Another New Toy

[DWRoller]

.019 / 1.414 = 0.0134 ?


Just guessing, but it doesn’t look like it would drop more than the amount of metal removed. With the angle it moves some laterally and some vertically so it would move less vertically than the amount of metal removed... wouldn’t it?


Danny

 

[strokersix]

Noob here reading this thread with great interest! I have a 1954 11" Sheldon 56" bed that is in need of repair. It looks like someone attempted to fill craters from dropping chucks on the ways with electric arc welder and a grinder. I wish they had just left it alone. Fortunately its the inner ways near the headstock so it mostly looks bad but is not much of a functional problem. I'd like to do a proper repair someday.

Suggestion to address sled drive smoothness: You could drive the sled with a long strip of synchronous belting and timing sheave instead of roller chain.

 

[rke[pler]

I'm not sure if you remember, but your suggestion when I did my 10EE was to use the leadscrew as a reference to the base of the saddle. That allowed me to use the unworn portions of the saddle where it connected to the apron and the cross slide ways to get the height and level right (for some definition of 'right', anyway). From there I was able to estimate the wear with feeler gages, but not really do much in the way of figuring if one side of the V dominated the wear picture since there's really no alignment front to back.

You might be able to do something of the same with shim stock and such on the flat and Vs until the saddle is the right height over the leadscrew. That would give you the cut at the ends of the saddle and let you set it up on the mill.

 

[beckley23]

Russ,
That's how I've been checking the saddle/leadscrew relationship..
Anyway, I got my act together this morning. The most plausible explanation for why I wasn't able to reconcile the numbers was in the leadscrew thrust bearings. I had to pull the leadscrew in order to get the apron off, and instead of reattaching the leadscrew the way it's supposed to be, I just inserted the leadscrew in the bearings so that I could get the saddle/screw relationship. Then it dawns on me that those bearings are either taper roller or angular contact, and need to be preloaded, thus the readings I got yesterday and much earlier when I ran the same check. Maybe a minor dope slap was earned, but I'm now satisfied that I've explained some, if not all of the .017", even though I still didn't preload the bearings. I did find .008-.010" elsewhere.
Harry

 

[beckley23]

Did a little math today to figure how much to remove from each face before epoxing the Multifil. I'll clean up the sketch and post a picture tomorrow, in order to lessen the confusion of the next paragraph.
The Multifil is .030" thick, I'm allowing .005" for the epoxy, and .006" for scraping stock. For the inside face; .006" stock + .012" wear in face + .007" for amount scraped off bed + .005 filler= .030" Multifil thickness. .005" epoxy + .005" filler= .010" to machine off the face, or .01414" straight down(this is the number I really need due to the way I'm setting it up on the mill). For the outside face I need to remove .021" from the face, or .029694 straight down, using .008" wear factor for the slide face. I'm not figuring any wear factor for the outside face of the V way. For the rear flat slide, I'm not figuring any wear factor on the slide or the way, so a .030" material removal will used.
Since the vertical head is still on the mill, I'll machine the flat slide first, then remove the head, not an easy task, it weighs 300 LBS approx. The V slide will be done with the horizontal spindle. The #4 Cincinnati was chosen because of the 18" wide table, and the carriage is a 25" square.
Harry

 

[beckley23]

Here is the sketch for the inside face of the V slide. A sketch for the outside face would look the same, but with some changes there is no bed scraping, and the amount of wear in the face is different. One term may be a bit confusing, "filler". You can call it gap filler, but it is Multifil 426, and it's there to account for the .030" thickness of the material. I could eliminate it, but that's an extra .005" to scrape off. If you're wondering about the amount of scraping stock, it's there because it is very hard to get all the bumps out in the application, and I need some of the scraping stock to align the cross slide ways. This stuff is not CI, it is Teflon impregnated with bronze and etched on one side for adhesive, and it is very easy to scrape.

The carriage is on 3/4" X 1" CRS parallels, cut from bar stock.

Sorry, the camera batteries need charging. The cutter is in the cut on the rear flat slide. The carriage bridge is approx 8-1/2" wide, and from measurements I took after the cut was completed, there was no deflection. The carriage was indicated in using the rear vertical face, where the TA mounts and the bottom surface, where the rear gibs attach, and there were no shims used for alignment. I was pretty much stuck with this set-up, in order avoid interference between the clamps and the spindle.

Harry

 

[beckley23]

Milling the outside face with a 45* cutter. The inside face has been previously done. I made the extended mandrel so I wouldn't be butting up next to the column. I've got a pair of these cutters that I use in gang set up on an in house job.

The Multifil has been epoxied, now it's watching the epoxy cure. Thankfully it's the end of the day, and this should ready in the AM.

There are 3 sections of weights on each slide, here's the reason. I put lightly greased tubes in each of the oil holes to keep the epoxy out.

The Multifil hasn't been trimmed yet. I use a wood chisel, pushing against the CI base to trim. This is the V slide and you can see the 4 tubes sticking out of the oil holes. The last time I did this, on the CK in 2003, I didn't grease the tubes, and had a bit of trouble getting the tubes out. The grease seems to have done the trick, all the tubes came out easily.

The last trimming operation is done on the mill, cleaning out the V slide's groove. 20 HP mill is bit of overkill, but it was the easiest to set up, including remounting the vertical head, which goes on a whole lot easier than it comes off.

The 3rd spotting cycle. The red medium is being used because it's easier to see on the black Multifil as the scraping progresses. After this was scraped, I did an alignment test with the cross slide, and it's time to start some selective scraping in the V slide. As it is now the CS is about .004" convex.

Harry

 

[strokersix]

Thank you so much for posting these photos. Fascinating! Please continue.

 

[beckley23]

A small progress report.
The picture below is about the 20th spotting, as you can see a lot of progress has been made. I now have a 0-0 reading on the cross slide, actually it's been fluctuating between 0-0 and .0003" concave for the past 8 or so cycles. The balance of the cycles have been to improve the bearing area, and I'm trying to the drop front approx .002" for cross slide parallelism to the bed. I did a couple more cycles after the picture and the bearing area has been further improved, especially on the flat slide. Also consider that the total bearing area of the slides has been increased approx 20%, due to the continous surfaces of the slides. Originally, there was a gap, in the surfaces, under the bridge.
If you look at the V slide, there are 2 yellow dots. These are reference markers I used to tell me which end of that surface got scraped to turn the carriage for the cross slide alignment. It gets very confusing when the carriage gets turned over, remembering which end to scrape. Each of those faces got scraped from the dot to a couple of inches past the bridge.

Harry

 

[beckley23]

3 more scraping cycles on the V slide resulted in the lowering of the front approx .0015". I ran the accuracy checks on cross slide alignment, 0-0, transverse level .0005"/12", high in front, L-R level, low L approx .001"/12" and carriage/leadscrew relationship is 3.250" +- .002", this is the dimension I was looking for earlier. I can try to improve the L-R level readings, but everything else is good to go. The next thing I have to do, is cut the oil grooves in the slide faces, then this carriage is going back together, and I take some test cuts.
The carriage/leadscrew check. This is the method Russ and I were discussing earlier.I found out the problem I was having earlier was caused by the gear key riding up the keyway slot, not allowing the leadscrew to seat properly in the angular contact bearings, removed the key and the problem was solved.

Rechecking the cross slide alignment. You can also the area where I was spotting the carriage.

Harry

 

[beckley23]

At Last

The lathe has been completely reassembled and test bars cut. Considering the machine is 55 years old, and what I just finished on the inside face is far from a complete reconditioning, the results are very pleasing. What follows is what has happened since my last post, in the order it happened.
The oil grooves have been cut using the end of a round file ground flush. The cutting action bears a very strong resemblance to the scraping motion. The patterns of the grooves were measured, recorded prior to the milling of the slide faces, and then drawn out on the Multifil after scraping. The machine was then reassembled.

35" and 14" test bar cuts after the reassembly. Unfortunately, I have lost my notes on the recorded dimensions, but for the 35" bar there was .0014" differential between the high and low. IIRC, the biggest difference was on the TS end, with the middle being relatively constant, and a minor variation on the HS end. I could have worked withese results, but I decided to do a modified "2 collars test" on the 14" bar. A .004" difference, small at the HS, convinced me otherwise.

Before I could get to the scraping, I had to use the lathe for a job, turning a 10" D X 22" long piece of aluminum. I was the only lathe I could load the piece in, I can't get my little forklift in position to load the 16" CY. Please note the small piece of plywood C clamped to the TA guard in an attempt to contain the coolant. At the speed I was running, it was mostly successful. The turning accuracy was very encouraging, the diameter didn't vary over .0005" the entire length.

The 14" test bar mounted, and the inside face has been spotted. I'm concentrating my efforts on the section(s) nearest the HS. At this point, I'm very close to being finished.

This is the same spotting as above, from a different view. The spotting may look like I'm hust starting, but it's due to the way I'm recharging the medium on the straight edge. The is spotting in the nearest "bare" area, but it is extremely faint.

The final 14" test bar results. The results from yesterday afternoon are; HS 2.2016", 4"- 2.2015", 7"- 2.2018", 10"- 2.2022, 12"- 2.2024", 14"- 2.203". Remeasured this AM the only changes are; 10"- 2.2018", and 14"- 2.2028". Maybe there is a reason the 2 collars test is only 12" long.

The final 35" test bar results. The results from late yesterday afternoon are; HS 2.2427", 6"- 2.2425", 12"- 2.2423", 18" 2.2427", 24"- 2.2427", 30"- 2.2426", 35"- 2.2427". Remeasured this AM; HS 2.2423", 6"- 2.2423", 12"- 2.2423", 18"- 2.2422", 24"- 2.2424", 30"- 2.2426", 35"- 2.2425".

Keep in mind the remeasurement points are not the same as the first points, but very close approximations. I could move the micrometer in between and get slightly different measurements.
There are still a few minor things to be done to the lathe, an oil seal replaced on the gearbox, a rear splash guard, etc., etc., but for all intents and purposes I'm done, and ready to get back to the "Wreck".
Harry

 

[Steve in SoCal]

Harry,

This has been a epic saga, I guess you don't need a gym membership after all that. The wreck may seem like a small project after this?

Steve

 

[Cal Haines]

Excellent results Harry! Do you have any idea how much time you put in to the whole project, from the time you brought the lathe into the shop?

Cal

 

[beckley23]

Cal,
A WAG would be approx 300 hours. I don't keep track of time on these projects, I use these projects as filler in between the paying jobs.

Steve,
In many respects the "Wreck" is, or will be, more difficult. The only saving grace about the "Wreck" is the physical size, the SE 60 was pushing my limits, weightwise. I'm not really set-up to easily do what I did.
Harry

 

[Pete F]

Another Noob question for you Harry. The Multifill you're using for the slides, how does this compare to the original material in terms of wear? Also, I'm assuming that there is a minimum thickness in its application, so you couldn't, for example use it instead of shims in the TS? Or maybe you could if you machined the TS base down to provide the minimum thickness, built it up with Multifill, then scraped that flat again.

I can only agree with everyone else here this has been a terrific read and thanks for taking the time to put it up here. Makes my little "project" look like a bit of a joke you'd knock off on a quiet afternoon

 

[beckley23]

Multifil 426 is a Teflon way material that has been impregnated with what appears to bronze particles. One side has been chemically etched to accept epoxy adhesive for attachment to other materials. It is available in thicknesses from .015" - .125".
It is very similar to Turcite and one grade of Rulon.
http://www.garlockbearings.com/products-detail.asp?MaterialID=58&Group=19&LangID=2
One definitely does not want to put this stuff on the bottom of a tailstock, don't ask how I found out. For shimming tailstocks, I use sheet shim stock placed between the top and bottom castings, as I did to this lathe's tailstock.
Harry

 

[Pete F]

One definitely does not want to put this stuff on the bottom of a tailstock, don't ask how I found out. For shimming tailstocks, I use sheet shim stock placed between the top and bottom castings, as I did to this lathe's tailstock.
Harry

Thanks Harry, I was thinking more along the lines of using the material between the TS base and TS top, in other words where you're presently using the shim rather than on the TS slides. If the multifill is too thick, machining the top of the base down slightly.

I've read comments from people here that say that shims are the work of the devil himself (which of course wasn't going to stop me using them ), and the only "proper" way to get a low TS on centre was to scrape the HS down. Absolutely zero chance of me scraping the HS down so just wondered if a fill material like you use could be a compromise?

 

[beckley23]

Any of these bearing materials would have to be scraped. Personnally I would not use this stuff for the TS application to build height. I have successfully used steel shim stock in this application. I've got shim stock in my 16" CY's TS, which I've had for 20 years, and haven't ever had a problem. I attempt to figure how much I have to raise the TS to match the headstock, and order the appropiate thickness. Generally I like to get my TS's about .0005" to .001" high.
I know what Connelly, and others, say about shim stock, but I'm not about to scrape a headstock if I don't have to. I think where people get into trouble is using multible layers, and improper/sloppy fitting. In all the years I've been a member here, I've only seen one comment about the heresy of using shim stock, and that was last week.
Harry

 

[DWRoller]

I have two lathes that I added laminated shim stock between the top and bottom of the tail stock and it seems to work okay.

http://tinyurl.com/ye7ft6o

I like it because you can buy it a little thicker than you need and peel it down in .002-.003 increments to the exact thickness you need. I would like to know what people who actually know what they’re doing think of this fix but for my purposes it worked really well. It cured some huge problems. (When you start with 1/8” shim stock you have huge problems, right?)

One lathe had the bed scraped in 1986 and something akin to Multifil 426 glued to the bottom of the tailstock. I wasn’t used very much before it wore thin and started breaking off and folding up on itself. After I cleaned up the mess the tailstock was almost 1/8" low.

 

[jockofthelowveld]

36 inch test bar

Harry;

re: the 36inch test bar in post #191 above. Did you turn the test bar with or without a follow rest?

Best Regards;
Steve