biodieselpictures.com

[JohnO]

Thanks to inspiration and a long ongoing dialog with Rick at B100 supply, this eductor evolved to the point of finally testing. Hoky smokes, does it work!!! That's a "blue" generic clear water pump, pumping cold svo (about 50 degrees), at an unknown rate and pressure, other than it's typical of the kind used for biodiesel processors up to about 50 gallon capacity.

The eductor is a 1/2 inch pipe tee, fed by pressurized oil from the pump, through a length of 3/8 copper tubing. The compression fitting that connects the tubing to the tee has been drilled through with a 3/8 inch bit, so the tubing can pass on through. It's the same arrangement used for making a HIH fuel system for SVO conversions.

I've also tested the same setup using a 3/4 inch tee, with slightly worse results! It still worked just fine, but developed slightly less suction. I'll play with it some more. It might work better with a larger nozzle.

Crimping the tubing onto a 1/8 inch drill bit shank in a 3-jaw chuck makes a reasonably effective nozzle. I haven't tried other diameters, other than straight tubing, which didn't generate any suction at all, although it was able to get to a "neutral" point where suction and pressure just balanced each other. I suspect that a little larger orifice than the 1/8 inch work work quite well, but I reserve the right to be wrong.

When sucking on a length of 1/4 inch tubing, it draws cold oil from the flask at about 200 cc/minute. That might be a good starting point for a methoxide eduction system. When sucking on 1/2 inch tubing, it moves about 500 cc in 15 seconds!!! That's 2 liters/minute. When deadheaded against a gage it indicates up to 4 inches Hg vacuum. The "best" position for the nozzle is even with the middle of the tee or "below" (closer to the compression fitting). Suction ability declines noticably if it's "above" the middle, although it doesn't go away unless the nozzle gets right up to the outlet fitting. Moving it down doesn't actually decrease the suction much.

An alternative nozzle would be to fill the end of the tubing with solder and drill it.

The outlet is a 3/4 inch hose barb x 1/2 inch pipe. It might work differently (worse) when directly connected to an Appleseed processor.

I'm amazed how much oil this thing pumps, even with the dinky restrictive nozzle. There is likely to be a best compromise between suction and flow rate.

Cheers,
JohnO

[JohnO]

Pumping rates:
1/8 inch nozzle - 0.17 liter/second
3/16 inch nozzle* - 0.24 liter/second

Suction using 3/16 nozzle was only about 1-1/2 inches Hg. taking 100 seconds to suck up 50 cc oil through the 1/4 inch tube. A larger tube would certainly flow faster, and this is still plenty of suction to draw methoxide from a container, but is also seems like a reasonable lower limit.

addendum: I tested the 3/16 nozzle with the 1/2 inch suction tube, and it's still awesome - drawing 11 cc/second vs 0.5 cc/sec with the 1/4 inch suction tube.

*A note about the 3/16 nozzle: when I tried to crimp the copper tubing onto the shank of the 3/16 drill bit, the 3 "wings" don't close. This means the actual flow area is larger than a simple 3/16 inch diameter hole would have.

3/4 inch tee tests: Suction improved with the 3/16 nozzle, to a bit more than 2 inches Hg. "Best" position for the tip of the nozzle was pretty close to the outlet fitting, about 1/4 inch from the edge of the hose barb entrance. The bent piece of wire indicates the relative position of the nozzle tip.

Flow rate of the oil (at 14 degC) is a steady 236 cc/sec or 1.7 gallons/minute. I've included a picture of what that looks like.

JohnO

[JohnO]

I switched the 3/8 copper tubing to 1/2 inch copper tubing, to get the least resistance up to the orifice. All tests used the 3/4 inch tee.

The 1/2 inch copper tubing is easy to bend into a U-shape around a 3-inch diameter pipe. This opens up interesting possibilities for different plumbing arrangements, including using a bypass for full flow rate. The following figures show that won't be necessary:

Orifice tests:
No orifice, 0.34 l/s, no suction although a neutral point was found.
5/16* orifice, 0.35 l/s (yes, it flowed slightly more), up to 1-1/2"Hg.
1/4 inch orifice, 0.36 l/s, up to 3"Hg.
3/16 inch orifice, 0.34 l/s, up to 4"Hg.

I stopped there since the suction is more than enough, and further restriction is likely to reduce flow rate for no suction advantage.

The "best" position for the nozzle was vaguely anywhere between the middle of the Tee to about 1/4 inch from the outlet. However, there was plenty of suction at ANY position within the Tee, and even some when the nozzle was pushed up into the beginning of the outlet fitting.

More suction can be had at the expense of lower flow rates, by crimping the orifice smaller.
Note that this will be a different nozzle shape than obtained with the 3/8 inch copper tubing crimped down to 3/16, because the 1/2 inch tubing will have bigger "wings".
The "best" position for the 3/16 inch orifice was rather near the beginning of the outlet fitting. It was also sensitive to misalignment. I'm assuming the "best" location was centered on the outlet fitting diameter, and the suction dropped off when it was moved to the side a little. When the nozzle was positioned closer to the middle of the Tee, the alignment didn't seem to have much effect on suction.

In conclusion, an effective aspirator/eductor/venturi can be obtained this way, without any machining. However, more than 4 inches Hg suction is beyond the capability of this assembly.
Cheers,
JohnO