Tuesday, April 27, 2010

The Unlikely Gradational Stratigraphy

I described the problem briefly in my last post about drilling at the Northumberland caldera: the volcanic formations and intracaldera units of waterlaid volcaniclastic and volcanic deposits — from the post-caldera Hoodoo Tuff, down into the intracaldera sediments, then into the caldera-forming Northumberland Tuff below — had sharp to relatively sharp contacts in the field. The chip boards from the previous year's drilling, however, showed gradational contacts from top to bottom — with Hoodoo Tuff chips present from the top all the way down, with chips of the uppermost unit 6 of the intracaldera sediments present from the beginning of unit 6 all the way down into unit 2 and below that into the Northumberland Tuff, and so on. At first examination of the chip boards, I simply didn't know what to make of the apparent change from sharp to gradational contacts, and thought that possibly it was the result of sloppy chip-board construction. Upon close examination, however, I finally realized that there were rock types from the top of the hole present all the way down to the bottom of the hole.

This kind of contamination is usually bad for sampling and assaying, but because we were looking for uranium, we were relying almost completely on a calibrated gamma-ray probe and not planning to do much assaying. Still, I wondered why the drilling was resulting in such strong contamination. The conventional answer, so to speak, was that we were using conventional rotary drilling, not reverse-circulation rotary drilling. Conventional rotary can easily result in contamination because the rock chips from the bottom of the drill hole come up the outside of the drill stem, and can pluck chips and material from rocks and formations above the bottom of the hole, all the way to the top. Reverse-circulation rotary drilling prevents this by having the chips go up the inner tube inside the drill rods, where the chips can have no contact with other rocks the drill has already drilled through.

I arrived on the Northumberland caldera drilling project, somewhere southeast of Austin, Nevada, and generally due east of Kingston Canyon — where I happened to be staying in a company-rented tiny cabin — in the fall of 1979. At first, we had a couple core rigs and one rotary rig running. I asked the rotary drillers how they usually did the sampling, and they told me that they usually laid out piles of chips and ground up rock for every 5-foot interval, in a nice line on the ground. Having been told to "ask the drillers," I thought, okay — odd, perhaps, but okay. After they laid out the chips in these nice, neat little piles, I was supposed to put them into the sample bags that I'd labeled. ("Hole 1, 0-5" or something like that; maybe we had a scheme, it's not one I remember.)

Now it's kind of awkward, even under the best of circumstances, to grab handfuls of piled, ground-up rock and somehow get them into 7 x 12½ white sample bags — and these weren't the best of circumstances. When I came back early the next morning to start bagging up samples from the first drill hole, I found the mounds of chips frozen into solid, cone-shaped piles, with the piles frozen to the ground. At first I kind of freaked, like OMG I should have done this yesterday, I am so screwed, etcetera. Yeah, the sun melted them later on — way after noon — but this sample-layout method really put a kink in things (at least for me). It was fall, and temperatures were prone to drop below freezing at night in the high valleys of central Nevada, suggesting that chip freezing would be an everyday occurrence.

I talked to the drillers and asked them if they could just put the samples into the bag instead of piling them up so neatly on the ground. They hemmed and hawed, indicating that they didn't *do* that. I told them we'd have to figure something out, because their on-the-ground method wasn't going to work unless my company was willing to wait around for next summer to get the samples. They finally agreed. I don't remember that it was hard to convince them, and I don't know whether they'd *done* that before, or were just trying to slip something by me. I have seen piles of chips lying out in nice lines as recently as 2005.

Okay, that problem is solved: the drillers are putting the samples in the bag. I'm numbering the bags — which is still a common but not always thing for a geo or tech to take care of — but they are sampling! Yay!

Now, let's watch what they are actually doing. On a conventional rotary rig, one that happened to not to have any kind of sampler or cyclone or splitter, the chips come up the outside of the drill pipe and just pile up on the ground around the hole. The drillers drill 20 feet of steel down at a time: that's the length of each piece of pipe. Every five feet, the helper (only driller and helper on the rig in 1979, not a third person for sampling) sticks a shovel into that pile and puts the stuff into one pre-labeled sample bag.

The pile around the drill hole just stays there, always being contributed to by more stuff coming up the outside of the pipe. Every now and then, when the pile is deemed too large by the driller, the helper pushes some of it aside with the shovel. I watch this process for a short while (having not seen it the day before), and eventually, between rods, tell them about the problem: every single rock type that has been drilled is still lying there in that pile every time the helper sticks a shovel into it to get the next 5-foot sample. Oh sure, the bulk of what goes into his shovel is the most recent stuff that came out of the hole — the stuff on the top of the pile — but not entirely. No wonder the formations glued to the chip boards from the previous year looked gradational.

I point out to them that they are taking a sample of everything that's come out of the hole for the last couple hundred feet, not just the last five. No, they insist that it's just most recent stuff, they are careful and just getting what is near the top. I tell them that it isn't really working that way, though, even if they think it is, and that I'd like them to clear away — with the shovel — that pile every five feet.

Well, no, they can't do that, they are on a footage contract and that would slow them down, having to stop every five feet to clear the pile away. They would be glad to clear the pile away every rod — every 20 feet. I tell them I really can't tell what the rocks are with all that stuff mixed together (although I sorta can, and the down-hole logs are diagnostic for many of the intracaldera contacts), and I tell them that the assaying will totally suck (although maybe we won't be assaying much of anything — and we didn't say "totally" like that back then, and maybe I used another cuss word, and maybe I hadn't fully developed my talking-to-drillers attitude or technique, which includes just the right cuss words at just the right time).

They finally agree that the helper can take a sample with the shovel, put the sample in the bag, then clear away all the rock debris that's piled up during five feet of drilling. It's not perfect, because the drill keeps drilling — or maybe it stops like it should, but I lost that bit of detail somewhere between then and now — but it's so much better, and the not-supposed-to-be-gradational contacts return to the clean sharpness that we'd mapped in the field.

And that's why you should never ask the drillers how to do the sampling.

And that's why it's a good idea to answer a young geologist's nagging questions, even if they seem really dumb to you.

And that's why you should have an hourly contract rather than a footage contract with your drillers: then you can slow them down anytime you want, and they can't bitch about it. (They might still bitch some, because a good driller wants to make footage, but a good driller should also recognize the importance of the sample he's getting.)

5 comments:

Anonymous said...

I hear you. Sampling is one of my two biggest headaches (the other is wireline logging: don't get me started). I've never had the luxury of uncontaminated RC cuttings. Your "gradational samples" issue is standard for oil & gas (O&G) drilling. On an O&G rig, cuttings and drilling mud come up the annulus (space between the borehole wall and the drillpipe--as you described) to a "bell nipple" beneath the drill floor, then flow by gravity down a "flow line" to one or more "shale shakers", which are giant, vibrating sieves. The cuttings go across the shaker and fall into a shale pit; the mud goes through the screens into a settling tank and is ultimately recirculated.

The aggravation is that nobody has ever invented a consistent, reliable way to catch samples. The companies that build shale shakers have never clued in to the fact that someone might want to take samples of the cuttings, so every time I arrive on a rig, I have to reinvent the wheel. Usually this consists of hanging or otherwise propping a 5-gallon pail off the end of the shaker, or placing a plank across the end of the shaker for the cuttings to collect in or on. Then you have to train the roughneck--a low-ranking rig worker who is assigned to catch samples: this is part of the problem IMHO, as the job is given to the lowest man on the totem pole, hence is considered a form of punishment for newbies.

Issues:

1) Making sure the samples are caught on time.
2) Making sure the samples are caught!
3) Making sure the pail (or board) are cleaned off after each sample is taken.
4) Making the roughie understand that you need a sample that represents *the entire 5-metre interval*, not just a tiny fraction of each interval. This means placing the pail so that it doesn't fill up in the first 2 minutes, and the rest of the sample interval falls off the top into the pit; it also means *not simply grabbing a sample off the shaker as it goes by!*

I've always tried to work on the principle that you catch more flies with honey than with vinegar, so I try to make the roughneck's job as easy as possible, and in most cases this works fairly well.

Cheers,
--Howard (Calgary, Canada)

Silver Fox said...

Howard, between this blog and Accidental Remediation, you have quite a bit of info about oil & gas drilling and sampling and logging - would you like to do a guest post, somehow? If so, contact me via email.

On the RC rig, the sampler is also the lowest person, and when that person is very new, it can be difficult to convince him that sampling is what we are paying to have a 3rd person for! Some drilling outfits were quite proud of their sampling procedures in the 80's when RC and the sampler were new. Now it's commonplace for the drilling companies to have the sampling built in to the contracts, but a really new person can screw things up and can be well worth watching. That was the problem we had in 2004, when exploration was just coming back from the gold crash of 1996-97, and all competent drillers had left the industry for parts unknown. Samplers, and even the 2nd-person helper rodman, were inexperienced, and many times didn't like the job, didn't care, and some I dealt with left before a week was up!

That's almost like another story.

Yes, it's good to make the sampler's job easier if possible (or to complement on technique, etc.); on RC rigs, possibly that person moves up more rapidly to driller than on O&G, so maybe they often have a better outlook.

Silver Fox said...

In fact, many drillers were inexperienced after the gold crash of 1996-97, as recently as 2004.

Concrete Core Drill said...

Sampling involves a lot of risks in it.. I agree with that point..Your "gradational samples" issue is standard for oil & gas drilling.

Silver Fox said...

Dear CCD, a whole discussion ensued about sampling problems (see above). These issues are unrelated to oil & gas drilling, as that is not what we were doing. And we were *way* less than even 2000 feet in the hole, with issues generated from zero feet on.

Meanwhile, the core drillers couldn't drill through the night without getting stuck, but that's another story.