July 10, 2004

4:28 PM

Question for people who know something about hardware, and who might come from more icy climates than I do - what's the best way to saw through frozen soil cores (4 inches in diameter)? I've got several hundred frozen cylinders of soil I have to slice up for my research, and I'm worried that any standard saw blade I use is going to go blunt in 5 minutes. Circular saw or band saw? Blade material?
posted by Jimbob to (10 comments total)

What do researchers working with ice cores typically use?

(what kind of research are you doing? I've got a degree in soil science myself)
posted by mathowie at 4:33 PM on July 10, 2004

i don't know anything about what it is you're doing, but i have decent google-fu. these folks use a gas-powered concrete saw with a carbide-chip blade....and that sounds like a decent investment if this is your thing...it's only about 900 bucks (compared with the price of the blades, 400 a piece!).

anyway, 2000 bucks is a lot of money, but perhaps you can get your department to go in on it with you.
posted by taumeson at 4:40 PM on July 10, 2004

taumeson, I think they use the concrete saw to cut the samples out of the ground - not to cut a long cylindrical core sample into shorter cylinders as Jimbob must.

Matt, I think frozen soil would be much more damaging to a saw blade than ice.

These guys used "an automatic feed horizontal bandsaw with a diamond-segmented blade." Yikes; I know tools and I've heard of diamond-segmented circular-saw blades but never of those.

And, "[l]iquid nitrogen vent gas is used to cool the blade and prevent volatilization of pore fluids." Double yikes. Although I guess if you're doing morphological analysis rather than chemical this isn't necessary?

Anyway, good luck!
posted by nicwolff at 5:03 PM on July 10, 2004

I think you should use lasers!

But barring that, why not use a diamond circular saw like nicwolff mentioned? Works for stone, so it should work for frozen soil. Also, they start out blunt, so no worries there.
posted by undecided at 5:22 PM on July 10, 2004

I think a good tungsten carbide blade would do the trick. It's what they use at the end of rock-drilling bits. I happen to have one sitting on my kitchen counter right now: I went at the one of the "teeth" with my dremel tool and quickly discovered none of the cutters and grinders would touch it.

Heat doesn't weaken t.c., so it's ideal for the job: just keep pumping water through to bring the rock dust back up out of the hole, and let it grind, grind, grind on down. It can get hot as hell without softening.

Here's some general info. My guess is that you can do perfectly well going into a general machine shop (House of Tools in Canada would be one) and using a generic TC tile-cutting type of blade.
posted by five fresh fish at 5:32 PM on July 10, 2004

My buddies at the National Ice Core Lab use band saws. Look at this page at #2 at the top graphic: it says Hz Bandsaw.

Email me (see my profile) and I'll ask what kind of blade they use. I'll get ahold of my friend monday or tuesday, if that's enough time.
posted by jazon at 7:36 PM on July 10, 2004

A diamond blade would probably be ideal. This is what is used to cut tiles, which would be at least as hard (and not disimilar in composition) as frozen soil. You can get diamond blades for bandsaws, which would be the ideal tool.
posted by dg at 10:49 PM on July 10, 2004

You might want to try a carbide ("metal cutting") blade on a plain ol' chop saw-- I'm guessing, since you don't have access to whatever is state-of-the-art in frozen-soil cutting, you're trying to keep from breaking the bank. You can pick up a nice half-horse Ryobi for a couple hundred or less from Home Despot.
posted by notsnot at 11:51 PM on July 10, 2004

Thanks for the replies people, some good ideas here.

Mathowie, I'm a botanist who studies root distribution, and I've developed a quite accurate technique for looking at the quantity of roots at various depths of plants grown in plastic tubes - saturating the soil, freezing it, slicing it, letting it defrost, extracting the roots, staining them and then scanning them into a PC to measure the length. In my trials, I just used a hand-saw, but that was slow, laborious, and for the real study I want something a bit more efficient. Since I'm the first guy to do this in my department, no, we don't have any specialized equipment, but I should be able to buy some if it's the price isn't too outrageous.

Tungsten carbide blade sounds good, I'll see if I can find one for a band saw.
posted by Jimbob at 1:21 AM on July 11, 2004

TC blades for a bandsaw might be a challenging search, but you should be able to get them for a reciprocating saw.
posted by five fresh fish at 9:09 AM on July 11, 2004