quote:

Congratulations on a very tidy job - I wish I had the knack. My kitchen table is covered in the world's supply of reducers, T pieces and various offcuts of pipe as I try to figure a way to make a venturi without recourse to "copper bashing"

Cheers

Nick

Since Graham originally posted this excellent design, I started reading up on venturi valves and pumps. Many of the commercially available venturi pumps have the main line pump's fluid intake restriction coming out in the middle of the pipe with the incoming vaccum line coming in and surrounding the incoming jet. What about putting a funnel like restriction into the ends of the tee which would allow the incoming vaccum line to surround the jet. That might allow for better mixing and may create a higher vaccum.

Stainless steel cake decorating tips might be a good item to use to easily create this valve for a number of reasons.

• They are stainless steel.
• Being thin walled metal, they could be easily modified by grinding or filing.
• They could be shortened up to fit properly in the valve.
• They have different sized openings which, depending on the size of your pump, may be benificial
• They even have different shaped ends which may or may not enhance mixing. The teeth might allow the fluid to mix without cavitating.

I thought that this design, see picture below, would be even more effective in its ability to pull a vacuum. Plus it pulls the liquid or air all the way around the opening so it mixes from all sides of the oil stream. The teeth of the cake tip might further break the methoxide into streamlets that might mix even better and fast with the oil.

Also I think this design would be easier and take less time to build than hammering out a piece of copper and may even be faster to make than Graham's first version. I do like the shape of Grahams latest creation. It has a great look to it and its shape is built more like an actual venturi.

I haven't built this yet, so I don't know how easy it would be to pre-tin and solder the stainless tip to the brass tee, but that may not even be necessary. A flaired type tee fitting might allow you to just compress the tip into place with no soldering needed at all. I would probably solder it anyways, just so that there were no pockets for debrea or water to hide in and it would create a smoother flow of air and liquid. I would solder B first, placing rings of solder around the tip. Then drill out the opening, if necessary, so that tip A will fit into the opening with proper clearances (what-ever they might be). Then with rings of solder around tip A, solder that tip into place as well. I would probably try the teeth on tip A to see if that would enable the material to mix better / quicker.

If the cake tips don't work for you or if you can't find them, there are also many different copper plumbing fittings which might be used as well. The copper flair cap I believe comes in larger sizes so that it may work better for those running larger lines. See pictures below. The size of the hole in the end of the flaired cap could be drilled to your specs depending on your line and pump sizes. Also, copper is very easy to solder into place. The fitting could easily be pre-tinned with solder which would make the internal attachment fast and easy. Again, a flaired tee fitting may make soldering unnecessary.

I did a quick web search and found several copper fittings that might work or which if used in combination with each other, might make constructing the valve easier. The copper gasket when combined with the flaired cap for example.

See drawings / pictures.

Home Made Venturi Elements

I am guessing that you probably would not even need to have constriction B for the venturi to work, but having B in place would most likely get a better mix and a stonger vacuum. This is because it would enable the mixture to Flash and mix slower into more oil because of the slower rise in its pressure back to the pipes normal operating diameter; but I'm guessing it would likely work fine without it.

Both types of fittings could be modified easily, the cake tip, because it is so thin, and the copper, because it is so soft, could each be filed, or ground down so that they could be fitted easily into the tee.

I have used cake decorating tips a few times for for several things in the construction industry, my line of work, which had nothing to do with decorating cakes. Having kids, I have decorated my share of cakes with them as well.


These are just a few of the many different shapes of cake decorating tips that are available.

Stainless Steel tips

My guess is that the third one from the left, the simple funnel would work well, but I would probably try the one with teeth as well. Tips are cheap, 50 - 80 cents a piece, and a different shape may help speed the mixing process. The wrong shape might just create a lot of cavitation. But at 80 cents a piece, it couldn't hurt to play, I mean experiment.

One other thought, if the size of the valve restricts the flow too much, what about putting a second venturi in line, which would help those of you that are running larger pumps and larger intake lines.

I'm sorry I can't help with the actual building / testing, I don't even have a processor yet. I have been researching and learning for the past several months, so I can't help with the actual building / testing of the design yet. Heck, I don't even have a diesel vehical yet, but I am looking into that as well, it is only a matter of time.

quote:

Originally posted by GrahamLaming:
Hi Cover,

That's a very interesting suggestion - thanks.

I've not tried processing above 60C, so don't get much condensate during the period, but it would be interesting to see how the process time is affected by running say at 65C and returning condensate.

I have a sneaking feeling may not be feasible to go much hotter than that, because the vapourisation of the methanol will tend to keep the temperature down.

Here's an analogy...

Imagine trying to boil a kettle of water, and you want to get the temperature above boiling point - it's just not going to happen - unless you pressurise the system.

I would be very reluctant to pressurise my system, as it is not rated for running at more than a few psi.

There may also be an efficieny tradeoff, as you'd need to keep condensing vapour, and your condenser may lose heat energy to the environment.


The venturi should be fairly free flowing if well made, because although there is a restriction, is is not impeding flow by much at all whilst not pumping, and even when it is pumping, we are moving fairly low flow-rates.

I have hydrostatic indicators (manometers) on the test system, so I can see backpressure and so on.

Sounds wrong, you'd expect a narrowing from 22mm to 8mm to have an eneormous effect, but here's why it is less than you'd expect ...

We are not really blocking the pipe, we are gradually encouraging the flow to speed up in one section, and gradually then slowing it down.

'Gradually' is important - the tapers control the gradual acceleration and deceleration.

So the flow rate is not affected nearly as much as if we had a classical sudden restriction in the pipe, like from a valve.

A really good venturi mat reduce flow rate by only 10% or so.

The back pressure does increase when the venturi is sucking a flow of liquid or gas into it, but this is only during dosing and dewatering and distilling, and we can turn the suction flow rate down to quite low levels during these processes. The back pressure reduces as you reduce the speed of drawn-in material.

Best regards,

Hi Graham, all
Here is a great explanation of the different types of venturis, sonic nozzles, etc and the rationale for each that might aid in understanding the issues. http://www.flowmaxx.com/cavitate.htm
Best, Jackmac

Hi All,
Just thought you might like to see the three way breeches piece mentioned in my post of 24/03. There are no doubt some who might not know what I was waffling on about.
The object of the execise is to reduce back pressure from my triple static mixer (which should cope with 1 1/2 times the flow rate of the pump) and reduce efficiencey or even stop my venturi from working.
Any sudden change in direction results in additional back pressure so I have removed all elbows both 90deg. & 45deg. from the out put side of the pump and replaced them with long radius pulled bends using a pipe bender. This has also been done on the pump supply line.
Have a look and let me know what you think.

Jim

Breeches_Piece.JPG (267 KB, 338 downloads)

Jim. That is sculptural aesthetics of a high standard. The fact it is also functionaly efficent adds to the elegance.

Let us know how it all works out.

quote:

Originally posted by Ant:
Jim. That is sculptural aesthetics of a high standard.

It is magnificent - truly a work of art! Hope it works as good as it looks...

Nick

Going to Soap removal a bit. At what temp should soap removal be done? Could centrifuge be used to remove soap?


thanks.

Hi ppusopa

I find it best to remove soap at less than 35C if filtering.
I usually do it between 15 to 20C.

Too cold or you'll block the filter with high melting point esters.

Too hot and your soap will be soft and may extrude thru the filter.

I believe some folks have used centrifuges, though I have no experience of this myself.

Take a look here...
http://biodiesel.infopop.cc/eve/forums/a/tpc/f/719605551/m/3421061071/p/1

Hope that helps.

Graham

Any update to setting up your Eco-System processor using an apple turnover? I really want to use your system but also want to use the apple Turnover as I do not have a cone botton tank to use as a processor.

Thanks

Ron

Graham,

If in summer the temp stay well above 35c overnight and day, the soap may not settle well.
I am seeing that effect now. I have a few waterless batches that after heating over 200F, I let it settle over night. Not much soap at the bottle of tank.

If that is the case that is a quite a problem for me.

Thanks.

Hi vorvon46

Not forgotten - I've made a start on the dwg. for you. Wish I could finish it for you sooner.

---------------

Hi ppusopa

It will settle eventually on its own, you may need a few settling tanks, use the oldest one first. Depends on your rate of usage and available storage space.

If you use a lot and have not much space, it may be best for you to filter in re-circulation. It won't take too long.

Start with 100 micron, then 10 or 5 micron.

Hope that helps,

Hi Graham

What volume (water heater)in litres would be the largest and most relatively cost effective to set up using this process that would work?

Thanks Peter

Hi Peter

I've only used a 115 litre tank myself.

The process should be scaleable, so if you increase the pumping rate, venturi size, heater power and the condenser size, in proportion to tank volume, I would expect you could use a larger tank.

Hello everybody,

I finally finished my fully flowed mixer, hopefully it will be undergoing full testing at the weekend. I will try to do a Graham type movie of it working with the venturi at full throttle (If I can).

Free_flow_mixer.JPG (118 KB, 246 downloads)

Graham,

Regarding the McMaster's part I suggested before - after reading your review, I gathered up the technical papers prior to installing one myself. I learned the part is designed to operate using steam in the primary, to inject liquid. Here's a cutaway:



I'm wondering if shortening the internal nozzle (labeled E) would improve performance with liquids.

I'm going to attempt to make a venturi from a standard NPT "T" using your solder-fill technique. What type of solder do you use, and is it chemically inert?

About centrifuges, take a look at www.simplecentrifuge.com ..Some innovation is going on there...

quote:

Originally posted by GrahamLaming:
Hi ppusopa

I find it best to remove soap at less than 35C if filtering.
I usually do it between 15 to 20C.

Too cold or you'll block the filter with high melting point esters.

Too hot and your soap will be soft and may extrude thru the filter.

I believe some folks have used centrifuges, though I have no experience of this myself.

Take a look here...
http://biodiesel.infopop.cc/eve/forums/a/tpc/f/719605551/m/3421061071/p/1

Hope that helps.

quote:

Originally posted by jamesrl:
Hello everybody,

I finally finished my fully flowed mixer, hopefully it will be undergoing full testing at the weekend. I will try to do a Graham type movie of it working with the venturi at full throttle (If I can).

I have a space in my art museum for your creation jamesrl

Hi ebtz

Thanks for the dwg. I think shortening the nozzle may have the opposite effect, because you want as much velocity as possible out of it, to give momentum to the inlet material.

The suction increases with the square of the velocity, so faster is better. Widening the nozzle will slow the flow.

Provided your pump can shift fluid rapidly, this eductor should work with liquids too.

-------------

Hi jamesrl

You should show your work to some of the formula 1 racing teams - you could turn out some neat exhaust manifolds!

Hi All,

I've tested my proposed setup with both venturi and mixer in line, the results are as follows.
All test were done with 50 measured litres of water and the time it took to deliver that amount into a seperate drum with a pump head of 4ft (9" higher than on my reactor).
My pump is a Stewart Turner with a flow rate of 20 galls. per min. I fitted a valve to the suction side of the venturi to control air and fluids.


1. Open 25mm pipe 33.33"
2. with venturi closed valve 33.47"
3. with venturi open valve 35.78"
4. Mixer with pipe 33.71"
5. venturi & mixer closed valve 33.78"
6. venturi & mixer open valve 34.46"
7. venturi drawing water, mixer 34.43"

I then measured the rate of input through the venturi, it delivered 1lt in 38" and would draw up 5ft vertical.

As my plant only needs a 32 inch lift to the venturi I will be able to draw methoxide no problem. The head on my plant is only 39 inches so everything should work better than on the test rig.

The object of the exersise was to design a mixer that would not effect the performance of the venturi.
It passed with flying colours.

I've attached a photo of the rig.

P.S Graham I used to make parts for Lotus when Graham Hill drove for them.( been there done that).

Jim.

test_rig.JPG (485 KB, 267 downloads)

Jameserl, can you give a newbie some hints on how to weld some of the joints you make? I have successfully done a but joint of 3/4x1/2reducer to a 1/2"pipe back to a 1/2x3/4 reducer for the start of the venturi. That was pretty easy. How do you hold the tiny pipe that becomes the suction pipe to the larger piece and ten sodder it in place? I feel like I need an extra set of hands. Is that the trick?

quote:

Originally posted by jamesrl:
I've tested my proposed setup with both venturi and mixer in line, the results are as follows.

1. Open 25mm pipe 33.33"
2. with venturi closed valve 33.47"
3. with venturi open valve 35.78"
4. Mixer with pipe 33.71"
5. venturi & mixer closed valve 33.78"
6. venturi & mixer open valve 34.46"
7. venturi drawing water, mixer 34.43"

I then measured the rate of input through the venturi, it delivered 1lt in 38" and would draw up 5ft vertical.
Jim.

I looked at your picture. It looks like you have a loop of flexible piping that starts at ground level in a bucket, rises to some height - that you can vary - and returns to near ground level where it enters your pump plumbing. If that is correct, and if this is the set up you used for testing, then your lift numbers are probably OK, but your flow rate numbers may be off.

The "lift" numbers were probably determined by starting with an empty loop and measuring how high the liquid would rise in the loop with the pump pulling a vacuum.

Liquid flowing down the "down-leg" of your loop will create a vacuum or siphon effect due to gravity. Therefore, the actual head of the liquid with a full pipe is a lot smaller than the negative pressure found inside the pipe when the liquid is being drawn up the "up-leg" side of the loop.