[Neil_J]

Hi,

I have a stupid question... I tried to search through the forums but I didn't find what I needed...

I am currently shopping for a new mill to retrofit... The problem I see with CNCing a Bridgeport-type knee mill, is only having 4-5 inches of quill movement. I will going between large and small tools (i.e. 1/8" end mill to large drill bits and chucks).. I can see large tools eating up my quill depth which would require repositioning... Is this correct? I have used benchtop mills but never a knee mill.

And are most bridgeport retrofits on the Z-axis quill only? Is there anyway to accurately control the knee movement up and down?

[Al_The_Man]

I personally would not recommend the knee as a controlled axis, too much mass etc for rapid moves, I have an Excello with Z quill and a stepper motor operating a Ballscrew type jack under the knee with air counter balance cylinders, which is semi manual operated, i.e. push-button, it has a graduated indicator for close positioning.
Al.

[Neil_J]

I personally would not recommend the knee as a controlled axis, too much mass etc for rapid moves, I have an Excello with Z quill and a stepper motor operating a Ballscrew type jack under the knee with air counter balance cylinders, which is semi manual operated, i.e. push-button, it has a graduated indicator for close positioning.
Al.

So automating the knee can be done? I wouldn't use it for rapid movement, just for repositioning between tool changes.

Do most knee mills have graduated handwheels on the knee? I was under the impression you lost your Z reference after moving it.

[Al_The_Man]

Most have a hand crank, this one was converted, you do not lose Z zero ref as the reference is at the quill extreme up position, so the table is just moving away, the only thing it will affect is the Z work co-ordinate, so some way has to be established to overcome this, you could either move the knee a known number of steps and reset the tool offset or I just usually re-zero the Z part zero, but I'm not a machinist!
Al.

[HuFlungDung]

Neil,
Available quill travel is a problem alright. It is tough to beat the convenience of a VMC bed mill for that very reason.

However, it is often possible to purchase toolholders that will extend the reach of small tools. And, it is also possible to purchase morse taper adapters to fit your spindle. Then, you also purchase what are known as 'drill drivers' which are a simple morse taper socket with a split cut lengthwise. A person can then buy one of these drill drivers for each size of straight drill shank that you will use. The advantage of this method, versus using a chuck to hold larger drills, is that the overall length of the toolholder plus drill will be 2 or 3 inches shorter than when using an arbor mounted drill chuck.

[Neil_J]

Neil,
Available quill travel is a problem alright. It is tough to beat the convenience of a VMC bed mill for that very reason.

However, it is often possible to purchase toolholders that will extend the reach of small tools. And, it is also possible to purchase morse taper adapters to fit your spindle. Then, you also purchase what are known as 'drill drivers' which are a simple morse taper socket with a split cut lengthwise. A person can then buy one of these drill drivers for each size of straight drill shank that you will use. The advantage of this method, versus using a chuck to hold larger drills, is that the overall length of the toolholder plus drill will be 2 or 3 inches shorter than when using an arbor mounted drill chuck.

I've heard of those, but would like to avoid them if possible...

HuFlungDung, in your opinion, what type of setup should I buy/build? Here are my requirements...

* Material cut: aluminum, steel, plastic, wood, (i.e. most everything)
* Material shape: Mostly flat stock that will be engraved/cut, but also the ability to hold blocks and odd shapes, ranging from very small to big.
* Work performed: Milling/drilling (straight from CAM or thru Mach3 conversational), engraving (text or PCB routing), 3d contouring, possibly digitizing using a probe, possibly mounting a plasma/laser if using a Gantry type)
* Machine construction: Very heavy cast iron/steel if possible. Must be very rigid (the main reason I would consider retrofitting a larger machine)
* CNC specs, after retrofitting: ballscrews, Geckos, servos, 100+IPM rapids, 0.0002" limit/home switches, auto coolant/spindle, etc... 3-axis to start, but will be adding a rotary table later on
* Accuracy: 0.003 or better
* Table size: X=20"+, Y=8"+, Z=10-20" (without losing z position!)
* CNC engraving (10K+ RPM) is a must
* Budget: $3000-6000 total, over 1-2 years time
* Experience/knowledge: very high in electronics/mechanical areas (but pretty new to machining)


If I could find one, a used gantry or small VMC would be perfect... However, modifying a manual mill or starting from scratch might be all my budget can handle.

[NC Cams]

It would seem that you really want a high speed engraver of some sort, especially with an RPM spec of 10K as a "must".

A Bridgeport is not going to run that sort of spindle RPM and if it could, not for long.
The generic bearings used to make the 1119238 oem spindle bearings used in most BPT milling machines are rated at only 9500 max and that is in an un-preloaded "stock" condition.

The oem bearings are custom preloaded at about 60 lbs which makes them stiffer for less tool deflection BUT it also reduces the allowable peak speed. Good engough for a mill that runs at the design speed but definitely not at 10K.

Simply put, a Bridgeport mill is NOT what one would/should consider a high speed machining center. Other (and surely more costly) machines will do the high spindle speeds that you want but these generally must operate at lower infeed rates. Othere than the spindle speed requirement, you could retrofit the BPT to do MOST of what you want.

Your 0.003 tolerance should be easily held with rolled thread ballscrews, good servo tuning and properly adjusted gibs. However, don't plan on doing fast inches/min. while trying to hold tenths in accuracy and repeatability. Used iron with work gibs/non-roller bearing ways won't facilitate it - simply too much slop and/or friction.

We hold error to within 0.0001 as compared to CNC ground part on our BPT EZTRAK but we take very light cuts (0.002/0.005 and travel at 1-2 inches/minute (this reduces effects of table friction). Moreover, we've got a pristine machine with chrome ways, several thousand dollars worth of ball screw and custom ball screw bearing work/rework and several weeks of maching tuning.

Result: we can cut the pattern we want but it takes us nearly 2 hours on a well tuned up BPT. The neighbor does the same thing with comparable accuracy and in only ~30 minutes with his HAAS machining center.

The key to NOT being disappointed with the results of a project are to embark upon it with realistic expectations while concurrently considering what the machine was designed and intended to do in the first place.

Hope this helps....

[HuFlungDung]

What he said

Good reply NC Cams.

If Andrew McCalip happens to come along here, maybe he can relate what he's had to spend to get his mill retrofit going. I think what he's done should be fairly representative of the least one would spend, starting with a decent used machine.

Every retro I've done, I've spent anywhere from $10k up to $30K, plus the cost of the used iron, which sometimes is not all that much

[Neil_J]

It would seem that you really want a high speed engraver of some sort, especially with an RPM spec of 10K as a "must".

A Bridgeport is not going to run that sort of spindle RPM and if it could, not for long.
The generic bearings used to make the 1119238 oem spindle bearings used in most BPT milling machines are rated at only 9500 max and that is in an un-preloaded "stock" condition.

Yep, sorry, I should have clarified that. I will either be retrofitting the spindle/motor to something high-speed/high torque (3-phase with VFD), OR use the mill as-is to build a seperate gantry-type machine out of steel. Right now I would seriously consider either.

Your 0.003 tolerance should be easily held with rolled thread ballscrews, good servo tuning and properly adjusted gibs.

Good, that's probably all I can afford Would a rolled double-circuit preloaded ballscrew have zero backlash? Same question for single-circuit preloaded?

However, don't plan on doing fast inches/min. while trying to hold tenths in accuracy and repeatability. Used iron with work gibs/non-roller bearing ways won't facilitate it - simply too much slop and/or friction.

Yea, I would be using the high feed rate for rapids and rough cutting only.

The key to NOT being disappointed with the results of a project are to embark upon it with realistic expectations while concurrently considering what the machine was designed and intended to do in the first place.

True. My reason for wanting to start with a decent knee mill is rigidity. (My CNC adventure started earlier this year with a Sherline 2000-CNC.. Taking a 3/8" cut at more than 0.05" or so will knock the whole thing out of tram. When it is cutting, you can feel the entire head vibrating, and see the effects in the cut).

I would start building a machine from scratch today if I had a few hundred/thousand pounds of steel and a steel furnace. I can't imagine building anything in my garage that would be accurate, and rigid, and cheap.

Any suggestions?? Wait for a used VMC?

[jdelaney44]

I would wait for / go find a used Series II with a dead BOSS or other controller and replace it with a PC based controller and the appropriate stepper / servo motor drives. If you want to see my retrofit story it's on this board. Pretty sure I'm at about $8K including the machine.

http://www.cnczone.com/forums/showthread.php?t=8749

Now you said you just want the big Z travel for the ability to change tools? Is that right? I think if you use tool holders for your drill bits you'll avoid the problem. Also, use shorter bits. Also, you'll probably end up doing a lot of "drilling" with center cutting end mills. I have had times when I've had to assemble the drill rig a piece at a time. First the tapered spindle, then the chuck, then the drill........ So yeah, it would be great to have 20 inches of travel and not lose reference.

You could put a DRO on the knee and hand crank it. Or fab up some kind of optical home finding "switch" or something. If the machine you get has a DRO you could move the DRO to the knee. OBTW, the knee on mine has a dial indicator. But I have not found it to be as reliable as I would like.

[Neil_J]

I would wait for / go find a used Series II with a dead BOSS or other controller and replace it with a PC based controller and the appropriate stepper / servo motor drives.

There is a dealer near Daytona Beach that has a few of those, used... I will have to stop by and check them out. What are the main things to look for on the bridgeports?? I know a lot of older ones are worn out, but how do you tell that?

If you want to see my retrofit story it's on this board. Pretty sure I'm at about $8K including the machine.

If I'm patient and do it all myself could I do it all for $5k? If it costed more that would be OK, it would just take longer.

Now you said you just want the big Z travel for the ability to change tools? Is that right? I think if you use tool holders for your drill bits you'll avoid the problem. Also, use shorter bits. Also, you'll probably end up doing a lot of "drilling" with center cutting end mills. I have had times when I've had to assemble the drill rig a piece at a time. First the tapered spindle, then the chuck, then the drill........ So yeah, it would be great to have 20 inches of travel and not lose reference.

That and also the ability to go between large pieces sitting in a vice and small plate material clamped to a sacrifial plate. Using 0.0002" limit/home switches and a Tormach R-8 fixed-Z tooling kit, and an electronic edgefinder, I would basically like to take my 3d part from SurfCAM and have the machine start doing its thing... I want any cranking/locking to be minimal.

At this point though, it doesn't look like I could ever get a Bridgeport to dance at 10,000 RPM, so I don't know what I'm going to do. Can you think of any way to attach one of these to the side of the Bridgeport spindle? (middle) It would do 30K+ and has a built-in pneumatic tool release.

[Hood]

If I'm patient and do it all myself could I do it all for $5k? If it costed more that would be OK, it would just take longer.

What it will cost will depend on what Iron you have in the first place. I have done a manual Bridgeport and am now in the process of Retro fitting a series 1 CNC.
For the manual I needed ball screws, motors and all the rest, the CNC already has the ballscrews and motors although I have opted to change out for modern motors.
My costs are:
Series1 CNC = £400
916oz/in steppers = £270
Gecko G202 = £230
PMDX 122 breakout board = £45
Nemicon MPG = £45
Limit switches (Industrial Hobbies optical)= £85
power supply (could be salvaged from original controls) = £50
Various bits and bobs (switches,wires, materials for brackets, etc) £200
17inch Touch screen = £260
Keyboard with touchpad =£40
Computer = £150
Mach3 software = £85
Time = free
Sold parts of the CNC that I didnt need and got £300 back
total = £1500 = approx $2600

Obviously it will depend on how much you get your mill for, whether you decide to use the old motors and any of the power components of the mill. Also some of the bits are not a necessity such as touch screen and MPG, so costs could also be reduced there.
Hope this gives you some encouragement as it is very rewarding to DIY and another advantage is, if in the future you have problems you should be able to fix it because you built it in the first place.
Hood

[machintek]

Just a reminder that the original NC machines by Bridgeport had the Z axis as the knee using a stepping motor but also making its work lighter by the air assist feature. The quill used the air over oil Spindle Wizard. Consider 4 axis with the 3rd being the quill and the 4th being the knee.

Conversly, I have seen PROTO TRAKS and MILLTRONICS with sliding heads to overcome the quill travel limitation. Maybe, price out one of those to see if a purchase of a used machine would be cost effective. I have seen new MILLTRONICS, enclosed with a tool changer selling for what my R2E4 cost new 20 years ago.
Still, none of these have your spindle speed requirement.

I installed about 14 EZTRAKS at Boeing in Oak Ridge, TN. They had a requirement that each pass a acceptance/positioning test. With a step guage and some effort, I could get these tuned to where they would position/repeat within a few tenths. But, rapid generated heat, and since these ball screws are not under tension, with the ball screw securing bearing on one end, they did grow and it could be measured very easily. A used machine is always a gamble as to what shape it is in.

Some great answers above!

George

[Neil_J]

Still, none of these have your spindle speed requirement.

Would there be any way to mount on of these to the spindle? It does 30K RPM and has a pneumatic tool release....

[jdelaney44]

There's probably a way to rig it off to the side, but that will introduce a lot of flex in my estimation. But for high frequency milling this might be OK?? Not sure.

Have you considered scratch building a gantry machine like the wood routers but for aluminum? I think the router forums probably have folks doing this. You may just be trying to get a Bridgy to go too fast.

[NC Cams]

Follow-up to Neil J: You can put any high speed motor you want on the BPT head BUT, you are still going to be speed limited by the speed rating of the spindle bearings.

My original post cited some specs of the OEM p/n bearings as I have some real familiarity with them. They just won't run/live long at 10K rpm as the system wasn't designed for that type of service.

As far as comparing the capabilities/expectations of a Sherline to a BPT, it really isn't a fair or reasonable comparison. One is a hobby, table top system designed for light duty, small part machining.

The other will cut 0.100 of 0.375 steel plate material (shuddering while doing so) but cut it it will and never raise a sweat doing so and remain straight and true after doing so.

I'd strongly suggest some research into the manufacturer's perfomance specs and qualifications for any machine tool you have your eyes on. The OEM's are generally conservative and realistic for liability and warranty reasons.

However, when it comes to cutting and travers speeds, you can bet that they have been tested and proved capable at the published speeds and a bit more perhaps.

If you insist on trying to turn your BPT into a 10k high speed system, you may find it to be an expensive and difficult if not impossible task - especially if you do not have access to the necessary engineering assistance.

Having formerly been a machine tool bearing engineer, I"ve done some "hot rodding" of my machine using prior knowledge of MT's and bearings used therein. There are places where you can tune up the maching by replacing bearings. But, due to space and design limitations of the OEM spindle design, this is one place where it is difficult and quite costly to turn a Ford into a Ferrari.

My prior suggestion remains the same: try to find a machine that, in factory trim, meets 90% or so of your realistic expectations. For the small percentage of time that you may need 10K or some other minor feature/capability cited on your "spec list", it may simply be often cheaper and faster to outsource the work to someone with a Haas or Mazak or.......

I wish you well with your endeavors....

[Neil_J]

Follow-up to Neil J: You can put any high speed motor you want on the BPT head BUT, you are still going to be speed limited by the speed rating of the spindle bearings.

My original post cited some specs of the OEM p/n bearings as I have some real familiarity with them. They just won't run/live long at 10K rpm as the system wasn't designed for that type of service.

As far as comparing the capabilities/expectations of a Sherline to a BPT, it really isn't a fair or reasonable comparison. One is a hobby, table top system designed for light duty, small part machining.

The other will cut 0.100 of 0.375 steel plate material (shuddering while doing so) but cut it it will and never raise a sweat doing so and remain straight and true after doing so.

I'd strongly suggest some research into the manufacturer's perfomance specs and qualifications for any machine tool you have your eyes on. The OEM's are generally conservative and realistic for liability and warranty reasons.

However, when it comes to cutting and travers speeds, you can bet that they have been tested and proved capable at the published speeds and a bit more perhaps.

If you insist on trying to turn your BPT into a 10k high speed system, you may find it to be an expensive and difficult if not impossible task - especially if you do not have access to the necessary engineering assistance.

Having formerly been a machine tool bearing engineer, I"ve done some "hot rodding" of my machine using prior knowledge of MT's and bearings used therein. There are places where you can tune up the maching by replacing bearings. But, due to space and design limitations of the OEM spindle design, this is one place where it is difficult and quite costly to turn a Ford into a Ferrari.

My prior suggestion remains the same: try to find a machine that, in factory trim, meets 90% or so of your realistic expectations. For the small percentage of time that you may need 10K or some other minor feature/capability cited on your "spec list", it may simply be often cheaper and faster to outsource the work to someone with a Haas or Mazak or.......

I wish you well with your endeavors....

The KaVo spindle I was asking about attaching to the spindle is actually a small-ish high-speed/low-torque type. I was wondering if there was a way to fit it into the taper or attach it to the side and use X/Y offsets. After talking to people, both in the forum and our machinists at work, I'm not sure that's the way to go.

It looks like I will be buying a knee mill of some sort in the next six months, retrofitting it, and use it to start building a decent Gantry/mini-VMC with it... I doubt it would ever be sturdy enough to accurately mill steel, but it would handle my engraving / panel making / PCB manufacturing / aluminum cutting needs.

[jdelaney44]

OH, and on the cost thing, Hood's right, a big driver is the cost of the machine. I could have saved $1.5K right there if I'd have waited and looked around more. Next time I will search every used machinery place and auction I can find in the local area. Being that I live in the L.A. area, that's a lot. Not as much as it used to be.

Getting a CNC machine to begin with is a big cost savings. You get screws, motors, motor mounts, the power supply, and cabinets at the very least and probably don't have to pay too much of a premium for the privledge.

The other cost driver is over engineering. I bet I could save another $1.5K on the next go 'round just cause I know what can be made lighter and how to do things differently.

[Cold Fusion]

I've actually had Z axis issues on my mind, so this is the perfect thread. Some Series II Bridgeports have their Z axis on the knee. While the 15" travel is nice, it's a LOT of weight to move around. The air piston counterbalance helps, but don't expect much speed in your rapids. What I am considering right now is keeping a powered servo on the knee, and adding a gearhead servo on the quill. The 5" travel on the quill will be my live Z axis, with the powered knee as my A axis. During tool changes I'll be able to lower the knee to accommodate longer tools, then lock it during actual machining.


Speed and accuracy don't really go hand in hand on a retrofit. I've got about a thou backlash in my ballscrews, but with compensation this isn't a problem, under 30ipm. After that, things start to stutter and evidence of overshoot is seen. I can hold .001 if I really try, but I have to kick the feed way back. Normally I run near the limits of my 3/8 HSS endmills and can hold .002-.003 without a problem.

[Neil_J]

I've actually had Z axis issues on my mind, so this is the perfect thread. Some Series II Bridgeports have their Z axis on the knee. While the 15" travel is nice, it's a LOT of weight to move around. The air piston counterbalance helps, but don't expect much speed in your rapids. What I am considering right now is keeping a powered servo on the knee, and adding a gearhead servo on the quill. The 5" travel on the quill will be my live Z axis, with the powered knee as my A axis. During tool changes I'll be able to lower the knee to accommodate longer tools, then lock it during actual machining.

yep, I've been convinced that buying myself a knee mill and motorizing the knee is the way to go. Definately post back to the forum if you make any progress... I'll be posting my stuff in a month or so....