My 1910 Mitchell "parts car" project

[JV Puleo]

I think he's already left...

[edinmass]

28 minutes ago, JV Puleo said:

I think he's already left...

Next time.......👍

Edited November 18, 2020 by edinmass (see edit history)

[JV Puleo]

I finally cut the slots in the end of the Cadillac pump shafts today which sets me up to cut gears next week.

First they were faced off and trimmed to the finished length. I'm not keen on cutting a face with so much of the work piece sticking out of the collet but the length was critical and I had to use a stop in the collet to make sure they came out identical.

 

 

I decided not to put in the flats for the set screws on the second shaft because until I take the other pump apart I don't know what I'll find...I may need to use different measurements.

 

 

Then 5/16 slots were milled in the ends... I went .025 deeper than the original shaft because I don't have the car here to check it against. slightly too deep makes no difference. Slightly too shallow would be a major headache.

 

 

You'll notice the new shafts are longer than the old one...this is because of the changes I made to the inside of the pump. They have more bearing surface now than they originally did.

 

 

I also drilled out the last remaining grease fitting hole and primed the other casting.

 

 

I'm away tomorrow so I wanted this done today so it could cure over the weekend. There is one piece left to make, a plug for the top of the pump to take the place of the water connection you see here.

 

 

These pumps had a connection that circulated hot water through the carburetor jacket. This was a compensation for the poor gas that was available in 1920 but it's not only not needed today, it is actually detrimental since, with the new ethanol-laced gasoline you certainly don't want to heat the gas. I'll make an aluminum plug that will look official ...

 

 

Edited November 18, 2020 by JV Puleo (see edit history)

[JV Puleo]

I hadn't planned to come in today but the project for this morning took a lot less time than I'd anticipated so I decided to make the plugs for the top of the water pump. Unfortunately, in order to grip the piece and have room to work it has to be about twice as long as it will be finished...

 

 

Not surprisingly, I thought of a better way to do this after I'd bought the stock and started the job. Still, the aluminum bar is relatively inexpensive - its actually the time it takes that makes this a lot of work for a very simple piece. I turned it down to fit inside the top of the pump housing...

 

 

Then flipped it around and took the excesses off... It would have been easier if I'd put a hole in the center and turned it on a mandrel or arbor but I don't want to have to plug that hole.

 

 

So, one down and one to go. When this is done, I should be able to assemble the first pump and, of course, I now have the parts to fix the second one.

[JV Puleo]

I trimmed the other plug first thing this morning and then set up the small dividing head to drill jp;es for the attaching screws. Rather than measure it, I used the original water connection.

 

 

Then drilled the holes...

 

 

 

Because my cam is round, it partly covers the holes for the bolts that attach this to the pump so I set it up in the mill to cut a half-circle relief for the heads of the bolts.

 

 

 

I also painted the other casting. Now I have a few small things to do... I have one gasket to make but I'm going to try using o-rings for the other two connections. I suspect this will work just fine and I don't like the idea of "gluing" the pieces together with gasket cement. I like to be able to take things apart if I have to...

 

 

Edited November 21, 2020 by JV Puleo (see edit history)

[JV Puleo]

Yes, that is precisely how I view the situation as well.

[JV Puleo]

There is an excellent book titled America Adopts the Automobile. It's not a "car book" in the conventional sense and I suspect it was the author's Doctoral Thesis but I've found it extremely useful in assessing what was really going on in the industry in the early 20th century. Yes, around 1910, for a short time at least, the luxury car market was saturated. The middle level makers faced a huge problem in that the very rich has plenty to choose from so increasing quality was not an option, even if they could have done it. The only market open was, as you say, the hardware store owner and other small businessmen and building a reasonably "big" car for a price they could sell it for in that market was a challenge. Aside from the chassis itself (which is pretty light), I don't see much on the Mitchell that is as poorly made as the engine so I also suspect that they were well into the design and production of the 1910 cars when they realized that they couldn't meet the necessary price point and took the path of cutting corners on the engine. Interestingly, the advertising material must have been in preparation before the design was finalized because in several places it does not agree with the final product and in a few others details are  left out.

 

Heldt makes it clear that the technology for making a better car was available and much of what we take as being the result of not knowing the best way to do things was, in reality, a matter of expense. They knew how to do it, but couldn't reconcile the cost with their goals. Also, cars were not expected to have a long life-span, at least not by modern standards so they could realistically ask themselves "will it last 5 years" because, if it did, it would be virtually worthless by then in any case. None of the great makers, Pierce, Peerless, Packard, Locomobile, RR and a few others thought this way but they dominated the "price is not an object" market. Ed's White is a great example...made by a company that was easily capable of making a fantastic car but logically assessing the market and realizing that they could do much better with trucks - which had to be durable because they were expected to work hard and last longer.

Edited November 22, 2020 by JV Puleo (see edit history)

[TonyAus]

For those interested in an informative read the author of America Adopts the Automobile 1895-1910 is J J (JIm) Flink, then Professor of Transportation and Cultural Studies at the University of California.  Some of his other transport related stuff is also worth chasing up.  Unfortunately I understand that his interests subsequently moved on to the history of jazz.  

[JV Puleo]

My plan is to assemble the first Cadillac pump...then move on to the gear cutting. I'd like to get the parts of the Cadillac pump off the bench and put aside so this morning I shortened 4 of these stainless cheese head screws to hld down the cap I made Friday.

 

 

Then started the assembly. Rather than a paper gasket - which would be a devil to cut neatly, I'm using a thin, flat 0-ring that is only about .120 thick with some RTV silicone "form a gasket".

 

 

This part went smoothly...I also put in the Zerk fittings and the drain,

 

 

To seal the shaft, I'm using 2 layers of 3/16 square, graphite impregnated pump packing and a modern seal...I suspect the packing would do the job just fine so the seal is just added protection against leaks.

 

 

Here I ran into an unexpected snag. There is a brass ring that presses down on the packing. When I tried to insert it, the shaft is slightly out of line with the housing. I suspect this is related to the braze repair. I know the shaft is mechanically straight because I lapped the bushings together. It wasn't out much, probably only about .010 but enough so that I have to make a new insert sleeve. I'd just turn the old one down a bit but the ID is too worn for it to grip on the expanding arbor.

 

 

Little as I like surprises like this, it really didn't take much time... I may add another layer of graphite packing too as I think there is room if I shorten the insert a bit.

 

[JV Puleo]

I finished the new retaining retaining ring for the pump packing and assembled the pump this morning. I then mounted it on an angle plate and filled it with water. This isn't a real test...but if a leak does show up it might as well be fixed now. To my surprise I did get a leak - a tiny one and not in any of the places you'd normally expect. It's weeping a tiny bit of water at the junction of the bronze piece and the main casting. This should be easy to fix as it just needs a better gasket.

 

 

 

I'm going to let that stand for a few days. Terry offered to make some gaskets so I'll get the measurements to him...

 

I then went on to the gear cutting by cleaning up a lot of the bits around the shop, putting things ways and taking a  look at the foot stock.

 

 

I hadn't examined it closely before. It was gummed up with old, hardened oil so I took it partially apart and got it working again...not a "restoration" but functional.

 

 

And mounted the test gear blank on a mandrel in the dividing head...

 

Edited November 24, 2020 by JV Puleo (see edit history)

[edinmass]

Looks like black magic to me! ✌️

[JV Puleo]

It's nowhere near as bad as that insane Gemmer worm gear...I haven't a clue where you'd start to make one of those and I suspect, just as you have said, that it was designed to be made on a highly specialized machine. I am curious as to what the machine was and who built it. It's not impossible that there is another one around - perhaps not even being used or used for something else entirely. I once bid on a fantastic gear making machine with all of the tooling of about that vintage...I was the high bidder at $150 but the seller was approached by someone who actually knew what it was and what it could do so they cancelled the auction. I wasn't all that disappointed. I've no good place to put it and don't really want to go into the gear business but I hated to see it go for scrap.

[JV Puleo]

With no one else in the building today, it was quiet and I had the plat to myself, perfect for switching the mill over to horizontal mode. The first step was to unbolt the vertical head as swing it around to the right. This thing weighs about 400 lbs. so it would be a real challenge without the built-in crane. Even with the crane, it isn't something I do very often. One of my goals is to get my vertical mill back together so I can leave this one in horizontal mode.

 

 

It is then bolted up to the holding plate that serves as a cover on the right-hand side.

 

 

Then I fitted a 1" arbor with the involute cutter for the gear teeth.

 

 

There is a chance the arbor is too short to accommodate the 20-degree swivel. If that proves to be the case I have a longer arbor but the shorter one will be stiffer. Having done this, I realized I have a problem. The table does not travel to the right far enough to cut the teeth. I can't move the dividing head (because it's attached to the driving gears). I'll have to insert the water pump shafts so I have a much longer piece to work with...and mill the teeth from the opposite end. That raises some issues that I want to check with a friend who has a lot more experience with gear cutting so I'm putting this aside for today. There is no question it can be done. I just want to do it the best way.

 

The White people would have had the same problem since we can be pretty sure the gear was cut on a similar machine.

 

Edited November 25, 2020 by JV Puleo (see edit history)

[JV Puleo]

To deal with the fact that the test gear blank is too close to the dividing head I made this mandrel. It's actually a piece of the feed rod from a long dismantled lathe...a piece of 3/4" shaft with a key way.

 

 

Threaded 3/4-16 on one end for a nut.

 

 

 

And a key way broached in the blank.

 

 

This setup approximates the water pump shaft. It's critical that the blank be held tight because if it were to move, the teeth would be ruined. There is a set screw collar on the shaft for the nut to tighten everything against. In this position I center it under the cutter...a fussy operation that worries me...

 

 

After it was centered I swiveled the table 20-degrees. It just made it with the short arbor. This is the first time I'm moved the table like this.

 

 

There is more to do but it's near the end of the day and I'd prefer to do the rest, and make the first test cut, when I'm rested and my mind is clear.

 

 

 

[edinmass]

Joe.........you missed this parts car! 

[JV Puleo]

It's going to need more than some Diamond tires. What is it? I looks as if it may be an earlier Mitchell but I've never seen one and I'm too lazy to look through my old MoToR magazines for an ad.

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

[edinmass]

1909 Mitchell........should make a good parts car if you can find it.

[JV Puleo]

Completely different engine...I don't know about the chassis parts. I have a set of jugs and one valve cage for one but nothing interchanges. The parts came with mine...All the jugs have cracked water jackets and one has a piston stuck in it but if someone needed them they could be saved and used. In any case, "they are too good to throw away" and don't take up much room so they'll sit until I find someone who needs them.

[edinmass]

Interesting.......I was guessing the one year difference wouldn’t have had many changes. 

[JV Puleo]

The 1910 Mitchell was slightly more than half the price of the 1909 but wasn't smaller. It was a drastic attempt to appeal to a larger market. I attribute most of the cut corners in the production of my car to that. You can't built a car of the same size, with roughly the same umber of parts for half the price unless you make some compromises and it looks to me as if they made lots of them. The new, cheaper design must have had a lot of problems too because it was only used for two years, 1910 and 1911. In 1912 the engine was completely redesigned again...so the '10-'11 cars are a design unto themselves. There probably are chassis parts that interchange since there is evidence they bought some of those out but never having seen one, I don't know. I do think the engine took the worst hit...in retrospect, I almost regret not finding and fitting a better engine but I think when I'm done, what I have will be fine. This is a good example why "as the factory made it" is largely pointless if you want the thing to go down the road reliably...they were probably suitable for 30MPH on dirt roads in Wisconsin and 99% were worn out and junked by 1915.

[edinmass]

Joe........when you're done with it, it will be fine. The new oiling system, better machine work, and better materials will make it a decent runner. Once it's running, I'll come up and help you dial it in........maybe we can do a few tricks to make it go better. 👍

[JV Puleo]

Thanks Ed. I'll look forward to that. I expect there will be a lot of adjustment after it actually makes engine noises.

 

 

[JV Puleo]

Today I reached the moment of truth on in cutting the spiral gear. I added the two additional gears to the gear train and addressed a couple of minor issues with the dividing head.

 

 

And took a test cut...it wasn't right and I couldn't see why but a friend stopped in and made a suggestion which proved to be the answer. There are times when you can't see the nose on your face...I've often asked people who knew nothing about what I was doing (my late father was one) to look at the job and see if anything didn't look right. Quite often, they see something I'd missed completely. Actually, I made three errors in the setup but when they were addressed the cuts looked very good.

 

You can't see them here because they are on the other side off the blank.

 

 

I could have gone on but I was so pleased with how they looked when everything was right that I decided to make another test blank and tomorrow I'll try to get a perfect one...I confess that if it does come out right I'll save it...probably as a paper weight on my desk!

 

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

[weathered1]

Joe, Let me guess as how this works. So as you crank the table the spindle rotates so many degree's thru the gear reduction. What is the function of the dividing plate?

Thanks 

[JV Puleo]

I'll try to explain it...but keep in mind that until a few weeks ago I didn't understand what was involved.

 

Think of the gear teeth as a giant screw thread. The teeth aren't actually straight although this isn't immediately obvious when you look at the gear. They twist slightly. The amount of twist is called the "lead" and it is a function of the number of teeth, the diametrical pitch (DP) and the helix angle of the teeth. This gear has a DP of 8, 20 teeth and a helix of 20-degrees. The lead is something like 23 inches...(I forget the exact number) so each tooth would make one turn in that length.

 

The dividing head does two things...the gear train twists the gear as it is milling each tooth. When the cut is done, you stop the machine and move the cutter back to the starting point. To line the gear up for the next tooth you lock the dividing head plate in place and turn the handle twice. It just so happens the dividing head has a 40:1 reduction ratio so 2 turns accounts for 20 teeth. This is the simplest number to cut. If it were more or less teeth it would be a bit more complicated but, in any case, when the next tooth is aligned, you unlock the plate and make the cut. When it is running the dividing head plate turns around so it's quite something to watch. If I knew how, I'm make a video of it...

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

[edinmass]

Ok, I have five years of higher education from a very good university...........I have been a professional mechanic  for forty years, and while I understand this, I could never get it set up on my own. 

[edinmass]

AJ should come to the shop and make a video. 

[JV Puleo]

And, the math involved is way beyond my capacity. Fortunately, this was all figured out at the beginning of the 20th century and there are published tables that tell you what gears to use to get a specific lead.

[edinmass]

Joe......I can’t remember, did you get timing gears yet for your car? Or are you now going to make them yourself? 

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

[JV Puleo]

He'd be welcome any time...we've spoken about that before but lets make sure I've got it working correctly!

[JV Puleo]

2 minutes ago, edinmass said:

Joe......I can’t remember, did you get timing gears yet for your car? Or are you now going to make them yourself? 

 

No.. but the gentleman who has them is a long-time friend and I know he's been very busy with other work. He's just about to quit doing outside work so I expect I'll see them soon. I did tell him I was in no hurry and he's been a big help with advice on how to do things - I was on the phone with him yesterday talking about this job.

[weathered1]

Joe, Great explanation, hope to see a video!

[JV Puleo]

I finished the second test blank this morning but then decided I had half of the surface of the original one that hadn't been touched so why not see if I could cut the teeth to the full depth. The biggest value to this, aside from looking at the finished piece is to work out a sequence of operations that can be followed on the finished piece.

 

Rather than each tooth to a given depth and go on to the next,,,rotating the gear several times, I decided to do each tooth to the finished depth. This minimizes the number of times I have to rotate the handle on the dividing head.

 

 

 

The teeth do "look" right...we won't know if they are right until the part is installed but, in as far as I can judge,  they are fine.

 

 

 

I'll set up the 2nd blank now but won't cut it until tomorrow. This is very intense work...I'm even thinking that I'll do the finished pieces on the weekend and turn my phone off while doing it.

[edinmass]

Is the lead figured by degrees? How do you adjust the machine to be at the exact angle? Do you measure it or do you work off the pointer on the head? I'm just clueless on this one.

 

Did Gary's gear mesh with it?

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

[edinmass]

OK, I see it now, the table is turned to get the angle............how is it set?

[JV Puleo]

I'll take some pictures tomorrow. Yes, Gary's gear meshed perfectly.

 

 

[weathered1]

The machining on the teeth looks great for such a deep cut.

[Gary_Ash]

Good to hear that my CAD-designed, 3D printed gear meshed OK. It says that TurboCad Pro can properly describe the surface of the spiral gear and generate the code to drive the 3D printer. In principle, it should be possible to generate code to drive an NC mill to cut the gear, but would require (at least) a 4-axis mill (X,Y,Z, rotation) to do it.  I think these are expensive beasts.  It’s entertaining to see Joe’s 100 year old equipment do the job smoothly and correctly. A lot of neat, accurate technology came out of the Age of Steam - and they didn’t have any computers. 

[JV Puleo]

This is the formula for calculating the lead...

 

 

Once you have the lead, you go to the lead tables which will tell you which gears go on the dividing head. Most, but not all, dividing heads have a 40:1 reduction. That was standard for Brown & Sharpe and I use the tables in the B&S manual, "Practical Treatise on Gearing" (1906 edition), where you select the gear train that will give you a lead closest to the theoretical number. In this case it is 22.93 so the difference is .033 meaning that my lead is 33 thousandths shorter than the perfect theoretical number...which on a gear that is only 1.25" long is less than .002 – such a small amount as to be inconsequential.

 

There are also charts for sines and cosines so finding that figure is simply a matter of looking it up...math was never a strong point for me, always a struggle to learn and remember so if I had to do the necessary math every time I'd never be able to do this.

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

[edinmass]

With that formula it’s now all perfectly clear to me! I could never do the math either. I probably would have had a better chance to get my head around it back when I was taking my math classes in college. Too many decades in the distance to really remember much of it. Sure wish I could fly up to see it done in person........which normally I would do.........hopefully this Chinese Kung Flu will start to die down with the vaccines coming out in the next week or two. 
 

I have decided to cast up a new impeller, in bronze. The aluminum one was rather pitted and in need of attention. There is an art foundry near the shop who will pour it. Is a very simple casting........but I haven’t worked with a foundry or done pattern making in over thirty years. 
 

I was supposed to get on the Stearns Knight this week, but a local friend down here suffered engine failure on his toy.......and I am getting elected to solve it. I rather not get involved, but owe him too many favors to walk away. I think if I supervise the project closely I can avoid doing the entire thing myself. 
 

This is the first time I have ever seen a dividing head in use.........the few I have seen sit on shelves buried in other seldom used items in the shops. Forty years ago, I could think of twenty five people I knew who a decent selection of machine equipment in their homes and garages. I also clearly remember a home about six places down the street where the owner ran a small custom machine shop out of his cellar.......and supported a family of eight with the work he was doing. I now only know of two hobbiest machine shops anywhere........and yours is certainly much better equipped than the other I am thinking of. 

Edited December 2, 2020 by edinmass (see edit history)