I have what I believe to be some basic questions about my condenser setup. I should start by stating that I have zero experience or knowledge in this field.
The SETUP:
I am using a modified GL push pull apple turnover with approximately 130L volume (100L batch size). I am utilizing a rotary vane pulley driven vacuum pump instead of the venturi with a 44" plumbers delight condenser. I welded a 3/4" union to the top edge (was bottom from here on in to be called the top) of the tank to be at the uppermost part of the tank. I have a 45 off of the union with about 3' of black pipe with some steel wool in it to break up the foam up to a 90 then about 6" of fittings down to the top of the condenser. The bottom of the condenser attaches to a glass carboy then to the vacuum pump. I have insulated the crap out of the entire setup.

I have brewed 1, 100L batch so far and passed the 27/3 after 2.5 Hours However once I started to recover the meth from the bio (I drained the glycerin for the first batch instead of using GL's PP method)the pump seal failed allowing air to enter the pump causing cavitation. I only managed to recover a little over 1L of methanol before the seals failed. I then drained the reactor, replaced the pump seals and then sucked the bio back in to test the seals.
PROCEDURE:
Start vacuum, pump, condenser water flow and heat all at once.

Once the vac reached about 15hg I could see bubbles forming in the bio via the sight tube. I believe these to be methanol as I did not introduce any water. However after heating and circulating for around 1 hour to about 200F there was almost no methanol condensing out! I believe this is because the 3' piece of pipe to the top of the condenser was condensing the meth thus allowing it to drip back into the processor. Sooo... I acquired some process heat trace to wrap the 3' pipe in that will reach about 250F.
QUESTIONS:
I have a spare PID I can use to regulate the temp of the pipe to the top of the condenser however I am not sure if this is necessary or if I should just let it get up to 250F? If I should regulate it what is the ideal temperature for recovering methanol? What about for dewatering?
I have a ramping PID for the reactor temp but I am not sure of the ideal temps Max/min/avg over what time frame? I have heard 150F tossed around but when you have separate heat for the pot and the pipe to the top of the condenser I am not sure what to do?
How cold does the condenser water need to be? I am looking for a refer unit to cool my 55 Gallon drum of water as after the dewatering the water is not cold any more.

I welcome any and all help and comments!

TIA!
Jon

Hi Jon

As a complete layman this is a best guess as to why you got no condensation of the methanol you evaporated using combined heating and vacuum pressure reduction.

Have a look at table 1 on page 28 (or 32) in this document http://www.methanex.com/products/documents/TISH_english.pdf

If you were pulling a vacuum below atmospheric pressure of 15in Hg that equates to about 7.4psi absolute (the third column). Reading left that shows the methanol was boiling off at about 115*F or 45*C.

Put another way, at 15" Hg if you heat the bio above 115*F it will start to boil. It doesn't matter how hot you heat the bio, at that pressure the temperature of the methanol vapour will be 115*F.

But getting it to condense again is another matter. You'd think that if you cooled the vapour below 115*F it would condense.

You're finding it doesn't. Something Rick posted in the GL push-pull reactor thread the other day suddenly starts to make sense. There is a lot of ordinary air in your tubing, and most of that is nitrogen. It takes extremely cold temperatures to make nitrogen condense so as a rule air does not condense.

And your methanol is only a proportion of the gases in the tubing. Rick referred to "partial pressure." What that means is that the air exerts its own pressure, and the methanol exerts its own pressure too. Combined, they add up to 15" Hg.

So, in fact, the pressure the methanol itself is exerting is likely to be a lot less than the 15" Hg you perceive. And correspondingly to get the methanol to condense you will need to cool it well below 115*F...and I mean well below...to whatever the saturation temperature (dew point) of the methanol is at that partial pressure. What that temperature is, I don't know, not knowing the partial pressure.

In the thread on the push-pull reactor Rick was responding to someone who was having trouble squeezing the last amounts of methanol out of his bio. Notice that in the push-pull design, the methanol evaporates in a moderate vacuum but condensation takes place at atmospheric pressure. What happens when you compress a vapour (ie shove it from a vacuum to atmospheric pressure) and cool it? - it condenses. This is how refrigeration works.

My guess is that with the push-pull design a fair bit of the methanol would escape to atmosphere due to the partial pressure effect, but seeing as both glycerine and bio are being distilled together it's not so noticeable. Maybe also there is plenty of methanol to condense and hence the partial pressure is greater.

Perhaps GL himself can comment further.

is the vacuum connection in the 'pot' or after the condenser ? some how the vapors must move from the 'pot' to the condenser. it sounds like your whole system is under a vacuum with no way to move vapors from the 'pot' to the condenser.

I run a temp of 152F at the top of my reflux column, but I'm not pulling a vacuum either. I would suspect that under a vacuum you'll need to have a lower temp. too high a temp and purity will be lousy.

-dkenny

I just run the vacuum pump after draining the glycerin ala the GL1day process. I don't add heat, and I have found that too much vacuum and the pump will not circulate (not sure of the physics on that one). My residual temp after draining off the glycerin is about 125-130F - again I add no additional heat. I have a 3 foot plumbers delight coming off the top of my reactor and cool it with a 55 gallon barrel of water pumped with a sump pump. I usually recover 1 gallon of 90-93% pure methanol in 45 minutes to an hour. I have no reflux/steel wool, etc. I run the vacuum pump right down to 28" gauge reading. I don't see any methanol coming out of the condenser until the gauge reads 20". It seems to run best at 25"

Thank you for the detailed response Paulus!
How do you convert the Hg to psi?
So I guess, other than for saving energy the temp of the pipe to the top of the condenser isn't critical as long as its above the boiling point of the methanol.
Thanks to your post I can now visualize the methanol vapor trapped in the head space of the reactor instantly condensing and falling back into the bio after the vacuum is released and the reactor is rapidly brought back to atmospheric pressure.
It would be nice to have some sort of inert liquid not soluble with the bio that could be pumped in to take up the head space for meth recovery and then easily separated and removed after recovery... Pipe dream maybe...

Jon

dkenny,
The vac is after the carboy attached to the condenser. It was recovering like crazy until I blew the pump seals and everything cooled down, in particularly the pipe to the condenser.
Purity should not be an issue as I am not doing the 5% prewash.
Jon H, (feels like I am addressing myself )
How big of a batch are you recovering the gallon from? Are you using GL's dry wash or water washing? If dry washing are you getting enough meth out to allow the soaps to drop out in a reasonable amount of time?
On my setup the pump started to cavitate on the recovery stage it got worse fast as the vacuum increased to the point the pump wasn't pumping. Once I replaced the seals I could bring it down to 29Hg and it still pumped fine.
On the first run before the seals went I noticed the same as you, methanol flowing between 19 and up to about 25Hg. I could not get down much lower than 26Hg while recovering I believe due to the meth evaporating.

Thanks for your input guys!
Jon

I wonder if after the meth stops dripping, without removing the vac I suck in clean and dry bio nice and slow from the bottom of the tank till all the head space is full, if it will force out the vapor remaining in the head space or if it will just mix back into the bio...??
Jon

Oh well I was almost right .

Convert can be d/l from http://joshmadison.com/article/convert-for-windows

quote:

Originally posted by Jon Heron:
dkenny,
The vac is after the carboy attached to the condenser. It was recovering like crazy until I blew the pump seals and everything cooled down, in particularly the pipe to the condenser.
Purity should not be an issue as I am not doing the 5% prewash.
Jon H, (feels like I am addressing myself )
How big of a batch are you recovering the gallon from? Are you using GL's dry wash or water washing? If dry washing are you getting enough meth out to allow the soaps to drop out in a reasonable amount of time?
On my setup the pump started to cavitate on the recovery stage it got worse fast as the vacuum increased to the point the pump wasn't pumping. Once I replaced the seals I could bring it down to 29Hg and it still pumped fine.
On the first run before the seals went I noticed the same as you, methanol flowing between 19 and up to about 25Hg. I could not get down much lower than 26Hg while recovering I believe due to the meth evaporating.

Thanks for your input guys!
Jon

My batches are 50-55 gallons in size. There is some methanol left when I am done with the vacuum distillation (not sure how much - but I can smell it.)

To drive off the remaining methanol, I spray circulate it on to the surface of itself in an open top 120 gallon cone bottom tank. Once it passes the sniff test, some soap will actually clump and float - I remove that with a kitchen strainer. I dry wash - have never used any water.

Lately I've been using a centrifuge to remove the soaps. On a good run, I get from start to finish (load processor to soapless BD) in less than 30 hours (one weekend). I really like the centrifuge - it speeds up the gravity settling and there is no soap scum to remove from the bottom of a settling barrel.

A quick note on circulation pumps and vacuum.

When using a pump with mechanical seals under vacuum, it will draw air in through the seals. If the air flow (vacuum) is high enough it will cavitate the pump and stop the flow of the liquid being pumped. If this is a multi use pump that is operated both under vacuum and at atmospheric, then it will weep liquid past the seals when operated under atmospheric pressure. The longer the pump is run under vacuum the more air it will suck at a given vacuum.

There are some solutions.

• Run a mag drive pump as it has no mechanical seals to leak.
  
• Reduce the vacuum below the point it causes cavitation.
  

It it's a multi purpose pump, then you'll need a drip pan when used under atmospheric.

If it's a dedicated pump, used under vacuum applications only, it will not leak, but will degrade over time, drawing more air as the seals open up from running dry.


If I were doing vacuum distillation, I'd do it like jon h. describes, only with a mag drive pump to circulate the biodiesel.

I just got off the phone with a pump expert. I stand corrected. The cavitation is caused by the methanol turning to vapor inside the pump. The little blue pumps (regenerative turbine) are better at handling the vapors without cavitation than most pumps and where the experts choice for vacuum distillation.

It will still blow out the seals over time causing the pump to weep, you still have the option of either using a drip pan or dedicating the pump to vacuum use only.

I stopped circulating for that reason. My pump probably leaks, but I valve it off on either side now.

Just started hooking up one of Murphy's pumps. I'll valve that off from day one to prevent busting the seals.

jon h.

When you do the vacuum distillation, what does the temperature of the biodiesel fall to? With a 50 gallon batch, there's about 1-1/2 gallons of methanol in it. So, you'r getting about 2/3 rds of it out.

Paulus,
Thanks, I already have that great program I just never thought about it!
Jon,
About how long does it take spray drying to pass the sniff test? Do you use heat while spraying?
Rick,
Thanks for the good info! Joe Street (nonprofitfuel.ca)told me the same thing about the pump. On the suction side of the pump the vacuum actually increases causing methanol bubbles to form in the impeller causing cavitation. He says you can mitigate some of this by throttling the output with a valve and stopping and starting the pump, I will try this this weekend.
I don't see how the vacuum will cause the seals to leak after time or for that matter weep when under normal conditions. Do you have any insight into this? However a mag drive pump certainly makes good sense I wonder what the cost of one for this purpose might be?

Thanks again all,
Jon

Dont take my word for it. Call up any pump company and ask them about it. Vacuums are hard on pumps.

Mag drives that can take the abuse start at $300.

Well, I am doing glycerin pretreatment so my methanol is down to about 17%. Not sure if I should get 1.5 gallons or not. I do get more than a gallon, but not a gallon and a half. I usually spray for about 4-6 hours as it cools, to drive off the bit of methanol that is still in there. I've seen reports on here where guys heat and run with venturis for 10-12 hours. I just don't have that kind of time, and if am losing 1/2 gallon - its worth it for me.

I've never paid much attention to the end temperature. I know its cooler, but its still hot to the touch.

Jon H,
Do you dewater the same way, without circulating?
If so how hot is the oil when you apply the vacuum? For how long does it take?
By the sounds of it my new pump seals are not going to last very long and if I where to go purchase a mag pump my baby girl will likely end up with a step father!
Jon

I mostly dewater with upflow settling as described elsewhere on here. But,.... since I am doing the glycerin pretreatment with glycerin that has had a 5% water wash in it, I have gone back to dewatering. I dewater without circulating. If you stick a valve on either side of the pump, you can valve it off so it won't see any vacuum.

I heat and circulate, turn off the pump, close the valves off then apply vacuum. I dewater at around 140-160 depending on when I get to it (no thermostat in system, manually shut off heat). This may be considered dangerous, never leave it unattended without the thermostat - it could overheat/boil, etc.

Thanks allot Jon! I am going to try this today and I will post back my results...
Jon

Hi Jon

You definitely want your vacuum pump between the processor and the condenser, so that the vacuum pump is feeding methanol directly to the condenser.

Don't try to pull vapour thru the condenser.
Push it thru.

That's why I opted for the venturi - it is very tolerant of the temperatures and has no seals as such to fail.

If you use a mechanical pump, you will likely have methanol condense in it as soon as the outgoing vapour hits atmospheric pressure, unless you keep your pump hot at over 65C, which is why I heat the rapeseed oil in the venturi.

Avoid a piston pump in this application (your vane pump should be OK if the seals can be made to last) because condensing methanol can find its way into the pump's sump and contaminate the lube oil, with possible friction damage being the end result.

Hope that helps,