Hi Paulus,

In the push-pull processor, the venturi is only there to reduce pressure, it is being used as a vacuum pump to lower the BP, so that we can reduce back reaction.

In the eco-processor, yes, I agree it was mainly to circulate gases at atmospheric pressure. The methanol was removed from the bio by evaporation also, not just boiling.

The push-pull processor gives a higher saturation of methanol vapour, so the condenser should work more efficiently. The vapour concentration in the processor should be at or near saturation, with no introduction or circulation of air.

I'm not sure what effect the water vapour will have. I get around 95% purity methanol.

Again, because we run at reduced pressure, the water removal is feasible at lower temperature, so your pump should be OK if you can get down to 1/2 an atmosphere or less.

Thanks for the clarification, Graham. Have you measured the vacuum you are getting?

Hi Paulus

My vac is pretty 'soft' !! Around 500mmHg or 10psi absolute - this is because my venturi is very rough - Making a new one which should be much better.

But even this level of vacuum is of benefit.

With better vacuum, we can go to lower temperatures and get less back reaction.

I have run recovery at atmospheric pressure up to 120C but got a little back reaction after neutralising. This may really be only an issue if you are water washing, because the MG and DG might cause emulsion.

With my lousy vacuum, I have no problem water washing my samples. With better vacuum, it should be safe to water wash, although I only water wash samples, I never water wash batches.

Hope that helps

Graham

GL I'll repeat my question also to save you wading through.

What is the least acid you have needed so far to neutralise a batch and how many batches have you neutralised so far?

I've got one question and a comment.

First question, in the posts it says the glycerol falls out in a couple seconds, is this due to the acid neutralizing the catalyst, or did i just misread or something?

Also, when using a conventional vacuum pump, why not just put the vacuum pump after the condensor and liquid trap? This would eliminate vapor or liquid getting into your pump and in theory you could route the exhaust of the pump back into the reaction vessel after you've reached whatever vacuum you want. This would effectively just be driving vapor through the condenser, basically just moving methanol from one area to another via a closed system. This would also allow for a much stronger vacuum if you had a nice pump.

quote:

why not just put the vacuum pump after the condensor and liquid trap?

I've been thinking this too and trying to think of a way to say so without being horribly long-winded.

1) There don't seem to be many cheap vacuum pumps that can deal with methanol vapour in a safe and unexplosive way.

2) If you pull a much deeper vacuum the methanol will boil off at lower temperatures. Refer page 13, 3.7 "vapour pressure vs temperature" of the methanex document. If you achieve a vacuum of say 4 psi absolute the methanol will boil off at 35*C. That's feasible, but your condenser water will have to be colder than this to get it to condense again.

From this table, there seems to be a point under 1.5 - 2psia where the condenser will need to be refrigerated to get under the MeOH dew point, and will probably not be worth it.

On page 32, 11.0 Appendix, table 1 sets this out well.

Graham, your comments here welcome please.

Paul

quote:

Originally posted by yaric:
Also, when using a conventional vacuum pump, why not just put the vacuum pump after the condensor and liquid trap?

I've been doing this for about a year now. It does work. I do get methanol contaminating the oil lubricant in my vacuum pump. The reason for this is that there is a vacuum on the liquid trap, so the methanol in there is boiling as well. I've been meaning to try to freeze my vapor trap, or at least cover it with ice to reduce this. I am getting about 2% of the methanol back from the finished BD.

I used to just run well water through the condenser, that worked well but was wasteful. Then I tried a 20 gallon barrel of water and circulated that through. I found that the water became hot and needed ice. Last weekend I filled a 55 gallon barrel with water, and it seemed to stay cold enough to not cause me any worries. I am using a simple sump pump from the hardware store to circulate my cooling water through the plumbers delight condenser.

I am not currently using a venturi, just the vacuum pump. I'd like to find a vacuum pump that I can pull the methanol vapors through and put the catch can on the output..... I like Graham's idea here, just trying to determine the vacuum generated from such a system.

quote:

Originally posted by jon h.:

quote:

Originally posted by yaric:
Also, when using a conventional vacuum pump, why not just put the vacuum pump after the condensor and liquid trap?

I've been doing this for about a year now. It does work. I do get methanol contaminating the oil lubricant in my vacuum pump. The reason for this is that there is a vacuum on the liquid trap, so the methanol in there is boiling as well. I've been meaning to try to freeze my vapor trap, or at least cover it with ice to reduce this. I am getting about 2% of the methanol back from the finished BD.

I used to just run well water through the condenser, that worked well but was wasteful. Then I tried a 20 gallon barrel of water and circulated that through. I found that the water became hot and needed ice. Last weekend I filled a 55 gallon barrel with water, and it seemed to stay cold enough to not cause me any worries. I am using a simple sump pump from the hardware store to circulate my cooling water through the plumbers delight condenser.

I am not currently using a venturi, just the vacuum pump. I'd like to find a vacuum pump that I can pull the methanol vapors through and put the catch can on the output..... I like Graham's idea here, just trying to determine the vacuum generated from such a system.

You need a liquid ring vacuum pump to do that effectively. They're not exactly the cheapest things in the world though, but it is essentially made for distilling stuff.

Jon

What sort of vacuum pump are you using? Also, someone who knows a little about refrigeration could set up a tube-in-tube coil heat exchanger very simply to cool the vapour (before it hits the vacuum pump) to below its condensation temp.

Ok 3 questions - all because I am trying to avoid drilling and welding more holes to the top of my pressure tank.

1. In the foam trap, do you really need a 2nd hole (the one that is submerged.) Couldn't you just make it a vertical column of 3 inch pipe full of steel wool? Wouldn't the liquid from the bubbles just drip back down? If I made the column 2 feet, that's π*1.5^2*24(inches) approx = 170 cubic inches. Do you get that much foam? Would the foam build up so much it would reach the top?

2. Would the vertical have any effect on methanol recovery (as long as it is insulated) or as I might guess, its strictly a volume issue and vertical versus horizontal makes no difference here?

3. Anyone want to point me in the direction of what do you weld into the pressure tank to create the 2nd whole?

Thanks
Doug

Hi folks

The ideal place for the vacuum pump is between the processor and the condenser. This is because the condenser will be acting at atmospheric pressure, where the condensation temperature will be higher, so the condenser efficiency will be better.

---------
If you put the pump after the condenser, you risk pumping out uncondensed methanol fumes, unless you chill the condenser appropriately.

You also need a vacuum-proof collecting vessel.

And the collection vessel will preferrably be chilled also, so your vacuum doesn't cause boiloff at room temperature.

Plus it would be nice to be able to see the condensate, to monitor collection quantity and rate, and to detect foam-over if it occurs. A see-through vacuum-rated collection vessel is not likely to be easy to find.
---------

This all led me to the conclusion that suction after the condenser was more challenging than I'd like to deal with.

Now, the risk with putting the vacuum pump between reactor and condenser is that you can get condensation in the pump, if it is cooler than 65C. A mechanical pump such as a piston pump would be bad news in that case, and affordable liquid ring pumps are hard to come by. Don't use a refrigeration pump, it just won't last in that location.

So, I chose to use a venturi, being driven by hot liquid above the BP of the distillate, so that condensation in the pump wouldn't be an issue. I use a home-heating circulation pump which is rated to 100C and costs US$100 new (available second hand in abundance for much less - google "Wilo Gold"), and this moves 60 litres a minute thru the venturi. This seemed an economical solution, with little risk of pump failure, and no methanol in an electrical pump. The main gasket can also be replaced with home-made cardboard ones which will stand up to 100C oil forever.

It really isn't necessary to pull a hard vacuum, we are just trying to keep the temperatures low enough so that back reaction is reduced to acceptable levels.

If you can get down to 15" Hg or 7 psiA that would seem to be enough. Lower may be better, but really low pressures may make foaming more likely.

Hope that helps.

Hi Doug

You may be able to do away with the 2nd tube as you say. I'm sure there is plenty of room for improving that design, which still isn't perfect - it can still be overwhelmed with foam.

It may not be necessary even, if you can monitor the contents of the tank with a sight window for example, and cut back heat when foaming develops. It is just a matter of reducing the vapour boiloff rate so that the foam can collapse to below the outlet pipe.

The foam trap can add excessive reflux if it isn't very well insulated, which can make the distillation process slow.

Lots of room for improvement!

quote:

Originally posted by GrahamLaming:
Hi folks

Plus it would be nice to be able to see the condensate, to monitor collection quantity and rate, and to detect foam-over if it occurs. A see-through vacuum-rated collection vessel is not likely to be easy to find.
---------

I got that problem figured out. I added a 3" section of the clear braided hose (3/4" diameter) at the output of the condenser. It does crush down at 30"Hg, but you can still watch the methanol flow through it. I think the braided walls keep it from crushing absolutely flat and sealing off the condenser. I've had that in there for about 10 months now and it still works very well.

quote:

is this due to the acid neutralizing the catalyst, or did i just misread or something?

No its the meth removal after you neutralise the catalyst.

quote:

Also, someone who knows a little about refrigeration could set up a tube-in-tube coil heat exchanger very simply to cool the vapour (before it hits the vacuum pump) to below its condensation temp.

A beer cooler from a pub is a cheap simple way to chill your vapours. They use stainless steel coils immersed in water that is refrigerated.

quote:

The foam trap can add excessive reflux if it isn't very well insulated,

Squirty builders foam from a can does wonders. You can stick shiny silver insulation to it as it sets to add reflective insulation to complement the conductive insulation. Works very well.

quote:

As far as using an indicator, many of us use tumeric as an indicator instead of phenol or phenolphthaline. Both of the latter are clear before their pH trigger, and pink after.... tumeric is bright yellow before and brick red after.

I guess my question is whether or not this even matters? or does the indicator color go away?

I did a few tests yesterday with adding phenolphthaline and tumeric indicator to about 200ml of biodiesel.

Graham,
Do you see the color change in your site tube clearly enough to determine when you've added enough HCL? Pre-separation, my biodiesel is always relatively dark. In fact, the only time i could see a color in the sample was when i sloshed it around in the container.

To answer my own question about the tumeric, yes it does stay in the biodiesel and while i don't think it will hurt your fuel system, it will stain everything yellow.

Hi ProjectPurity

I squish the sight tube flat so the light can pass thru the biodiesel easier. It is certainly hard to see thru a 1" diameter clear tube, and darker biodiesel makes it even more of a challenge!

Have a white sheet behind the tube and a bright light to illuminate it.

Sloshing it continuallt in the flask is important.

Remember - just add enough HCL to make the colour lighten briefly, then stop. It will darken again - don't add more HCL.

You'll get a feel for it after doing it a few times.

I use one of those keyring white LED lights - works just right!

All the phenolphthalein makes it to the glycerol layer, none remians in the de-meth'd biodiesel.

graham

i am in the process of building one of these and collecting my list of materials. i would like to ask you about a couple of details:

* how many gpm circulating pump are you using?
* do i need a refrigeration grade vacuum pump or would something lesser work? if so, do you have a couple of recomendations?

thanks in advance. i can't wait to make my first batch.

Hi Darren

If your batch is X gallons, use a circulation pump rated at least X/2 gallons per minute.

I have an 80 liter batch , I use a 60 liter/minute pump.

Don't use a refrigeration vacuum pump. Use another circulation pump, rated the same as your main circulation pump, to flow the vacuum liquid through the venturi, as in the diagram.

That way, your pump will never see methanol fumes or liquid and will last many, many batches.

We want the pump to be between the processor and the condenser - a refrigeration pump would have a short life there.

Hope that helps,

Graham