Here is the new pump, pretty much done and ready to mount on the bike. The lower fitting goes to the radiator and the upper feeds the engine. The studs holding it together are extra long because they will also go through the mounting bracket. As you can see, I also cast and machined the pulley. It was a regular deal, no need to elaborate. I got all set up to cut out the bracket and realized I was out of acetylene. After a trip to the welding supply I will get with it. I hope to have it all done and pumping by the end of this week.

Persistance Is It's Own Reward

Yesterday I poured , yet, another housing. I am happy with the result. As you can see, it is a bit different from the original one. I added a bump out on the back for the inlet instead of the flat back. This seemed to be a good idea, but it introduced a few new problems. The first was the issue of shrink defects. I actually poured another one a few days ago and it had a moderate divot on the rear bump. It was a usable casting as is. The defect was more visual than anything else. It upset my eyes to look at it, so I decided to make another. The biggest problem was that the bump made it, nearly, impossible to chuck it in the lathe. It needs to spin on center in both the horizontal and vertical planes to properly finish it. A huge PITA. My solution to both problems was to drill the back of the pattern and use a removable dowel for a sprue. This allows extra metal in the area that is prone to shrink and, since I drilled it on center and perpendicular to the mounting face, the sprue can be used to chuck the housing in the lathe. It worked out just fine and the machining process was a snap. My big problem now is to find the proper bit to drill for the 3/4 NPT fittings. You don't just go to The Borg and pick up a 29/32" blister packed, Chinese cheepo bit off of the pegboard. I think I will try to bore it out. We'll see.
Well, as usual, I screwed up and ruined my beautiful casting. I tried to bore it out and something went awry. The hole ended up egg shaped and too large at the same time. So much for experimentation. I went straight to Ebay and ordered the proper 29/32" bit. I should have it by the middle of next week. At least I have the casting process down to a science. I'm busy tomorrow, so maybe Sunday I will give it another shot. I think I will go ahead and pour a couple of housings, while I'm waiting for my bit to arrive, so I can have a spare. I'm normally not a very patient person so maintaining my cool has been a challenge. So far there has been little cursing and no thrown tools/parts. I can last for only so long, though. I've lost track of the number of pours that I've made.
Today is Sunday. I poured another housing today and it came out very nice. Probably the best, so far. Maybe I'm getting better or luckier. I'll take it however I can get it. I faced it off, bored the cavity out and generally cleaned it up. I will resist any urges to try to drill the inlet and outlet openings until my drill bit arrives. I can ram up and pour one of these things in about 2 hours, but I'm not anxious to do it again. I still have to make the pulley and a mounting bracket. The pulley will be cast aluminum and I will probably cut the bracket out of 1/4 or 3/8 CRS.
The first pic is of the sprue being trued up so I can turn the casting around and chuck up the resulting protrusion and face the housing and bore it to size. The next pic is of the semi finished pump. I just have to drill and tap for the 4 studs that will hold it together and the inlet and outlet. The sprue will be cut off since I no longer need it.

Introspective Thoughts and Self Flagellation

I've thought about all of the failures that I've experienced while trying to cast the pump housing. Each one was different....sand too wet, melt too cool, core shift, sprue too close to edge causing fall out, etc. I feel that I've educated myself and the next pour will be a good one. Part of the problem was that my flask was too small and I couldn't ram the sand properly or keep the sprue holes far enough from the outside of the sand. Today I built a larger flask, re tempered my sand and, generally, got ready for tomorrow. It's supposed to rain, so it will be a good day too fire up the furnace and pour some metal.

Overlooking the Obvious in Search of the Obscure

Often, I find that my superior intellect interferes with my ability to see what is in front of my face. It is the curse of being a Boy Genius™, I suppose. The #3 housing would have worked perfectly well had I not drilled it in the center for the inlet fitting. I did it because it was expedient, since I had it chucked up in the lathe, anyway. Had I drilled it off center, into the area where the impeller vanes are, it would have worked just fine. The coolant doesn't care how it gets in. In the stock, MoPar configuration, the entire backside of the impeller is in contact with the coolant. All that really matters is that the outlet be at the periphery of the housing, since the centrifugal action of the impeller forces the flow to the outside. In the next couple of days I will pour, yet another, casting. As long as I don't try to reinvent the wheel, I expect the next one to be the last.

"You can't always get what you want, but if you try sometimes, you just might find, you get what you need." Micheal P. Jagger

*&%$*$@!!!!!!!!!

Well, tomorrow is another day.
I failed again, but it was entirely my fault. I stopped in the middle of a pour and got a huge gas bubble that didn't show up until I was boring out the back for the inlet fitting, which was the last step. I couldn't have found it before I spent several hours machining it. NOOOOO, that would have been too easy.

Snatching Defeat From The Jaws of Victory

Figuring that the third time is the charm, I rammed up the mold and poured another housing today. The casting came out very nice except for a few ragged edges at the parting line. No big deal, a few swipes with a file and we're lookin' good. I chucked it up, faced off the front and back surfaces and bored it out to accept the pump. So far, so good. As I stood there looking at it and feeling pretty smug about my limitless talents, I realized that there wasn't enough clearance at the rear of the pump shaft to allow coolant to flow in. The shaft nearly touched the inside of the housing. If I counter bored the housing to allow room for coolant to pass by, it would be too thin. Rats!! What I did was to attach a disc of 3/4 plywood to the back of the wooden form. This will give me enough meat to counter bore for coolant flow and the housing will still be thick enough to drill and tap for the 3/4 pipe threads needed for the fittings that will connect to the radiator. I forget how many light bulbs Tom Edison built before he got one to work, but it was more than four. Tomorrow, I will have a working pump housing

Prometheus Tamed

Today I rammed up the pattern for the water pump housing, fired up the foundry and chucked in a couple of old pistons and a top fork crown from an old Honda ATV. In 25 minutes or so I had a nice crucible full of molten aluminum. As I poured it into the mold, my gut feeling was that it was going to be a bad pour. I was wrong. It was a horrible pour. I think the sand was too wet and steam filled the upper part of the cavity instead of the alloy. No big deal. I'll allow the sand to dry out a bit and try again later. It's been about 5 years since I messed with the hot stuff, so I'm a little rusty. I have plenty of time and the cost is next to nothing, so I'll keep on trying.

The second pour was another failure. Not as spectacular as the first, but a failure nonetheless. I didn't allow the aluminum to get hot enough and it began to freeze in the crucible as I poured.
Though I melted enough metal, I came up short and the mold didn't fill completely. All was not lost, however. The casting is unusable but, other than a large hole in the back of the housing, it came out rather nice. Let's hope that the third time is the charm. If I had a pyrometer, I could avoid the guesswork but it's not in the budget. They didn't have fancy stuff like that 100 years ago and they did OK.
It's too late today to do another, so I will have to wait until tomorrow. Stay tuned.