JV Puleo

My 1910 Mitchell "parts car" project

Recommended Posts

Joe, Im on the edge of my seat waiting for the demonstration of the pump "PUMPING" I hope you rig it up with a supply and a drill motor and show off your work! Its great!

 

Share this post


Link to post
Share on other sites

That's the plan. Actually, I'm going to rig it up to run at the maximum engine RPM. First I'll see if it pumps. If it does (and I can't imagine why it wouldn't) I'll set it up to replicate pumping into the bearings. I need to do that to regulate the pressure relief valve I designed...

Share this post


Link to post
Share on other sites

I appreciate all the work you have put into this pump, and the other machining work you do. I've learnt a lot from your posts and you have given me the courage to be a bit more ambitious with my machining work. Keep up the excellent work, text and photos. 

  • Like 1

Share this post


Link to post
Share on other sites
10 hours ago, Mike Macartney said:

I appreciate all the work you have put into this pump, and the other machining work you do. I've learnt a lot from your posts and you have given me the courage to be a bit more ambitious with my machining work. Keep up the excellent work, text and photos. 

 

x2

 

Share this post


Link to post
Share on other sites

Today I drilled and reamed the holes for the dowel pins. I had to set it up in the drill press in order to get enough vertical travel.

 

IMG_0934.thumb.JPG.20d27f6e0cac34f0a3d242185ece7031.JPG

 

After I set it up I rotated it to the positions I'd calculated and one of them just didn't look right. I've too much work in this to take a chance so I opted to go home to look at the original of the drawing. Sure enough, I'd misplaced one of the holes about 3 degrees...Also, I had one of those immensely frustrating errands to run to see our local internet provider who charged me $200 for a "free" modem. Once back in the shop, I drilled and reamed the holes.

 

IMG_0935.thumb.JPG.2cbbfe8bef8e66150564a64b6a7321f8.JPG

 

I've no idea why but one turned out perfect and the other is very slightly oversize. It isn't enough to make an operational difference but it is frustrating. I want the dowel pins to be tight in the aluminum piece and a slip fit in the steel pieces because I'm concerned that one could come loose and jiggle out. There are two possible fixes for this... oversize dowel pins or I could use some Loctite "press fit" glue. I'll probably go for the oversize pins but, in any case, it has no effect on continuing the job. I don't plan to put the pins in permanently until nearly everything else is done.

 

IMG_0936.thumb.JPG.dccf3c932b981360adf4004aad5d975b.JPG

 

Since I was working on the cap, I cut down a bushing to go in, replacing the very long bushing that served as a guide for the counterbore pilot.

 

IMG_0937.thumb.JPG.8a93af8d3b3b65af6d1fb3b636a67767.JPG

 

I got this pretty close but the nature of this pump requires that it be absolutely flush with the inside of the cap so it was back to the lapping plate to finish it. This is tedious but it does do a very good job.

 

IMG_0938.thumb.JPG.d5dada68180997fe3e4adee2e18e18d2.JPG

 

If anything, a lapped surface is slightly flatter than a ground surface. Theoretically, if I lapped all the surfaces perfectly I could assemble it without gaskets. This is a good example of a very old-fashioned technique that works well but is time-consuming. Aside from gage making, I doubt it is done at all any more but or the sort of one-off "I'll never have to do that again" projects with old cars, it is easy and accurate. It occurred to me that, having no exposure to professional metalworking, about 90% of what I try is gleaned from books, virtually all of which are as old as the car. I'm probably about the last 19-century machinist. This is certainly suitable for my machines... most of which are at least 70 years old and some are quite a bit older. I have, however, succumbed to one heresy and own a vertical mill (not as yet running) built in the 1960s.

  • Like 3

Share this post


Link to post
Share on other sites

 

I've accomplished quite a bit more since my last installment...

First, I counterbored the cap for lock washers. I don't have a really good way to do this... my non-running vertical mill would be perfect but for now, I have to make do with this antique (and worn out) little drill press.

 

IMG_0939.thumb.JPG.ec68ec0ab0d4b3122fa1e9f8bfb24732.JPG

 

In order to get a uniform depth, I set the part up so that the pilot is in the hole and the teeth of the counterbore touch the surface. Then I use a set screw collar as a stop on the top of the spindle with a spacer (in this case a .125 gage block) to set the gap. It's a vertical version of my method on the lathe and gets you within a few thousandths of the desired depth.

 

IMG_0940.thumb.JPG.0f2b34900e30ec84bd95b797bd390c2c.JPG

 

The cap came out good... the lock washers will be invisible when everything is tightened up.

 

IMG_0941.thumb.JPG.09f6d0618bd1187c3afd0548c939b519.JPG

 

Then I went on to the part I'd been avoiding. I want the pump outlet to be vertical on the back of the engine. I'd already calculated the offset for the attaching holes but felt I should double check them. The first step was to put the original piece in...

 

IMG_0942.thumb.JPG.e7f146167751d8bee758c0c71313b990.JPG

 

Then I cobbled up a disc with a line scribed down the center mounted on a piece of aluminum rod that fit the bushing.

 

IMG_0943.thumb.JPG.65598b0ec4c3090feeb3e073e0b46a09.JPG

 

I used a square to get the line as vertical as I could see and then transferred the center line to the original piece.

This went in the fixture I'd made to drill the holes in the perimeter. I used this to check the location of the attaching holes. Sure enough, true to the Mitchell-Lewis companies lack of interest in precision, the holes were not 120 degrees apart but I was pleased to find that I had calculated the location of the first hole perfectly. I decided to use these measurements rather than take a chance that the crankcase was perfect and the part was off... this was probably a mistake but I knew that this piece fit, although not perfectly.

 

IMG_0944.thumb.JPG.01489a955e2dc074ec80424e8a481484.JPG

 

The next step was to drill center holes. I can't drill the actual hole because there isn't enough vertical travel with the rotary table and chuck in the mill.

I drilled the holes on the drill press using the center holes to locate them by aligning them with a center I made some time ago.

 

IMG_0945.thumb.JPG.3b9bcb81ea4e5dc35dd021982b2b3b2f.JPG

 

IMG_0946.thumb.JPG.3b7898852fb160e3f92b4d717bca5a08.JPG

 

Before proceeding with the holes, I counterbored the pump base for a flanged bushing.

 

IMG_0947.thumb.JPG.b77815b860e6c5803409535b91212a5b.JPG

 

Then pressed out the guide bushing and pressed the new bushing in. The base then went back on the lathe to trim the bushing to size.

 

IMG_0948.thumb.JPG.0a5083235d4cf059682f5a8b2686d1bb.JPG

 

IMG_0950.thumb.JPG.d3934c94d70d39d2bdfcceaa5b65f061.JPG

 

All of this went smoothly... here's the pump base with its bushing in place.

I was going to surface grind this but no sooner did I start the grinder and the belt parted again. So, since the tool I ground last weekend is giving me a near mirror finish, I faced off the bushing and went back to the lapping plate. An hour later... I counterbored the holes for the attaching screws and here I nearly messed things up. Despite working from the original piece, it didn't fit perfectly. I enlarged the holes to almost the size of the original and enlarged the counterbores slightly. One of the three screws still refused to go in all the way. It was then that I finally noticed that the difficult screw did not go through the back of the crankcase. It screws into the wall of the crankcase and the cap screws I'm using were just slightly too long. I feel like a bit of an idiot but no harm was done.

Here's the base finally attached to the crankcase.

 

IMG_0951.thumb.JPG.dd5ddece0b8405df487828843583834e.JPG

 

Then, just because I could, I roughly assembled the pump to see if everything looked as if it fit. Thankfully, it does.

 

IMG_0952.thumb.JPG.0b51835b3e3b28a899d0d986698b2323.JPG

 

 

 

 

 

 

 

 

IMG_0949.JPG

Edited by JV Puleo (see edit history)
  • Like 1

Share this post


Link to post
Share on other sites

Look great Joe!  When I was younger and something didn't want to go all the way in I used my muscle power to make it fit... only to find out later that there was some reason it didn't fit and my muscle power had just created a new and much bigger problem.  That's the real definition of "idiot".  :)

 

 

Share this post


Link to post
Share on other sites

I like the idea of hiding the spring washers by counterboring. I must admit I have never thought of doing that. I agree with Jeff above - as we get older we do get wiser. It's all looking good.

Share this post


Link to post
Share on other sites

There will be a paper gasket between the pump body and the crankcase. The pump itself will be self-contained with thin paper gaskets between the three sections. It is because that joint is so critical that I've been lapping the surfaces but whether it works is anyone's guess. I'll only know when I test it. Under no circumstances will the pressure be very high so I have a good chance it will be tight. I'm making no effort to seal the inside of the pump off from the crankcase thinking that the small amount of oil that will move along the camshaft will only lubricate the bushings. I had planned to groove the bushings (and still could) but I'm wondering if that is necessary. Some material I've read recently suggests that, with modern oils,  grooved bushings are not as important as they were in the past.

 

Share this post


Link to post
Share on other sites

Now to a really fussy part... the sliding vanes in the pump. I don't have much experience making really small things. My machines are really more suitable for larger items but I can't avoid this one. I need two pieces of bronze 3/8" thick and about 1" long. I couldn't find anything close in the shop and I'm not prepared to spend a large amount buying 10 times what I'll need. In the end, I opted to make them out of a piece of bearing bronze I have. I'm wasting a lot of metal but at least the job is progressing. The first thing was to drill a 5/16" hole in the center. This is just for reference purposes so when milling it down I'll be able to see when I'm close to the correct size.

 

IMG_0953.thumb.JPG.93ffeccd20b9cb0f992bb9910f74fbda.JPG

 

I faced off the ends so that it would grip securely in the milling vice and milled off most of one side.

 

IMG_0954.thumb.JPG.fea1e465db8afb92d26f61b4bb8176e8.JPG

 

Then flipped it over and milled the other side using the flat portion I'd made as a reference point. That way the two sides will be relatively parallel.

 

IMG_0955.thumb.JPG.74969caaed06c7d1adcffa08f9d17d9f.JPG

 

I also milled the ends square.

 

IMG_0956.thumb.JPG.1ea645f3b77ab34707c1d224ebbd3e58.JPG

 

I've got the thickness relatively close, about .050 oversize. Now I need the surface grinder again so I spent most of the day replacing the broken belt. This time I used a piece of conveyor belting I must have bought for this job and never used. In fact, until I started looking I'd forgotten I had it. Three hours of frustration and struggle finally got it on. I don't bend as well as I once did and this is a very dirty job done largely in the dark behind the machine with my back up against shelves of parts. However, for the moment it seems to be working.

 

I did get to use this little grinder vise, having only recently realized what it was for. You can hold small parts or parts that are non magnetic and grind the perpendicular side perfectly square without taking the piece out of the vise.

 

IMG_0957.thumb.JPG.5b9e3390dd16e7f3eac1c68b413316d3.JPG

  • Like 2

Share this post


Link to post
Share on other sites

We call it a "permanent magnet chuck" but I don't know what the British term is. It might be the same. Another trick was to glue the piece down with pitch. I've only read about that and can't see how it could be terribly accurate. Years ago, when I was making lenses at the Peerless Optical Company, we glued the glass to an iron holder with pitch.

Share this post


Link to post
Share on other sites

I made a little progress today. This part is VERY fiddly...take off a tiny bit too much material and you have to start over but so far, It's worked better than it has a right to. I took the width of the vanes down to about .005 wider than the roter. This is so, when they are assembled, they can be ground and lapped together. The thickness was another issue. First I tried putting the piece in the little vise but it isn't designed to hold something at the upper edge so it cocked the piece slightly. I ended up trying this method...blocking the piece on the magnetic chuck. I have done this before with not-so-good results but having no choice, I tried it again.

 

IMG_0958.thumb.JPG.7071055dc2339736f4e8982714360d92.JPG

 

Fortunately, it worked a lot better this time. With both sides ground, I lapped the surfaces. Grinding is very good but even here, a lapped surface is a little flatter. You can see, from the shiny spot, the slight variation in the surface. This was literally invisible to the naked eye and would probably defy measuring with anything I have.

 

IMG_0959.thumb.JPG.106d70425791d89b1da7c5cce9715ba3.JPG

 

I lapped it until the vanes fit tightly into the rotor. They have to slide easily when finished but I will leave that fitting until all the pieces are ready to go together. The slots are really perfect... the end mill leaves the upper end about a thousandth wider than the bottom because as you deepen the cut the cutter is passing by that point much more often. I doubt this is or any consequence...in most circumstances, it wouldn't even be worth mentioning. Whether this degree of precision is necessary is questionable but I have a rule that if you work to the closest tolerance you can, and fail to hit it right on, most of the time you're still good to go.

 

IMG_0960.thumb.JPG.45440ae1caa726b35c34ac34444e401b.JPG

 

The vane piece now needs to be cut in half and I have to make a fixture to turn the outer radius which will match the inside of the pump liner rather than the diameter of the rotor.

 

I'll keep my fingers crossed but it looks as if I may have really fixed the grinder belt too.

Edited by JV Puleo (see edit history)
  • Like 1

Share this post


Link to post
Share on other sites

I am still finding your reports and detailed information fascinating and very informative. I remember hearing or reading about pitch for holding parts for machining before. I found I had a spare Clarkson Autolock chuck for my Archdale mill. I am going to modify a drill chuck to fit into this so that I can drill with the milling machine. With the link you sent me about the practical machinist forum I may be able to find a instruction book for my Archdale mill which may help me understand the machine a bit better.

Edited by Mike Macartney
wrong word typed (see edit history)

Share this post


Link to post
Share on other sites

Good. Quite a few of the PM members are in the UK. The site gets people from all over the world (just like the AACA) but aside from Americans UK residents are the most often heard from. I've never heard of a Clarkson Autolock... your mill must have a tapered bore to the spindle. Do you know what the taper is? It's most likely NMTB 40 or 50... the standard adopted in the late 1930s. I am reasonably certain that it was commonplace in the UK also but I admit to being weak on British machine tools. The premier site for general machine tool history is, however, lathes.co.uk.

Share this post


Link to post
Share on other sites

Now to finish the sliding vanes...

The first step was to make a fixture that will allow me to turn the leading edge with a radius of 2.128 – the actual measurement of the inside of the pump body. I cut a piece of aluminum bar, faced the ends and reamed it to 1".

 

IMG_0961.thumb.JPG.150dfc4647dd35f4294d0fc157d40ea0.JPG

 

It was turned down to 2.2" – slightly smaller than the finished size.

 

IMG_0962.thumb.JPG.172d131539c5e05ff85014ebad12b590.JPG

 

Then slots were cut, exactly as I did for the rotor but, in this case, a little deeper. I want the pieces to come out a bit too long so they can be fitted after the radius is in place.

 

IMG_0963.thumb.JPG.520069dfda9b372d97339f85dca9ddf4.JPG

 

They were then drilled and tapped for small set screws to hold the vanes in place.

 

IMG_0964.thumb.JPG.19986599d6e75e519f99ea8559e81e7a.JPG

 

After attaching the vanes to the fixture I set it up on an expanding mandrel. This will be a little stiffer than the expanding arbor I used earlier.

 

IMG_0965.thumb.JPG.e4ab24c42a312307d9228cb857836d9e.JPG

 

And started turning it down. I ought to have milled about 3/8" off these first. The turning took a lot longer than I'd anticipated because I was concerned about the ability of tghe fixture to hold the pieces secure and didn't want to stress it any more than I had to.

 

IMG_0966.thumb.JPG.fe4b70368a258427ae30e23c663fdd6c.JPG

 

Eventually, I got there. Of course, while doing this I thought of a better way to do it but it was too late to go back and this seemed to be working.

 

IMG_0967.thumb.JPG.97958679943c7289247d919ec106d2d0.JPG

 

Here are the vanes with the rotor. They are still both too long and a bit too tight. They have to move freely in the slot but that will be a matter of lapping the vanes and the slot. In a real industrial setting, the slot would have been finished by grinding but I haven't the skill to do that and my grinder is pretty worn. Maybe when I get my better grinder back together I'll try something like that.

 

IMG_0968.JPG.40d82dd17093a3dd35f50609b624617d.JPG

 

Now I have to fit the rotor to the pump body. It was purposely made a few thousandths oversize because I need to do the final fitting with the bushings in place.

Edited by JV Puleo (see edit history)
  • Like 1

Share this post


Link to post
Share on other sites

Fitting the rotor to the pump...

Here it is at its original diameter, just touching the bronze liner.

 

IMG_0969.JPG.bbfd3b1b9cf2eef29611c31e5ef2a4ca.JPG

 

I set it up in the lathe with the dial indicator and decided to take it down .002 at a time. I haven't any good way of measuring just how much oversize it is so I figured that I couldn't get in much trouble with a measurement that small. I doubt I'd be able to do this without the dial indicator.

 

IMG_0970.thumb.JPG.e4a36414dabe83da992b02e1ec3f0d57.JPG

 

In the end, I took it down .022. I suspect there is a slight taper to the bore of the liner but I doubt that will affect it too adversely. In any case, all that is left now is to fit the vanes into the rotor and assemble the pump. Then I get to test it and see if it actually works.

 

IMG_0971.thumb.JPG.af098d6d1d547ba3b30a73ea2d57e860.JPG

 

The gap here is not as large as the picture makes it look. The rotor is slightly longer than the pump body to allow for gaskets and very thin thrust washers on either side. In fact, it may be too short and I'll have to reduce the thickness of the center portion but I'll figure that out when the time comes. I'm off for a few days on a trip so I won't get back to this until next week.

  • 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