Making a compound column for perfect separation of fractions

VM CM LM Flutes Plate

Moderator: Site Moderator

Making a compound column for perfect separation of fractions

Postby Tropic » Wed Dec 06, 2017 9:40 pm

I’m still considering my first build and will be for the next few weeks so please excuse my way out ides for a bit. It seems like stills are like pancakes - the first one is always not quite right, but I’d like to get close. So if you’re inclined, please let me know if this would work and if not, the physics of why not.

Anyway, what I’m considering is running a still with a condenser in the middle of two lots of packing. I’d run the coolant in the central condenser at 79 degrees C. This is pretty achievable with a high temperature pump circulating the coolant from the condenser and using a solenoid adding small amounts of cool water based on a temperature controlled thermostat. What I’m thinking is that this will run like two stills. The bottom half will reflux the heavier fractions and mostly pass up vapours that condense below 79 degrees. The top will then run like a normal still (I’ll be using a LM head) but the source vapours will be much more pure than coming straight from the boiler. In effect, I’m trying to force perfect separation of fractions.

The other thing I’m considering is making it in parts so I can remove and swap bits around if it doesn’t work or I want to run only the head as a pot still. The head, top packing, central condenser and lower packing will all be able to be separated. I plan to take each section pipe to a muffler man to have 120mm ends turned down so they can slide into the next section below (with Teflon tape to seal it). Is this viable?
Attachments
Separation column.png
User avatar
Tropic
Novice
 
Posts: 9
Joined: Thu Nov 23, 2017 10:17 pm

Re: Making a compound column for perfect separation of fract

Postby greggn » Thu Dec 07, 2017 10:38 am

I don't believe that's a very efficient or practical design. The 300mm section is too short to provide meaningful reflux and, in order to create robust reflux in the 600mm section, you'll need to throw so much heat into the boiler that you'll probably overwhelm the reflux condensor.

Make it easy on yourself ... buy a few stainless or copper spools with tri-clamps and make yourself a CCVM.
________________

I drank fifty pounds of feed-store corn
'till my clothes were ratty and torn
greggn
Swill Maker
 
Posts: 424
Joined: Sun Jul 27, 2014 11:59 am

Re: Making a compound column for perfect separation of fract

Postby zapata » Thu Dec 07, 2017 12:27 pm

Its not a terrible beginner idea. But I think it shows either a lack of understanding or faith in the natural functions of a reflux still.
Take out your middle condenser, and fill the whole thing with packing. Thats 1.2 meter packed. After reaching equilibrium at full reflux with decent packing and a reasonable power input you will have PERFECT fractionation. Not limited by the precision of your thermometers or valves, or whatever. Just laws of physics perfect fractionation. As in infinte precision. Science is like that, have faith until you have experience. You're just replicating the experiments thousands of really smart people have done before us.

It is after all the only reason a reflux still works at all. The fractions stack up the column by being constantly redistilled. If you really don't trust it, might as well take out both your packed sections and try some other "improvement".

Or maybe try this way of thinking about it. The point of your partial condenser is to stop higher bp components from traveling up the column, right? Well that is exactly what reflux is already doing in a packed column, it is condensing high bp components and sending them down, while vaporizing lower bp components. And it isnt doing it at one spot for one temperature but at a million spots at every fraction of a degree that matters, and "magically" skipping the degrees that dont matter.
zapata
Swill Maker
 
Posts: 379
Joined: Fri Jul 07, 2017 1:06 pm

Re: Making a compound column for perfect separation of fract

Postby zapata » Thu Dec 07, 2017 12:41 pm

Hold on, my non metric brain just caught up, this is 120mm diameter pipe? I dont think 1.2 meters will cut it unless you are using SPP for packing. And at almost 5" diameter, you're likely to have problems with any packing, as well as feeding the thing with enough alcohol to work.
zapata
Swill Maker
 
Posts: 379
Joined: Fri Jul 07, 2017 1:06 pm

Re: Making a compound column for perfect separation of fract

Postby StillerBoy » Thu Dec 07, 2017 4:03 pm

This setup is something in line to what Darek Bell, at Corsair Artisan Distillery, uses on some of his production units.. he states that it compresses the vapours closer to the boiler, and the vapours that do move on, have a higher proof.. which is also the concept that the ethanol auto industry uses..

And you would not need to have 120 mm long pipe flair, that over 4", 70 - 75 mm long flair would be plenty.. the problem I see in flaring is to be very careful with the fit.. if the gap is too wide, teflon tape alone with not seal, then you will need to use 3/4' tape around the joint area, then a thin sheet of copper metal, and held in place with a 3/4" wide muffler clamp.. and you may want it closer to the boiler, that is the first condenser up from the boiler..

Mars
StillerBoy
Site Donor
Site Donor
 
Posts: 467
Joined: Thu Dec 01, 2016 6:27 pm
Location: Ontario

Re: Making a compound column for perfect separation of fract

Postby Tropic » Thu Dec 07, 2017 6:13 pm

Thanks everyone. I have a few years’ experience with an Essencia Super Reflux CM head and really understand why it doesn’t work well. It is obviously too short to properly reflux. It uses cold water as LM which is a balancing act trying to get the vapour down to the right temperature and flow rate to pass on to the condenser.
I think that by using a much larger internal LM condenser at the right temperature it will compress the vapours as I dictate it. I have a science background, and thus my need to measure and control the variables. I am thinking of doing this to make it more controllable, faster and better tails. Towards the end of the run in a normal column the ethanol content in the vapour runs low and more and more heavier fractions get carried through to the head. I would think that the internal condenser would knock those down low in the column allowing for the top to keep processing higher quality product later in the run. I’m glad to hear that others are already using this so will try to check them out.

And sorry for the confusion. I’m using 63mm (2.5 inch) diameter pipe and will use stainless steel scrubbers as packing. I was thinking of making the turned down ends 120mm (5 inches) long. I’ll drop that down a bit as advised and check the fit of each pipe as it is made.
Last edited by Tropic on Fri Dec 08, 2017 11:14 am, edited 1 time in total.
User avatar
Tropic
Novice
 
Posts: 9
Joined: Thu Nov 23, 2017 10:17 pm

Re: Making a compound column for perfect separation of fract

Postby zapata » Thu Dec 07, 2017 11:16 pm

stillerboy, do you have a reference to Corsair's still you are referring to? I poked around but couldn't find anything about them having a packed column still. There are a few things on the tour photos I honestly don't know what they are though, but I did see several pot stills and a plated column. Are you perhaps referring to a more traditional "purifier" on a pot still like in some scotch whisky stills? That would be a similar goal, but totally different from being in the middle of a packed column. Barring Corsair having something novel I didn't find I don't think there is anyone doing anything similar to OP.
zapata
Swill Maker
 
Posts: 379
Joined: Fri Jul 07, 2017 1:06 pm

Re: Making a compound column for perfect separation of fract

Postby RedwoodHillBilly » Thu Dec 07, 2017 11:28 pm

Tropic, the concept isn't uncommon. There are those that use a packed column above a plated column. The idea is that you feed the packed column with higher ABV vapor to make the packed section more efficient. If I read your post correctly, this is what you are advocating. If you are trying to get a high ABV, low flavor product, I don't see a problem. There are many ways to skin a cat, and this is just one of them.
John Barleycorn must die.
"and little Sir John in the nut brown bowl proved the strongest man at last.
The huntsman he can't hunt the fox, nor so loudly to blow his horn
and the tinker he can't mend kettle nor pots without a little barleycorn."
User avatar
RedwoodHillBilly
Site Donor
Site Donor
 
Posts: 756
Joined: Tue Sep 01, 2015 7:30 pm

Re: Making a compound column for perfect separation of fract

Postby zapata » Thu Dec 07, 2017 11:39 pm

Tropic, by all means try it if you are convinced. I understand your motivation, physically suppress the tails so they can't possibly rise in the column. I do question the need to do it at all, a good reflux still just sits at azeo all day long before spouting out a bit of tails and being shut down (or automatically shutting down if a good VM). If your still works according to theory it'll also sit at azeo all day and then just shut down without spouting a drop of tails. Some benefit I guess, though I've never really looked at a few jars of tails out of 20 as major burden, it's not like they are smeared into the hearts. Truthfully, I'm usually super tired by then and just turn it off and walk away at the first hint of tails so I barely see them at all.

But just to help you think it through, what is your hypothesis on specifically how this dephlegmator will work any better than the reflux that would naturally be there if it were just a packed section? I don't monitor column temps that low in the column, but I do monitor them at about 1/3 down. I've got azeo there for hours, that temp is the main thing I run my column based on after heads are handled. The reflux keeps the tails suppressed below there until it's time to shut down. Sometimes after a .1 degree rise I'll close the takeoff to restack the column and take another good jar, but often I'm just tired and stop. Even with your still functioning as hoped, production is going to slow down to the same degree as less and less vapor makes it to your upper column. I guess the question remains, why do you really care if you have tails at the lower 1/3 of your column if they don't come above the top 1/3 until your last jar?

And how do you manage the upper LM head with no way of really knowing how much vapor is being fed into the upper column? On a trad LM head you set the reflux ratio by taking off the inverse of known maximum production (based on boiler contents and power input). I guess you can run it blind with unknown reflux rate just by assessing output, or maybe watching a mid column thermometer somewhere?

What if your dephlegmator isn't run perfectly? Will it knock down some heads, trapping them in the lower column longer than necessary and just smearing them longer than needed? Theoretically it shouldn't, but how closely will practice align with theory? How cold is your input water, and how long before it heats to your set point, is it condensing heads and ethanol until it heats up? Will it condense too much and flood the lower section? Again; shouldn't, but in execution?

Your goal however is not a bad one. Though I think it is already solved pretty well in hybrid reflux stills which put physical plate/s below a packed column. Physical plates really do seem to present a physical barrier (rather than "just" a theoretical plate) to higher BP components rising in the column. Which would make them a very good design to also compare your proposal to. They also improve the efficiency of the packed section by feeding them a steadier diet of higher ABV which allows the packing to function at a lower (better) HETP. And with a sight glass they provide a visual cue of the onset of heads (fogging or running dry) Just in terms of height, if you used that 300 mm for physical plates (but placed at the bottom of the column where they are proven to perform best) you could fit at least 2 plates (3 if you give it just a few more mm). A hybrid still with 2 plates and a meter of packing is a proven workhorse in both theory and practice. (I'm not really certain how well plates work in 2.5" pipe, I've not seen smaller than 3" used). Granted it's a more expensive and troublesome build, but so is a novel 1 off, especially if it doesn't work as hoped.

Honestly, I'm of the never done building and experimenting camp so I will applaud you for building it and letting us know either way. The worst that happens is you "waste" some time in the shed, learn a bunch, make some spirits, and maybe have a few holes to patch when you make your next pancake. I've pointed out some theoretical questions but I think you'll hit some practical ones too if you get to the building and running stage. If it's of any help to you, though I'm veering off topic, if I were making a new perfect pancake I would focus on gadgetry for dealing with heads more efficiently as that pays dividends to me that dealing with tails doesn't. I can shut off early and never see tails, but I HAVE to get through heads, and the faster and better you can do that, the more time you save every run and the more and better your hearts are (assuming you still cut in time for the tails). I'm currently playing with a variable volume heads trap to deal with heads during the standard equilibrium phase and not have to sit through bleeding them off. Just food for thought.
zapata
Swill Maker
 
Posts: 379
Joined: Fri Jul 07, 2017 1:06 pm

Re: Making a compound column for perfect separation of fract

Postby kimbodious » Fri Dec 08, 2017 3:06 am

Great discussion!
I didn't know there was an Essencia Super Reflux LM still, I thought their reflux still was a CM?
Good point Zapata about always having to deal with heads but being optional about having to deal with tails. I am the same as in choosing to stop as soon as tails appear.
Tropic's design will require two condensers in the column, what are you going to use coils or shotguns? I've got all the components to build a VM still similar to this design but I think I'll stick with my CCVM :thumbup:
kimbodious
Swill Maker
 
Posts: 318
Joined: Fri Apr 17, 2015 3:57 pm

Re: Making a compound column for perfect separation of fract

Postby StillerBoy » Fri Dec 08, 2017 9:22 am

zapata wrote:stillerboy, do you have a reference to Corsair's still you are referring to?

Darek makes reference to it in his book,"Alt Whiskeys", and also has a picture of the setup.. the build is in the style of a dephlegmator, located just at the base of the column..

Mars
StillerBoy
Site Donor
Site Donor
 
Posts: 467
Joined: Thu Dec 01, 2016 6:27 pm
Location: Ontario

Re: Making a compound column for perfect separation of fract

Postby Tropic » Fri Dec 08, 2017 12:17 pm

Hmm, my hypothesis… My original thought was to run a water jacket with a coldfinger the full length of a 1.2m column both held at 79 degrees. Pack s/s scrubbers inside the column around the coldfinger. I theorised that if I ran an infinite length column of this at 78.0 degrees for an infinite amount of time I could pull out all the heads. Then if I ran said infinite column at 78.2 degrees on the remaining wash for an infinite time I’d get pure azeo. There would be no need to reflux anything.
A standard column (or even a column on top of purifying plates) is a balancing game of how much to reflux and how much heat energy to apply to the wash. The longer the column, the easier this is to keep near-pure ethanol coming off the top of the column. The reflux is used as the cooling agent. Instead of using the reflux as the coolant, I thought I could use an intern temperature controlled condenser to as a method to cool the vapour and stop the tails from entering the column at all. Because I can’t have an infinitely long column, I would originally settle for trying to eliminate as much of the tails as I could then refluxing to cool the top half in this design.
But at first, I would run the lower condenser at 77.5 degrees until it ran dry. Theoretically this would remove the heads. Or more amusingly, I’m going to create this in components so could move the temperature controlled condenser to the top of the column, just before the take-off top to ‘polish’ the product, and allow the rest of the column to reflux as normal.
If I can’t remove all the heads then the LM head could be calibrated exactly the same way as normal – run the thing at full take off to assess the theoretical maximum, etc. I’ve heard a constant, definable reflux ratio is one of the holy grails of still design. How about this keeping with my theory of controlling everything? You take off ALL the distillate produced and divide it up manually. Eg, you collect it into a holding container with ten holes in the bottom. You then have two sliding trays – one for collection and the other to return as reflux back into the column that can be moved to collect each of these streams.
Hi Kimbo, the Essencia is a CM, sorry for the typo. I’d use a coil packed with s/s scrubber inside and out for maximum efficiency. You ccvm indeed looks impressive.
User avatar
Tropic
Novice
 
Posts: 9
Joined: Thu Nov 23, 2017 10:17 pm

Re: Making a compound column for perfect separation of fract

Postby Saltbush Bill » Fri Dec 08, 2017 1:02 pm

zapata wrote: I do question the need to do it at all, a good reflux still just sits at azeo all day long before spouting out a bit of tails and being shut down (or automatically shutting down if a good VM).

That is the crux of the whole thing, what ever you do you wont build anything different that works any better. So why not just stick to a known design?
User avatar
Saltbush Bill
Trainee
 
Posts: 910
Joined: Thu Mar 17, 2011 2:13 am
Location: Northern NSW Australia

Re: Making a compound column for perfect separation of fract

Postby Tropic » Fri Dec 08, 2017 5:23 pm

Saltbush Bill wrote:
zapata wrote: I do question the need to do it at all, a good reflux still just sits at azeo all day long before spouting out a bit of tails and being shut down (or automatically shutting down if a good VM).

That is the crux of the whole thing, what ever you do you wont build anything different that works any better. So why not just stick to a known design?


To experiment and have a bit of fun with it.
User avatar
Tropic
Novice
 
Posts: 9
Joined: Thu Nov 23, 2017 10:17 pm

Re: Making a compound column for perfect separation of fract

Postby The Baker » Fri Dec 08, 2017 8:01 pm

Saltbush said, '
That is the crux of the whole thing, what ever you do you wont build anything different that works any better. So why not just stick to a known design?'.

When I first joined distilling forums Alex Bokakob was doing different things with reflux stills.

Then later people were experimenting with plate stills....

Geoff
The Baker
The Baker
Distiller
 
Posts: 1307
Joined: Sun Aug 27, 2006 4:48 am
Location: Australia

Re: Making a compound column for perfect separation of fract

Postby greggn » Sat Dec 09, 2017 1:39 pm

I think the prevailing attitude here, with regard to new ideas, can largely be summed up as "Give it a try."

... but usually novices are encouraged to use proven designs and recipes to develop a feel and understanding of how things work before experimenting.


>When I first joined distilling forums Alex Bokakob was doing different things with reflux stills.

I assume that the still design commonly referred to as the "Boka" was not his first build. In this case, the OP is building his first.
________________

I drank fifty pounds of feed-store corn
'till my clothes were ratty and torn
greggn
Swill Maker
 
Posts: 424
Joined: Sun Jul 27, 2014 11:59 am


Return to Column Builds



Who is online

Users browsing this forum: No registered users and 2 guests