Flame Tests

[Tscarborough]

I have one area around my flue that is too oddball shaped to cut, so I need a small amount of castable refractory. Being the cheapskate that I am, I decided to formulate my own and test it. I have a propane heat gun that gives the following results:

at 6" from nozzle: 1120 F
at 12": 650 F
at 24": 390 F


I also wanted to test the perlite cement samples I had already made.

The first batch I used was:

1 part portland
1 part fire clay
1 part fine silica sand

It was very sticky and workable and dried with no shrinkage cracking.

It only took 30 seconds at 6" F for it to begin spalling. I got tired of holding it at 12" with no spalling (about 2 minutes). It appears that the sand was popping, probably from expansion.

Second batch is:

1/2 part portland
1 part fireclay
1 part fine coal slag

It is also very sticky and workable, I will dry it today and test it tomorrow.

Third batch is:

1/2 portland
1 part fireclay
1 part marble dust (1/4" to fines)

It is also very sticky and workable, I will dry it today and test it tomorrow.

The perl-crete sample was an 8 to 1 mix formed into a 5x7x2" slab. I heated it at 6" until it was glowing red (about 45 seconds), and the backside, while warm, was touchable. No material degradation at all after a total of about 3 minutes.

[david s]

Re: Flame Tests

You'd be better off using the sand, lime, fireclay, cement, 3:1:1:1 brew. Better still get a bag of castable refractory that contains calcium aluminate.

[kebwi]

Re: Flame Tests

Hey, I responded to this thread five hours ago. What happened? Rats!

I was thinking it would be great to conduct a small suite of test with 'cretes of various ratios and thicknesses. This is actually an idea I had a while back, but I don't have a precise heatsource like you. It would be fun to try to put together though. I'll have to give it some thought.

Cheers!

[Tscarborough]

Re: Flame Tests

The sample with coal slag did not spall at 6", so after 1 minute I put the direct flame on it and it spalled slightly.

The sample with marble dust did not spall even under direct flame, so I will use that mix.

I will also make a sample of the standard mix: sand, lime, fireclay, cement, 3:1:1:1, but I can tell you already that it won't perform as well as the
1/2 portland/1 part fireclay/1 part marble dust mix for this application.

Pics attached:

sand mix, coalslag mix, marbledust mix, glowing perlcrete

[Wiley]

Re: Flame Tests

Tscarborough,
Here's a link you may find interesting:
Marble Stone,Calcite Marble,Marble Suppliers,Marble Stone Information,Marble Stone Properties

Scroll down until you reach Chemical Properties of Marble. Depending upon your sample it would appear that the lime and magnesium oxide (both refractory) could be quite high in your sample.

A quick read of the info on that link would make it appear that alot depends upon the sample of marble and its composition.

I would think crushed up refractory bricks (brick dust) and calcium aluminate cement might yield satisfactory results. Or perhaps substituting crushed refractory bricks for the sand in your test sample.

Hope this helps,
Wiley

[Tscarborough]

Re: Flame Tests

The homebrew sample tested fine, even with 2 minutes of direct flame. The problem for my application is that it does not have enough body (i.e. ability to be built up). The sand is too fine, which is what it needs to be for firebrick use.

Wiley thanks for the link, but I have data for all these specific products I am using

Anything else anyone wants me to burn up, er, I mean test?

[Wiley]

Re: Flame Tests

Tscarborough,
Rado has a recipe for home made firebrick/castable, see link. For the aggregate in his mixes he uses crushed firebrick. Perhaps crushed firebrick aggregate on the size of a pea or slightly smaller would be make for a mix with enough body to be built up and still meet your requirements for flame, not spalling etc..

Here's the link;
traditionaloven.com: Wood fired brick ovens for baking bread making pizza pictures.

Hope this helps,
Wiley

[kebwi]

Re: Flame Tests

Well, this may be slightly off topic, but I would love to see a spectrum of vermicrete insulation responses: 4:1, 6:1, 8:1, 10:1, 12:1 (maybe not all of them!) at a few thicknesses: 2", 4", 6" perhaps. Expose them to a prescribed heat source on one side (same temperature and duration across all experiments), then simply report the temperature on the backside.

Of course, what I just described is a 15-item experiment. I doubt you would want to make 15 different vermicrete bricks...perhaps a subset would suffice.

That would be cool. I don't expect you to do it; it's a lot of work to satisfy someone else's curiosity...but it would be super nifty.

I might try to do something like this myself if I can ever find a pure source of heat.

[Tscarborough]

Re: Flame Tests

If I thought it mattered I would, but the fact is that a 2" slab of 8-1 perl-crete was heated to glowing hot on one side and was only warm to the hand on the other, so that is good enough.

[kebwi]

Re: Flame Tests

BTW, did you dry mix your perlcrete (perlite and portand first, then water) or slurry mix it (portland and water, then perlite)?...or something else?

[Tscarborough]

Re: Flame Tests

I tried both ways. Using the mixer it didn't matter a whole lot.

[david s]

Re: Flame Tests

Kebwi,
I love the idea of your experiment and would be keen to see the results. Each of the slabs must be perfectly dry before doing the experiment because any moisture present will act as a conductor. I suspect that the material density would correlate directly with its conductivity.
Dave

[MK1]

Re: Flame Tests

I think I read a post about density vs. conductivity in cal-sil boards and I believe an authority stated it was opposite.

Mark

[Tscarborough]

Re: Flame Tests

Correct, MK1, as a rule it is inversely proportional.

[david s]

Re: Flame Tests

If you add vermiculite to the mixture it will make it less dense and a better insulator or poorer conductor.