The Home Machinist!

I have been trying to use some of the sodium silicate (liquid glass) core binder that reacts with co 2 gas and am having no luck so I need to see what the heck that I am doing wrong. Here is a little prelude of what I am doing and how I am doing it to give everyone an idea. I am mixing the binder in my mifco 65# muller but to be honest I have not paid any attention to how much that I am putting in (does anyone have any suggestions on the amount)? After I have mixed it I take some out and put it into the wooden corebox pack it gently and then strike it off level (it is a 2 part core). After this I then gas the corebox with co 2 and let it set for about 5 or 10 minutes just to see if I had pulled it out too soon. Every time I have the same end result ( a crumble) and I have no clue as to what's going wrong. The binder is in date as it was just bought last friday. The sand that I am using is an oklahoma 90 and the co 2 cylinder has a horn on it to gas the core that actually came off of a fire extinguisher. Anybody got any ideas as I am obviously clueless. One other thing that may contribute to the problem is that I have some slight moisture in the sand don't know if it could snafu the chemical reaction between the sodium silicate and co 2 or not. Let me know if you have any ideas; I know big dave knows this one (or at least I would put money on it) as he's been around this stuff a time or two.

Thanks for the ill gotten credit.

Roll your cores out by using a turning board, or cookie sheet. They should come out like sandy playdo with almost no strength. they wont have until you harden them off.

Sand size has a lot to do with the bond, as does the amount of binder used. Silicate cores can be quite hard and non porous with enough binder and cure time. In other words a problem for venting if too 'tight'. Your core mold sand should respond to the hand squeeze test the same way as any casting sand, maybe less tight, but not WET. In my experience, stiff enough to get it on a cookie sheet. A shot of CO2 is good to set it up some, but you are dealing with a chemical reaction, not drying. So enough CO2 must be present to complete the reaction and develop full strength. That is why I bake the cores in the home oven. It eliminates the water problem, You dont want moisture inside your poured metal for grain structure and blows due to even a little steam. And the oven flame, creates a chamber just loaded with CO2 to fully harden off the binder. Sodium silicate is the binder in most water based paints, and pulls CO2 from the air to harden, after water evaporates. This is the same 'crack' sealer used in engines that reacts with the exhaust gas to harden and -hopefully- seal a pressure leak.

If nothing else, use a shoebox, or a small storage tote to create a gas chamber, and flood your finished cores in CO2 overnight. It should make a diference. Cores should be stable enough to handle without breaking, but not be concrete either. That is controlled solely by the amount of binder and sand/dust media it holds together.

Having been asked the question in a Guru and go to form of limelight, I find myself short here. My casting notes, exposed in the mostly abandoned garage shop #1, has been destroyed by nesting mice. This I know some time ago.

In consulting 3 of Ammens books, a Gingery pamphlet, and Edge of the anvil, There is only one reference to Silicate Coremold, and no formula, for volume or weight. I am sorry. That reduces you to seat of the pants guesstimating.

That means a little work on your part, to take a food scale, and measure and mix, and note what you did, to get where you are at. Do not waste time trusting to memory and chance, write it all down in a collegiate notebok or similar, and bring your conclusions forward. Like shooting artillery, or the F-stops on a studio Camera, bracket your shots, that is arrive at a formula for so many oz sand ,liquid binder, and raise or lower the binder on two more small batches 10%, and work them to finish. See what is best. Fine tune later on % if need be, you just need to land quickly in the ballpark. Dont snicker. They did this all the time in the Sand lab, around the foundry I worked, you dont risk 100 ton castings on some clowns feelings that day. You will have to know, so get the information. Whatever I would refer you to, would start with a weight of a particular sand and screen size, and that alone would be taken from about 20 variables without adding any dry fillers. So please understand, many things are very complex, and while I am not doing 100 ton castings, rules apply, or you can produce a whole day of unusable scrap for your efforts. (Did that, dont want to again, I want fun, and thats not).

Thanks for the advice big dave and I hope the weather isn't as darned hot as it is down south. We have had 3 days of hades to say the least 102 yesterday and 101 today. I gave it up for the day and came inside; the humidity in south carolina is terrible. Oh well, thanks for the advice and I will keep everyone posted. By the way; if the core boxes weren't so expensive I think that the shell core machines would be the way to go. I still want to go over them with a fine tooth comb get them running and see how some cores would turn out.

We have had no trouble with our cores. I believe I follow this formula.
" Budget Casting Supply has a good description of this on their web site. Basically it says to use 3 to 4% sodium silicate by weight mixed with sand:"

http://www.budgetcastingsupply.com/Sodium_Silicate.html

I weigh out mine with a scale and the mix, Make sure you get a good mix, our sand is dry that we use. It is a Silica 90 IIRC. We gased in a plastic bag at first but now have gone to a plastic dish with a cover. I mixed up our last batch on Saturday and let them sit in the dish till we needed them on Monday.

Tom C.

I have been using Sodium Silicate CO2 since the 1970's. I have made both cores and molds with it. I still use a commercial SS that used to be made by Ashland Chemical Company called AdcosilNF, (They gave me a 55 gallon drum at one time, to stop me from asking for 5 gallon samples.).

There are a number of companies that still produce foundry grade SS.

I found that if I used 400 ml per 25 pounds of AFS 105 sand, I got the best results. For mulling, I use a lab muller that can mull only 25 pounds at a time, it worked best that way. I mix up about 150 pounds and store it in a plastic bag inside a garbage can. I get several months out of this tightly closed can.

I usually gassed with CO2 ~ 15 seconds in several places depending on the mold/core.

If you have an old copy of the November 1984 Live Steam Magazine you can see how I used this with "Lost Foam".

Here is a photo where I'm gassing a mold to be made up partially of SS sand. Only an inch or two of the mold will be SS sand, the rest Petro Bond sand.