Molecular Sieve Filter

[SkyHighLer]

I thought I had no water in my tank after purging atmosphere and then leaving it upside down in a fridge for 36 hours and the tank was down to 30F.

Then i was super low in the tank and a friend had borrowed my scale and I needed to keep moving so I said "fuck it" took the tank to a safe location turned it upside down and just dump the couple pounds that were left, I mean it was extremely.low I couldn't even flood a tube let alone get a wash.

I came back to discover A SERIOUS AMOUNT of frozen water all on the ground and on the weeds. I said to myself "there's no way that's water I'm using 2-600 gram pure molecular sieve 48 cubic inch filter driers!" So i sat there and watched it all melt, yep it was water for sure!

Hope this story helps

I was doing it every time it got low before I filled it up. It's a 100# tank so it big and heavy. I don't think 30F was enough to seperate the water but I've heard people leaving it overnight at ambient to bleed out the water. That was as low as the fridge thermostat would go. I think maybe you have to reach a certain saturation point.

One of the few good things that came from my battle with parker is that they told me the drier filters only will take out about half of the moisture in the system on every pass depending on saturation. Our setups are a bit different than a large AC setup. I think a longer drier filter would be great. I kinda wanted to go a 3x12 route as I mentioned before.

The actual properties of butane saturated with water has been nagging me, I've been thinking through a discovery test, adding a measured amount of water to my test bag and finding if it sublimes off or remains in the bag after the butane boils off. A saturation test at the boiling point of butane would also be simple, I suppose I'll just keep adding water until...???

If this info is all well known and understood I'm not seeing it being posted up. Theory is one thing, but seeing a phenomena for yourself in your own situation drives the point home.

Please read this one:

I just did the simple experiment mentioned above, I added about five grams of water to about 125g of cold openly contained butane, and it instantly formed a couple of large, solid, floating 'icebergs.'

I wasn't prepared for this, and therefore my camera was inaccessible, I'll do it again with pictures... big, three dimensional 'icebergs,' they contained large frothy bubbles from the initial action. They remained solid until the butane was gone, I never let the heated dish the test bag was sitting in get over 60F.

All the water seems to have been left behind going by the numbers.


If all the water drops out of butane and freezes if you lower the temperature, it would seem to me the way to remove all of the water in your tank/solvent would be to freeze the tank below 32F, and drain off the liquid solvent leaving the rock hard chunks of ice behind...

or most crudely, place your butane in an open container with the temperature down below 32F, and scoop out the ice with a strainer...

So, what about the azeotrope factor? I don't know, but Sunfire's observation seems to point in the direction of if there's azeotrope action going on it's probably minimal.

I've got a couple of pages of numbers, I measured and weighed everything... nothing much to note, though I did finally discover how much of a can of butane is blown by when squirting it into a quart Mason jar, more than I thought, over an ounce (159.9g actual net in the can, 126.5g in the jar, 32.8g lost.)

My next test I'll add the water while the butane is still in the jar to confirm the buoyancy of the 'icebergs' in a deeper pool of solvent, and take pics!!!

This was interesting!


So, GW, do you and WW get ice or whatever in your moisture trap, how does it ever get a chance if it's forming an ice build up to drain? I assume you'd do whatever you could to remove the last of the butane from the trap without heating it past freezing, and then melt and drain the ice/water?? If you've already explained this, or it's just supposed to be obvious as xxxx, pardon me.

One more thing, if there's a specific temperature you're aware of when the moisture drops out of the solvent, I don't think I recall the post.

 

[Gray Wolf]

The butane vapor steam from the Haskel is over 100F when it enters the heat exchanger coils on the first pass and around 27F leaving it to the trap, which is heated to 50F, so as to not retain butane.

The stream is pushed through the mol sieve drier, through the third heat exchanger coil, which self drains to the tank.

The amount of water that butane will hold in solution is a function of temperature. It will hold about .061 grams/L at 20C/68F, with a boiling point of -0.49C/32.12F. It is obviously under pressure to be that warm.

 

[SkyHighLer]

The butane vapor steam from the Haskel is over 100F when it enters the heat exchanger coils on the first pass and around 27F leaving it to the trap, which is heated to 50F, so as to not retain butane.

The stream is pushed through the mol sieve drier, through the third heat exchanger coil, which self drains to the tank.

The amount of water that butane will hold in solution is a function of temperature. It will hold about .061 grams/L at 20C/68F, with a boiling point of -0.49C/32.12F. It is obviously under pressure to be that warm.

Betcha I'm not the only one that would like that in more detail! Dirty pictures!

So, you don't freeze the solvent to get the water to drop out, I kinda get that, but does it all drop out?

It came to me after posting up my ineffective test that what I needed to do to have a real test was to use a sealable container. (If my Mason jar and lid/sealing ring could take the pressure, which it won't) I'd squirt a can of butane into the jar, add the weighed amount of water which would immediately turn to chunks of ice, and then seal up the jar and let the temperature rise until the ice melted, at that point the butane I assume is going to absorb as much of the water as it can, then I can release the pressure, let the butane boil off, and weigh how much water remains, i.e., if any water is being carried off with the boiling butane I'd have the weight of it.

Thanks for putting up with me, this is how learn things at what I believe is approaching base level.

 

[Sunfire]

I think the problem with your experiment is that the butane is boiling away which freezes the water. Inside a pressurized tank the situation would be different.

 

[SkyHighLer]

I think the problem with your experiment is that the butane is boiling away which freezes the water. Inside a pressurized tank the situation would be different.

My thought is if you actually let the water fully freeze out, and keep it frozen out while removing the butane, you'll remove all of the water.

I don't want to clutter up this thread anymore, thanks for following along sunfire, and without getting pissed off!


Here's the pics of about 5g of water added to a can's worth of butane. This was done in what I believe to be a safe manner with plenty of ventilation, on a grounded anti-static mat with zero potential ignition sources.

When I added the water to the bag (back a couple of posts) it formed clumps and floated, I assume due to the trapped bubbles, because in the jar it went right to the bottom, and was hard as a rock and stuck like glue to the glass...and when I finally got a few pieces pieces dislodged, they definitely don't float. I had a tiny little piece I broke free that was composed of frozen bubbles, and it floated, so I think I've got it right.

Quick thought, if you're running sub-zero, how does the moisture from the material get by the filters and into the collection pot if it's rock solid ice? Maybe when you heat the column to recover the last of the butane from the spent material?

 

[Gray Wolf]

Betcha I'm not the only one that would like that in more detail! Dirty pictures!

Dirty peeectures of? The outside of the closed system?

So, you don't freeze the solvent to get the water to drop out, I kinda get that, but does it all drop out?

I'm thinking we only get out the water that is below its dew point at a given temperature, but that appears to be enough.

It came to me after posting up my ineffective test that what I needed to do to have a real test was to use a sealable container. (If my Mason jar and lid/sealing ring could take the pressure, which it won't) I'd squirt a can of butane into the jar, add the weighed amount of water which would immediately turn to chunks of ice, and then seal up the jar and let the temperature rise until the ice melted, at that point the butane I assume is going to absorb as much of the water as it can, then I can release the pressure, let the butane boil off, and weigh how much water remains, i.e., if any water is being carried off with the boiling butane I'd have the weight of it.

Thanks for putting up with me, this is how learn things at what I believe is approaching base level.

Tongue in cheek food for thought. If we are able to remove enough water from the butane for our process to work well, is it necessary to remove 100%?

 

[Sunfire]

I personally doubt one would ever remove 100%. Just gotta do your best.

Sky - from what I've noticed is the properties of the water and butane are different when under pressure. My observations have led me to believe that the higher the pressure in the tank, the less likely the water will freeze and drop out of the solution. This is why I'm recommending purging the atmosphere first, then purging the water. Even still I was unable to get it all.out after 36 hours at 30F. Seems to me the best thing to do is sacrafice about 5 pounds of solvent at the end and just purge the whole damn tank. I was blown away by how much water was still in there. So blown away that I sat there for like an hour watching the ice melt just to make sure it was in fact water. 15 psi compared to atmosphere seems to make a big difference.

 

[SkyHighLer]

I personally doubt one would ever remove 100%. Just gotta do your best.

Sky - from what I've noticed is the properties of the water and butane are different when under pressure. My observations have led me to believe that the higher the pressure in the tank, the less likely the water will freeze and drop out of the solution. This is why I'm recommending purging the atmosphere first, then purging the water. Even still I was unable to get it all.out after 36 hours at 30F. Seems to me the best thing to do is sacrafice about 5 pounds of solvent at the end and just purge the whole damn tank. I was blown away by how much water was still in there. So blown away that I sat there for like an hour watching the ice melt just to make sure it was in fact water. 15 psi compared to atmosphere seems to make a big difference.

If I had a simple six by six sanitary fitting container with a ball valved port, view ports, and a pressure gauge I'd be doing some further moisture test observations on exactly what your conjecturing sf. It would have taken me a million years to come up with GW's moisture trap, but I think the moisture freeze out phenomenon I'm observing is so simple you guys might not have given it proper consideration when it came time to search for solutions.

Thanks again for putting up with my crazy approaches, peace.

 

[NorPene Solutions]

Water Trap, pre mole sieve

Water Trap, pre mole sieve

Hey GW, do you have any recommendations for what type of valve I should use on my water trap's, 1/4" dip tube?

 

[Gray Wolf]

Hey GW, do you have any recommendations for what type of valve I should use on my water trap's, 1/4" dip tube?

We used 1/4" Swagelok valves with compression fittings.

Although the system worked well in the shop and beta testing, the first one in the field didn't fare as well, in that the new owner kept flooding it.

They believe they operated with the traps PID controlled heat turned on, which was working when we recovered the system and prevented flooding at all times that I've been on site, but seemingly has an Achilles heel.

After two field service calls over the hill and far away, we removed the trap and put the heat pad on the drier, which we relocated to hang between the heat exchanger and the storage tank.

The whole purpose of the trap, was to reduce the frequency that the drier had to be serviced, not make it more prone to service in actual use.

No service calls since then, so it is ostensibly a more robust and tolerant arrangement.

Of interest so far however, is that both times that we recovered the flooded system, we also checked the tank for water, and found none either time.

 

[NorPene Solutions]

We used 1/4" Swagelok valves with compression fittings.

Although the system worked well in the shop and beta testing, the first one in the field didn't fare as well, in that the new owner kept flooding it.

They believe they operated with the traps PID controlled heat turned on, which was working when we recovered the system and prevented flooding at all times that I've been on site, but seemingly has an Achilles heel.

After two field service calls over the hill and far away, we removed the trap and put the heat pad on the drier, which we relocated to hang between the heat exchanger and the storage tank.

The whole purpose of the trap, was to reduce the frequency that the drier had to be serviced, not make it more prone to service in actual use.

No service calls since then, so it is ostensibly a more robust and tolerant arrangement.

Of interest so far however, is that both times that we recovered the flooded system, we also checked the tank for water, and found none either time.

Flooded the system with collected water or liquid butane?

 

[Gray Wolf]

Flooded the system with collected water or liquid butane?

Liquid butane

 

[NorPene Solutions]

We used 1/4" Swagelok valves with compression fittings.

Although the system worked well in the shop and beta testing, the first one in the field didn't fare as well, in that the new owner kept flooding it.

They believe they operated with the traps PID controlled heat turned on, which was working when we recovered the system and prevented flooding at all times that I've been on site, but seemingly has an Achilles heel.

After two field service calls over the hill and far away, we removed the trap and put the heat pad on the drier, which we relocated to hang between the heat exchanger and the storage tank.

The whole purpose of the trap, was to reduce the frequency that the drier had to be serviced, not make it more prone to service in actual use.

No service calls since then, so it is ostensibly a more robust and tolerant arrangement.

Of interest so far however, is that both times that we recovered the flooded system, we also checked the tank for water, and found none either time.

Hey GW, any thoughts on putting the water trap and drier prior to the heat exchanger or does the process work more efficiently on liquid, post condenser?

 

[Gray Wolf]

Hey GW, any thoughts on putting the water trap and drier prior to the heat exchanger or does the process work more efficiently on liquid, post condenser?

It would be less likely to flood before the heat exchanger. We had it between heat exchangers.

 

[HG23]

Hey GW,

Is there any new news on the water trap? It sounds to me like you were dealing with operator error in the described instance. Is the trap continuing to operate normally for that person since the issue?

I think I am ready to add one to my system. Can WW help me out with that? I know Wayward is a busy guy and I don't believe I have a current contact for him.

I am also interested in adding more filtering in general to the hydrocarbon stream. I know you were working with VICI Metronics, I believe, but I'm having a hard time digging up your posts on the subject. Do you remember which model filter you were using and where you had it located in the system? I thought maybe you had mentioned a guy you were working with over there, are you still in though with him? I want to add one just as an extra precaution.

 

[HG23]

GW, I saw you posted the VICI info the other day, I will be calling him next week.

How about the water trap, did that customer get it to work for them properly?

 

[Gray Wolf]

Hey GW,

Is there any new news on the water trap? It sounds to me like you were dealing with operator error in the described instance. Is the trap continuing to operate normally for that person since the issue?

I think I am ready to add one to my system. Can WW help me out with that? I know Wayward is a busy guy and I don't believe I have a current contact for him.

I am also interested in adding more filtering in general to the hydrocarbon stream. I know you were working with VICI Metronics, I believe, but I'm having a hard time digging up your posts on the subject. Do you remember which model filter you were using and where you had it located in the system? I thought maybe you had mentioned a guy you were working with over there, are you still in though with him? I want to add one just as an extra precaution.

Following field beta testing, WolfWurx only built two traps for customers, which they didn't charge for, and removed them both from the field after having flooding issues with the first one. The second customer had not yet tried it.

As we tossed in the traps for free, there was no issue removing them and we did so because of the distance we were having to travel to recover flooded systems. Whether it is a design issue or a customer issue, it is too touchy for a production unit, which needs to be bulletproof.

Wayward now has at least one sitting around, which he could no doubt be talked out of it and can be reached by PM on this forum.

The filters from VICI have all been custom, and have not assigned part numbers yet. I've been working with Isaac and their general number is 877-697-9199.

 

[Gray Wolf]

GW, I saw you posted the VICI info the other day, I will be calling him next week.

How about the water trap, did that customer get it to work for them properly?

We gave up on the trap as too finicky for general use, requiring more labor that it was going to ostensibly save.

 

[HG23]

Ah, too bad on the trap.

The trap part was simply meant to extent the service life of the mole sieves, right? Sounds like it might be easier from a processor standpoint to change out the sieve more often.

Are you going to improve the design, maybe a larger base? Was their post pump chiller perhaps too efficient, if that's even possible?

 

[Gray Wolf]

Ah, too bad on the trap.

The trap part was simply meant to extent the service life of the mole sieves, right? Sounds like it might be easier from a processor standpoint to change out the sieve more often.

Are you going to improve the design, maybe a larger base? Was their post pump chiller perhaps too efficient, if that's even possible?

It would have been easier for the customer to service the mol sieve from our standpoint, because of the four hour drive plus recovery time, aka known as (2) man days, plus a $100 in fuel.

We took out the second one, before it came on line, cause it was a man day just getting there.

The simple heated mol sieve after drier has performed flawlessly for both operators thus far, so we're liking it from here.

We are also experimenting with fore line filters, because of how much damage ingesting butane with resin it it can do to a pump.

It is more efficient to filter and dry after the pump, but if you add more area you can reduce loses.

It is hard to flood a fore-line filter, but also harder to keep water in it, once it is under vacuum, so we are looking closely at filter/absorbant medium, to see if we can combine a drier function as well.

PS: The system was installed with the center 50' coil of the heat exchanger after the trap/drier, so it is possible some back feed occurred after pump shut down and trap heater cooled down. The center coil is actually designed to be the chiller coil, for a refrigerant plant, or liquid N2, but we shanghaied it when he decided to start with dry ice.