Waltham Machine Works Lathe Restoration

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

steamer

Well-Known Member
Joined
Nov 30, 2007
Messages
5,388
Reaction score
29
Hi All,

I posted this on the PM site, but I thought some of you might like to see it...so here it is

Some of you may know of my Waltham Lathe (http://www.lathes.co.uk/waltham/index.html)

The lathe takes a W20 collet. OK they predate Schaublin....so is it really

In reading the 1909 Brochure, it states that the thread is a "Ratchet Type" with a pitch of 1.6667 mm and a diameter "equal to the diameter of the body of the collet"

see page below
Waltham1909catalogpage6.jpg


As I have 1 original collet, I can confirm that the body is 20 nm but the tips of the threads are worn and difficult to measure. It is made of cast iron...consistant with the brochure

Now I scored some W20 collets on the bay recently and they measure thus
ReferenceforW20Collet.jpg


You will notice the thread is 19.7 mm, not 20mm

Well, looking at the sorry threads in the draw bar, I came to the conclusion that the bar had to be repaired to make it functional again. 100 years and soft threads = alot of wear.

I Decided to be pragmatic and thread the bar for the "modern" W20, and see were this all ended up.

I decided to insert a piece into the bar via silver solder and then thread the original draw bar. I could have just made a new draw bar....well ..I didn't.

Making the insert, just prior to parting off. Material is 1144sp

P1040003.jpg


Setting up the draw bar in the lathe with a steady rest. The handwheel end is pressed on and is too big to go into the spindle. Additionally, a shaft collar was made up to grip the bar in the 3 jaw. The collar was turned up in the 3 jaw and marked of the number 1 jaw. Run out at chuck was less than .001". The far end ran true to within the roundness of the bar...which wasn't real good. I spend some time lining it up with an indicator as best as I could.

Here is the bar bored out to fit the insert. I lightly knurled the insert in two locations ot center the insert in the bore of 0.800" with .002 clearanc at the knurl and .003 elsewhere...on diameter The insert is not pushed in flush yet.

P1040006.jpg


I then cleaned, fluxed and soldered the insert into position and fed the solder until rejection.

I then mounted it back up in the lathe and faced and bored the end to final size. Notice the silver line all around showing good penetration.

P1050009.jpg


Now for the thread. The thread is a 0.775 x 15.3 threads per inch 45/5 buttress thread.
To do this I set up a 37/47 metric transposition gear in the lathe thus
the lathe as is
P1050010.jpg


moved the banjo
P1050011.jpg


The gear pod

P1050012.jpg


gear pod installed

P1050016.jpg


Now with a 37/47 in the train and the box set to 12 threads per inch we get

12 x 47/37 = 15.243 threads per inch = .0656" pitch = 1.6663 mm pitch....off by .0004 mm

I won't tell if you don't....:cool:

Threading it was a pain, but doable single point. The compound was set to 45 degrees and backed out to put the cut on. This results in the cutter only cutting on one side...always a good thing.

Here it is finished with a "modern" W20 in place

P1060017.jpg


Now when I went back to check the old collet for fit, it was clear that the thread on the collet was too big for the draw bar. So the sales brochure was correct.

Waltham thead 20mm x 1.6667 mm pitch

Schaublin thread 19.70 mm x 1.6667 pitch

A mystery solved....and my lathe one step closer to running again..

Dave
 
Thanks for the neat write up, Dave. I find the watch/instrument lathes from the MA makers to be pretty interesting.
Though not in the same league as this big (relative term) Waltham, I have an original maker Webster Whitcomb here that I often use for bitsy parts. It's a different kind of turning, and I think I change personalities when I put on an Opti-Visor and grab a graver.

Do you have pictures of your own lathe you can show us?

Dean
 
"Do you have pictures of your own lathe you can show us?"

The headstock and tailstock on Tony's site are of my lathe.
(http://www.lathes.co.uk/waltham/index.html)

I wish I could show you another lathe like it, but in 10+ years of searching I have not found another one. This lathe was intended for export to the Swiss and French watchmakers, very few stayed on this side of the pond. They stopped building them in 1914. This one was a "works" lathe ie WMW used it in their own plant. At least I think so...

My everyday lathe is a 12" Logan.
Dave

.....and thanks for posting!...as I get more of the lathe restored, I'll post pictures...if people want to see it... ;D
 
YES I want to see it, this has been excellent reading so far. Thanks for sharing it!
 
Thanks Vernon....maybe more today....we'll see what is on the Honey do list....

Dave

 
Hi All,

An update on the Waltham.

Now that the VN is back up and running, I can start to continue with the Waltham Restoration.

I need to rebuild and rescrape the cross slide. It had .008" "belly" in it from years of abuse. Thus far I have it flat within about .0015" via rough scraping. I can get this close with an indicator stand, a gage block as a scraping block and a surface plate, but to get any closer and bring the bearing up to snuff, I will need a master straight edge. The problem is the dovetail is odd. 55 degrees. So...I'll have to machine and then scrape a straight edge first. Thus far I am following "Machine Tool Reconditioning" practice. ( pg 264 and 265)

Here's a shot of the roughing out of the straight edge blank.

PB250015.jpg


PB250018.jpg


"Norm"...my VanNorman #12, is happily taking a 5/8" DOC x 5/8" diameter roughing cut in a piece of Durabar.

I next need to cut the bevel. To do that I will need to measure the dovetail , which will require another tool to be made...... ::)

Stay tuned...

Dave


Oh and Vernon...I hope your reading this........ :)
 
Well I got the two masters done..

Wait a minute.....2!

YUP....I needed a master to scrape the angle master.

The problem with the angle master is that trying to mark it results in me tipping it during the mark up. The result of that is an influence on the marking. To get around that, I made up a flat master, and scraped it flat.

I then used that to mark the angle master, as now its a flat on a flat. Much less likely to influence.

OK that went well...here's some pictures of the masters.

P2050079.jpg

P2050078.jpg

P2050081.jpg

P2050080.jpg


Now to mark the part up.

P2050086.jpg


Damn Bubba...we got work to do!
P2050087.jpg

P2050088.jpg

P2050090.jpg

There was a very har line right up the side of the part where the iron was chilled. It was pretty hard and didn't cut like the rest of the part. PITA!

Many cycles later
P2060093.jpg


And still more.

P2060095.jpg


Best I can do anyway.....The marking is VERY thin here....and is just showing light spots.

Parallelism is about .0003" over 8 1/2" of length with respect to the swivel base. Both objectives (20 spots per square and the parallelism tolerance) obtained simultaneously.

Dave



 
:bow: :bow: Wow Dave - great job!

Kind regards, Arnold
 
Thanks Arnold...it's been a learning experience.....and I still have a way to go. It's nice to be able to do even with my mere modicum of ability. I can't machine anything this flat....it's another tool for the box.

Dave
 
Three tenths in 8.5 inches.. Yeah, that sounds pretty good, Dave! Interesting post, this last one. It's been a while since your first. Good to see it pop up again!
 
3 tenths of a thou in 8.5 inches is staggeringly good - excellent workmanship :bow:

Dave
 
The parallelism tolerance is more "tedious" than difficult...Once you start into finish scraping...your only removing maybe a tenth per pass...0001"...you can sneak up on it....its actually more difficult to measure repeatably, as if anything gets between the part and the parallel cylinder I have it mounting on, and it's out a mile!...Then you wonder...did I twist it? is it dirt? did I measure wrong in the first place?...and you clean everything again, wipe everything down with your clean hand, check for burrs and measure again...usually you find things come back to where you thought they should be.....though sometimes ...no.

Dave


Oh and PS...I still have to do the other flat side. It's parallel to this but lower by .106" That is to accomodate the leadscrew nut attachment to the slide. Once they are flat and parallel to each other and parallel to the original leadscrew axis, I can work on the angle surfaces....those are better alread after I rough machined them....they shouldn't be as difficult...

Then I'll us this as a master to scrape in the carriage....This is WORK.
 
oh and the guys at work who have been doing this for 30 years.....they hold .000020" over a couple of feet....

 
Hi Guys,
I have or had 1 1/2 Ames Lath. One has a lever Collet closer and about 15 - 20 Collets.
The second one I think I only have the head stock left. I threw the bed out to the Junk.
Hilmar
 
I think.....I'm done scraping the slide.....I'll fit the nut to the slide now...probably fit a new gib too.....

P5070132.jpg


Dave
 
It's been quite a labor of love, Dave. I'm always glad to see this thread pop up. : )

Dean
 
Thanks Dean for your support!

I'm sneaking up on it. Every step is something new, so you think about it and go kinda slow.

I'll tell you this...these cross slides were no fun to scrape in and fit. I'm glad I'm not doing it for a living!

I can't put any part in contact with a primary reference ( like a known surface plate). Everything is a constructed reference, like a custom scraping master.

All you can do is cross check one way with another, if they don't agree, you have to investigate, and it's not always obvious. If your not in a rush, it can be fun.

Dave
 
Nice work with the scraper Dave :bow:

Without a datum this could be a very looooooooooooong process. Still as you infer, it's a labour of love.

Best Regards
Bob
 
Thanks Bob for your kind words and support!

The flat portions of the dovetail are parallel. BUT.....they are offset from eachother by .106". One is higher than the other. This was to allow for the lead screw nut to peak out the side at the apex of the dovetail and attach to the cross slide. To get them parallel to each other, you have to use the bottom of the compound swivel as a reference, parallel to the surface plate, and then compare the two surfaces, in two directions, to the surface plate with a dial indicator for parallelism, and with a master for flatness and bearing. This has to be carried out this way so that the cross slide itself, which also has this .106" offset, can be compared to the flats just verified It becomes it's own gage. This makes everything sequencial .......and tedious. I'll post a picture to show what I mean....

Dave

PS Here are the pictures I was talking about.

P5080120.jpg

P5080122.jpg

P5080123.jpg


You'll notice the swivel post on the bottom. That's part of the casting. So your parallel riser has to be able to accomodate it. Additionally, the slot down the side tends to make this part move around on you if you clamp it tight....a real PITA.

Dave
 
Back
Top