JV Puleo

My 1910 Mitchell "parts car" project

Recommended Posts

Nice job Joe!  I wish I had a neighbor friend that was in the Casting business!  Good for you!

Al

  • Like 1

Share this post


Link to post
Share on other sites

Joe that looks very nice!

 

Question, in regards to the impeller what takes-up the end thrust? I know you have bronze bushings in both the front

and rear covers does the impeller run against those? (aluminum on bronze) or will the impeller be setup with

A bushing to act as a thrust bearing?

 

T.

 

 

  • Like 1

Share this post


Link to post
Share on other sites

I have given that some thought... the OD of the bushings is 1-1/2" and the diameter of the hub is the same. I don't know if it needs more... I thought of putting some in there but I don't know what would be appropriate. Steel would rust and I'm not sure bronze will be any different than what I have. There will be fittings on the pump shaft on both ends - for the timing adjustment coupling in the front and the magneto in the back so there will be very little movement. I may put roller thrust bearings on both ends on the outside. It's certainly isn't "original" but it would make certain that the impeller can't move inside the pump.

  • Like 1

Share this post


Link to post
Share on other sites

First thing this morning I pressed the bushing out of the inlet side of the pump and put it back in the lathe to remove the threaded portion where the seal retainer went in. This gained me a whole 1/2". Since all it did was retain the seal, I'll make a plate about 1/8" thick to attach to the hub of the pump.

 

IMG_1852.thumb.JPG.818c20e851f56f9690382ed5d3e991f1.JPG

 

It does look neater...

 

IMG_1853.thumb.JPG.a05feff1acad97ff1e855de58b8e17da.JPG

 

And makes a significant addition to the critical space between the pump and the magneto.

 

IMG_1854.thumb.JPG.be6604f6153aa29f44c5277920039ac9.JPG

 

I also drilled and tapped the impeller for a set screw. It will get a woodruff key as well but that isn't necessary for testing.

 

IMG_1855.thumb.JPG.a50320e4428930fa91cc04814a9131d4.JPG

 

I then assembled the pump. It bound up very slightly...you could force it if you wanted to but it was easier to put the impeller and the shaft in the lathe and take a very light (about .003) cut.

 

IMG_1856.thumb.JPG.bbaf6c5d258b96afa5880804a98a3637.JPG

 

When I reassembled it there was no rubbing at all so I pressed in the seals and set it up on the mill table to get an idea of what I'll need for a test stand. I'll be making that today and tomorrow so it's possible I'll be able to test  it early nest week. I will probably make another one as I've found three or four details that could be improved but as I've already made all the special fixtures I'll need and I now know exactly what I want, it should go much faster.

 

IMG_1857.thumb.JPG.cff53ec6c27555583fed9fba49a86124.JPG

  • Like 2

Share this post


Link to post
Share on other sites

I spent the day on the test stand, using up scrap materials around the shop. The worst part was having to make a pair of "U" shaped pieces to hold the motor. I'm using the motor from the oil pump test stand since it's easily moved and has it's own switch.

 

IMG_1858.thumb.JPG.a71f0459341c2eeff7ab7f151cbcfd27.JPG

 

IMG_1859.thumb.JPG.9479ba5e8c4d52afe0e0f9f5e224c99d.JPG

 

Tomorrow I'll rug up some sort of water tank and get some clear vinyl tubing so I can see the flow. When I tested it, it was blowing air so I'm optimistic it will work. I'm going to have to run it outside as I'm pretty certain that if it leaks I'll have a mess to clean up in the shop.

  • Like 2

Share this post


Link to post
Share on other sites

I have to admit joe, I would have just drilled a couple of holes in hardwood blocks instead of milling the Aluminium ones but I also admit it wouldn’t have been quite as much fun making them!😀

  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)

I did think of that but the motor jumps around a bit and I'm worried about it slipping or breaking free and tipping everything over. As it is, it's a Rube Goldberg contraption. Today I'm trying to figure out how to plumb in a big plastic water tank I have left over from my print shop days... Besides, I want to lift it on an off so I can put it back on the oil pressure test stand.

 

jp

Edited by JV Puleo (see edit history)
  • Like 2

Share this post


Link to post
Share on other sites

The test stand is proving to be one of those "easier said than done" jobs. In order for this to be a valid test I need the water to enter the pump as it will on the car. That calls for a 1" ID tube and a 1" outlet from the tank. I wasted an hour at Home Depot looking for some sort of fitting I thought I could modify but found nothing so I decided to make one from this piece of Delrin. It's a machinable plastic, very easy to work but I don't have a lot of experience with it. Gears are made from it and I'd thought of making the timing gears out of it but I backed off on that idea as I've no idea how durable they would be and it goes against my goal of trying to stay within the technology of 1910-1915.

 

It does machine easily. I set it up to drill and ream to 1"

 

IMG_1860.thumb.JPG.93662d57cb531e041feffc5d0024a0ab.JPG

 

IMG_1861.thumb.JPG.e56413aea964617c5036babc3a5f2eed.JPG

 

Then bored one end to take 1" PVC pipe (which has an OD of 1.130)

 

IMG_1862.thumb.JPG.07b704347873bc9421d64034e80cb480.JPG

 

Then I threaded the other end to match the nut that was on the original fitting. This was about the easiest thread cutting I've ever done.

 

IMG_1863.thumb.JPG.1eec96f8147113f95a0d1dae6a508e4f.JPG

 

I glued the pipe in with superglue gell and the ball valve with normal PVC cenent.

 

IMG_1864.thumb.JPG.aa68b8577086ecb3d69a16a0d865f879.JPG

 

And put the whole thing back together with the original rubber gasket.

 

IMG_1865.thumb.JPG.a84d7ff91a25d5cdbcd0ed4902d21eef.JPG

 

To connect this to the pump input I'm using a piece of 1-1/4" ID clear vinyl hose and to make that fit over the pipe I had to turn it down slightly. I'd no idea how this stuff turns but it proved to be rather good.

 

IMG_1866.thumb.JPG.b442ff02d6b4ed9806e7b9f90685b976.JPG

 

All that is left is to attach the 1-1/4" tubing but it's flattened and hard to work. I've had that problem before and solved it by warning it with a heat gun. When I went to do that, the heat gun died so I'll take it home and using some boiling water...I think that will work. With any luck, tomorrow we'll find out if the pump works.

 

IMG_1867.thumb.JPG.381a9a2daf7ed42f303f130557519b3a.JPG

  • Like 3

Share this post


Link to post
Share on other sites

This morning I put the big hose on and tried the pump.

 

IMG_1868.thumb.JPG.30e48599ca685f551bcfe1878b1826e3.JPG

 

There are a few minor leaks where I didn't do a perfect job soldering the fittings. I wasn't worried about that. To say it worked is an understatement. If anything, it works too well but there is a flaw in this test in that it is spinning much too fast. I'll have to see if I can find a bigger sheave for the pump so I can slow it down but, for the moment, everything seems fine. Its pretty clear the pump is bigger than it should be. I would take a photo of it pumping but the motor was jumping around and the outlet tube wanted to come out of the tank. It took both hands to hold on to everything.

  • Like 4

Share this post


Link to post
Share on other sites

Most excellent Joe! Another success!

 

On the Wisconsin the brochure talks about how the water flow to each cylinder block is "carefully proportioned"

so each block is cooled evenly.  Its interesting that though both the inlet and outlet pipes are 1-1/2" o.d. all the fittings

have the the flow restricted at the flange by smaller diameter openings. For instance the top water manifold fittings are restricted

down to 1-3/16" dia. while the lower water manifold outlet opening is only 3/4" dia.

 

Here is a photo of one of the lower water manifold fittings to help illustrate what I am talking about:

IMG_1215.thumb.JPG.0327ee7676ff186ae3c352fd50a0adc4.JPG

 

My thought is they needed to restrict the flow to allow the water to remain in the block long enough.... or in the radiator (no thermostat here!)

to allow for efficient transfer of the heat.

 

T.

 

 

  • Like 2

Share this post


Link to post
Share on other sites
Posted (edited)

Yes...I was going to say I'll bet it isn't thermostatically controlled. Before the adoption of the thermostat calculating water flow and radiator capacity was very complicated, made more so because no matter what they did they couldn't control the ambient temperature. The usual "fix" was to make it cool adequately on a very hot day which means that on most days they ran cold. I'm not even going to try to do those calculations...I'm just going to incorporate a thermostat.

 

After I posted the pictures of the test rig I looked around for another sheave. I finally took one off my small drill press. The motor has a 4" sheave and now the pump has an 8" sheave. It's a 1725 RPM motor so before it was probably running at close to 2000 RPMs which is probably the absolute maximum of this engine. With the 8" sheave the speed is closer to 860 - probably 30 MPH. At that speed it was a lot better.

 

IMG_1870.thumb.JPG.5ee85ac33a81297f76bbc3641de8f239.JPG

 

 

Its still moving a considerable volume of water.

 

IMG_1869.thumb.JPG.be1ace1911c0aa17d8e42463698fd2f1.JPG

 

I don't think there is any danger in making the pump smaller. This water was coming out under a measurable (If I had a way to measure it) pressure.

Edited by JV Puleo (see edit history)
  • Like 3

Share this post


Link to post
Share on other sites

Terry... those restrictions in the water lines were calculated, at least in part, on the capacity and flow rate of the radiator. I'm guessing that unless you use a radiator with the same specifications as the original, they won't work as intended.

  • Like 1

Share this post


Link to post
Share on other sites

Below is a photo of the middle T-fitting for the lower water manifold. They bored the casting for the OD of the pipe clear through, soldered it in place than

punched a hole the wall of the pipe for the outlet  - so much for precision!

 

Did your Mitchell have any fittings sized like these?

 

IMG_1378.thumb.JPG.881e209a05db6aee1e70da6fb48e4139.JPG

 

Unfortunately for me the original radiator is a beast!

IMG_1263.thumb.JPG.d5239d27fa27465a56e090b84bd0a047.JPG

 

 

 

 

  • Like 2

Share this post


Link to post
Share on other sites

I had no water lines at all - which is why I had to make all that stuff up. I will say that if there was a cheesy way to do things, that's the way the Mitchell Lewis Company did them.

  • Like 2

Share this post


Link to post
Share on other sites

Was Mitchell-Lewis always like that Joe or was it a product of trying to compete?

Did quality improve with the later products or did it keep sliding?

 

I hear a lot of references to the marketing disaster of the "Drunken" Mitchell but I can't help

think that there was an underlying root cause such as lack of quality.

  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)

This just a guess but, as far as I can see, the problem came with the 1910 season. The market for expensive cars was saturated for the time being. Mitchell had been selling for about $2500, putting it on the high end of inexpensive or the low end of expensive but that market had temporarily dried up. The only large, untapped market was in the $1,000 to $1,500 range so they built a car to meet that price point but, in order to keep it in line with their previous cars, it had to be a reasonably large car. It looks as if the the engine is the worst part. although there were shortcuts taken with the chassis I suspect much of that was bought out (although they didn't admit that). The company was undercapitalized so they didn't have the financial strength to survive a bad selling season - the problem that eventually killed them. This engine was only manufactured in 1910 and 1911. I suspect it was always a headache...certainly they went for a complete redesign after only two years of production. The saving grace is that the design is quite pedestrian with nothing particularly unusual and the really cheap materials like pot metal hadn't been invented yet. The savings came from the fit and finish which is very "slapdash" at best... It sold for about $1,250-$1,300 and didn't come with headlights or a top...(those were extra).

Edited by JV Puleo (see edit history)
  • Like 2

Share this post


Link to post
Share on other sites
Posted (edited)

There is an excellent book titled America Adopts the Automobile that gives a good insight into the business dynamic of the first ten years of the 20th century. It isn't a "car book" in the sense it talks about the technical specifications of models but it provides insights that I've found nowhere else. I suspect it was written as a doctoral dissertation.

 

[EDIT] Author is James J. Flink, published by the MIT Press. I looked it up on Bookfinder.com

The cheapest I could find was 2.98 so there isn't much excuse for not reading it if you're at all interested in the history of the automobile in America.

 

 

Edited by JV Puleo (see edit history)
  • Like 3

Share this post


Link to post
Share on other sites

When I knocked the test stand down this morning I discovered I'd forgotten to put the o-ring seal in one side of the pump. It still had only a very minor leak so I guess that part of my design is working well.

  • Like 3

Share this post


Link to post
Share on other sites

I started on the "improved" water pump today. As much as I don't like doing things over its wort noting that the second time around is always much easier. You already know what the problems are so you can avoid them. The first step was to make the smaller of the two hubs. Drilled, reamed and turned to size and then threaded.

 

IMG_1871.thumb.JPG.20cff24b80de89a88dbfc4d4424df661.JPG

 

IMG_1872.thumb.JPG.14f6d0e4457734da9e5e7dcfe03084f2.JPG

 

With that done, I started the plate it attaches to... bored out to 1.702 (the hole size for the 1/3/4-20 thread) and then threaded.

 

IMG_1873.thumb.JPG.24c59e999da8996b8a78d781400ca2c1.JPG

 

IMG_1874.thumb.JPG.0895751f25343c253e1dc5329683d8b2.JPG

 

If this design works out, I will be reducing the outside diameter by 7/8" but the diameter and height of the impeller by only 1/4". That's still bigger than the original water pump and based on the circulation I got from the first pump I am confident it will be adequate.

  • Like 2

Share this post


Link to post
Share on other sites

I don't know what that is but it's only 33 pages of a 200 page book. It may also be a copyright violation since the book was published in 1970. If they put the entire text on line without the permission of the copyright holder is is clearly a violation. Off hand, I don't know if the copyright belongs to the author or the publisher, it could be either.

  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)

The front plate assembled.

 

IMG_1875.JPG.d27992512589605323a3319a4c99d2ec.JPG

 

I then started on the water inlet hub.

 

IMG_1876.thumb.JPG.303e41b011b45f4094fdde7f5af7d176.JPG

 

I used my 20TPI to thread it. 1-3/4-20 isn't a real size but these parts aren't intended to ever come apart. This is just a way of getting around making a complicated casting.

 

IMG_1877.thumb.JPG.d088bf7c17201369c29029e8914827ba.JPG

 

I faced and bored the inlet side plate to 2.70 - the size for the thread.

 

IMG_1878.thumb.JPG.a1d9a56185430053344459b06953879d.JPG

 

It was a little fussy fitting and, in the end, I had to lap it slightly to eliminate some burrs but it went together OK.

 

IMG_1879.thumb.JPG.e2575c92a04e4e42c9c35e3bf6bf9c50.JPG

 

The inlet side getting faced to 1/2" thick.

 

IMG_1880.thumb.JPG.338d1b3fdc34f066f94de72427ef249e.JPG

 

Both plates with their hubs...

 

IMG_1881.thumb.JPG.fe26e77aec9a27ec9e64f39a6d3fa699.JPG

 

The small hub was quite a bit too long. I must have remembered one of the dimensions incorrectly so I cut that one down to 1". I also discovered that the roller thrust bearings I intend to use were, in fact, available at the time but weren't used much because they were very expensive.

 

IMG_1882.thumb.JPG.1facfa3e0d65fec192a8599a266ce9cf.JPG

 

While making one of these isn't the work of a day or two, the second one is going much faster and, so far at least, coming out better.

Edited by JV Puleo (see edit history)
  • Like 3

Share this post


Link to post
Share on other sites
Posted (edited)

I also reduced the height of the input side hub. this is still .100 long but I don't think I'll take off the extra metal until the hole for the input tube is bored.

 

IMG_1883.thumb.JPG.8386452d9166ea1c71003b0075478622.JPG

 

I now have to put in the slots that will position the central body of the pump. These are 3/16" thick and have to be arranged so that the inside diameter is 4/1/4". The math is simple but for some reason it took me some time to decide what the right way to figure it was. I was doing that last night and it is another good example of why I shouldn't do that sort of thing when I'm tired. The last time I got the number wrong. I was able to compensate by making the body fit the slots but this time I want to get it right.

 

IMG_1892.thumb.JPG.a338a37be02da77d4eabb60d2cf1ffef.JPG

 

Then, I wanted to follow Spinnyhill's suggestion and recheck it another way... so I painted on a little Dychem and scratched the surface. I calculated what this measurement should be and set my dial vernier to that, then check it against the scratch marks.

 

IMG_1893.thumb.JPG.2ab87082af00e931492b046b9cd919b4.JPG

 

It was right on, or close enough so that it won't make a difference. I went out today to look at a lathe for a member of the Practical Machinist forum so it was almost 4 PM before I finished setting this up and decided to let milling the slots wait until tomorrow.

Edited by JV Puleo (see edit history)
  • Like 2

Share this post


Link to post
Share on other sites

I milled the slots in the end plates. I'm trying to make these without taking them apart to insure that all the machined surfaces are true with each other. The slots are very slightly more than 1/8" deep.

 

IMG_1894.thumb.JPG.d04659561a5dc6f2be53b05c9322f0b1.JPG

 

IMG_1895.thumb.JPG.502b3976fb55704c25cf7f63268eccb5.JPG

 

I got very close to the dimension I want so I now have to make the center section. for this I need to turn the OD to match the boring fixture I made and the ID to match the inside diameter of slots. I made the end plates first because I couldn't be certain I'd get that measurement perfect. I didn't so the ID of the tube is going to be about .020 larger than my drawing specified. To do this I put the big chuck on so I could hold the tube securely.

 

IMG_1896.thumb.JPG.c33266cb62e3b4033c4e76fc17295fa5.JPG

 

The aluminum tubing isn't perfectly round and it took about .015 to get the low spots out. All that was anticipated...so I was pleased when it went ii had a nice surface and into the fixture.

 

IMG_1897.thumb.JPG.e36c098537bc4f645a735e53e32c47b4.JPG

 

This time around I'm going to make this piece quite thick and mill flutes in it for the cap screws. This is all an attempt to make the pump look more like a finished and well thought out part as well as reduce the outside diameter.

  • Like 2

Share this post


Link to post
Share on other sites

Boring the center of the pump out to 4-1/4". I had to get this measurement just right because it serves as one of the surfaces that go into the slots in the end plate and if I got it even slightly too small it wouldn't go together.

 

IMG_1898.thumb.JPG.25b7733d4b0b22cdbe04719c5a105325.JPG

 

the piece that goes into the end plate. The slots are .135 deep and the tab is .125 allowing .010 extra space should I want to put a little bead of silicone gasket sealer in there. The actual seal for the pump are on the inside surface so the major function of this tab is positioning the center in relation to the end plates.

 

IMG_1899.thumb.JPG.f09588fdd7018fe2b425e986009f8ed1.JPG

 

And, it fit well. It's a little tight but I'll lap it to get the surfaces smooth.

 

IMG_1900.thumb.JPG.3f119ee7aa9f852786cdb67a6ab9b286.JPG

 

Now I have to turn it around and do the other end.

  • Like 2

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now