JDs BP J-Head CNC Conversion Blog

OK, well I'm over a year into this so it's a little late to get started, but I have pretty good notes and the machine is there as a record of what's going on.

I've decided to post here as I can keep it all in one thread -vs- the deal on Yahoo where it's all chopped up. I can post photos in line here as well. That will add to the context. Some photos later.

Here's where I am at.

First off what am I doing? I want to enter the insanely competitive world of kitchen gadgets to supplement my income. We live in "The OC" and things are just crazy expensive here. Everyone I know has something on the side going.

Why on earth would I do this? Well, I need a business partner and after 20 years of marriage I think my best bet is my wife. She likes to cook and loves kitchen gadgets. So, given that everything is competitive, this seems like a way to go. Maybe not the way, but a way.

2nd, these are relatively low tolerance products. +/- .005 is fine 90% of the time. Heck +/- .01 is OK some of the time. You don't have customers Micing a bottle openner.

3rd, design is important. I have a knack for design. I get 50 design ideas in my head every week. It's easy for me. Something I do well by nature. I also know that if I am successful that I will be ripped off, but I don't care. I'll just keep moving. My angle is to keep changing it up with short run products. If one of them goes hot, I'll license it to someone who does the high volume thing and keep going.

Ok that's the plan, we'll see how many ways I can screw that up.

The machine:

Given all that a three axis CNC machine seemed to be a natural fit. Part of this is about buying some U.S. made stuff so I decided to go with a Bridgeport.

So on about December 4th, 2003 I found an '82 J head on e-Bay and bought it. This was after visiting it at a more or less local machinery dealer and hour away. I'm convinced I paid too much. $3425.00. 48 inch bed, power feed on X, and a DRO, which I am not using at present. In good shape. Everything works pretty good on it. The power down feed needs some adjusting and there's something goofy in the vari speed, but it runs and is square. Not to bad for a Noobie I suppose.

It was delivered buy a nice guy named Gabriel. He hoisted it off his truck and rolled it into my Garage where it has been holding down the concreted since.

BTW - I have a Grizzly G1005 mill drill too. That was my first mill. I learned all about rigidity on that one. I also have a 7 x 12 mini lathe. Again, not too rigid. Turning AL is OK. Bronze starts to get interesting and steel is a real crap shoot anything better than +/- .005. On the Bridgy even I can eek out +/- .001 and I think I'll get better with time. So mass counts for a lot.

So now for the CNC stuff.

All the motor calc stuff went right over my head and I just did not have the patience for it then. I may yet be paying for that decision. But I winged it with some Reliance Micro 100 motors I found on e-Bay. 1/3 HP. 100 volts. The 100 V is going to be the issue.

I then got the Gecko drives and the Campbell / Sound Logic break out board. ( http://www.geckodrive.com/ http://www.campbelldesigns.com/ )

I have spun these motors succesfully under Mach2 and will go to Mach3 when I get back 'round to that. So I'll be writing about that real time.

One of the first things I decided to tackle was the electronics cabinet. I found a surplus cabinet at Youngs Surplus. ( http://www.youngssurplus.com/ ) It came with a bunch of relays and stuff in it too. Quite the find. It is big enough to house my mini tower clone PC, the DC power supply the relays, and the Geckos.

I did a whole DC power supply on my own. That's a whole story in and of it self. More on that later.

At this point the Bridgy has yet to be turned on. This 2200 pound paper weight was just a nice home to a bunch of electrical components. So how to power it up with my Dryer circuit? Well I did the whole research dance with phase converters, changing out the motor, rigging some crazy jack shaft thing, and the VFD. I bought a Leeson VFD in the end. I've never been happier with a power tool in my life as a result. The VFD not only takes care of the single phase to three phase problem it makes operating the machine a piece of cake. I can't wait to see how cool it can be with Mach2 controlling it.

Fortunately it powered up and ran good enough. I was afraid of some $1500 repair bill lurking in the J-Head. It was a little noisy, but once I got it greased and oiled it settled down. Before I go full production I need to take a deeper look in there.

Then I tackled the ball screws. I went with Rockford. I am happy. Rolled screws were fine given the products I want to make. Replacing them was a chore. My Dad helped me. It can be a one person job, but having someone there for safety is important. That table is really heavy.

We did discover that if the table is supported with some adjustable roller stands that it does not have to be taken all the way off. You can slide it to within about six inches of the end and avoid the whole hoist thing. We did the hoist thing on the first atempt.

I had purchased a shorter Y-screw to use for the Z-Drive and mistakenly installed that first. Then I ran the nut off the end of that and we had to redo the job and re load the other screw. Photo's of that eventually too. Not as bad as I thought.

So at that point I had power, computer, motors spinning, and X-Y screws installed. Yay!

It is now November 2004. Thanksgiving weekend. Now to mount the motors. I'm thinking now that I'll get the project in under a year.

Then a major set back. My z-Drive design was totally wrong. I was going to bolt this big ol' 1/2" AL plate to the tram bolts and mount everything on there. No go. WAY too much material and WAY too much flex as a result. I also started burning time making the bearing blocks.

This is when I started to learn about machine work for real. All of the sudden my world became about getting something, anything, within +/- .002. Well, several micrometers and several paper weights later, I can at least make the parts I need to.

So I am currently redesigning the z-Axis. Well that's all for today. Hopefully I'll get some pics and drawing and such posted here shortly.

Just added pics of the electrical cabinet and the big z-Drive plate of design number one.

Very interesting so far! A true labor of love. How much money invested so far? And then add to that a figure of your hours put in and attach a monetary value to that as well. I understand that you must enjoy the challenge, but when you add it all up, if you had just went out and bought a small used 3 axis knee style CNC machine, you would be making parts to sell right now. I have many projects laying around my house (and in the driveway) and to the regret of my wife, I think I will get them all done in a reasonable amount of time. It has not been so for years. But keep us posted. Have you looked at some of the designs already out there? You mentioned the aluminum plate. BPT (on the original Z drive EZTRAK design) used a 5/8 inch steel plate sandwiched between the head and the knuckel as a mounting surface for the drive components. That required longer "T" bolts to hold it all together. The latest design was made by ELROD and mounted to the front of the head. Very similar to an old BANDIT design I saw many years ago.

George W.

Money so far? Well, $3200 for the machine, $400 for the motors, $300 for the power supply after all my screw ups, uh.....misc stuff $1,000.00, break out board $100.00. Kurt vise $400.00. Cutting tools $500.00. I'm not done yet $2000.00. Wozt that? About $8K.

Yeah, after I value my time, I will probably be over the cost of many other options. But my time is only worth something if I can make money doing something else AND if I have the cash on hand to buy the machine. So it's a hard decision at that point. Do you try to save for the machine? I prefer to get going on the project then I have a drive to put the cash into that instead of the 18 other things I want. That's my oddness.

Heck a new CNC'd BP is $27K. Not that much really. I looked for used CNC machines. I probably should have looked a little longer and harder. Anyway.........here I am. The other objective here is to learn how to do this. I'm getting plenty of that.

Well, today was a bust. I'm working on the lower bearing block / mount for the z-Drive. I've decided to go Elrod style and grab the pads for the quill engagement lever and the quill speed. So it'll be a three point mount. I've been working for a week on the knuckle mounted bearing block. Today I over bored the bearing journal by a hair. I may be able to save it yet with a further over bore and a sleeve. Done that a couple times. :-) Also, I discovered that the top of the screw is going to hit the oil fitting up top. So I may need to back off the screw some. Which means a whole new part. Well, it'll take less time on the second go 'round. Could be worse.

In the attached pics you can see the block from the top and the bottom. Note the 5/8" allen bolt. I reduced the thickness of the cap on that and counter sunk it into the block.

A few more pics.

One is setting up the 7x12 to turn the ball screw.

The other is the VFD control panel.

Old School Power Supply Build Pics.

So the chaos on the dinning room table there is the operational circuit after I got it all bread boarded. Note the toroidial transformer on the right and the big 18,000 uf caps up towards the center. In between is my high voltage bread board. I just got a bunch of binding posts and mounted them to a panel. I did not like the idea of hot 120v A/C flopping around.

The next pic is a close up of the circuit that controls the relays. Note the heat sink on the zener diode. In the end after much circuit tweakage it cooled down but by that time I had already done the heat sink and a little fan to keep the sucker cool. :cool:

Then there is the final assembled and installed supply. There are two toroids on this supply. I stacked them one on top of the other.

Finally there is the circuit with a parts list.

Warning, warning, warning, this electrical stuff will start fires and can kill you. OK, I feel better now. Seriously, be careful. 18,000 uf caps do some crazy stuff. :eek:

Why is this Old School? Well I chose to use relays to cut in the resistor that controls the in rush current. The higher tech methods are to use an active circuit with an SCR or a thermistor. The thermistor seems to be the component of choice for these unregulated supplies. But I didn't want any residual resistance hanging around so I went with the relays. I could not figure out the SCR until much later. Next time maybe.

So the last problem I had was that the bearing block was over bored and the bearing at the top of the screw on the Z-Drive was hitting the oil fitting for the J-Head.

After looking at it some more tonight with the ball screw nut shells loosely in place I determined I could reduce the length of the screw by ~ 0.5 in. and still be OK on the travel.

So I put it back on the mini lathe and turned it down. I'm getting a lot better. I managed to get it right! 2nd time on that one. I've determined that I need to use a "rule of on half". That is I take off half of what I need to once I get within about .010 of the diameter I am looking for. That's a quarter of the change in diameter I am looking for. Given my flexy bendy mini lathe that allows me some room incase it does something stupid when it hits a hard spot in the metal. It also gets me to the point where I am taking off less and less with each cut thus stressing the lathe less and making for a more accurate cut.

I also got started on the new bearing block. Just got the next chunk of 4 x 1.5 bar stock cut on the band saw. I'm going to make the new block just so I feel better about getting it "right". I could loctite the bearings in there and it would work, but I want to get the boring process wired better. Practice makes perfect. I'll start milling the new block tomorrow night, maybe.

Pics below.

In the first one you can see that it now clears the oil fitting. It will be ~0.25 lower than that when I remake the bearing block.

Second pic is the partial assembly with the nut shells on the screw. No balls in there, just the brush seals holding them up.

Third pick is the bar stock getting cut. Real exciting that one! :tired: But if someone has never seen a metal cutting band saw, it might be interesting.

Got it done! Had one more "paper weight event" and a near miss but got it done. The process was about the same.

1. Cut the block and face all the sides to make it square
2. Drill the bolt hole (I bored it last time, drilling is close enough)
3. Notch the block
4. Bore the bearing journal

The boring was the hardest as it has the real tolerances. I ended up with a pretty good press fit. I needs to be pressed in but it's not too hard.

The bearings are 1.850. The hole ended up being between 1.849 and 1.851 depending on where you measure it. I was shooting for 1.849. I've seen this on about eight attempts to bore things to fit bearings. The lathe is more accurate by nature I have determined. On the mill +/- .002 is fantastic.

What I am doing is stopping at about 10 thou and then going a thou (.001) at a pass from there. On the boring head and the lathe that means half a thou. Of course the graduations are .001 so the .005 is a bit of a guess.

The tooling marks are often about .003 deep from what I have observed too. A .002 cut will get most of the previous pass. .003 will get all of them. Again on the lathe or boring head that's .006 net.

I got my hands on some tooling for free. A guy I worl with who used to be a machinist full time before he got into computers gave me a box-o-stuff. One of the tools was a freshly sharpenned 3/4 inch roughing end mill. I have been using 1/2 in finish end mills. This 3/4 inch roughing tool can really remove some stock. I was able to take out about 0.40 at once and I think I could do more. Note, make sure the collet is really tight. It pulled itself into the work about 1/16 inch during the cut.

I'm going to get a 3/4 inch smooth end mill. I think this will help the stiffness of the rig and make for better cuts where I can use a 3/4 in. tool. Anyway, it was a really enlightenning experience.

Along the way I fixed up some splash shields for the coolant. This is working pretty well now. I ran across some pieces of plex that I got for free. It's actually too thick at 1/4 inch. 1/8 inch would be better. It doesn't need to be bullet proof.

I then used my $20.00 heat gun from Kragen to warm the pieces up and bend them into "L" shapes. Now they stand up and can be clamped together in any number of configurations.

I also leveled the machine. It was just sitting on the garage floor. The coolant was running to one end and the back. This caused overflows on the coolant channels on the Kurt vice. With it level the coolant is running off the vice much more evenly now. Big difference. Still getting some on the floor, but a lot less.

The KURT is one of the best investments I have made in the past two years.

To go with is I just got some 0.5 thick parallels. These and the vice have cut down my set up time by probably hours a day. That's a lot when you only have weekends.

It was pretty easy. Just used a 24" crow bar to pry up the front or back and put some aluminum shims under the corners to make it level.

I was looking at a Tree Journeyman mill on e-Bay. This was the first time I noticed some channels cut into the edge of the table on a CNC mill. That would be nice. The new Bridgeports have the same thing I think.


Pictures:
Some pictures of the bearing block

The KURT with the parallels holding up the block before tightening

Milling the notch in the block with a 3/4 in. roughing mill.

Thank you for sharing your effort jdelaney44, nice work.

Anyone one that machines and does not get a little pumped up over your work is one of two things, already done that or dead :D

The true BPT CNC iron such as on the BOSS machines do have the coolant "gutters". The series 1 machines and the converted knee machines (EZTRAK and V2XT) do not have this feature. Unless you have an enclosed machine, you will have coolant on the floor. Even if you have an enclosed machine, it may still leak.
It looks great so far! I wish you well.

George

George & Ken thanks for the positive remarks. I'm glad to hear my work is of interest and that it looks good from there. As a new guy it's hard to tell if you are on track or not. These forums really help a great deal.

Equally glad to hear that coolant on the floor is normal. Well....maybe five gallons of epoxy paint and mop would be the next logical step. The garage would look prettier too!

Best,
-jd

Just a quick hint. A long time ago I was a R & D machinist in a prototype shop with linoleum square tile floor. It turned black rather quickly from oil and chips. I found that FUTURE acrylic floor polish on the floor AND THE MACHINE worked great. The chips slid right off the machine, and the floor lasted a lot longer before it turned black.

George

Been busy this week. Sorry I haven't posted. Work, family, and my HOA. Jeez, too much going on.

The Z-Drive is taking shape. The top bearing block is done and the side plates have been surfaced. I've got some centering issues between the two blocks. Oops. But I can fix them. Just about got the boring thing down pat. No paper weights on this one!

Temporary diversion on some coolant problems. I got too much oil in it and it started to stink real bad. I've introduced an aquarium bubler and a little bleach and things are OK again. Thanks to the folks at DIY-CNC for advice on that one. I found a $149.00 skimmer called the Skimpy. It's on order and should be here in a few days.

http://www.skimpyoilskimmers.com/4436/23932.html

Should take care of the small amount of oil I have to deal with.

I ordered some of those Loc-Line splash shields and modified my coolant sheild rig in general. Trying to reduce the bulk. I bolted a couple of pieces of plex right to the KURT vice. That was a big improvement. Steadied things up. There are 1/2" bolt holes standard on either side of the jaws. Seemed to be a perfect place to bolt up the splash shields. So between that and the Loc-Line shields I think I'll be even better off. This coolant thing can be messy. But I'm gaining confidence cutting at higher speeds and now the coolant is becoming a real requirement.

Also, found a local guy to sharpen up about a dozen end mills I have collected. 1/2 in and 3/4 in. Charged me $50.00 for the whole batch. Seemed good based on his rate sheet. They look OK to me. I'll try them out in a few days.

No time to post pictures now. This weekend will be all family stuff, no machine work.

Well, it looks like I'll be spending around $8K for my Bridgeport Clone conversion also, but via a very different path.

The basic machine cost me $250 (the head was making noise and the brake didn't work, but $100 worth of parts fixed that).

Today I ordered:
1 -- Elrod X, Y, Z ballscrew and axis conversions with replacement yoke (around $4500).

2 -- pico systems drive electronics (say $600).

3 -- $175 worth of connectors

4 -- Capacitor from Mouser (around $40), 1000KVA transformer from Plitron ($112.) Rectifier and bleeder resistor from Digikey.

I'll scounge a case and maybe a computer -- or get a used computer for $100.

I'll be running EMC (which is FREE).

I'll start assembly when the ballscrews get here -- next week, I hope.

I've got to fix the lighting in the shop so I can get some good pictures.

Ken

Left out the servos -- $600 for reliance motors.

(Also, it sounds like jdelaney44 is having a lot more fun, but I need the machine in a hurry.)

KL

Ken, sweet deal on the machine! I wish I had shopped more or bargained harder. But $250.00 is a stroke of divine intervention. I'm jealous! :D

Anyway...some additional pics.

Current stat of affairs with coolant control
The two blocks on the screw
The parts for the Z-Drive so far
The material before I started

On milling the plate. Took two passes on each side minimum. The first pass was to relieve the stress. It was curved to begin with. It was curved differently after that. The next pass(es) began getting it true. The amount that metal can move by simply removing the outter layer has always been amazing to me.

It's easy to understand why coolant is important with dimensional stability too. Getting a piece like this hot would really deform the metal under milling and yield very odd results once it cooled.

Happy Easter!

-jd

Here is a tip from an old machinist who had a S1STD on a vinyl tile floor for 8 years where he worked making prototype molds and machine parts. The floor turned back very quickly due to the cutting oil I used (typically applied with a small acid brush). We have all been there. But we were part of the engineering department and we had to look good for visiting customers. Here is what I found:
USE FUTURE ACRYLIC FLOOR FINISH. After cleaning the floor. Apply the Future. Believe it or not, it worked so well that next the machine was cleaned and Future was applied to it. The chip slid off! The oil stopped staining the floor and machine. True that eventually the Future had to be reapplied. Fly cutters and walking on chips can damaged a finish. But try it!

George

George, thanks for the tip on the Future. One of these days I'm gonna tile the garage floor. That'll help the cause.

So, this weekend is going to be a bust too. Got paper work to do. BUT, last weekend was great. I got the major parts for the Z-Drive nearly done. It is now taking shape. See photo.

The tooling I had sharpenned is working great. Alpha Precision in Huntington Beach. When they are sharp I am getting great results in the 1018 steel. The carbide work he did was good too. As long as I don't screw up they are staying sharp.

I crashed one of the 3/4 in roughers into the work at like 20 ipm & 1200 rpm & .30 deep. Little too much too fast! Things got red, sparks flew, stuff like that. Oh well.

But everything is fitting, is coming together square, and looks OK.

You can see in the picture that the ball nut is not together. That's just for layout purposes. But that's what it's going to look like.

When I get this done I'll post some kind of drawings with the as is dimensions. They are all wierd 'cause I'm designing as I go here. That saves a little time, but not much. You end up having to work to what you did on the first part. So all the odd dimensions just follow you around.

More progress on the quill drive.

This weekend saw me get holes made for some rolled pins providing some alignment at the corners of the frame and holes drilled and tapped for the bolts. 3/8 in bolts and 1/4 in pins were used. This is the first time the whole thing has been bolted up to the Bridgeport without any clamps holding it together.

Pretty stiff as is. I'm going to attach the top of the frame to the machine via the 3/8 in studs that hold the top and bottom half of the J-Head together. Need longer studs. Hopefully next weekend I'll get the brackets made for that.

Pictures below.

Got the "chasis" for the quill drive done tonight. Instead of attaching to the old quill feed lever on the right and the quill feed speed shift housing on the left I went for the 3/8" studs that hold the head together. Not sure I've seen this approach before, yet. Nothing new under the sun usually.

Major accomplishment for me. I got all the parts made, they fit and the screw turns fine. Now onto the motor bracket and the link between the nut and the quill.

This week's experiment with materials was hot rolled steel. Not sure of the alloy. Behaved better with respect to deforming when one side was milled or cut.

Pictures of the mounted drive chasis below.

Never mind........... Questions? Please contact me off list.

Thanks,
-jd