My 1910 Mitchell "parts car" project

[JV Puleo]

This will be the flange on the inlet side of the pump that the lower water fitting butts up against. It was aluminum on the previous pump but this time I'm using a brass tube . I will eventually solder this to the tube. The piece of metal is marine bronze, a piece of a drive shaft from a fishing boat. It is actually a mistake I made when I was rebuilding the milling machine but it looks as if there is just enough metal to make something of it...its been on the shelf for a good 3 or 4 years.

 

 

It was threaded and I had just enough material to bore the threads out.

 

 

Then tapped 1-1/4-20.

 

 

I then put it on one of the fixtures I made earlier and turned the OD and knurled the larger portion.

 

 

Then back in the mill to put in slots for a hook spanner.

 

 

The finished piece. This took most of the day because there are so many different steps.

 

 

But I wasn't done yet. I put a piece of 1-1/4 brass tube in the lathe to thread it. This is the end that will screw into the pump.

 

 

I used the flange as a threading gauge...

 

 

Obviously I have to shorten this but I want to set it up on the engine to get the exact measurement. This tubing was not particularly easy to thread...it isn't made from the "free machining" 360 brass that threads easily. Still, it will work just fine.

 

[JV Puleo]

I took yesterday off to do the laundry and work on my book and because I wanted to give the slip fit Locktite time to completely set. This morning I put the inlet end of the pump in the lathe to turn off the extra material.

 

 

It seemed to work perfectly.

 

 

Then I put the pump back in the engine to mark the length of the brass inlet tube.

 

 

I cut it off...

 

 

And threaded the end.

 

 

 

I then assembled it and tried it in the engine. The angle iron sides of the engine stand are the same material I intend to use for the sub-frame so if it fits here it will fit when the engine is in the chassis.

 

 

 

I don't think there is any question that this looks a lot better than the welded pieces. Its stronger too. Next I'm going to replace the socket head cap screws with brass rods and special nuts I've dreamed up. It certainly doesn't look like the original pump but I think it does look like "1910" – or at least a lot more so than my previous efforts.

[mikewest]

Joe It looks great . One suggestion... Where the pump bolts to the case , soften up the corners  to match the case. (radius the corners) ,this will make it look more period.  Great job buddy.. Mike

[JV Puleo]

I intend to but haven't gotten to it. Its one of those jobs I'm leaving for a day when I'm waiting for materials to arrive.

[JV Puleo]

I made some little brass studs to replace socket head cap screws...the first step was to cut them to length. The trick is to make them identical so I did it in the lathe with a cut off tool and the piece of rod in a collet with a stop set.

 

 

 

Then both ends were threaded. This threading device was critical. I'd never have been able to thread them straight by hand and they are too small to single point.

 

 

This is what they are supposed to do. I'll make special nuts for the end but I'm waiting on the hex stock.

 

 

 

I then drilled and tapped the holes for the cover plate on the small end.

 

 

And pressed in the bushings.

 

 

With the plates screwed together, the fit is just about perfect. However, when I put the middle section in the pump it tightenes up just a little so when everything is done and bolted together I'll hone the bushings a little.

 

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

[John S.]

Outstanding work !

[Matt Harwood]

I haven't said recently how flat-out amazing and inspirational your work is, Joe, so I'm saying it now. This thread gets me back out in the shop more often than you'd believe.

 

Beautiful job!

[edinmass]

The entire pump is very well done............👍👍👍

[JV Puleo]

Thanks guys. The pump has been a bit of a struggle but I'm happy with the way it is coming out. There are still a few tweaks to do but overall it would work just as it is. This project has made it clear why people build prototypes. When the job is done you ask yourself "why didn't I think of that the first time" but it seldom works that way.

[Mike Macartney]

8 hours ago, JV Puleo said:

I made some little brass studs to replace socket head cap screws...the first step was to cut them to length. The trick is to make them identical so I did it in the lathe with a cut off tool and the piece of rod in a collet with a stop set.

 

When you get a chance Joe, please would you mind taking a photo of your 5c collet stop? I have a number of studs to make to bolt the two halves of the Humber crankcase together. I am delighted with using the 5c collets that you suggested I bought.

 

I am in agreement with all the above great comments you are getting. Without your help and advice my big lathe and milling machine may still be sitting idle in my workshop waiting for me to pluck up courage to use them.

Edited January 21, 2020 by Mike Macartney
Added a bit more (see edit history)

[chistech]

Following joe’s thread, reading his text, looking at his photos, and taking his advice is inspirational to so many of us.

[JV Puleo]

Cut it out guys... this will go to my head...I'm just a Yankee mechanic.

To answer Mike's question, this is the collet stop.

 

 

It is just a plug that screws into the end of the collet with another threaded piece inside that can be adjusted to the desired depth. This one has been modified by adding the 1/4" dowel pin for a job that needed a smaller diameter pin. The thread is 7/16-20 so I have a few extra bolts and nuts to use for different setups. It wouldn't be too difficult to make but, if I remember correctly, it was inexpensive. To cut small pieces to a uniform length, I set the stop and also set a stop on the lathe bed for either a turning tool or a cut off tool. I usually face one end of a short piece square then use the collet stop to match the exact length of the piece multiple times. I'll use this to make the studs that hold the jugs down and the studs for the main bearing caps. If you needed more travel, I don't see why you couldn't make an extended version.

 

Edit: The pin is actually screwed in backwards in this picture because I needed extra depth.

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

[JV Puleo]

I had several distractions today but still managed to make two threaded sleeves. I have to make 30 of these so it is important I make sure they will work. 12 are for the main bearing studs. Those will be 9/16-18 on the OD and 7/16-20 on the ID. 18 are for the sump. They will be 9/16-18 OD and 3/8-16 ID. Cutting a 9/16 thread with this tool is pushing the envelope but it seems to work, probably because this is free-machining brass.

 

 

The long one is for the main bearings and the short one for the sump.

 

 

I also got 4 packages in the mail at the end of the day, including these nice Franklin brake-line fittings from my friend Mike West that I'll use to conduct oil to the center main and the center camshaft bearing.

 

 

The hex stock and the collet I need to make the final parts for the water pump also came in so I've gone from having to think of something to make to having plenty to do.

[Mike Macartney]

16 hours ago, JV Puleo said:

To answer Mike's question, this is the collet stop.

 

Many thanks Joe. I thought I would have to make one. On searching the internet, I found a new one on the UK eBay from Gloster Tooling at about 11 dollars including postage and have ordered it. I am slowly learning a bit more every day about machining and the associated tooling that is available with your excellent help and posts.

[JV Puleo]

Today I made nuts to hold the pump together. Why, you may rightly ask would I bother to make nuts? Because I want them to look appropriate for the period. It isn't that they look like originals, it is that they don't look as if they came from the hardware store. The first step was to cut some pieces of 7/16 brass hex stock.

 

 

It was faced off and drilled with a #25 drill.

 

 

 

Then I set up the radius turning tool and put a slight crown on one end. This was done with a collet stop in place so all the pieces would come out exactly the same length.

 

 

 

Then the other end was turned down to 3/8"

 

 

 

And went back in the lathe to be trimmed to length. I then threaded them (although I forgot to take a picture of that).

 

 

I only need 5. The 6th was my insurance in case I ruined one. I have to admit it doesn't look like much for most of a day's work but my feeling is that it is small details like this that disguise the fact that the entire unit is new.

 

 

Edited January 23, 2020 by JV Puleo (see edit history)

[Luv2Wrench]

Certainly worth the work and more.  Modern hardware looks so bad on a 100+ year old car.  Job well done.

[JV Puleo]

I am now in the final stretch of the water pump job. I put the studs in...to keep them from turning when I tightened them I put a drop of super glue on the threads.

 

 

Then assembled the pump without the shaft.

 

 

I then ran a barrel lap through. It took only two light passes for the pump shaft to turn very easily.

 

 

The next step was to put a slot for a Woodruff key in each end of the shaft.

 

 

And mill a flat opposite the key.

 

 

Then the seals were pressed in.

 

 

I need a gasket to go under the seal cover on the inlet side because the screws go into the water passage. These were made by putting a 3/4" hole in the gasket material with a Forstner bit. I then put them in the lathe sandwiched between the two pieces of aluminum that I used to make the seal cover. They were trimmed down with a razor blade and the touched up with a bit of sandpaper. I actually made 4 of them.

 

 

The inlet tube was screwed into the bottom of the pump using this old-fashioned thread sealer.

 

 

Then I put it all back on the engine to make sure everything lined up as it should...which it did.

 

 

I still have four gaskets to make but the pump is now assembled and ready to go on the shelf until it is time to assemble the engine. Here's the finished product.

 

 

 

[Guest]

Absolutely fabulous.

[Terry Harper]

Very nice Joe! Love the use of the barrel Lap!

 

Its interesting to compare your first attempt versus the last and see the design progression

and refinement.

In the engineering it would be the the sixth step of the design process..... design improvement.

[JV Puleo]

The whole process makes it clear how much work is required to design and make something. I can't imagine why I didn't make the casting first...it all seems so simple from this end. It is one of my shortcomings that my design progression always seems to go from complicated to simple.

[mikewest]

Joe , Its a very handsome pump! What is next??? Con rods?? Pistons??

[Luv2Wrench]

8 minutes ago, JV Puleo said:

The whole process makes it clear how much work is required to design and make something. I can't imagine why I didn't make the casting first...it all seems so simple from this end. It is one of my shortcomings that my design progression always seems to go from complicated to simple.

 

I don't think it is a shortcoming, rather just who you are.  My path to simple always goes through complicated.  

 

[JV Puleo]

Main bearings....actually a set of temporary mains made of aluminum as a test for the machining processes and to use when I'm fitting the con rods. Then the rods and then the pistons. With those done I'll do the bronze shells for the mains. That's the plan in any case.

 

Actually, I'm getting ahead of myself. I also have to finish the oil passage for the center main and make the parts to secure the center bearing of the camshaft and provide oiling. I should do that before I bore the case.

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

[Terry Harper]

Quote

The whole process makes it clear how much work is required to design and make something.

 

(LOL) Yup there is a lot too it! My students are finding that out with the impulse coupling. 

They are moving forward slowly but they are not liking the vagarity of patent drawings.

 

However... they are hooked on the challenge.

 

In regards to the bearing shells.... do you have the formula for calculating the "squish".

I have it buried somewhere if you need it.

Edited January 25, 2020 by Terry Harper (see edit history)

[JV Puleo]

Your students aren't alone in that. I also find patent specifications difficult to work with.

 

On a completely different note, last night I found a reference to aluminum pistons in 1908 or 09. I'd always thought W.O. Bentley, with his DFP race cars was the first to use them. I'd guess that Bentley didn't know about the earlier car, which wouldn't be surprising because it to was a race car and I doubt the builders were going to advertise their idea. Bentley didn't either. He did share it with the Admiralty at the beginning of WWI which is how he came to designing the BR1 and BR2 engines for airplanes.

 

I do have the formula for determining the crush. The challenge will be to work to an extremely close tolerance. That's why I think I should make a test set before cutting into $200 worth of bearing bronze.  The job will probably require more special fixtures, one in particular to hold the shells together while I machine the thrust faces.

[Gary_Ash]

There is a reason why patent drawings and descriptions are hard to follow:  the goal in writing a patent description and making the drawings is to convince the patent examiner that you have something new and patentable, but not giving enough information to let the competitors reproduce it easily.  Once someone understands how it really works, another method can usually be found.  Many ways to skin a cat!  To beat a patent, you only need to eliminate one of the described components. 

[Mike Macartney]

5 hours ago, JV Puleo said:

my design progression always seems to go from complicated to simple

 

Many years ago, when I was doing engineering design at college, I was told to to sketch out and write down as many ideas to as possible to achieve the same objective, even if they seemed very silly at the time. Then pick the most suitable solution to the problem. Do you like my economical watering can.

 

 

It never needs refilling! On a more sensible note; it is more difficult with complex items like your water pump. So many variables come into play, I think that you have done fantastically well with the pump, its making and development. You should be very proud of the work you have achieved.

 

 

 

 

[edinmass]

Joe, how are you going to lubricate the pump and seal? Are you going to run cutting oil in the coolant? 

[JV Puleo]

Yes. I lean toward plain soft water with a little water-soluble oil in it. I also used oilite bearings and they are easy to replace if they wear noticeably. The original pump had no means of lubricating it. It's important that it not leak because I'll probably put anti-freeze in it for winter storage. A heated garage is beyond my means.

[JV Puleo]

Ed, what oil do you use? One of the characteristics of water-soluble oil is that it decays and stinks to high heaven. That is the reason I don't use it in my mill even though it has a built in coolant pump. I seem to remember the product that was sold as "water pump lubricant" years ago was just that but didn't stink.

[Terry Harper]

Hello Joe,

 

Anyway you could drill and tap for a grease cup or Alemite fitting?

The Wisconsin water pump has grease cups and the bronze bearings

have simple broached slots for grease grooves.

 

T.

[JV Puleo]

The problem is that the front bearing is inside the bracket that holds the pump - that one might be possible but the hole would have to go through both pieces. The rear bearing is much more complicated because it is surrounded by the inlet water passage. I did think of a way to do it but it rapidly gets complicated and creates the risk of a water leak. Since the original pump didn't have either I thought it would be safe to rely on soluble oil in the coolant, the oilite bearings and the fact that I can replace them easily.

[edinmass]

The modern seal will burn from lack of lubricant............oilite will be a problem. You need to figure out a few things......cutting oil will protect the oilite, but the seal will fail, rather quickly. The seal needs to have a pocket of grease behind it , I usually run two seals, back to back with a pocket of water pump grease in between. Then twice a year I give the cup half to 3/4 of a turn.

Edited January 25, 2020 by edinmass (see edit history)

[edinmass]

Joe, can you bore the pump to take two seals, and make a grease passage and a grease cup to lubricate it? Looks like it’s a possibility. Your going to have to run a fair amount of cutting oil to keep the pump from tearing itself apart. It’s not a problem if you run water and oil in the summers and antifreeze in the winter. The seal will be a bigger problem than the oilite. Been there, done that. 
 

On my cars now I run a ceramic seal......pain in the ass to engineer.....but I have 25k on my pump, and have never had to grease or service it. The bearings are modern with modern seals........also “sealed” and non serviceable.  Works great, but not easy.

Edited January 25, 2020 by edinmass (see edit history)

[JV Puleo]

I can revisit the problem in the future if it does prove to be a problem. Adding grease cups would be a challenge, mostly because it would require a tube passing through the inlet water passage to the bearing in the back and passing through the bracket that holds the pump in place in front. I could shorten the bearings and put seals on both sides...that would make it easier to lubricate but would decrease the bearing surface by 1/2". That may not be an issue - the total surface is slightly less than 3" now which is a lot for a 3/4" shaft. If I did that, I'd be inclined to put a seal on each end of the bushing and a small "button" type oil fitting in the middle. I actually those since they were incorporated in my original design. I bought a pair of ceramic seals but there simply isn't room for them. The width of the pump is limited by the bracket and the magneto and I've got it as tight as I dare go now.

 

I wonder what the original thinking was since they didn't have water-soluble oil. The original pump shaft was badly corroded from electrolysis but the bushings  didn't show much wear at all. They may have picked up some lubrication from the packing material.

[Terry Harper]

Joe,

I am sure you will figure it out! I am using a water soluable cutting fluid at school and it

doesn't seem to stink after sitting all summer.  Durakut I believe is the brand.

 

T.

[JV Puleo]

That's a good sign. I'll make a note of the name.

This discussion has given me some ideas that may work. I'll have to make a small part and adapt a tiny grease cup but the good part is that I can make the part first and then, if it looks as if it will work, add it to the pump.

 

Thinking a bit further on this... maybe I'll put a seal on each end of the bushing and lubricate it by connecting a small Gits "twist" oiler using way oil. I like way oil because it much stickier than motor oil (which isn't all that great a lubricant for anything other than motors).

Edited January 26, 2020 by JV Puleo (see edit history)

[Terry Harper]

Joe,

Durakut is available as a synthetic and semi-synthetic. You probably want the synthetic.

1:20 is the mix they recommend.

[JV Puleo]

Does anyone here know when "flush" grease fittings were invented? This is the type that used a pointed attachment or a needle on the end of the grease gun. I know the fittings with a little ball check valve were in use because there are several on the car in places where they must be original. I'd prefer to keep everything that shows on the car in my 1910-1915 time frame.

[JV Puleo]

3 hours ago, Terry Harper said:

Joe,

Durakut is available as a synthetic and semi-synthetic. You probably want the synthetic.

1:20 is the mix they recommend.

 

That sounds good and the synthetic is advertised as having no odor. At $50 a gallon it isn't cheap but with a 20:1 ratio and a water system that doesn't leak (I hope) 20 gallons of water should last a long time.