Ok much better this time! I made 2 100l batches this weekend with great success!
Jon your method with the pump off worked great!
I recovered about 4L of 98% methanol(this is according to an ethanol trellis??) for each batch out of the initial 22L. Is this about right?
The heat tracing on the pipe to the top of the condenser made a huge difference. I put a PID on it with the thermocouple on the top of the 90 that drops to the condenser and ran it up to about 180F. I am not sure what the ideal temp for this is but 180F seemed to work great, the methanol started condensing out almost right away once it started to heat up.
Here is what I did:
Drained glycerol and started to heat and circulate, I also turned on the vacuum and regulated it to 10Hg (so as not to hurt the pump seals) and ran the condenser and heat trace while heating. Once the bio started to reach 130 the methanol started to flow nicely. At around 150F the methanol was really pouring out. Once the bio reached 190F I shut down the pump and valved it off, then I applied vacuum in 2 steps first I upped it from 10 to about 18Hg and left it there for about 10 minutes until the flow slowed and then closed the valve and let her go to max about 28Hg. The methanol flowed nice and steady for about 40 minutes before it slowed then dripped for about another 15 minutes.
Just for the hell of it, once the meth stopped flowing I fired up the circ pump again to see if any more would come out... None! not even a drip more came out with it circulating so I am not sure if there is any advantage to being able to circulate or not... However I think not! I dewatered the same way and recovered about 1/2L of water in about half an hour.
The bio cooled to 172F after the recovery was complete.
You can really see the soap forming in the sight tube towards the end so once done I immediately pumped the bio in to the settling drum and you could see big clumps of soap all through it. I am surprised the soap didn't clog the pump! I then threw in a big bubbler I made from some copper tubing and valved into my pond aerator
I let it bubble for 4 or 5 hours in the sun outside until dark. In the morning the bio was crystal clear The second batch is bubbling now

I know this was a long drawn out post but I thought I would provide some details in the hopes it may help some one else...

Thanks for everyones help!
Best regards,
Jon

Hi Graham,
I think I could actually do this with my pump by adding an immersion heater to the oil bath. It has a large cast aluminum reservoir that I could drill and tap for a small heater and it utilizes packings instead of rubber seals that should hold up fine.
However I am not sure this will be necessary. Having the pump in between the condenser and tank will be allot harder on the pump then the small amount of methanol that it might see now. I am going on a long awaited fishing trip on Wednesday night and don't have any time remaining this week but when I return I will check the oil in the pump for methanol. Is there an easy way to do this? Is methanol soluble with motor oil?
I am pretty sure no methanol is entering the pump anyways. I have a loop in the vac line from the condenser to the pump that I placed in an ice water bath to see if anything would condense out... nothing. My glass wine carboy is wrapped in foil bubble wrap that extends about 4" up above its shoulder that allows me to cover the top with ice, I then wrap the whole thing in an old wool blanket, this keeps the whole carboy and the exit line icy cold. I also ramp up the vacuum slowly to try and avoid swamping the condenser...
I will post back when I check the oil.

Best regards,
Jon

Graham,
How much methanol is recovered on average from a 100l batch of bio using the venturi?
I have not attempted demething with the bio and glycerol together yet but will hopefully give it a try soon...
Jon

Hi Jon

Good work!

- What sort of vacuum pump are you using? If you put a chilled trap on the vac pump outlet you should get a better indication of how much methanol is escaping the system.

- Could you please explain what you mean by a "heat tracing" on the pipe to the top of the condenser? You mean you mounted the thermocouple there to keep track of the vapour temp? If that's where you were measuring how do you know the bio itself was at 190?

- Also, after you ran the temp up and then valved off the pump and increased the vacuum, did the methanol flow increase again and by much?

Hi Paulus,
It's a werie rietschle vacuum pump from an industrial sausage stuffer!
- The heat tracing is basically a 120v ribbon of heater wire made to wrap around process piping and vessels. I had a spare PID and relay kicking around so I used it to control the tracing to heat the pipe. I mounted the thermocouple for it on the 90 at the top right before the condenser. I have a separate PID in my main processor control box for my inline heater that heats the bio/oil in the reactor. I also have a process thermometer that I acquired somewhere in my pipe work right before the pump.
- Yes. As soon as I started dialing up the vacuum from the initial 10Hg the methanol started really pouring out! You could see the difference after only increasing 1 or 2 Hg Thats why I increased the vacuum in stages, just encase I was swamping the condenser.
I bought a ramping PID for the inline heater but I wont be needing it now as it seems that regulating the vacuum has the biggest impact on my setup and the rate of heating seems to be just rite. I am using a 3000 Watt heater at its rated 240V in a 1.5" piece of pipe. The rest of my plumbing is 3/4" black pipe.

Does that answer your questions?
When I get some time, hopefully next week I will take some pictures for those who may be interested

Best regards,
Jon

Update.
I finally made it back out to the bio shed last night to brew a batch.
I used a piece of 1/2" clear plastic pipe to take a core sample of the vac pump oil to see if any water or methanol got in there as I promised... The oil was uniform in colour with no signs of water or methanol, I also tried to light it on fire and there were no signs of flame propagation.

I would not call this test conclusive but its good enough to reassure me that very little or no meth or water vapors are getting into the pump.
Jon

This looked like a good post to ask about this. I posted elsewhere about a gast pump I can install between my plumber's delight and reaction vessel. It is a GAST dry vane style pump that has the four sliding leafs that move in and out axialy due to the vacuum chamber's eccentric geometry.
Has anyone installed this type pump and used it for methanol recovery? I had the whole pump open and there are no carbon brushes in the motor. Will this be safe with methanol vapors running through it?

I'm using two of those Gast pumps on my super sucker. Mine are rated for 1.5 cfm with a max vacuum of 20" Hg. In their documentation of lubeless sliding vane pumps, I did not see any with better specs. Based solely on their inability to pull a high vacuum, I chose to use something else for the processor. However they might work just fine.

That said, from a safety standpoint I would still be concerned. They tend to run quite hot and I don't know if the friction from the vanes, plus the transfer from vacuum to pressure would create conditions suitable for a fire. Also, when the flow of methanol isn't so great, the ratio of air to methanol in the pump + heat + possibility of spark, might make for interesting times.

Just speculation, but I wouldn't do it...