My 1910 Mitchell "parts car" project

[Terry Harper]

28 minutes ago, edinmass said:

I will check with the art foundry near me on Monday, but the web page says it’s closed due to COVID. Twenty years ago, foundry’s pouring brass, bronze, and aluminum were easy to find......cast iron not so much. Today with EPA regulations most of them are gone. A sad side effect of us not making many things anymore. 

So true! I had to resort to a gentleman that has a "backyard" foundry. Its a hobby but he can pour aluminum, brass, bronze and just about manages iron though its hit or miss. He loves teaching people about foundry work and Its an enjoyable hands-on experience working with him.

 

 

Edited December 13, 2020 by Terry Harper (see edit history)

[JV Puleo]

Average shrinkage for aluminum is 1/8" per foot so in a 4" diameter it's only about .032. I would allow about 1/8 in all directions for machining which will be plenty.

If the bronze place can't do it, I can take the pattern next door. I think we should make at least 2 of them - I always make at least one spare which is why I have the one I photographed on Friday...it's my insurance against something going wrong.

[edinmass]

Joe, I would like to make four castings.........and machine two of them.

[JV Puleo]

In which case, I think a pattern is the way to go. Then you also have the pattern to keep on the shelf for future needs as well - though 4 of these take care of all the extant examples!

[edinmass]

I figured I would give two rough castings to the two museums that have cars...............they will probably misplace them.........but I like to think in the future the cars may get back on the road.

[JV Puleo]

14 hours ago, edinmass said:

I will check with the art foundry near me on Monday, but the web page says it’s closed due to COVID. Twenty years ago, foundry’s pouring brass, bronze, and aluminum were easy to find......cast iron not so much.

 

When I was in business I used a place in Pawtucket, RI - the Rumford Brass foundry. He did brass, aluminum and bronze - a one man shop run by a gentleman named Raoul Goyette. He was the third generation in that location, the foundry having been started by his grandfather around the turn of the century. Raoul was about 5' tall and had forearms that looked like Popeye, the result of 60 years of pouring hot metal. He was also a veteran...

 

One day, while I was there, a "biker" riding a HD chopper with all the conventional tawdry fixings stopped to ask if he could make something. Raoul walked outside and noticed that the back rest on the chopper was a WWI veteran grave marker. He was so incensed by that that he walked over to the bike and literally wrenched it off the back with his bare hands breaking both of the base supports. He then used some colorful language to tell the biker - who was at least a foot taller than him and 40 years younger, where he could go.

[JimKB1MCV]

   As a young engineer I sailed out of the Boston Fish Pier on the trawlers and learned quickly that age and size really are not important in a confrontation.

  Good for Mr. Goyette.

[JV Puleo]

With the mill is set up for the helical gears we'll have to wait until Ed has a chance to test the test gear. In the meantime I'm working on the 2nd Cadillac water pump. This one does not look as if it's ever been apart – all of the original lock washers are in place and everything is very tight.

 

 

The other one was difficult to disassemble but this one is proving even more of a challenge. I tried pushing on the impeller shaft to break the two halves apart...this didn't work, at least with the pressure I was prepared to apply. I may have to heat the casting...

 

 

I was able to get the thermostat adjuster out by heating the casting with my torch. I really don't like using heat on cast iron but there may not be another way.

 

 

I'll think about it. I'm stuck at home tomorrow because I'm having the water pump replaced on my Blazer...odd that I am comfortable with these old ones but really don't like working on modern cars...and I have no garage and it's going to be cold so I'd rather have it done than suffer through doing it myself.

[JV Puleo]

In thinking about this I realized there is no gasket visible between the two halves of the pump. I suspect it was put together with something like the old Indian Head gasket shellac, in which case it would be truly stuck. When I get back to the shop I'll try heating all around the outside edge, probably with a propane torch first, in the hope of melting whatever is in there. Judging from the other pump, the fit of the two halves together is loose but close enough so that would work. It shouldn't be this hard to separate them... I also have an induction heater as recommended by Ed. That might be better still but I'll have to see if the longest wire will go around the pump.

 

We'll have to see, but there is always and answer...just sometimes it takes a while to find it!

[Gary_Ash]

If you think it's gasket shellac holding it together, something like Loctite "Chisel" paint remover may soften it enough to get it apart.  Maybe put the thing in a metal garbage can, spray well, and put the lid on for a couple of days.  The safety data sheet says Chisel is made from things like ethyl-methyl-bathroom-tile and other unpronounceable chemicals, including 30% acetone.  Also, a good bake at 500°F or more might break down the stuff.  An old oven set on the self-clean cycle could be used, but not one you cook food in.  Engine machine shops sometimes have ovens for doing this kind of thing to burn off all the old oil.

 

https://www.henkel-adhesives.com/us/en/product/general-industrial-cleaners/loctite_sf_7631.html

[JV Puleo]

That's a good thought...

I wish I had an oven...maybe a large toaster oven would be a good thing for the shop. As it is, my stove/oven at home died some time ago. I'm replacing it in the spring when I can move the stove and refrigerator I've been given from the family house we're cleaning out. I'll try gentle heat first...then go on to more aggressive means if necessary.

[Terry Harper]

If its a true shellac, denatured alcohol will work well as a solvent to soften it.

[JV Puleo]

The trouble is, I don't know what it is. It could be Gasolia - a hardening varnish that goes back at least this far and I'm sure would go a good job of sealing it. I think the puller used to push on the shaft is a good idea. Having seen the inside of the other pump I'm sure it's pushing the impeller against the other half of the casing so it's just a matter of softening whatever is gluing it together.

 

Terry, I'm going to take you up on cutting some gaskets...I'll make a print to the surface and send it to you.

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

[Terry Harper]

No problem Joe,

 

We will take care of it! A print, tracing or rubbing will work fine.

Edited December 15, 2020 by Terry Harper (see edit history)

[edinmass]

Go to Walmart and get one of those apartment pizza sized ovens that have a broil setting. I have been using one for years..........toss it in on broil for a couple hours......be sure to have removed all hardware possible. And, be sure to have the welding gloves handy. The inductor will work but it’s a fairly large casting. 

[Gary_Ash]

Ed sent the old White water pump impeller to me.  I've got a CAD file created, trying to sort out the last details.  The 8 vanes were machined off at an angle plus a tilt of about 10 degrees.  I'm not sure why they did this and which dimensions are critical, but I got close.  I'm guessing that they set up the impeller in the milling machine and tilted its rotation axis by about 10-12 degrees and rotated the part as it was cut.  But, why did they need to trim off the vanes at an angle?  So far, I can't get the CAD program to add the fillets where the vanes join the base, but I'll keep trying.  Once I get the actual part copied correctly, I'll add some more material for finish machining, then 3D print a pattern.  The vanes are about 1/8" thick and about 0.85" high.  I had to guess at their takeoff angle at the edge, say about 45°.  I think I should print out one copy exactly to finished size to check that it fits in Ed's pump.

 

The impeller was originally pinned to the shaft with a pin about 13/64" diameter (odd size!).  There is a straight section about 3/8" long for the pin.  Joe can indicate off this section for mounting in the lathe.  The wall thickness where the pin was located is about 3/16" - Joe wanted 1/4" for 1 or 2 set screws.  I can increase the diameter of the straight section of the hub another 1/8" to get the meat Joe wants as long as the hub will still clear the internals in the pump.  Will the impeller be too unbalanced by the set screw(s) or am I overthinking this? 

 

Joe, you were going to post that old paper about water pump impeller design.  Let's see it!

 

Here are some images rendered by the CAD program.  The impeller is about 3.5" o.d. x 1.5" high with a hole for a 3/4" shaft.

 

 

Side view of impeller.  Note how vanes are machined off at an angle to the axis.  There is also a slight tilt, ~10 degrees, in the cut to the side,

but I don't think this matters, just an artifact of how they machined it, not included in the CAD model.

 

Rear view of impeller showing recess and flat of about 1.24" diameter.

 

Top view of impeller showing curved vanes.  

 

Original 1917 White impeller, sand cast aluminum.  I see why Ed wants a new one!

[edinmass]

Everything on the “Great White” is strange and difficult to deal with.........especially the owner!

Edited December 16, 2020 by edinmass (see edit history)

[Terry Harper]

Was the pin a taper pin? That could possibly explain the odd size. 

Given the quality of everything else on the "Great White" I am surprised the impeller wasn't bronze but than again, given the 

time period aluminum was sort of the new miracle alloy of the age.

[edinmass]

No taper to the pin.............and it was a nightmare to remove............

 

 

In 1917, aluminum was still rather a rare and new metal to the public, and a "sign" of quality and a representative of above average engineering. 

Edited December 16, 2020 by edinmass (see edit history)

[Gary_Ash]

I had to beat the CAD program into submission - or maybe I figured out what I was doing wrong - and got the fillets on the vanes.  I ran the CAD file through the Ultimaker Cura 4.8 "slicer" software, loaded the resulting GCode file in the Creality ender 3 printer, then started the 3D printing process.  With the quality set to medium, It will print out in about 7.5 hours.  So, by this evening, we'll have a prototype impeller which I can send to Ed to check the fit.  I've got my fingers crossed.

 

 

[edinmass]

Went to the foundry today........they are closed down permanently...........back to the drawing board. 

[TonyAus]

7 hours ago, Terry Harper said:

Was the pin a taper pin? That could possibly explain the odd size. 

Given the quality of everything else on the "Great White" I am surprised the impeller wasn't bronze but than again, given the 

time period aluminum was sort of the new miracle alloy of the age.

The impellor was most probably fixed with a standard taper pin (1/4 inch to the foot taper).  I recently dealt with one on my 1925 Dodge - which was also offset to the shaft.  When replacing the shaft I turned up a stainless steel taper pin, reamed the hole with a matching taper reamer and fixed it with Loctite.  Should be good for some time. 

[Laughing Coyote]

Can it be machined out of billet aluminum on a CNC mill?

[TonyAus]

Another thought.  Some time ago I read an article in The Horseless Carriage Gazette about making a replacement water pump impellor by CNC for a very early car.  The make escapes me but the original had straight fins and the replacements were curved.  This article explained the rationale for the change in terms of hydrodynamics.  It appears to have worked.  Unfortunately I don't have access to the Gazette due to the current lockdown (club archives) but it would have been published sometime in the last 15 years.  

[JV Puleo]

Later tonight, when I get home, I will post what PM Heldt has to say about impeller design.

Today's efforts almost ended in disaster...and may still pose a major problem but I think I can deal with it.

I started by setting the pump up in a vise on the drill press table with a puller rigged to push on the impeller shaft...the idea being to push the halves apart without putting strain on the rather thin ears that the nuts pull up against. I heated all around the edge with my torch before doing this.

 

 

Everything was going reasonably well and a gap was opening between the two halves when I heard a disconcerting "snap"...

Much to my surprise, I'd broken the rear casting.

 

 

Needless to say, this didn't make my day.I stopped to think about it and then went on to continue the dismantling...after all, if it can be repaired it can only be done after I have the separate parts in hand. The rest of it went fairly well.

 

 

It's clear this pump was in better condition. When I did the other pump I added thrust washers on either side of the impeller and now I see that is exactly what was done originally. The washer on one side is fine. The one on the other side is worn paper thin but it hasn't eaten into the impeller as had happened the first time.

 

 

So, regarding the broken housing. It's cast iron under no measurable pressure, the break is clean - in fact it doesn't even go all the way around so lining up the two pieces will not be an issue. I think that it can be carefully brazed although to do that properly the entire piece should be brought up to nearly brazing temperature. I may do this myself but before I do I'll check with the gentleman in town I've used before...it has to be done carefully but there is no reason why it can't be fixed. Failing that, I've a friend who has another pump but I'd rather fix this one than sacrifice a good one. The first step is to get it antiseptically clean so I dropped it in Evaporust. It will stay there as long as it takes out get all the rust out then I'll see about getting it even cleaner.

 

Edited December 17, 2020 by JV Puleo (see edit history)

[JV Puleo]

I'm also wondering if stitching might be a better solution...I wouldn't try brazing except that it's a clean break and there is no need for tremendous strength. I'm going to have to think about it more a lot before I do anything.

[edinmass]

The casting will be too thin to stitch. I’m not convinced that brazing it will work long term. Your going to have to heat it to 1100 degrees and then cool it down very slowly...........it’s likely to be stressed unevenly along the crack. Addition cracking next to the repair is probable......and any inclusions in the casting would be a game breaker. It’s going to have to be so hot that it will probably require re-machining and surfacing of all mating surfaces. I would source another pump if possible. Since the owner only has one car and two pumps, the repair of the cracked casting probably isn’t necessary. I “properly” repaired a crack on a cast iron Pierce 12 head four times.......and it continued to crack .........this was all many years ago before I knew of stitching. Castings stress relieve in ways hard to predict after a heated repair........even at a high temperature. Often times skill and process when properly done just won’t hold. Nothing ventured, nothing gained. If a good casting can be obtained for say 400 dollars or less............it’s a much better option than rolling the dice. IMHO..........

Edited December 17, 2020 by edinmass (see edit history)

[JV Puleo]

That's basically my thinking on this too. But, if it can be repaired there's no reason not to try...it's useless as is.

The car does need two pumps though - one for  each bank of cylinders so I'll have to find a way. I also thought of having one cast...I'll show it to the foundrymen next door and get an opinion. My guess is that it would require too much pattern making to be worthwhile.

[JV Puleo]

This is from the 1917 edition of PM Heldt...

 

 

 

 

 

 

 

 

 

 

 

 

 

[Gary_Ash]

The 3D printer finished the first print just after 10:00 tonight.  The parts looks good, dimensions check well with the original, surface finish is very good.  I used black PLA for the print.  I made a revised CAD file for a casting pattern to add 1/8" to the surfaces that need machining, though not to the upper tips of the vanes.  I'm hoping the final casting will be close enough to size to not need the complex set-up for machining the tips.  It appears that the distance from the bottom of the impeller to the 1.24" dia. surface is used to set the position of the impeller in the pump, so both need to be machined smooth and to set the exact distance, which I measured at 0.450".  I've started the printer again to make the pattern, will be done in the morning after 9.5 hours of printing.  I added 1.57% to the part dimensions (3/16" per foot) for shrinkage allowance during casting in generating the 3D printing file.  Fortunately, the printer needs no supervision while it's working.

 

Terry, the software I am using is TurboCAD Pro Platinum, 2018 edition.  They'll soon force me to upgrade to the 2021 version for another $400, but that seems to be the model for software companies now.  If Santa would only bring me a new Creality printer that can print from 2 materials at once, like PLA and PVA, it would eliminate peeling support structures off some of the prints.  The good news, is that I haven't run out of things to ask Santa for yet.

 

I'll put the 3D print in the mail to Ed tomorrow, but will hold the original part for a few more days to make sure the pattern comes out right.  I'll chat with Joe about the best way to get a casting made.  I have used an art foundry in Providence for bronze parts and Joe has the aluminum foundry literally next door to him.  I'll have to look carefully at the old technical paper that Joe posted about how these centrifugal pumps were designed before 1917.  We owe a lot to steam engine technology.

 

Top view of original impeller and 3D print to same size.

 

 

Bottom view of impellers.

 

Impellers at angle.

 

Impeller dimensions.

[edinmass]

It never ceases to amaze me the talent of the people who regularly post on this forum. The pattern looks fantastic...........hopefully Santa is good to Gary and Joe!

[Mike Macartney]

 

22 hours ago, Terry Harper said:

. . . . given the time period aluminum was sort of the new miracle alloy of the age.

 

Was aluminium used later in the USA than in the UK? Below is the the front aluminium wheel from my circa 1900 Singer motor wheel tricycle.

 

 

Joe, I thought you took the cracking of the water pump casing very well. I don't think I would have reacted in such a matter of fact manner. Mike

 

[edinmass]

I had sent Joe a PM about the pump........but I will post it here in the open also..........when working on 100 year old components things happen.....and most "things" are not positive. Joe is a very experianced mechanic and machinist. More than 99 percent of the people working in repair and restoration shops. Put plainly and simply.............none of this stuff is easy, and anything can happen. Two weeks ago I damaged a very valuable part......it happens. I try and never have it occur ....but if you work on stuff, unpredictable things happen. Keeping this in mind.........

 

 

The failure of the pump was NOT Joe's fault. He's skilled and doing his best. He is not financially responciable for the failure(some people actually think that the shop should be, no one knows how things will end up.)........sxxt happens. And it happens much more to 100 year old lumps of cast iron. Ask yourself this question........Why did the pump get sent to Joe? Answer....because he's smart and capable and most likely the best person the owner of the pump could find........and I agree. Usually when something gets into my hands it's been to two or three shops that couldn't or wouldn't fix it. There is inherent risk working on anything........and fifty times more on very old cars. Ultimately it's an unforeseeable problem........and you take the ball and run with it. While working on my 1917 White .....EVERY DECISION....... is based upon several criteria ..........keep all possible parts and repairs in house......other shops can lose items, things disappear in shipping, and many people just don't treat irreplacable parts with respect. I always try and look at thing in a positive manner.......but also try and figure every down side to a repair ahead of time. 

 

FYI- My opinion of Joe's skill set? He is one of three or four people I TRUST WITH MY OWN PARTS............he's talented and has gasoline in his veins. An almost impossible to find combination today. And one other thing about Joe's work.............he never takes a short cut.........and he WILL FIX what he is working on. He doesn't give up or give in. Truly the result of being a highly skilled craftsman. 

 

Edited December 17, 2020 by edinmass (see edit history)

[JV Puleo]

5 hours ago, Mike Macartney said:

 

Was aluminium used later in the USA than in the UK? Below is the the front aluminium wheel from my circa 1900 Singer motor wheel tricycle.

 

 

The method of refining aluminum in commercial quantities was only developed around the 1880s. Up until then it was regarded as a semi-precious metal. Napoleon III had a dining service made from it to impress his guests and there are mid-western half-stock rifles that date from about the Civil War and slightly later that have aluminum inlays. As soon as it became commercially available it was treated as a "wonder metal" and used for just about everything it was thought suitable for. My 1897/1899 (we were never sure of the date) Panhard chassis had an aluminum transmission case and the engine (which was built n 1903) had an aluminum crankcase. Unfortunately, a few of its applications, like water pumps, really weren't suitable, at least for the alloys available at the time. But, in the short term, even poor alloys were better than none. It was great for aircraft engines in WWI that had a life expectancy of maybe 15 hours before being overhauled. W.O. Bentley installed aluminum pistons in his racing DFP about 1912 or 13. He kept it a secret until 1914 when he gave the information to the War Office.

 

The potential value of the metal was fully appreciated and the alloys got consistently better but I suspect that the UK was slightly ahead of the US in it's application. Immediately after WWI the American firm ALCOA hired Vauxhall Chief Engineer Laurence Pomeroy to develop aluminum alloys for automotive purposes. Pomeroy had been an early and enthusiastic promoter of the material because it lowered the reciprocating weight of engine parts. He published several papers on the subject in the SAE Journal and apparently had much to do with the development of Lynnite - the alloy used to make Franklin connecting rods and, I think, Dusenberg. Apparently they don't wear well in the Dusenberg and most have been replaced, probably as a result of the stresses they are exposed to, but I've never heard of a problem with the Franklin.

 

If you are wondering how I happen to know all this, it was part of my research for designing aluminum connecting rods.

 

Edited December 17, 2020 by JV Puleo (see edit history)

[JV Puleo]

As to my reaction, I wasn't always calm about things like this but 50 years in business, mostly printing and publishing, has taught me that the only worthwhile reaction is to step back, take stock and then go about doing whatever is needed to right the situation.

[Bloo]

1 hour ago, JV Puleo said:

As to my reaction, I wasn't always calm about things like this but 50 years in business, mostly printing and publishing, has taught me that the only worthwhile reaction is to step back, take stock and then go about doing whatever is needed to right the situation.

 

It is the most important skill that exists in any sort of auto repair, but one of the hardest to acquire.

 

 

 

[JV Puleo]

13 hours ago, Gary_Ash said:

 

Impeller dimensions.

 

 

That radius in the base is an interesting problem. What does it fit over? If the projection has a radius it may be necessary to have a special cutter ground or finding a corner radiusing cutter the right size. If the projection has parallel sides, it isn't a problem.

Edited December 17, 2020 by JV Puleo (see edit history)

[Terry Harper]

16 hours ago, Gary_Ash said:

 

Terry, the software I am using is TurboCAD Pro Platinum, 2018 edition.  They'll soon force me to upgrade to the 2021 version for another $400, but that seems to be the model for software companies now.  If Santa would only bring me a new Creality printer that can print from 2 materials at once, like PLA and PVA, it would eliminate peeling support structures off some of the prints.  The good news, is that I haven't run out of things to ask Santa for yet.

 

 

 

Hello Gary, 

We use Solidworks but we are an educational facility so its affordable (thankfully!) We have one 3D printer that is a dual extruder and I hate it with a passion. Its a Makerbot 2X that prints only ABS and a dissolvable filament. It's never worked right. The ABS warps and lifts so easily and trying to keep both extruders going without clogging is always a challenge. In fact often when using both extruders the supports will not be registered with the object. After countless hours trying every trick in the book its now pretty much a paper weight.

[edinmass]

[edinmass]