Retrofit for knee mill

[Chuck Pool]

Hey to all:
I am considering looking for a Bridgeport vertical bill to convert to 4 axis cnc. Maybe a NC machine or a Boss controller machine from the late 1970's to mid 1980's. Is it true these machines are good candidates for this conversion?
Do they come equipped to handle cnc, that is, ground ball screws etc.?
I plan to remove the old controller.
Chuck
poolchuck@yahoo.com

[machintek]

The old NC machines had ball screws on the X, Y, and knee.
The BOSS machines have ball screws on the X, Y, and quill.
Boss 6 and older are stepping motor machines.
None support a 4th axis the way they were made.

George

[jdelaney44]

Well, a way to go is to get one of those machines and put a PC based controller on it. TurboCNC and Mach3 are two I know of. I am using Mach3. EMC is another that is Linux based. They will all support a 4th axis. People often convert rotary tables for this use.

This is not a trivial project as I am sure you are aware. But getting a machine that was originally CNC is the way to go. Way less to do.

Thanks,
-jd

[Chuck Pool]

Jd: Thanks!
Chuck

[Klean-Power]

I have a Bridgeport series 1 knee mill that I am taking to CNC. bought it on ebay for 1500.00
I have to say, you are at the right place here. There are alot of very helpfull and knowledgeable people here.
I was going to spend the 8k and get a full retrofit kit from http://www.cncmasters.com/CNC%20Supr...ofit%20Kit.htm
This includes the 4th axis. But I think that since I have a wife & 3 kids my budget says that I should build my own. I am doing my homework. I will first completely rebuild my machine from the ground up. I will then buy the servo motors and the gecko drives and use mach3 software. I recently purchased Bobcad V20 and I will take my time and get very good at the software while I am rebuilding my machine.
I am a design engineer by trade so I will completely design my system on the cad.
I am hoping to be finished by late August this year.
I can share whatever I do here if you want.
hope this helps.

thug

[Chuck Pool]

Hello Thug:
Yes, please. I would appreaciate any updates on your system. At present, I don't know what to do. There is a big learning curve takig place and I am trying not to make a big mistake. Thug, thanks!
Chuck Pool

[Klean-Power]

Hello Chuck

Well, If you go for a bridgeport it looks like we will both go thru a learning curve together. The one thing that will be able to help us and others is by me fully designing a bridgeport in 3d cad. this should enable us to design and build a retrofit kit best suited for the machine. I will also have a friend who owes me a favor do a complete fea analsys on the parts to prevent any deflection issues.
if you send me your email address I can send you all data that I have weeded thru so far.

my email is thug6891@comcast.net

Thug

[jdelaney44]

Guys, couple of suggestions. If you are machinists by trade, pardon me repeating what you know already all to well and feel free to correct me. I am not, so I am learning all this the hard way.

I've found the words 'precise' and 'zero' are relative terms.

Don't get too tweaked on precision. +/- .005 is perfectly fine for almost anything but journals for bearings. +/- .002 is fine for that. My mill will deflect .005 to .010 with the table extended to one side or the other on a direction change. It's better in the center, but then the center is worn more. Still my parts are pretty darned good since the CNC upgrade.

But I am not building turbo pumps for the Space Shuttle either.

The machine changes dimensions as it warms up during the day. As do your measuring tools. ;-)

My guess is that everything is moving around about .005 to .012 or more, all the time. So bear that in mind. I think some of the claims of precision we see need to be evaluated more closely. A business card is .015.

Design accordingly. If you always design for peg A to fit in hole B you are going to have a lot of work. If you can find a way to do it that does not require a precise fit, it will go faster.

There is NO such thing as zero deflection and NO such thing as zero backlash.

When you get to building, build X & Y first. Then use your functional X Y rig to build Z.

Dedicate as much planning to your electronics cabinet(s) as you do the mechanics. I did the best I could with the cabinet I could get, but I am beginning to fight it.

Figure $500.00 - $1000.00 for measuring tools and another $500 - $1000.00 for other tools if you don't already have a ton of tools for this kind of work. End mills are $10 - $100.

Pay attention to small details. Turns out limit switches and e-stop circuits are a pain in the a@$!

Having said that, everything I have done is working well. Knock...on...wood. ! ! !

-jd

[Klean-Power]

Thanks John

Your right, There is no such thing as zero, I am more thinking of the fixtures and mounts to attach to the machine.
I take your word to heart, you did a nice job and you create nice products.
Thanks for the input.

Tony

[NC Cams]

Sorry, but I have this "thing" for doing bearing fit calcs the proper way. With all due respec to JDleaney's otherwise ULTRA SUPERB mill retrofit thread, +/-0.002" or +/-0.050mm for journal or housing sizing for bearings is rather crude, attrocious actually.

Considering that a bearing only has an internal clearance of 0.010 to 0.020mm at best, you could easily consume the clearance via shaft and/or housing tolerance stacks. Thus your +/-0.050 becomes +/-0.100 TOTAL all things considered. A bit too much for most bearings to tolerate. Negative clearance is not something you're looking to achieve, even by accident....

Actual tolerance and fit is a function of load and desired displacement potential of the inner and outer ring on the shaft/in the housing. Hence, there are al least 10 different fit conditions for the shaft and a similar set of conditions for the housing - each is dependant on load, duty cycle, etc, not to mention materials, thermals and housing/shaft surface finish.

Can't recall but there was a prior thread wherein a link was posted that led to the bearing fit table for virtually any size bearing. Try doing a "bearing fit" search on this website and it should show. With luck, someone will chime in and repost it.

I've seen any number of well meaning folks use the "give it a thou" press" to fit a bearing.

Sadly, this sort of logic falls into the "Gimme a wrench" - "What size?" - "Don't care, plan to use it as a hammer anyway..." category of applications engineering/design.

If you are an engineer and are going to design it in/on cad as you indicate, go the extra step and do the proper bearing fit calculations - you won't regret it - especially when things work right and for long periods. You don't have to grind shafts and housings to obtain fits BUT you can turn and bore them to tolerances better than +/-0.002"

If this is too much of a hassle, maching on the +0.002 side of things for housing and -.002 side for shafts and use (shudder) Loctite.

[jdelaney44]

"Gimme a wrench" - "What size?" - "Don't care, plan to use it as a hammer anyway..."

and

Maching on the +0.002 side of things for housing and -.002 side for shafts and use (shudder) Loctite.

Guilty as charged!

Green Loctite fills up a .003 gap amazingly well! I'm actually happy to see someone else might have tried that one.

It wasn't so much a matter of hassle as a matter of .002 just disapearing somewhere. I really, really tried to bore things right, but it just didn't wanna work. After four paper weights I decided to short it by .001 or there abouts and press it in there.

My point was not so much that this could not or should not be done, but that the practical tolerances of at least my old knee mill can be limited. Especially at the end of the table.

It will be interesting to hear back how much a total re-grind / re-scrape helps in the final analysis.

NC You ARE the bearing man!

[NC Cams]

John

I try, I really do try.

By the way, even RED loctite will loosen and refreeze if it sees just the right vibration frequency - long sad story learned by an OEM I used to work for.

If you have trouble holding tolerance, bore the hole with a ROUGHER finish, IE 125 microinch in stead of 25 micro and leave it a bit undersize. The peaks of the rougher 125 surface will "smoosh over" (highly technical, not often shared bearing technology term) and "self size" to some extent. The grooves hold the 'loctite as well. BTW, there are "sizing rings" you can find that "shim" the OD of bearings to fit in O/S holes. They make knurling tools to "fix" shafts.

If you're boring aluminum, WATCH THE TEMP. IT will shrink a bit when cool. Also, a thick radial will is stiffer than a thin one hence you can run a smidge (another precise machining term) more press on a thin walled part as opposed to a thick walled one.

My ooint is that when you quote a number, there are MANY folks who will read it and run with it. Why? I contend that some folks are NOt interested in learning how to do or are too lazy to look up how to do it right. The old "gimme the answer" as opposed to "tell me how to figure it out".

I'm from the "show em how, let them figure it out" school... That's how I was forced to learn by pencil pouch carrying, white sidewall haircut, old time engineers and it served me quite well. That was however before political correctness came in vogue, however.

Anyway, I misplaced my hammer goto go find a metric wrench to pound some SAE screws into place.....