BIODIESEL & SVO DISCUSSION FORUMS





Sponsors    Biodiesel and SVO Forums Home    Forums  Hop To Forum Categories  Biodiesel  Hop To Forums  Methanol Recovery    Interesting Observation during Methanol Recovery

Moderators: Shaun, The Trouts
Go
New
Find
Notify
Tools
Reply
  
Interesting Observation during Methanol Recovery
 Login/Join
 
Member
posted
I have observed this a few times and have waited to report on it till I was absolutely sure of what I saw as it appears to defy common sense. It appears that during methanol recovery, with the main processor vented to the atomosphere, the pressure in the main processor actually DECREASES. Now how is that possible? Perhaps it is limited to my setup (I will describe below.) But the bottom line is that if I crack open an extra ball valve that is is in the headspace, methanol fumes do no escape, but rather air goes into the processor and then into the recovery tank.



The line coming out of the processor is 3/8". The line in the condensing jacket is 1/4 copper. The line to the recovery tank and the vent line are 3/8".

This message has been edited. Last edited by: Doug Weiner,
 
Location: Los Angeles | Registered: March 25, 2008Reply With QuoteReport This Post
Member
posted Hide Post
quote:
Originally posted by Doug Weiner:
I have observed this a few times and have waited to report on it till I was absolutely sure of what I saw as it appears to defy common sense. It appears that during methanol recovery, with the main processor vented to the atomosphere, the pressure in the main processor actually DECREASES. Now how is that possible? Perhaps it is limited to my setup (I will describe below.) But the bottom line is that if I crack open an extra ball valve that is is in the headspace, methanol fumes do no escape, but rather air goes into the processor and then into the recovery tank.



The line coming out of the processor is 3/8". The line in the condensing jacket is 1/4 copper. The line to the recovery tank and the vent line are 3/8".
Your drawing does not show the main reactor headspace vented to atmosphere but actually the atmospheric outlet is shown downstream of the condenser. If this is drawn correctly then it is significant in answereing your question.

There are two things that are at work here, firstly heat transfer, hot always goes to cold which is creating a sort of flow but the biggest thing is that you are taking a large volume of hot methanol vapor and condensing it to a much smaller volume of liquid methanol. So the actual volume occupied by the methanol at the start of the process is much higher than at the end of the process. This is where you get the negative pressure from which causes air to go in when you open the ball valve in the headspace.

Hope that explains it clearly for you.
 
Location: East Yorkshire | Registered: January 14, 2006Reply With QuoteReport This Post
Member
posted Hide Post
Good point about the Processor not being vented, but I think the explanation you propose isn't quite complete for two reasons.

Think of an air-compressor with multiple outputs that are vastly different sizes 2" and 1/4". As long as the pressure in the tank is greater than the atmosphere, both holes will blow out, not the 2" out while the 1" sucks air.

Second and I failed to mention this in my initial explanation is even if I do not open the ball valve, I can see in my transparent methanol recovery tube at the top of the diagram, that methanol which has condensed in the upper hose, but has not made its way to the condenser yet, sometimes flows back toward the processor.

Also, the hot to cold answer you suggest although is surely correct, I 'think' is not strong enough to compensate for what I assume is extra pressure caused by the methanol boiling to the point that now the headspace is less than atmospheric pressure. (1 atm) However, you do point out something important. What is the effect on pressure of the gaseous methanol returning to liquid in the upper tube even before it hits the condenser? Could the condensation in the upper hose be enough? Or is something completely different going on?
 
Location: Los Angeles | Registered: March 25, 2008Reply With QuoteReport This Post
Member
posted Hide Post
A couple of questions. When are you opening the ball valve? During heating? What's the size of the tube in the condensor, and is it crimped at all?
 
Location: Santa Cruz, Ca | Registered: July 29, 2006Reply With QuoteReport This Post
Member
posted Hide Post
I use a PID and thermocouple to keep temps more or less constant and accurate. In this occurrence, I had been maintaing 165F. I think temps had dropped to 161 and the PID was turning back on the heaters.

The tubing in the water jacket (condenser) is 1/4" copper tubing, so it is smaller than the 3/8" tubing that delivers the methanol. The copper tubing itself is not 'kinked' like in a plumbers delight, but it is 'corkscrewed' wrapped downward. There is approximately 20 feet of copper tubing wrapped in corkscrew pattern over a 3 foot span.

Doug

PS. Also please note, that the upper methanol tubing (connecting processor to condenser) is not full of liquid methanol. Its vapor and some liquid methanol that has already condensed. Usually there is not even enough liquid methanol to fill the interior circumference of the 3/8" tubing, just some to fill about 1/2 the circumference. This makes it easy to see which way the methanol is heading.... toward the condenser or back to the processor.
 
Location: Los Angeles | Registered: March 25, 2008Reply With QuoteReport This Post
Member
posted Hide Post
Where in your system is the thermocouple? In the vapor(up top) or in the fluid/plumbing? Remember, that going from liquid to vapor takes energy. Since you have essentially a pot still, the minute you stop adding energy, you stop producing new vapor unless your vapor producing liquid very abundant(not the case), and you're over its boiling point.

Here's a theory: You're getting the back flow as the vapor in the head is cooling and hence taking up less space, and if you get recondensation, then the effect is even more so.
 
Location: Santa Cruz, Ca | Registered: July 29, 2006Reply With QuoteReport This Post



Member
posted Hide Post
The thermocouple is in middle of processor tank (brass tube through a drilled out fitting). Since I am boiling off at such a 'low temp,' the difference between 161F and 165F might very well be the breaking point at which I can keep the top of my tank and head space above 149F and thus have more boiling than condensation. I do have an analog thermometer in the recovery line, and I can see it reads about 143 during recovery, but its numerical accuracy is questionable. What it does tell is that as I recover, it holds the 143 temp pretty steady and after a few hours, when recovery starts to drop off, the temp on this thermometer drops below 140. During my unexpected "lower pressure situations" I don't see much change in the analog thermometer... again the scale is not so detailed on it.

At the cost of setting off another firestorm of hyrdrometers, so far I think I have worked out at way to recovery reliably 97% pure methanol. In the past, using higher temps of about 185-195F, my recovered methanol purity measured about 94% and at times, it would also be polluted slightly yellow, which I am guessing was glycerin or more likely biodiesel. Currently I am getting clear methanol at about 97%.

The recovery technique is to add a another ball valve after the pump to add another path into the processor. In this path, I add a venturi and then at the end of the path just inside the tank, a spray nozzle. The venturi sucks up fresh air so that the headspace never becomes saturated with methanol while also creating a pressure differential to move the methanol to the collection vessel. The spray nozzle creates a lot of surface area for the methanol and air while also helping to heat the top of the tank as the hot mixture hits the top of the tank. Also, of course, the collection vessel is vented to the atmosphere to release the added air. This setup has a number of benefits, first and foremost, the clean air you add, seems to really improve methanol purity. Other benefits, lower recovery temps - so less energy and possible limiting back reaction and is also very safe. Two issues with it so far, is that it doesn't decrease recovery times and you are exposing the bio to more air which is causing some 'aging' of the fuel.

PLEASE NOTE: my initial question does not involve adding air. I observed the 'low pressure' situations when the suction port of venturi was closed as well as when I didn't used the venturi path at all for recovery.


In terms of safety:
In the past, I did implement methanol collection using an Automotive A/C evacuation vacuum pump during recovery. However I was always worried about how much methanol this pump was sucking in and if it might cause a fire. So when I set that system up, I plumbed in a collection jar after the vacuum pump. I simply had a mason jar with two holes in the lid. One was plumbed so that is collected the output vent of the Vaccum pump and the other whole was plumbed to the atmosphere. After 5 or 6 batches of testing, I was collecting about 400 - 500ml of methanol, some was still left in the vacuum pump and unknown amount of that now re-warmed methanol escaped the mason jar to the atmosphere. I was estimating in total 800ml - 1L had passed through that vacuum pump. Way too much for my worrisome head. In terms of purity and time, I feel I didn't fine tune that process enough to officially report on it. All I know for sure was it never performed magic, like boiling off all the methanol at 97% pure in 2 hours. But others are doing it, and have yet to have any issues, so the jury is still out.
 
Location: Los Angeles | Registered: March 25, 2008Reply With QuoteReport This Post
Member
posted Hide Post
On the non-venturi active setup, which I'll call the pot still, I'll go with the contraction & condensation causing the suck back. If I visualise it correctly, the recovery thermometer, only gets heated via the vapor fumes on the way to the condensor. One factor that would affect the temp. reading is how fast the fumes reach it from the top of the chamber. With the pot still, early in the process(assuming constant temp) you have more vapor traveling through there vs. at the end which is why your temp reading goes down.

With the venturi set up, I would suck my 'air' from a tee at the bottom of the condensor(letting the liquid drop into the collection container), or from inside the collection container itself. That way you wouldn't loose any not condensed methanol.

Does your themocouple control the heater on/off(via a controller)? If so, then you might consider putting another thermocouple up at the vapor exit, and hook up a two input controller which I believe should let you choose which one to use to turn the heat on/off while still monitoring the other.
 
Location: Santa Cruz, Ca | Registered: July 29, 2006Reply With QuoteReport This Post
Member
posted Hide Post
quote:
Originally posted by Doug Weiner:
Good point about the Processor not being vented, but I think the explanation you propose isn't quite complete for two reasons.

Think of an air-compressor with multiple outputs that are vastly different sizes 2" and 1/4". As long as the pressure in the tank is greater than the atmosphere, both holes will blow out, not the 2" out while the 1" sucks air.

Second and I failed to mention this in my initial explanation is even if I do not open the ball valve, I can see in my transparent methanol recovery tube at the top of the diagram, that methanol which has condensed in the upper hose, but has not made its way to the condenser yet, sometimes flows back toward the processor.

Also, the hot to cold answer you suggest although is surely correct, I 'think' is not strong enough to compensate for what I assume is extra pressure caused by the methanol boiling to the point that now the headspace is less than atmospheric pressure. (1 atm) However, you do point out something important. What is the effect on pressure of the gaseous methanol returning to liquid in the upper tube even before it hits the condenser? Could the condensation in the upper hose be enough? Or is something completely different going on?

Doug

Can you clarify your analogy.
You mention multiple outputs but according to your drawing there is only one outlet from the system (assuming the ball valve at the top left is closed) and that is via the condenser. Is there an extra nozzle one your reactor that you have not drawn which may be causing confusion? If not then where is this second outlet that you are hinting at?

Regarding the condensing of methanol, you will be suprised at how fast methanol condenses and what a significant change in volume you get. This is assuming that the condenser has a heat transfer rate bigger than the heating element. What are you using for a heating element ie how many kW and what sort of condenser are you using?
 
Location: East Yorkshire | Registered: January 14, 2006Reply With QuoteReport This Post
Member
posted Hide Post
Bernhard & Fuzznag, I updated the diagram. Its a little messy but perhaps more clear.

I am controlling heat with a PID and a thermocouple submerged in the bio.

As for the condenser, its about 20 feet of 1/4" copper tubing wound into a corkscrew that now is only about 3 feet inside a 4" water jacket. I circulate about 40 gallons of water with a small pond pump.

I have experimented with having the thermocouple in the the collection line, and it works great. However, the issue I was having was that my methanol purity would decrease, down to 94 to even 89%. However, this decrease was before I added the Venturi and perhaps I was using much higher temps - can't recall as that was very early on in learning curve. Perhaps for a later project I will add an additional thermocouple in the recovery line and test things out. My goal right now for recovery is to get out 97% methanol and then air evaporate the rest. From my understanding, what you could get out of the bio alone is about 4% volume of Methanol. So for a 40 gallon batch of bio (that's finished bio) you should be able to recover about 1.6 gallons of methanol. I am getting about 1 gallon of 97% pure in 5 hours.

I have tests devised and when I do the next batch in 2 weeks or so, I will report back.

TEST1: During normal processing, for 20 minutes, process with all valves to the atmosphere closed. Then open valve at the head space (far left of diagram) I think I once observed the suck occur under these circumstances. (As for safety, my processor is an Air pressure tank and since I can keep temps under 140F, should not cause too much danger.) This will test if the biodiesel reaction itself increases or decreases volume.

TEST2: During recovery at 165F, close all valves to atmosphere, but for only 3-4 minutes. Increase time as necessary. This should be very apparent that there is no suck in this test. (NOTE, no venturi used in this test) I assume you will only get blow, but it might be possible that all you get is an equilibrium and get neither suck nor blow.
 
Location: Los Angeles | Registered: March 25, 2008Reply With QuoteReport This Post
Member
posted Hide Post
Yup, it's better. Verifies what I pictured.

My thoughts for your experiments:
Exp. 1: I think a sight glass/tube would be better. Since as I recall, the reaction is slightly exothermic, the pressure increase due to a higher temperature, and more methanol possibly evaporating during the time it's closed. Idealy to test the volume change of the reaction, one would have the methanol in a balloon which is submerged in the oil. Pop the balloon, and watch what the volume does - yeah I know this isn't exactly good from the mixing point of view.

Exp. 2: These will be your result. Heat on, blow. No heat, suck.

I have an inline heater & thermocouple following it in my setup(which is a modified appleseed setup). I also have found that I have much better recovery amounts and speed when using the venturi setup vs. pot still with circulation setup. This is from before I added the thermocouple and was relying on a surface temp reading(under the insulation) on my pipes.
 
Location: Santa Cruz, Ca | Registered: July 29, 2006Reply With QuoteReport This Post
Member
posted Hide Post
while I cannot answer you question..
I have some suggestion to improve your recovery..both purity and recovery time

first..add a 2nd thermal couple to to top of the condenser. during recovery READ this couple. ignore the 'pot'..this should be set for 150-151F,,this improves purity..97%. I don't find good enough. I like 99% or higher.

2nd..inject air into the pot..help move the methanol vapors out..reduces recovery time..

if you don't have one already..a reflux column..then move the temp monitoring to the top of the column. mines 5' tall and made from a 2" diameter copper pipe mostly filled to stainless steel scrubbies..

-dkenny


'84 bluebird school bus, DD8.2L turbo( 4/2011, the bus tranny has died..Frown 8.23.11 bus driven to scrap yard Frown )
2006 Jeep Liberty CRD Smile - the wife's
99 dodge 2500 5.9l 24v..-mine Smile
everything run B100 when its warm enough Smile
 
Location: RTP, North Carolina | Registered: December 15, 2004Reply With QuoteReport This Post



Member
posted Hide Post
I inject air using the the Venturi. I do find it makes a tremendous difference. Not to belabour the point, but what is the scale of your hydrometer? I woul love to get 99% pure.

Doug
 
Location: Los Angeles | Registered: March 25, 2008Reply With QuoteReport This Post
Member
posted Hide Post
scale or calibration?

what I did was to use a standard ethanol trellis..the kind that reads % ethanol at 59F.
then I started with a known purity methanol..recorded where this measured..
all future references are to this line not the 'scale' on the trellis.

for my known standard..I used the yellow ISO heat methanol..before they changed the formula. I would think that a new drum of methanol would be close enough..

like I mentioned already.
I use my recovered methanol for 27/3 testing
I also use it to AE processing. its known the water slows/stops the AE process..this leads me to believe that the measurement is pretty good.

-dkenny


'84 bluebird school bus, DD8.2L turbo( 4/2011, the bus tranny has died..Frown 8.23.11 bus driven to scrap yard Frown )
2006 Jeep Liberty CRD Smile - the wife's
99 dodge 2500 5.9l 24v..-mine Smile
everything run B100 when its warm enough Smile
 
Location: RTP, North Carolina | Registered: December 15, 2004Reply With QuoteReport This Post
  Powered by Social Strata  
 

Sponsors    Biodiesel and SVO Forums Home    Forums  Hop To Forum Categories  Biodiesel  Hop To Forums  Methanol Recovery    Interesting Observation during Methanol Recovery

© Maui Green Energy 2000 - 2014