Mistakes were made, part three

This post is from my series on building a backyard foundry.

You remember back when I said in part two of this series that I was temporarily using a flimsy stainless steel tub as a crucible until I managed to obtain a 3 1/2 inch (nominal) pipe nipple? Turns out that when you think “I can probably get one more melt out of this thing before it is destroyed”, that is the time to throw it away. The crucible failed. Fortunately, the crucible was still in the furnace.

As I also mentioned before, there was a design problem in that I re-routed the tuyere to come in the bottom of the furnace, which meant that if there was a crucible failure in the furnace, then the molten metal would head down the air pipe to the fan. Which happened. Fortunately the fan was blowing cool air and the melt froze in the air pipe. The remaining small amount of molten metal in the furnace could be easily scooped out with a long handled steel spoon, so no significant harm was done. I was going to be rebuilding the air pipe anyways.

I’ve done so. I’m still not very happy with the design, but it is better. Basically I have a T:

              | =========== air input
              ||  <— aluminum foil plug that will melt

An immediate problem was that ash and coals are going to fall into the pipe and fill up the drain. I solved this problem by putting a pipe cap with twenty or so holes drilled in the sides on top of the air pipe. That certainly lets enough air in, and, bonus, directs it towards the charcoal surrounding the pipe. Whether it will let the metal out in the event of a crucible failure, I hope to not find out. I think it will.

I’ve been trying to come up with ideas for better designs for future furnaces; should there be multiple tuyeres to distribute the airflow around better? Should the floor of the furnace slope towards the drain? If I ever build another furnace I’ll experiment with these ideas.

About these ads

Big boxes

As someone who owns an old house and likes “do it yourself” projects, I spend a lot of time in “big box” warehouse stores. I try my absolute best to interact with as few employees as possible when I go to these stores because it never seems to go well. Here are a few conversational highlights from over the years:


—— Holy trash bags, Batman ——

Me: Hi there, you probably don’t have these but Western Safety is closed today. Do you have large four or six mil tear-resistant trash bags?

Big Box Store Employee: I don’t think so; what do you need them for?

Me: I’m tearing up a hundred-year-old sub-floor and the test for asbestos contamination has come back positive. The toxic waste dump won’t take asbestos contaminated waste unless it is properly bagged and labelled.

BBSE: Well, I’d just bag it and throw it out in the regular trash and not tell anyone.

(I went to Western Safety.)

—— Circular is the round one ——

Me, speaking to the guy at the tool counter: Hi, I need an eight inch abrasive cutoff wheel suitable for cutting thin, soft steel with a chop saw or circular saw.

BBSE: You mean these? 

Me: Those are reciprocating saw blades. Circular saw blades are circles.

BBSE: Oh, so you mean these?

Me: Those are ten inch wood cutting blades.

BBSE: Hmm. You mean these?

Me: Those are concrete cutting wheels.

BBSE: How about these?

Me: Those are metal cutting wheels but those are four inches wide. I need eight.

BBSE: Maybe you should try Lowes.

(I tried Ace, successfully.)

—— It’s not a nuclear reactor, it’ll come back online easily enough ——

Me, fifteen minutes before the store closes: Can I have this twelve foot board sawed into two six foot boards? I need two six foot shelves, and a twelve foot board won’t fit in my car.

BBSE: Sorry, the saw is already shut down for the night.

Me, speaking to the store manager 90 seconds later: I have a question for you: is it the policy of this store that the saw “shuts down for the night” at some time before closing?

Manager: Uh, no… who told you that?

Me: I think it was the guy who just vacuumed up the sawdust and doesn’t want to do it again.

Manager: I know just who you mean.

(They sawed my board but boy, were they not happy about it.)

—— That’s just smurfy ——

Me, talking to a guy in the electrical aisle: Can you tell me where to find one-inch diameter flexible electrical conduit? It is made of thin, ridged plastic and is sometimes called “smurf tube” because it’s that colour of blue.

BBSE: Sorry, we don’t carry anything like that.

As I turned to leave I realized that of course the smurf tube was directly behind me; the BBSE was looking at it as he was telling me he didn’t have it.

—— How useful! ——

Me, talking to a (different) guy at the tool counter: Where are the rivets?

BBSE: Rivets?

Me: Rivets.

BBSE: I’ve never heard that word before; what’s a rivet?

Me: A rivet is a metal fastener usually used to attach metal objects together. You insert the rivet through the objects you wish to fasten together and then deform one end of the rivet by peening it with a special tool. If you have access to both sides of the objects you can use solid rivets, otherwise you can use hollow rivets.

BBSE: Wow, that sure sounds useful!

(Another employee knew what rivets were, and, bonus, where in the store they were.)

—— Take a number ——

Now, I understand that the people hired at big box stores have no experience whatsoever using any product that they sell, and, as we’ve just seen, often no knowledge of what they sell in the first place. I know that if I want knowledgeable conversation about a tool with an expert I should go to Hardwick’s, which is like paradise for hardware geeks. The trouble is that they don’t have convenient hours; they’re closed by the time I get home from work, and not open Sundays. I try to go to local small-box stores as much as I can. Which is why this experience I had at my local small-business lumber yard yesterday was so disappointing:

Me: Hi there, I need three dozen eight foot two-by-fours and three sheets of quarter inch drywall.

Cashier standing by the front door: I think we have those.

Me: I’m quite sure that you do, since this is a lumber store. Are the two-bys and sheet rock in this building, or in the warehouse across the street?

Cashier: I don’t know. I think you’ll have to ask someone else.

Me: You don’t know where the two-by-fours are?

Cashier: This is only my fifth day on the job. Take a number and someone will help you.

I would have thought that “where are the two-by-fours” is the kind of thing you’d sort out on day one at the lumber store, but, whatever.

At this point I note that I am the only customer in the store. Behind the counter there are five employees. Three are talking amongst themselves. One is typing on a computer. One is on the phone. As instructed, I take a number, and walk over to the paint aisle to browse spray paint while I wait for one of the five people behind the counter to call my number.

They do so immediately. The moment my number is called, the three employees who were talking amongst themselves immediately leave the building by the back entrance, and the guy on the phone hangs up and leaves by the front entrance, leaving in the building me, the guy on the computer, and the cashier who does not know where the lumber store keeps their two by fours. I point out to the guy on the computer that my number has just been called, and he says that someone else will help me shortly.

I waited ten minutes watching him silently ignore me, typing away, and then I left and went to the big box store at the other end of town; I knew where the two-bys were there. 

Attention small business owners: I am doing my best to give you my money. Stop making it so hard.

Attention big box store owners: You run vast multinational corporations with huge profits. You can afford to hire and/or train employees to familiarize them with the products you sell and their basic functions.




I emailed the last portion of this blog entry to the owner of the small business involved, and:


I really appreciate the opportunity to address such an egregious example of poor service. I won’t bore you with the details but it was a bad intersection of shift changes, yard service people hanging out at the counter and too few sales people. We watched the tape of your arrival and departure and have talked it over with everyone involved. Please let me tell you we’re embarrassed and ashamed of the way we treated you. Please accept my sincere apology.


The owner also offered me a discount on my next order and free delivery, which was I think a very nice gesture. As I have said often, you can tell the quality of customer service at an organization by how they deal with mistakes. Good service means recognizing the mistake, taking ownership of it, identifying the structural problem that allowed it to happen, and making a gesture of goodwill to the customer; this is an example of really excellent customer service, and I appreciate that very much.

Static analysis of “is”

Returning now to the subject we started discussing last time on FAIC: sometimes the compiler can know via static analysis[1. That is, analysis done knowing only the compile-time types of expressions, rather than knowing their possibly more specific run-time types] that an is operator expression is guaranteed to produce a particular result.
Continue reading

Reduction and oxidization

We all have a basic understanding of “oxidization” I think: when a metal like iron is exposed to oxygen, either in the air, or dissolved in water, the metallic iron turns into iron oxide, which has quite different properties. In particular, iron oxide is brittle, flaky, and expands away from the underlying metal, which means that oxidization destroys the metal. Copper produces a green oxide. Aluminum oxide does not flake off; it actually forms a protective “passive” layer on top of the aluminum protecting it from further oxidation. Same with the chromium that is in stainless steel; the surface is actually chromium oxide, which does not flake off. 

That chemical reaction can of course be driven in the other direction; it is possible to turn an oxide back into a pure metal by a variety of means. My favourite technique for doing so is to put a tarnished silver object loosely wrapped in aluminum foil in a plastic tub full of boiling water with baking soda dissolved in it. The silver oxide is reduced to metallic silver, and the aluminum is aggressively oxidized far more than it would be when exposed to gaseous oxygen. Essentially the oxygen is moving from the surface of the silver to the surface of the aluminum. The cheap aluminum can then be discarded. This technique is not only less work than polishing by hand, it saves the silver. (Conventional silver polish is simply a chemical which dissolves and removes the silver oxide, destroying the silver.)

Now, that description of polishing silver with chemistry should be making you scratch your head (provided that you do not remember your grade twelve chemistry): what’s the baking soda for?

It’s to help move electrons around; it make the water more conductive.

But wait, what do electrons have to do with it?

Oxidation is in these specific cases the reaction of a metal with oxygen to form an oxide; 18th century chemists knew that, and hence gave it the name “oxidation”. And those same 18th century chemists also observed that when oxidized metals were turned by chemical means back into pure metals, they appeared to lose mass, and were hence “reduced”. Of course the “lost” mass was nothing more than the oxygen leaving the metal and going into solution or into the atmosphere, which those same clever chemists did manage to work out for themselves as well.

We now know that the hallmark of an oxidation is that the oxidized substance must give up some electrons in order to bond with the oxygen, and the reduced substance must accept those electrons. The key fact for the chemist to understand is that the number of electrons is always conserved. If the oxidized substance is giving up, say, two electrons per oxidized molecule then some reduced substance must be accepting those two electrons. There is no oxidation without reduction. (*)

I said last time that what I wanted to talk about the air blast into the charcoal furnace. A charcoal fire is essentially oxidation of the fuel at a tremendous rate, and the thing oxidizing the fuel is in fact oxygen.

Suppose the air blast is set so high that the rate of oxygen arriving is outstripping the speed at which the oxidation reaction can occur. What characteristics will such a fire have?

First off, it will be incredibly hot; the heat-producing chemical reactions are happening as fast as they possibly can.  Second, the atmosphere, despite having been reduced enormously by the oxidation of the fuel (remember, if something is oxidized then something else has to be reduced, and that would be the atmosphere) there will still be oxygen available for more oxidation of something else. Such as the metallic body of the crucible itself, or the metal in the melt. Third, there will be a strong flow of air out the chimney, carrying the heat with it. Fourth, the exhaust gasses will be extremely hot and quite “clean” for a charcoal fire, because more of the combustion will be “complete” combustion.

Now suppose the rate of oxygen arriving is just barely enough to run the oxidation reaction, or even slightly less. What characteristics does such a fire have?

It will be less hot because the chemical reactions are not running at peak speed. The atmosphere will be entirely reduced, leaving no oxygen to oxidize the crucible or melt. In fact, the atmosphere could be so reducing that it starts pulling oxygen out of the crucible and melt, unrusting them. The airflow will be lessened, and the smoke is likely to be dirtier. 

What I think will work, and I’ve been trying to do, is to run the furnace so that the air blast produces a reducing environment until the metal melts. Once I’ve melted as much as I want, I turn the air blast up to rapidly increase the temperature from the melting temperature of 1220F to the pouring temperature of 1400F. 

When it achieves this temperature the melt should be glowing with a red heat, and it will not stick to a steel rod used to stir the melt. 

Lacking a pyrometer, it’s going to take some practice to figure out exactly what the right moment to pour is. I’ll continue to report my results as I practice.


(*) There can be oxidation without oxygen; perhaps the oxidized substance is giving up its electrons in order to get together with sulphur, rather than oxygen.

Mistakes were made, part two

This post is from my series on building a backyard foundry.

My third mistake was building the furnace before I had obtained the crucible. Post construction I went to a number of thrift stores looking for cast iron pots, tall stainless steel tubs,  and so on, to try to find something that would fit the 6 3/4 inch bore of the furnace. It would have been better to obtain a good crucible first, and then ensure that the furnace fits it. Remember, for a charcoal furnace you have to be able to pack charcoal around every side of the crucible; the maximum outer diameter of the crucible should be about 2 inches less than the bore of the furnace.

Thus far I’ve been using a cheap 4 inch diameter stainless steel tub made of pretty flimsy steel. Though it has almost no thermal mass and therefore heats up red hot very quickly, the thin steel will (1) be dissolved by the molten aluminum on the inside, and (2) will oxidize on the outside, and will eventually fail. I’ve therefore obtained a 3-inch inner diameter black steel pipe nipple and pipe cap from Ballard Hardware. It needs some modifications before it will be a useful crucible; more on that later. I wish I had obtained the crucible first, because then I would have chosen the larger 4-inch inner diameter nipple, which does not fit well in the bore.

My fourth mistake was one of operation, and was very simple to fix: I did not use nearly enough charcoal the first couple of times I tried to melt aluminum. Once I started putting in a good four inches or so in the bottom, and more on the sides, it melted nicely.

My fifth mistake was also one of operation: when I went to go for my first pour, when the metal actually melted, I got too excited and poured too soon. I was not molding anything other than ingots in a muffin tin; the metal froze in the crucible when halfway poured into the muffin tin. The combination of pouring too soon and having not enough fuel was not good.

It is somewhat dangerous to have a mass of solid metal in the crucible, because when it heats up again, the metal will expand and possibly break the crucible. Fortunately, after I allowed it a day to cool down, the metal came out of the bottom pretty easily in a big lump. Of course, I will melt it again. One of the truly nice things about metal casting work, as opposed to, say, fine woodworking, is that you take only a small loss of materials for mistakes. 

The problem of getting the metal up to pouring temperature is, once there is enough actual charcoal in there, essentially becomes an oxygen supply problem. More on that next time.