My 1910 Mitchell "parts car" project

[JV Puleo]

I found an exploded view of the Cadillac water pump in the 1917 edition of Heldt. It looks as if I can fit a modern thermostat inside in place of the original one...which may still work but at 100 years old I wouldn't count on it.

[edinmass]

3 hours ago, JV Puleo said:

I found an exploded view of the Cadillac water pump in the 1917 edition of Heldt. It looks as if I can fit a modern thermostat inside in place of the original one...which may still work but at 100 years old I wouldn't count on it.

 

Joe.....it’s just an old car......they are easy to fix..........shouldn’t take more than ten minutes and some bailing wire.

[JV Puleo]

Bailing wire is hard to come by these days. Just last week I rescued a roll of it from my cousin's fallen in barn where I found the motorcycles (that's really true!). It seemed "too good to throw away." But, I do think the modern old car mechanics rely on Duct Tape and JB Weld. That said, I've actually found a worthwhile use for the JB Weld steel putty...it's just the right consistency for making fillets in aluminum patterns. (Having said that, I've started using wax fillets which are even better.)

 

j

[JV Puleo]

With the OD of the gear and the area that rides in the bushing done I had to trim the ends to match the measurements I'd taken. This involved taking about .065 off the front and back of the piece. The front was relatively easy.

 

 

I also put the chamfer on the edges.

 

 

Holding the piece to do the other end proved easier than I'd expected...I used the hex collet to hold it from the screwed-in projection.

 

 

Both of them trimmed. One came out .002 short but that is hardly measurable.

 

 

I then drilled and tapped these for the set screws that will retain the shaft.

 

 

There are two set screws in each piece. They will align with notches milled in the shaft so once tightened the shaft is going nowhere and, should someone want to renew it in the future it will be easy to remove it and press a new one in.

 

 

There is one more operation aside from the gear teeth...there is a groove 1/8" wide and 1/8" deep located 1/8" from the end. On the original it has a round bottom. I'm guessing that it held string packing to keep oil from the timing case from migrating back through the bearing end of the water pump. ED...is that right and do you think the groove has to have a round bottom? I ask because that will require grinding a special tool but if a square bottom will work it's easy to do with a 1/8 grooving tool. I'd be inclined to put a piece of square braided packing in the groove, in which case the flat bottom will be a little better.

 

Tomorrow morning I'm off to pick up the gear driven dividing head I bought. If all goes well, I'll be back to the shop before noon and can get something done on the test piece for the taper on the end of the shaft.

[edinmass]

Joe.......a square groove is fine, it’s actually a oil gallery scavenging grove.......the large surface area rides in a bronze bearing fed with four oil gallery ports......more of the “White” over engineering.

Looks good! 👍

 

PS- the set screw must sit lower than the bearing surface........obviously!

Edited September 28, 2020 by edinmass (see edit history)

[edinmass]

[JV Puleo]

9 minutes ago, edinmass said:

👍

 

 

PS- the set screw must sit lower than the bearing surface........obviously!

 Yes... obviously it has to. The wall thickness is 5/16. I'll use 5/16 set screws and the notches in the shaft will bring the top to just below the surface.

[edinmass]

I knew you had it covered.......just wanted to show that when one makes modifications you need to think of every detail when you change something. Also, I have had the experience of installing and removing it from the car several times........it looks fantastic so far.......there is a lot of things to consider, and it’s 100 times more complicated than I would ever attempt. 
 

 

PS - I’m running cutting oil in the coolant system for rust prevention after the block cleaning, and for water pump lubrication. The temporary shaft fix is holding so far........👍

Edited October 1, 2020 by edinmass (see edit history)

[JV Puleo]

I drove to rural eastern Connecticut this morning to pick this up...the Brown & Sharpe Universal Dividing Head. The gentleman who sold it thought it dated from the 1950s and was fairly surprised when I told him it was more like the 1890s. This was such a good design that if you do an ebay search on "dividing head" you'll find numerous copies, all of this B&S type. It will need some work and I may have to dismantle it to get the dried grease and other grunge out. There are also several bits missing but as far as I know, everything that isn't there is something I can make.

 

 

Since I got back from my trip earlier than I'd anticipated I started on a "dummy" end of the shaft. The critical part here is the taper. It's quite difficult for me to measure a taper accurately and the measurements I cam up with are close, but not exactly the same as the standard SAE taper for this application so I'm going to make a test version using my measurements that I can mail to Ed. He can test it on the piece and get back to me regrading fit. I think the best idea would be to put some dychem or fine grinding paste on it and rotate it in the fitting...that would show if it's off and where.

 

The first step was to turn the end down to .625. This is the OD of the large end of the taper.

 

 

Then the last 1/2" was turned to 1/2" and chamfered.

 

 

I single pointed the thread about .025 deep and finished it up with a die.

 

 

I threaded it 1/2-20. The actual part has a 1/2-18 thread - a size I've never heard of. Rather than make something that there is only one nut for, it seems prudent to use a thread where the nuts are commonplace.

 

 

Tomorrow I'll cut the taper...I made a special fixture for that and we'll see how well it works.

Edited September 30, 2020 by JV Puleo (see edit history)

[edinmass]

Looks good Joe. The head sure looks heavy! 

[JV Puleo]

I was told it weighs 87 lbs. so it's about the same weight as my big lathe chuck. If it were much heavier I'd find it a chore to shift at the end of the day. This is a good example of a tool you will only use once in a while and you may go years without using it at all but when you need it, nothing else will do the job. I think all of us who fix old (i.e. really old cars) know all about that.

[edinmass]

I have my special Cadillac V-16 tools........I only use them every three or four years.........hopefully, it’s another ten before I use them again!

[JV Puleo]

I still have my box of RR tools for the Ghost & PI...chances are I'll never use them again but you can never tell.

[Luv2Wrench]

My L-W Chuck Co dividing copied a lot from the Brown & Sharpe design.  Looks like your's is built to run from left side of the table.  I think that was typical for the universal horizontal mills so they could cut helical gears and such.  Mine is a bit bigger and a bit heavier and you're correct, it is a real chore to move around.

[JV Puleo]

I has to from from the left if you are going to use the gear train. There will be a bit of a learning curve here but I've ordered a copy of the B&S treatise on gearing - I already have "Formulas in Gearing" which just about makes my head spin. It's no different from learning to thread...after I've done it a few times I'll wonder what all the fuss was about.

[JV Puleo]

Every so often I run into a problem that really requires some thought and today was one of those days. I set up the test pump shaft to cut the taper. You can cut a taper by offsetting the tail stock but that often isn't particularly accurate and its such a PIA to get it straight in the first place that I am extremely reluctant it mess with it when it's right on. I made this this fixture from a boring head. In theory I can adjust the offset very accurately and I was gratified to see that it worked exactly as planned.

 

 

I used the indicator to make certain the piece was perfectly straight and then used it again to set the offset which, I had calculated as .0275.

 

 

Then cut the taper...

 

 

It didn't come out right. I'm sure I made a math error but I've been over this a dozen times today and every way I figure it I come up with the same numbers. Clearly I'm doing something wrong. I tried again doubling the offset...

 

 

The measurements still aren't quite right but this is much closer. I then decided I'd better put it aside for today and think about it. When this has happened in the past (more than once) the answer often comes to me in the middle of the night and I find myself thinking "how the h--l did I miss that!" There is a technique I could use that is more or less fool proof but I'd need the original shaft and it's back in the car doing it's job but I will solve this problem eventually.

 

 

On the plus side, there are some pieces I need for the dividing head to hold the gear, including a reducing sleeve from #10 B&S taper to #9 since I have a set of #9 collets. One of the things I need to do that is a 9 B&S reamer - a very expensive item and I was trying to think of a way to do it without having to buy one. Well, when I went through the reamer drawer I found both a 9 and a 10, brand new and unused. I think they came from the shop where I bought my surface grinder...they were closing and everything was $10, so I bought anything that looked as if I might have a use for it some day. I'd say that has paid off this time.

 

I have another idea of a method to try so I'll make another one before sending them to Ed to test...

 

An afterthought... Ed, would it be too much trouble to measure the piece that attaches to this taper? I'd need the small end, the big end and the length. That might be more accurate than the measurements I took from the shaft since it's almost impossible to measure a tapered edge perfectly. there's no great hurry, in the meantime I'm cleaning up the dividing head. It's frozen with dried oil so that has to be done in any case.

 

[EDIT] I now think I know what I did wrong and how to fix it...

Edited October 1, 2020 by JV Puleo (see edit history)

[edinmass]

Joe, I’ll send you the piece by priority mail tomorrow. Since I don’t have the generator drive shaft installed, it’s just hanging on the end of the water pump. You can see it in the photo below. Best, Ed.

[JV Puleo]

Great. That will make it much easier to get it right. I didn't realize the generator wasn't on the car. With that in hand I can finish the shafts.

 

j

[edinmass]

19 minutes ago, JV Puleo said:

Great. That will make it much easier to get it right. I didn't realize the generator wasn't on the car. With that in hand I can finish the shafts.

 

j

Great thing about the magneto........don’t need the generator charging right now while we work out the electrical system. With the two wire system to the starter, it makes connections kind of funky. We replaced the entire starter wiring system, which is quite extensive and unusual. The generator is tied into the starter.....strange but in this case makes sense. Since we want to steam clean the engine, we left off the splash pans and generator. The latter to protect it from the water. It’s already overhauled as is the cut out, which we also two wires. I figure we will have the engine cleaned next week....and then we will install the generator. Best, Ed

[JV Puleo]

It turns out that turning between centers isn't practical in this application because the overall length of the work piece (it's 11-1/8" for the real shaft) effects the offset and there are limits to how much you can safely offset a piece. So, I've ordered an electronic edge finder and I'll use the "Mike McCartney method" to set up the compound. The taper itself is only about .760 long. The trigonometry Mike demonstrated when he made his C5 adapter should make this relatively straight forward.

 

Thanks Mike!

 

The center turning fixture is perfect for making the B&S 10 to B&S 9 adapter for the dividing head.

[Mike Macartney]

Joe, If I can be of help with anything, please let me know, as I am struggling to do anything apart from sitting at the computer or watching the TV. I have still not found myself a 'project' I can do apart from putting small stuff on eBay. Mike

[JV Puleo]

When I get the piece from Ed I'll take he measurements again and send them to you. Then we can both work out the numbers and if they match we should be right on. I always have second thoughts about figures...though I have to admit that I don't often get them terribly wrong.

 

I hack sawed the chain of the Royal Enfield today in the hope of taking the wheels off as it will have to be carried out of the place it's in. I doubt I'd ever get the wheels to turn because the brakes must be rusted stuck. I probably shouldn't be taking on another project but I'm an addict. I'm going to do this one at home though. There's not enough room in the shop and another pile of parts will only make it worse.

Edited October 1, 2020 by JV Puleo (see edit history)

[edinmass]

Joe.....the part is on the way....sent to the shop. 👍

[JV Puleo]

I did some experiments this morning to devise a better technique for cutting the short taper on the end of the water pump shaft. They seemed to work although thus far none have been perfect. Tapers are an interesting problem...in theory they are easy but in practice cutting an ACCURATE taper is not easily done, especially when you are trying to match something that uses what today would be considered an unusual taper and you don't have the original prints to look at. Even the old machinist's books talk about doing it several times and adjusting the measurements to get a proper fit. I suspect that all the factories that used them also had permanent fixtures so the workmen didn't have to experiment. That's not a luxury we enjoy...but with Ed sending the actual mating part I think we are getting very close. I started on the shafts...

 

As an example of how odd this is, based on the measurements I have the angle of the cut is 2.07 degrees...

 

 

The two long ones are the shafts. The short piece with both ends turned is for the setup and experiments. When I get it right on that, I'll do the finish cuts on the real shafts. I also had an idea of how to make the shaft length micro-adjustable...it's very simple but you'll see it all when I get to it.

Edited October 2, 2020 by JV Puleo (see edit history)

[JV Puleo]

I did the threaded ends of the water pump shaft and the test piece today. The piece that fits on the end of the shaft came in late this afternoon so I had a chance to measure it. It looks as if my original measurements were good – at least good enough to use as the starting point. With the mating piece in hand I can check the fit as I go which is a big help.I will do all the figures over tonight and send them to Mike M so he can check them. I'm pretty sure I have it right but, as I've said before, I don't have implicit faith in my math skills.

 

Not wanting to do that until I've checked all the figures, I started taking apart the Cadillac water pumps I'm repairing for a friend. I started with the questionable one...there was no packing under the gland nut (which I'm sure is why it was leaking badly) but, more significant, part of the housing was smeared with a J.B. Weld-like material. Sure enough, under it I found several cracks and a fitting, critical to being able to disassemble it, is broken off. I'm afraid these pumps have suffered the attentions of Ed's "tractor mechanic" though perhaps that isn't fair to the guys who do a good job fixing tractors. Hack might be a better word. In any case, I have a plan. The original Cadillac thermostat, built into the pump, was toast...it looks as if someone put a screwdriver through the bellows but at least I now understand how it worked and can't think of a good reason why I can't add an in-line thermostat for each pump. It's an unrestored car and the owner isn't even slightly interested in showing it. To satisfy myself, I'll do it in as unobtrusive a manner as possible. I plan to do the same thing for the Mitchell but in that case I'll make a special brass thermostat holder - maybe even attach it to the water return line so it looks "period"...even if it's a device that wasn't thought of in 1910.

Edited October 5, 2020 by JV Puleo
typo (see edit history)

[Mike Macartney]

13 hours ago, JV Puleo said:

I will do all the figures over tonight and send them to Mike M so he can check them. I'm pretty sure I have it right but, as I've said before, I don't have implicit faith in my math skills.

 

I better make sure I put my best'ist 'thinking head' on tomorrow morning!

[JV Puleo]

Today I finally cut the short tapers on the ends of the water pump shafts. I admit that I put this off a few days while I worked on something else... I didn't want to do it until I was relaxed and there were no likely distractions. The first step was setting up the taper. I started with this grinder mandrel and used my offset center and indicator to make sure it was perfectly straight. On the bed of the lathe you can see a 4" gauge block. I set the indicator at zero and a stop at the end of the gauge block.

 

 

Then removed the gauge block and moved the saddle down exactly 4". Based on the measurements I took of the original shaft, and using Mike McCartney's excellent idea of treating it as a right triangle, I calculated that, at 4", the sort side of the triangle would .1448". The saddle was then turned so that I got this measurement...the use of the 4" length is to mitigate the inevitable small fussy measurements. The actual taper is .750 long. It's it's out .002 at 4", its only out .001 at 2" and .0005 at 1".

 

 

I then used a piece of 5/8 stock to cut a test taper. It gave every indication of being good.

 

 

 

Then I cut a taper on one end of my 3/4" test piece. I made a small error here which I was able to correct but it's in anticipation of such things that I wanted to get a good result and work out a system before I made the finished pieces.

 

 

I tested this one with the mating piece. You can see that the striations are relatively uniform...the piece isn't polished yet but it went on with absolutely no shake or wobble.

 

 

 

I then felt comfortable machining the finished pieces, both of which I'm satisfied with. I will still have to put in a slot for a Woodruff key but that's a relatively easy things to do. With this done, the remaining serious challenge is cutting the spiral gears. I've been reading my B&S "Practical Treatise on Gearing" and will get on with making the remaining pieces I need to make the new dividing head operational and can now send the little brass piece back to Ed for the White...with the tapers cut, I won't need it again.

 

In between doing this I've been working on a pair of 1920 Cadillac water pumps but I'll save that for another post. It's a job that should be relatively easy, except that the first pump has obviously suffered from the attentions of someone who really wasn't qualified to do this sort of work.

 

 

Edited October 9, 2020 by JV Puleo (see edit history)

[JV Puleo]

Now I have to clean up the dividing head I bought and make a few bits for it...one of which is a #10 B&S taper to #9 B&S taper adapter for the spindle. Needless to say, it has to be accurate so the experience making the pump shafts has been invaluable. As it is, it's stuck from sitting for perhaps 50 years so I'm also taking it completely apart to get everything free and properly lubricated.

 

 

And, in between working on this (because I always end up needing something and don't want to stop and wait for it...I took apart the worst of the 2 Cadillac pumps I'm rebuilding. I was several hours just getting the two halves of the casing apart...gently because it was stuck with rust, schmuts and old gasket cement.

 

 

It's pretty obvious that the shafts are shot...it looks as if they were running sand in the water.

 

 

And I discovered that someone had done this ¶•ªº braze repair...why, I don't know but it will look familiar to anyone who does this sort of work.

 

 

There is a flanged bushing in one side. Getting it out was an all day job. These pumps have built in thermostats. They were non-operational and  stuck in place. Also, to remove them there was an adjustment fitting that has to come out and someone had broken the hex end off it. In the end, I drilled and bored virtually all of the pieces out of the casting.

 

 

 

To get the remains of the adjustment fitting out I bored it with and end mill...

 

 

And threaded the hole. I've slightly redesigned the fittings to eliminate the thermostat. I'll put in-line thermostats in the hoses, and I'll have to make the new fitting that goes here.

 

 

At the end of the day I was finally able to press the bushing out...you can see what condition it was in.

 

 

Then on to the other side...

I faced off the braze repair but didn't want to go too far because I've no idea why it's there.

 

 

Then, very carefully, bored out the bushing since I've n way of pressing it out from the other side.

 

 

And discovered that the engineer that designed Ed's water pump must have moved over to Cadillac because the bushing for this side is one piece with the housing that holds the water pump packing...this was a surprise ...

 

[JV Puleo]

Fortunately, I've thought of a simple way to fix this.

[edinmass]

Joe......you seem to get all the “easy” jobs. Your going to be the “king of water pumps” from now on! 👍

[JV Puleo]

To me, the ironic part is that I now know exactly how they went together and I'm reasonably sure I can make them as good or better than they were new.

I'm a bit concerned about that messy patch of braze though...I suspect that they literally broke the center boss out of that casting and then brazed it back in place. It's messy looking but I don't think I should do any more with it. The repair seems to be sound and nothing will show when It is back together.

[edinmass]

We have a saying in the shop.......”can’t see it from my house!”

 

[Mike Macartney]

5 hours ago, edinmass said:

. . . . ”can’t see it from my house!”

 

Or,

 

"A blind man would be pleased to see it!"

 

Yet again Joe, this is going to be another one of your masterpieces of repairing a part that many would consider was just scrap metal, to a working example, that is very often better than it was new. Good on ya, Joe.

 

Mike

[JV Puleo]

I'm not sure about "masterpiece" but thanks for the compliment!

 

 

[JV Puleo]

I was back to the White water pump today – or rather the dividing head that I'm getting ready to cut spiral gear.

 

 

 I had to make an aluminum drift to knock the spindle out out, after that, it came apart better than I'd anticipated.

 

 

The primary problem was that it was stuck but ultimately it proved that was just the result of a layer of dried oil. Needless to say, the tolerances in a tool like this are extremely fine. They are a joy to use when working properly but a little dried oil will seize them up. It's so clean inside that I won't go much further...just clean it up as is and reassemble it. Then I have to get an idea what I need for studs to hole the intermediate gears for spiral cutting – those are missing. I ordered some bits ($95 worth) from McMaster Carr last night for the Cadillac job so I'll be bouncing back and forth between these two jobs depending on what machines are needed and whether the setups are similar.

 

[edinmass]

The head looks brand new...........careful not to drop the chuck or head on your foot!

[JV Puleo]

There is some wear on the worm and mating gear that leads me to think it was used by someone who didn't properly engage them because the other gears show no wear at all. It won't have any effect on how it works though. There is always some backlash in these - much more in the gear train. I'm anxious to set it up and try it out...I'll make one or two test blanks and run them until I'm sure I've got the drill down. This is a big step forward. Making one-off spiral gears may be the sort of specialty job I can do on the side once in a while since I know it's very difficult to find anyone willing to make a single, unusual gear. I really don't want to get distracted from my main project but the occasional side job takes a bite out of the constant expense...

 

I found a number when I cleaned the crusty oil off... "1956" ... much more likely a serial number than a date. If I had to guess, I'd say this was made around 1900 - 1920.

[JV Puleo]

The dividing head is back together. There are some small parts that look really hammered but it also looks as if they interchange with my other B&S dividing head so I'll just swap them around...if that doesn't work it isn't a major problem in any case.

 

 

I was just about to leave for the post office to mail the 3-legged, tapered fitting back to Ed when I realized I'd forgotten something important. White used castellated nuts and cotter pins everywhere and the end of the shaft has to be drilled for a pin. With the tapered end and the threads, that is a lot easier said than done, at least as precisely as I'm trying to make these. So, I came up with a fixture to hold the shaft while I drill the pin hole. I'm off tomorrow but I'll be able to finish it on Friday.

 

Edited October 15, 2020 by JV Puleo (see edit history)

[Mike Macartney]

Joe, I saw this video on the Hobby Machinist forum and wondered if it may be of interest to you. Cheers for now, Mike

 

https://www.hobby-machinist.com/threads/how-is-this-guy-cutting-a-spiral-bevel-gear.87678/

[JV Puleo]

That is an even more complicated problem. On a spiral bevel gear the grooves between the teeth aren't exactly parallel. They are slightly wider at the outside of the bevel.  Even the B&S book that covers the procedure says it is impossible to do with extreme precision on a horizontal milling machine because, to get the widened grooves, you have to finish the teeth by hand filing. There were machines that could make them...something like a Gleason Gear Generator and I know there were some dedicated B&S Gear-making machines. I once bid on a gear making machine. It was set up and working and had a huge amount of the necessary tooling with it. I was the high bidder at $150 but the seller cancelled the auction before it ended. That was wise...it was such a specialized piece of equipment that e-bay was not the place to sell it. I did hear that it later sold for about 10 times that amount, still a bargain if you knew how to use it. With that one machine you could set up a small business making custom gears.