I have a non probe moisture meter. I'll give this a try. its calibrated for wood not oil..but hey it cannot hurt to try. I never thought about using one for this ??
I think calibration would be the issue but not much of one.


-dkenny

Try making the following sampler so you can see layers in the barrel, if you're worried about liters of free water in the bottom of the barrel:

-piece of clear hard PVC pipe (sold at USplastic.com) in a 1" diameter, the length of the barrel contents. Dont' use tube, you want it to be stiff but see-through.

-Put a valve on top end.



Slowly insert the sampler pipe into the barrel of oil, with the valve open, so that it takes a sort of 'core sample'. Close the valve so that core sample can't drain out quickly. Lift the core sample out, but not all the way out of the liquid. This will keep the core sample from draining out.

I think this method will work because we used to do something like this with tubing and a thumb used as a valve in the Berkeley Biodiesel Collective before we had good wash tanks. I think we were looking at layers this way, if I recall right. Maybe it'd be different with WVO/water than it was with biodiesel/water.

mark if he were to rig a mechanism to put the valve at the bottom he'd get a perfect 'core' sample..

-dkenny

I know, but this is easier.

Mark

Of course, it doesn't help if the oil is full of emulsified water and it separates into garbage and oil later on back at the plant, but it'll tell you if the restaurant has started pouring water in the barrel like the horror stories have it.

do you think a simple check value would work for a core sample?

-dkenny

It would have to be a very light-sprung one, but that seems like a good idea.

Flapper style check valve

Trc

Calorimetry.

This would be a different use of the technology, but I think it would be applicable since the specific heat of water is ~4.18 [J/g C] whereas the specific heat of rapeseed oil is ~1.97 [J/g C] (taken from Bosch Automotive Handbook 5th edition).

Comparing changes in DSC results is like comparing changes in specific gravity results.

Horn

Wow, just got back to work after the weekend, and I must say I am flabbergasted by all the responses!

THANK YOU!!!

I'll talk it over with our oil collection manager and see what we want to test out first. I like the "core sample" idea- if we're able to get a really accurate sample, including solids (the tricky part), then the real work could be done later back at the plant, and all the driver has to do is collect the core samples.

For the bottom of the core sampler, I'm picturing something like the trip mechanism for a lake water sampler (Kemmerer bottle is one). You send the sampler down open to the depth you want, then trip it to take a sample at that depth. Search for "Kemmerer sampler" for pictures. Maybe the core sampler can be wide enough around to capture some solids?

Okay,

So none of the pre-existing core samplers seem ideally suited to getting a sample of oil, water, and food that goes all the way down to the bottom of the barrel or dumpster.

(incidentally, we use mechanical suction tube liquid samplers all the time- they're really easy to make, and they work great if you're not trying pull up anything heavy, but they wouldn't be much good for this)

I put on my inventing cap, and came up with this nifty (or perhaps stupid- we shall see) idea, which I will happily put out here in the open source spirit, and perhaps someone can help me get one of these things built. (fabricator? murphy?)



complete "core sample" gets sucked into appropriately-sized mason jar, which is detached, sealed, labeled, and packed away to be analyzed later. Core sample procedure shouldn't take longer than a minute or so.

Popgun sampler.

Parts needed:
1) clear rigid tubing long enough to reach the bottom
2) cork, size to fit tube 1) above.
3) string.

Assembly: Punch a hole for the string through the cork and run the string through the tube. A leather stitch-awl works well.

Use: To take a sample, let the cork dangle on the string as you push the tube vertically into the oil, "cork" end downward. Pull the string to seal the sample inside the tube.

Kumar

I use to work in the oilfield. When I first started one of my jobs was to get the water levels on the large tanks (about 300 barrel tanks) When oil and water came out of the ground it would go to these tanks. After some time the condensate (oil) would separate from the ground water. We would use a specially designed tape measure with a plumb bob on the end that we would drop down the hatch and into the tank. We would apply "Kolor Kut", a paste, approximately where we thought the water level would be. There would be a definite change from where the water level changed to oil. The water changed the paste red. So given the dimensions of the container that the the WVO is in, and you know that you have water for the first 6 inches, you can determine how many gallons of water there is in the container.

Now for the device that takes a core sample. We called them thief's. I also had to use these if I needed to get a more accurate measurement. It is a tube with a rope and a spring loaded disc and when you tap the bottom of the container it would trip and close giving you a core sample. I will try and get some more information for you and post later. I am sorry if this info has already been covered I skimmed through quickly and did not see it discussed.

Cue

Ok here is the Kolor Kut

Here is the gauge line that we would apply the kolor kut to.

Here is the thief

Personaly I wouldnt transfer any core samples. This is the way I think it would work if you are using the thief.

1. Pull sample out of container.

2. Then record your levels. For instance you have 1" of solids and water (because you will have water within the solids). Then at 6" is your water level. After this you calculate based on your container size the gallons of water and whatever unit of measurement you would use for the solids.


Cue

Cue,

Those are interesting tools. The 'oil thief' would be great but it only comes in up to 24" deep.

Kumar, the core sampling idea is what I had mentioned above in this thread. Here's a way to easily 'close' the line when at the bottom. So you have a rigid tube the depth of your containers. At the bottom, put a connector and a 90 degree right angle adapter. Attach a brass ball valve to the right angle connector. So here's the trick, attach a rod to the end of the ball valve handle that is the length of the tube.

The idea is that you open the valve, & slowly lower the tube with the valve first to the bottom of the container. You then pull up on the rod, and that will close the bottom (orientate the valve so pulling up closes it). You then pull up the assembly & full without any fallout. You can size it to 'catch' solids. The reason for the right angle is that it's easier to close the valve, you could eliminate it but your leverage point will be more difficult to close, but it could work). Only disadvantage is that you won't see the bottom inch or two because of the fittings. Also, with a right angle you'll get a 'feel' for where the solids start in the container (at least for most types of solids I get). So that would tell you how much water is above the solids, but not if there is water at/below the solids level (but would you really care as I'd not credit the customer for the oil at/below the solids level)?).

Thoughts? This is a great discussion dealing with something that many of us are having troubles with.

-tony

TexasTony

I would be concerned about no being able to get the last two inches of sample. The oil thief may as well not get the last two inches either. In that last two inches could be the solids layer and the water level. Depending on the size of the container an inch of water could add up to a lot. Also, if you have the valve body itself in the solids layer he may have a hard time closing the valve because of the crumbs getting into the valve. If the thief does not get the last two inches, then maybe the kolor kut and gauge line would work. You would definately get a reading form the last two inches.

My .02. I love this. Being able to bounce ideas off of each other to find a solution.

Cue

Thanks cue, et al.

The thief sampler seems pretty good. As with everything though, the problems arise with the valve.

The simplest thief sampler would be a long, stiff "drinking straw". You'd tap the bottom of the container, and put your thumb over the top, lift, and observe. McMaster Carr sells one of these that's 42" long, high density polyethylene, with a very small inner diameter (literally like a straw). I'm going to try it out.

Might have to color the bottom few inches, to avoid lifting it all the way out.

A Simpler idea:
1) Clear 3/4 inch PVC tube, long enough to reach the tank bottom (McMaster-Carr 49035K24, $9.83)
2) Pipe cap, with large hole drilled through the bottom, glued to the bottom end. The hole may need to have a chamfer on the inside, to form a tight seal with item 3), below. (9161K154, $5.68)
3) Marble or ball bearing, big enough to seal the hole in item 2), above (9528K25, $11.45/50).

Use:
1) Push the pipe all the way to the bottom.
2) Drop the marble in and wait for it to seal the hole in the bottom.
3) Pull it out of the tank.

I love the idea, but it assumes all liquid. What happens if there's a bunch of charred bits of potato at the bottom?