Kevin - A V Carroll Horizontal Mill Rebuild

[durableoreo]

Started on the adapter. I don't know what's taking me so long. A bunch of new operations, maybe, lots of unknowns. Anyway, here's the plan:

1. Center drill, step up to 0.250, 0.400, 0.500, then bore to 0.650
2. Dust off compound, set to 12 degrees, and bore ID that will accept 3C collets
3. Set compound to 20 degrees (to match spindle) and machine OD
4. Turn flange
5. Part off
6. Heat treat O-1 and draw to R_C 58 (click here for history and details of the alloy)
7. Make a split-ring collar to set the depth in the chuck, holding the part by the flange
8. Grind the OD (check with hi-spot in spindle)
9. Install the adapter in the spindle with a tiny amount of Loctite 609 under flange and clamp with the draw bar
10. Grind ID in-place

The chicken scratching below didn't scan very well but it'll give you the general idea. So far, I've only done step 1. I did manage to find my compound, too.


Setting the angle on the compound accurately is my next problem. I decided to do the trig method where you measure the legs of a triangle. The ways are 1 leg and the other leg is perpendicular. I'll move the carriage x along the ways and adjust the compound until the tool post moves by the right amount in y. The tool post will be square to the chuck for this operation. Whatever I use for x, y = x * tan(angle). Tangent isn't too crazy for small angles so I hope this works. Other methods might be more accurate. If I don't get a good print, I may need to use a different method.

For now, I need some carriage stops to control the length of the x leg and I'll use an indicator against the took post to measure the other leg. I was going to buy a carriage stop but one of the first things I saw was this 3-d printed thing and he wanted 45 $ for it. Seems like a lot. Also, this is one of those basic projects that you're meant to do for yourself. So I decided to make a pair.





It's un-filled ABS with a 3/8 hole for drill rod or an indicator and a 1/4-20 clamp for the dovetail. The hinge is integrated into the body. It's 50% filled, which is fairly stout. I bought knobs from McMaster and some "press-fit" threaded inserts. After shipping and the extra parts, it was 42 $ for 2 stops.

BTW, I'm not a fan of the 3-d printing hype and various trinket printing that goes on. If I had a working mill, I would make this out of brass or aluminum and feel smug about it. But for now, the plastic version will have to do. I do not intend to over use the FDM printer (partly due to all the clucking) but I've been surprised at where it can be used effectively. I'm keeping an open mind while simultaneously resisting expectation of The Internet by not printing Groot. if you're quietly thinking about how this could be better done in metal---you're right. And I'll get there someday. Possibly by the end of this build.

 

[durableoreo]

Well, it's been a minute. Had some family stuff to deal with. But I escaped to the machine shop today.

The current plan is to make an adapter so my 3PN spindle will take a 3C collet. The spindle tapers at 20 degrees and 3C tapers at 12 degrees so it should all work out. Same diameter body, too, about 0.655 in. The adapter is O-1 steel. I've been using carbide insert tooling, mainly a boring bar, and getting along well enough.

After looking all around, I found my compound. I set the angle with some angle blocks from some budget manufacturer.

Here is the part. It's not tested against the spindle yet.

Later I inked the spindle and checked the fit

This is what I found (below). The pattern is not symmetric, so there is some wear in the spindle's surface. Not the best contact. Had to fiddle around for quite a while to straighten this out.

Here's the final fit, which shows better contact.

You can't see it but the adapter sits fairly high in the spindle taper. That flange does not sit against the face of the spindle. It's about 0.100 in proud. So I need to do a few more passes. But I was anxious to see if I could cut the taper angle correctly...


Here's the other side of the adapter. Had some trouble parting off so I eventually used the hack saw (for character) and dressed the edge with a file. Note the finish. I later mounted it in the 3-jaw chuck and took another cut to clean it up. But before that, I tested the fit against the 3C collet.

You can see the ridges from the adapter on the collet.

After monkeying around with this for a while, I remembered that I planned to do the final turning of the ID in situ. I'll just mount my compound on the mill's table. Come to thing of it, I don't need X or Y working to finish the spindle---only to lock in place.

As of now, the axes are not quite what I want:

• X lever
• Y lead screw
• Z lever

I had been planning to convert the levers (rack/pinion) to lead screws. But I've been watching Rotary SMP videos and now I wonder if I should convert to ball screws.

 

[durableoreo]

My first attempt failed. Turns out that I trimmed too much off the outer surface AND my angle was too large for the internal (3C) taper. It's hard to see but when I pressed the collet into the arbor, the adapter spun freely. The 3C taper was only engaging at the smallest part of the taper, which was actually on the surface of the arbor.



See the gap... The flange sits flush on the arbor's face when I push it in. It also makes decent contact on the taper of the arbor. But the collet bottoms out in the arbor and the adapter floats. See how the collet extends beyond the adapter (red dot above). Also, the angle inside the adaoter is too large so the top of the collet does not engage with the large end of the adapter spins loosly when the collet is pressed in (below, using the tail stock to apply pressure).


I measured the 3C taper I had cut. It's not a very good measurement. It depends on corners contacting the taper and the gages were almost too thick to fit through the part. Tried 24 deg, which was loose at the top. 23-1/2 deg was worse, 24-1/4 deg was better. This is consistent with the previous observations.



I cut another part. This time I had the hang of it and it took about half as long. I didn't bother trying to get the flange to seat against the arbor's nose and I took care to set up the 12-degree angle accurately. Also, I left a bit more meat in case I need to touch up the internal taper. I did cut back the face of the adapter so that the collet stood proud even when pulled in.



To make sure I didn't have the same problem as before, I pressed in the collet and checked if the adapter rotated. There's a set screw preventing the collet from rotating so I wasn't able to check contact in the arbor.


I was going to harden and grind the adapter but now I'm not so sure. I wish it was done, which makes me want to skip steps. That kind of laziness and impatience guarantees a bad outcome.

Also, I'm thinking about the risk of hardening a part with thick and thin features. But that's a fear more than a reason. If I get a bad warp, I'll just make another part.

To do the grinding, I got a fog-buster type system. I also ordered an air compressor (which I've needed in the shop for a long time). I already have a router type toolpost grinder with an AXA mount, various abrasives, soluble-oil coolant, etc. Even with all the equipment on-hand, grinding on the lathe will be a stretch. Maybe that's why I've got half a mind to call it done prematurely.

 

[durableoreo]

A few years ago, Tubelcain posted a series of videos about a drill-press vise. I made the vise but it basically broke my have-file-can-do attitude. Since then, I've been trying to bootstrap a machine shop. I bought this old mill because it seemed less expensive and a million voices on the internet have been preaching the gospel of old iron since I was young. Now I've been working on this mill, in fits and starts, for a year, and with little progress except for what I've learned. But I'm getting too old for this. So I bought a Rung-Fu type mill. It's the G0755, which is like a big brother to the popular G0704.



With this, I'm hoping to remedy the get-me-by things I've done. For example, I made a solid tool post for my lathe. It literally had saw-cut marks on 2 sides and mill finish on the others. Now the bottom is scraped, the sides are machined, and the edges are contoured so chips don't collect under the tool holders. It's de-burred and I even touched up the surface a bit with a Renzetti-type magic sanding block. It could be better but it's one of my first machined parts.



I've got to get the DRO installed. Have a butterfly-type tool changer coming, too. Then I'll tackle some parts for the horizontal mill.

The grinding project is still in process. I finally got an air compressor but it was damaged in shipping, didn't come with oil, etc. I have the parts for a fog-buster setup so I'd like to get that going for grinding the 3PN --> 3C adapter.

I've also been improving the 9x20 lathe. I added deep-groove ball bearings in the cross-slide assembly, which is a nice improvement. I also scraped the saddle to the ways. Here's the before and after photos. When I started, I was working in cramped quarters and with makeshift fixturing to hold the part at the right angle. After a while, I used one of those portable workbenches with a custom-cut 2x4 fixture. It was a huge improvement. The finished job is not perfect bearing and the cross slide is not aligned well. It it off by 0.005 as the cross slide travels 4". But this is too much to scrape in with my level of experience. At the time, I only had a single 1-1/8" wide scraper to this was really difficult. Now, despite it being mediocre bearing, it FEELS so much better.


After scraping the cross slide flat, top and bottom, and parallel, I tried to clean up the top of the saddle. (Didn't get a picture of the bottom of the cross slide.) It didn't go very well but I was getting quite tired of scraping at this point and was torn between alignment and fit.


I also worked on a 3" vise that I had purchased for the horizontal mill. It was around 100 $, off Amazon, HHIP or some such Chinese vendor. It needed some de-burring and the bottom was not quite flat. Then I made the rails flat and parallel. Then I scraped the moving jaw. Then, to my irritation, I found out that fixed jaw was not the same height as the moving jaw. But it turned out fairly well and I've used it on the new mill, despite the comic size mismatch


I also purchased a Kipp adjustable-tension knob for the carriage lock on the lathe. It was so nice that I wanted one for the tailstock lock. Unfortunately, nothing off-the-shelf would fit, so I made one myself. Here it is with cold blueing. I painted the knob (which is 3D-printed) black.


I was also inspired to improve the way wipers on the lathe. Robert Renzetti did this crazy rebuild of a D-bit grinder a while back. He made a wiper with a sharp edge which caught my eye. So I did that for the flat and V ways. Originally, the rubber was screwed directly onto the casting. I used the FDM printer to make some enclosures out of PLA. The rubber was then trimmed to make a sharper edge. There is a also 1/4" felt backing with an oil hole at the top of the enclosure. Those are installed and seem to be working. I also added silicone guarding (Renzetti style), which keeps chips off the ways and leadscrew.



One of the failings of the import lathes is the carriage gibs, which are steel plates with tortured arrangements of screws and set screws. I noticed that mine don't even touch the underside of the ways. So I'll eventually need to make tapered gibs. Like this, maybe:

https://www.hobby-machinist.com/thr...h-tapered-gib-conversion-on-mini-lathe.14406/

And speaking of gibs, the straight gib in the cross slide is a crime against machine fitting. It's like throwing the proverbial hotdog down a 2-lane highway. The gib is half the width of the gap AND it rotates badly under the pressure of the adjustment screws. I'd like to make something that's a little better sized. I'd even consider modifying the casting to accommodate a tapered gib. But for that, I might just buy a new piece of cast iron and make a new cross slide.

So, with a big mill to bring in the new year, cheers! My New Year's resolution is to make some money this year with these tools. Cheers!

 

[jwmelvin]

I haven’t really followed this thread but I’m glad I looked at it this morning! Good stuff. It takes some effort to document your work and I appreciate you doing so. I’d really like to do some scraping, like you’ve done to your vise. Please keep posting as I’ll be interested.

 

[jwmelvin]

Oh and can you post a picture of your silicone way covers?

 

[durableoreo]

Oh and can you post a picture of your silicone way covers?

Sure, see the photo below. Renzetti said McMaster had a rubberized cloth-backed material that was 0.020" thick but I couldn't find it. This is cloth backed silcone, 6" wide, (3635K64). It is barely wide enough to cover the ways and lead screw. Wider is better but the material is fairly expensive and I didn't know if it would work. I backed the carriage up to the tail stock to measure. Next time I'll use a shorter piece. In the picture, that's a 6" chuck and there's no problem working close to the chuck although there are now some rubbing marks. I'm planning to make a MT3-->3C collet adapter soon, which puts the carriage very close to the headstock but with no chuck in the way, there might be enough space for some folds.



Oh, and you can see the solid tool post mount before it was machined.

 

[jwmelvin]

Thanks. The folds and bunching has kept me from trying this style but I’m still letting it kick around in my thought soup.

 

[durableoreo]

I finally hardened the 3PN-3C adapter.

In the past, I would use a torch and try to judge temperature by color and quench in cold oil and use my wife's oven to temper. It's possible and perfectly serviceable for the occasional part but the uncertainty and inconvenience are too much for me. You see, I want to be a real machinist some day. So I picked up a Hot Shot 360. Today I spent about 6 hours breaking it in and heat treating my first part.



I followed the recipe for O-1. From memory, pre-heat to 1200 oF, heat to 1500 oF, quench in 120 oF oil, and temper at 1000 oF for 2 hours. The part should temper to R_C 35 or so. From what I read, that's what pre-hardened 4140 bars are sold as. You can cut it with a file if you bear down. I'm not sure if this is the right design choice for this part but it's a starting point.



Out of the oven, it look fairly bad. I might put it in the lathe and try to touch it up a bit with an abrasive pad. I considered tumbling it with abrasive media... that's probably better for my fingers. The one I have is cheap and loud loud loud. It's so damn loud that I want ear protection just to get from the door to it's power switch. But I'd rather not leave it overnight so that will have to wait for tomorrow.

The next step is in-place grinding. I've been working on this for some time. I got a Dewalt compressor---the one with 2 small horizontal tanks and a manifold between them. It was damaged in transit, I think. After making some repairs, I proceeded to completely overfill the oil and run it for a few minutes. I stopped it to dink around with the regulator. It wouldn't run after that. Lots of foam on the dip stick. I drained the oil to the correct level, waited a while until the air bubbles had dissipated, and tried again. It ran for a few seconds and then stopped. I have not idea what damage I've done but I'm hoping that oil will drain back into the sump overnight and it will work in the morning. Let me know if there's anything else I should do.

When I have compressed air, I can run the coolant system as I grind the ID of this adapter. Have a bunch of parts for a pressurized-reservoir import kool-mist type system. Cool-Moist.

 

[durableoreo]

Some advice on what to do when bearings don't fit tightly on a shaft. This guy had the some problem but he got professional advice, which was plating metal on followed by grinding.