is financially reasonable, can I pull something together that makes sense?

BTW, it might not make any sense at all. I am perfectly ready to accept that possibility.

Honestly the millpwr retrofit is probably the best way to do it. The control is super easy and can be used with gcode programs.

I have a ProtoTak and its a great machine. The controller has menus for circles, pockets etc and if you treat it as a manual machine
that can do profiles its very nice and doesn't require writing any gcode. Its never going to be as fast as any VMC (lack of horsepower and cast iron) but as a
improved manual machine it can't be beat.

If you can find a control like that I would highly recommend it.

plastikdreams said:

Honestly the millpwr retrofit is probably the best way to do it. The control is super easy and can be used with gcode programs.

Click to expand...

I like Millpwr G2 and got to work with it a bit when I was consulting for a local aerospace company. They had a retrofitted knee mill with the that control and I got to use it a bunch of times. That said, it is really expensive and hard to justify financially. If I remember correctly it's in the $20K range. I can't see dropping that kind of money into a knee mill.

My objective isn't to make it a full blown CNC rig but rather give it additional functionality on a reasonable budget. If all I accomplish is automate the X and Y axis with manual g-code entry or something like Mach 3/4 it could be very useful.

Example: The other day I had a job where I had to drill a bunch of 1/4 holes on a pile of 80/20 extrusions. I could have easily run that on the Bridgeport and free-up the VF2 for more critical parts.

I guess what I am thinking is that motorized numerical positioning on x and y would be a huge improvement and one that should not cost a ton and should not require going to ball screws.

BTW, and maybe this takes the conversation in another direction,...I don't even have a DRO on the thing. I actually have the old (30 years?) Sargon glass scales installed on x and y but the DRO head got stolen the last time the machine was moved.

Maybe if I can find a DRO that's compatible with the Sargon scales I can go back to being able to use numbers rather than counting wheel turns and that will be enough to give it greater utility?

If you're really just looking for a project, check out the Clearpath servos. I've heard they are pretty easy to use.

ClearPath Servos - Teknic

The problem is you are going to need ballscrews, without them the backlash will put your features all over the place.

Keep your eye out for even some older millpower stuff. My dad has it as a 2 axis on on a Sharp, for one off half simple stuff it's awesome. I am just starting to use a real VMC, and man do I miss just grabbing the handle and moving the table for setup, and the super easy to use conversational control.

If you are looking for cheap cheap and don't mind to tinker a bit, and arduinocnc/GRBL and some cheap drives is an option. It's really hobbyist stuff, but you could be up and running for $100 or something. I had never done anything with G-Code ever. Bought one, set it up on my dads Sherline that had a old pita MAXNC setup, and within 30min was cutting some 3d shapes from fusion. Thing ran for like 2hrs off a old cheap laptop without a hiccup. Out to be plenty good for some hole patterns and such.

Older Millpower I played with a bit was great with canned cycles. Trying to manually input special shapes was pretty much the same effort as writing g code by hand. Loading g code over a cable was a joke, like I was stuck back in the 80's.

I really wished I had looked into the Prototrak instead.

I have a Tree Journeyman 325 with a Dynapath Delta 20 control. About the same work envelope as a Bridgeport, weighs about twice as much, 40 taper. The control is still supported and parts are available. I have about 6 tool holders some are new as in never used. Ready to go for 4500.00 You will never convert a B port for that money. Nice tight machine priced to move.

Maxim said:

I have a ProtoTak and its a great machine. The controller has menus for circles, pockets etc and if you treat it as a manual machine
that can do profiles its very nice and doesn't require writing any gcode. Its never going to be as fast as any VMC (lack of horsepower and cast iron) but as a
improved manual machine it can't be beat.

If you can find a control like that I would highly recommend it.

Click to expand...

SWI makes ProtoTRAK retrofit kits for bridgeports. I'm not sure how much they cost (I have a vague memory that I heard 8,000 somewhere, but don't hold me to that). I rather like those controls for the most part and would be inclined in that direction myself.

wesg said:

Older Millpower I played with a bit was great with canned cycles. Trying to manually input special shapes was pretty much the same effort as writing g code by hand. Loading g code over a cable was a joke, like I was stuck back in the 80's.

I really wished I had looked into the Prototrak instead.

Click to expand...

Last millpwr unit I used in 2012 took floppy's as did the much older prototrak retro...now I believe the new ones take USB.

I'll probably sound a bit harsh, but - the title is misleading - it is exactly what everyone thinks when they see topic like this...

it is absolutely NOT worth it if you don't do everything properly, good screws, good thrust bearings, good servos, and if you leave acme thread screw backlash in it - it will not be of much use in cnc mode, maybe it would work if you'd use dro scales (assuming you have them) for feedback, but side milling will still be bad, so basically it will work ok for drilling holes, which you can do very well with a help of just a good DRO

best get rid of it for quick $$$ and add some to buy the machine from Moonlight post #11, take it apart completely if you want the learning experience, clean everything, put back together and use

I looked into this and found that for very little money you can throw some stepper motors onto it, grab an old PC and get a touchscreen for the controller, throw GRBL onto an Arduino an have a pretty functional machine. The problem though is the backlash in the lead screws. If you're just looking at drilling hole patterns with reasonable tolerances, that won't matter, but as soon as you start trying to do anything more complicated you're going to need to upgrade to ballscrews.

Two issues with ballscrews on a bridgeport, the first is a decent set for the X/Y will run about $1500-2k. That was more than I wanted to throw at my old bridgeport. The second is that ball screws will make the machine a pain to use as a manual, since you will need to lock each axis that you aren't using or it will move on you. IN addition the ball screws will likely be too coarse to make precision work easy.

At the other end of the spectrum there are also conversion kits out there for $7k-10k, but at that price I think there are better options out there.

If you've got the space, and want a fun project, you could build your own from scratch. Depending on the precision you are looking for its really not that hard. Then you don't have to worry about mucking up your Bridgeport.

jaguar36 said:

I looked into this and found that for very little money you can throw some stepper motors onto it, grab an old PC and get a touchscreen for the controller, throw GRBL onto an Arduino an have a pretty functional machine. The problem though is the backlash in the lead screws. If you're just looking at drilling hole patterns with reasonable tolerances, that won't matter, but as soon as you start trying to do anything more complicated you're going to need to upgrade to ballscrews.

Two issues with ballscrews on a bridgeport, the first is a decent set for the X/Y will run about $1500-2k. That was more than I wanted to throw at my old bridgeport. The second is that ball screws will make the machine a pain to use as a manual, since you will need to lock each axis that you aren't using or it will move on you. IN addition the ball screws will likely be too coarse to make precision work easy.

At the other end of the spectrum there are also conversion kits out there for $7k-10k, but at that price I think there are better options out there.

If you've got the space, and want a fun project, you could build your own from scratch. Depending on the precision you are looking for its really not that hard. Then you don't have to worry about mucking up your Bridgeport.

Click to expand...

Which is EXACTLY why this should be over at "the zone"

and NOT here.

Try following the rules.

If that even matters anymore.

martin_05 said:

I like Millpwr G2 and got to work with it a bit when I was consulting for a local aerospace company. They had a retrofitted knee mill with the that control and I got to use it a bunch of times. That said, it is really expensive and hard to justify financially. If I remember correctly it's in the $20K range. I can't see dropping that kind of money into a knee mill.

My objective isn't to make it a full blown CNC rig but rather give it additional functionality on a reasonable budget. If all I accomplish is automate the X and Y axis with manual g-code entry or something like Mach 3/4 it could be very useful.

Example: The other day I had a job where I had to drill a bunch of 1/4 holes on a pile of 80/20 extrusions. I could have easily run that on the Bridgeport and free-up the VF2 for more critical parts.

I guess what I am thinking is that motorized numerical positioning on x and y would be a huge improvement and one that should not cost a ton and should not require going to ball screws.

BTW, and maybe this takes the conversation in another direction,...I don't even have a DRO on the thing. I actually have the old (30 years?) Sargon glass scales installed on x and y but the DRO head got stolen the last time the machine was moved.

Maybe if I can find a DRO that's compatible with the Sargon scales I can go back to being able to use numbers rather than counting wheel turns and that will be enough to give it greater utility?

Click to expand...

Glass scales even with manual control are a huge leap from counting turns. :-)
However, my experience with glass scales for cnc has not been as good as encoders on the ballscrews; not enough counts. However, that is with 5 um scales. One um scales or .5 um scales could change that. The problem is simply not enough counts. Difficult to tune the servos. The fix for this is using an encoder on the ballscrew or servo motor for P,D, FF1, FF2 and then tie the glass scale to I. Something like linuxcnc will let you do this.

As for DRO display I recently bought a Chinese readout for <$100. It works beautifully but with single ended scales like the Chinese ones. Doesn't work with my
Accu-rite scales. I need to opto-isolate or maybe just buffer from differential to single ended. The plan is to use the machine with glass scales for drilling and use my Mazak V5 for contour work.
HTH

jaguar36 said:

If you've got the space, and want a fun project, you could build your own from scratch.

Click to expand...

No, thanks. I have a Haas VF2. Exactly zero interest in trying to build a machine from scratch. I used to dream about that stuff many, many years ago. Iread entire threads here and elsewhere advising against it, got intelligent and bought a real machine.

This is about spending a little money to make the Bridgeport a little more useful. I think a bunch of folks who are commenting are taking this way too seriously. I have a real CNC machine for the real stuff. Not trying to make another one.

I'm thinking, xy positioning on the Bridgeport would be nice for stuff like drilling holes. I don't even want to surface anything on it if I don't have to, it's a mess, stuff flying everywhere, etc.

I think I am going to see if I can find a DRO head that works with the existing glass scales, work with that for a while and see what I think.

My gut feeling is that it might be sufficient for the occasional job where I want to drill a few holes on a few parts. Anything that involves lots more holes and more parts really needs to go in the VF2.

Side note: I will never sell this Bridgeport. It was originally owned by a friend and mentor who unexpectedly passed away. His wife called me one day and asked that I come get the machine, that he wanted me to have it. I literally talk to this machine. It may sound stupid to some, but I really miss my good friend and that machine represents him being with me as I navigate life's many ups and downs. It has been nearly a decade and he is still with me. It isn't just a machine.