Please review the continuous still design

[rad14701]

Thanks, Nykter... That helps a bit but what, exactly, is going on up in the head section...??? I guess I'd be concerned about slicing a long column in such a fashion to be possible to resolder it with a plate running almost the entire length... From a practicality standpoint I'd think two columns would be a better way to go... You could still use a single heat source if you were creative...

[Nykter]

Yes, you're right,
You maybe could solder two pipes together, or put one smaller pipe inside the other. Both would be easier to build.

But you get the principle?
Have any solution to how to distribute the reflux evenly?

[manu de hanoi]

oh ! I just got it. but .... That's a strange setting.... usually the outlet of the stripper is connected to the bottom of the rectifier....

but why not... I see 2 uncertainties maybe,

1) the temperature gradient in the rectif section will be the same as in the stripping section, how would this affect the process ? I feel the level of the output pipe should be 78oC in both section .... And even then... If the temp gradient is the same , you wont be able to fractionate more accurately than you would in the stripping section....


2) the output of the rectif section will be gazeous, how would you regulate that output ?


All in all maybe having a tray for ethanol with a valved output plus a packed column and a reflux condenser on top of that would do the trick.

[Nykter]

My drawing is just the heads removal section. You will also need a rectifying section between this and the stripping section. You could also add a heads concentrating section on top of it. This wont give any cleaner product, but it will let you remove a more concentrated heads, thus not wasting that much ethanol.
Try a google for "divided wall column" thats the principle, but you dont need tp build the lower part since you wont need to concentrate the tails.

In your last drawing you will be able to remove heads in the top, but you will never be able to have a heads free product, since new heads are always inserted into the still with the new mash. This heads will always be on its way to the top, passsing the product outlet and polluting the product.

[manu de hanoi]

The still has been built and tested, here are the results :

-the wash inlet of the column makes it very difficult to insert packing in the section between the condenser and the wash inlet.

- instead of using the boiler as drawn i just used the bottom of the column pipe because it was easier. Only prob with it is that it's surface is too small for heating, resulting in an extremely slow operation

- with an 10% wash i get 70% purity .... Something is wrong, i want more

- using the jacketed cooler on top of the condensator makes it very difficult to set the wash flow in the narrow zone between between total reflux and little output with high purity.


TODO after tet holidays :
-make a bigger boiler that allows a bigger heating surface to speed up things
-use slanted plates below the condenser with an ARC ouput.


About ARC:
1) It seems very nice, but.... I fear it hasnt been fully tested. Riku advises for 80% of the column length for the sensor tube (in other words that means waiting for the column to be filled 80 % with 96% ethanol ) ..... but in a continuous setup this may not apply to the column section that is below the wash inlet (stripping part of the column).

2) There should be some distance between the wash inlet and the bottom of the sensor tube... if I take 20 cm, that leaves me with 20cm for the sensor tube. Would that be enough ?

3)Moreover I don't understand the purpose of the small loop on the top of the sensor tube just before the connection to the column, does it has purpose ?

4) If we want ARC to work well perhaps the sensor tube could be sligthly flattened and soldered along the column in order to ỉmpove heat transfer ?

[possum]

Wow.
I was just going to ask if anyone has made a Coffee type VM still for stripping runs (of clarified, all liquid washes/mashes), and here it is.

I would only use a rig like this for stripping runs, and then make cuts on a potstill, or run the rig in traditional VM style.


I don't really trust the automation designs, because I want to observe and sense firsthand what is comming out of my rig.

I'll give updates after I get it made and running.
My one concearn is that the flavors might get stripped out too much.

I would personally be happy with getting 30% abv out of the rig.
I'm planning on using gravity feed system.

[Nykter]

Congratulations!

- instead of using the boiler as drawn i just used the bottom of the column pipe because it was easier. Only prob with it is that it's surface is too small for heating, resulting in an extremely slow operation

I can imagine that if you use fire fpr heating. I have only tried electric, and as long as you can fit the element into the boiler, thats enough.

- using the jacketed cooler on top of the condensator makes it very difficult to set the wash flow in the narrow zone between between total reflux and little output with high purity.

I can imagine this is why you only got 70%. Did you get all the ethanol from the wash out as product, or did you loose some through the drain?

About ARC:
1) It seems very nice, but.... I fear it hasnt been fully tested. Riku advises for 80% of the column length for the sensor tube (in other words that means waiting for the column to be filled 80 % with 96% ethanol ) ..... but in a continuous setup this may not apply to the column section that is below the wash inlet (stripping part of the column).

Yes, I'm no ARC expert, but i think you should only count the rectifying part of the column. The stripping part is to be considered as boiler

[manu de hanoi]

i forgot to try the hydrometer on the output wash, but i tasted it regularly. All flavors (including alcohol) were gone, it just tasted like boiled water. Of course at some point i pushed the wash flow high enough so that alcohol and taste did come out of the boiler but otherwise it`s pretty safe.


Btw I`ve put 2 trays at the bottom of the column, just on top of the boiler.

[absinthe]

i just thought of a flaw in the AFC system..

the CO2 it will bubble up the feed pipe and push the wash out before it starts to boil and will change how the system runs depending on how much CO2 is dissolved in the wash..

I'm not sure how much this will affect the running of the still but it might be worth thinking more about

[manu de hanoi]

yup, the wash i tested with wasnt real one and the preheating of the wash was already producing gas that was making the gravity feed unstable......

AFC (wich is not used for this still unlike ARC) doesnt need to be extremely accurate, besides a good shaking of the wash priot to operation plus the dropping in the wash inlet should remove most of it.

[Nykter]

Yes, Absinthe, you're probably right about that. I have only tried AFC in "old" wash, that has been sitting for a while so it hasn't any CO2 left.

Another strange effect is that AFC works on water as well, probably because of dissolved gases.

Manu, I'd really like to see some pictures of your still. What kind of trays is it you have between the boiler and column?

[absinthe]

i thinking if you added a per heater to the system and a gas "burp" pip or similar then you should get rid of allot of the gas..

i know from my "in line stripping flange" you get bubbles about 1-2 inches long about every 10 to 15 secs..

but with the vent (burp valve) they are gone completely before the wash enters the still

[Nykter]

You might be able to combine the preheater and burp pipe with the first part of the product condenser. The burp will have to be close to the reservoir, not to the bottom bend, since the gas then would have to bubble up all the way.

[manu de hanoi]

Serviteur !
http://s249.photobucket.com/albums/gg22 ... filter=all" onclick="window.open(this.href);return false;" rel="nofollow

One more burp (a dripper right ?) is an excellent idea absinte, it will require one more valve though.

[absinthe]

as far as i can see from your setup if the mash is getting heated before it runs into your funnel then allot of the OC2 would excape as it drips into the funnel

[evilpsych]

I've reviewed it.. and here are my thoughts on the hobby-scale continous concept. Have the wash enter below the product output, but in a long continous spiral down through the reflux column -perhaps 1/8" tubing, slowly picking up heat. by the time it reaches the bottom, it will be ready to give off the alcohol without a lot of added heat, you could probably save on packing material this way.. also, on the overflow concept, use a standpipe, say 1/2", that connects into a larger pipe, perhaps 3/4 or 1" sort of a 'parrot overflow'. that way, steam pressure would have to fight the added volume of spent wash in the larger pipe. i would probably also cool the spent wash and perhaps recirculate it back into the main body of wash, probably at the bottom to prevent evap of any remaining alcohol.

what sort of volume are you considering for the boiler? I would think it to be lots smaller than traditionally held, but with a lot more headroom.. for a hundred gallon wash,, what.. a gallon? when the wash come out, i'd also expect the alcohol to almost flash off straightaway.

[manu de hanoi]

Hi continuous guys , i've tested the setup twice

1st test)
- Boka's head added,
- 5 liter boiler added really improves the heat transfer
- aquarium pump added and gives a more constant feed and ease of use
- 2 more trays were added on top of the existing 2, that increased the pressure in the boiler and steam instead of spent wash was going out of the siphon


one run with 60 L real wash, took about 9 hours in the cold with coal fumes and a leaking product output valve, without understanding why i would sometimes climb mysteriously to 85 degrees why other times struggle around 70 degrees ABV.


2 nd test)
- Only kept 2 trays instead of 4
- added one extra burp - thanks absinte - very useful because even after last racking some co2 still goes out in the reflux circuit

- Added one meter of packed column on top of the existing one, used the 3 reflux tubes going thru that column as holes for holding thermometers.

-the whole setup is 2,5 meters high, it once fell and broke the abv meter.
The chinese (for uncle jessie's information) aquarium pump cant even pump that high, the wash container had to be elevated.

Then things became much clearer. After stabilisation here are the metrics :

- atmosphere/ wash temp : 18 C
- the 3 thermometers located at 30 cm from each other from the top of the
column indicate : 79C-82C-85C (adding one extra meter of column was really helpful as u can see). Alcohol is 90 % ABV
- product output rate : about 1,7 drops/sec
- wash input rate: a very thin flow
- wash enters the column at 60C
- wash leaves the setup at 55C (go figure!)
- boiler temp : 99 or 98,5C
- it uses only a little coal , the air input is set to a minimum and the last
meter of column is not isolated

- as usual, no cooling water used !!!!!!!!!

.....Still too slow for a 3 inches pipe (7,5 cm).....



TODO:
- add a tension regulator to the aquarium pump cable, putting my cut hands in the wash to reach and make subtle turns to the pump valve is not fun anymore
- silicone tubes get soft with the hot wash coming out of the reflux coil,
wash leaks from the joints (and of course the tubes give a silicone taste away when heated)
- broaden the hot wash inlet of to the column, and perhaps put it 15 cm higher to handle increased wash flow

[Nykter]

Nice.
Another question that crossed my brain is: What wash temperature do you have between the preheaters, i e after the condenser but before the drain heat exchanger? I am thinking of building a similar setup, and was thinking of skipping one of the heat exchangers or maybe use them in parallel instead.

[manu de hanoi]

after the preheater (reflux/condenser boka coil) i get aroud 35 degrees depending on the wash flow (could reach 50 degrees if ran slow but then after a while cooling aint enough and steam goes out on top). That's why I am making a bigger copper coil instead of the current 5 loop stainless coil with the target to increase that temperature.
Took me 3 tries but the 3rd hand coiled tube is nice.

[Nykter]

Cool.
I might try building a wash cooled continuous stripper since you say it words, but the math suggests that wash cooling isn´t enough if you use stronger wash than 5% and take out weaker low wines than 37%, but i might have miscalculated:

1 litre of 5% wash needs 242 kJ to heat from 20C to 80 C,
it gives 0,135 litres of 37%, which leaves 246 kJ when condensating.

I have used

2.43 kJ/(kg* K) for ethanol
4,1 kJ/(kg* K) for water

841 kJ/kg for ethanol
2246 kJ/kg for water

_not_ corrected for temperature variations in density.

Anyone to check my math?


This wouldnt leave any room for reflux...

[manu de hanoi]

1 litre of 5% wash needs 242 kJ to heat from 20C to 80 C,
it gives 0,135 litres of 37%, which leaves 246 kJ when condensating.

Anyone to check my math?

- congrats, it seems like you found a way to create energy ! Ok to make it easier to understand, mix 0,13 l of boiling water with 1 l of cool water, and measure the resulting temperature ! Then consider that the specific heat of ethanol is lower than the one of water, you will conclude that 0,13 L of 35% contains less heat than 0,13 L of water, at the same temperature.

- the last meter isnt insulated
- alcohol goes out at around 40 degrees C
- I strip at 90 % abv not 37 %

My math as follow:
1 volume of 5 % wash distilled at 90 %abv has too condensate 5,5% of it's volume, that's more than enough, and the remaining is used for reflux.
Because the reflux is limited and because you already loose the bottom 0.5 meter of column for mash heating, u need a tall column if u want high abv otherwise if u want 50 or 60% abv 1.3 meters is enough.

[Nykter]

Ok to make it easier to understand, mix 0,13 l of boiling water with 1 l of cool water, and measure the resulting temperature !

Not quite, you should mix the resulting steam of 0.13 l of boiled water, thats a completely different thing. To boil water into steam takes a lot more energy than just to heat it to the boiling point. The same counts for condensing it, but reverse.

But, I wont argue any more before i have tried it myself, you have reality on your side!

[manu de hanoi]

Damn Nikter your pushing me into corners but you're right...

first, for info :
5% by weigh ->6,3 by abv
37% by weight ->44 abv
90 by weight ->93 abv

1 kg of fresh wash (5% alcool by weight)heated from 20 C up to 70 or 80 C will absorb
(2,43 * 5/100 + 4,1*95/100)*50=200,8 kj (distilled a 90%)
(2,43 * 5/100 + 4,1*95/100)*60=241 kj (distilled a 37%)

notes :
-37% by weigh i.e. 44 abv condensate at 95 C therefore you can allow the refrigerant to heat up to 85 absorbing : 261 kj
-wash is not only alcohol & water it contains other stuff that may increase the heat per mass.
-I aint sure but the product should condense much hotter than i though, I should lenghten the output pipe perhaps

condensation of the alcool in 1 Kg of wash:
50 grams*alcool evap heat + water weight *water evap heat

5/100*841 +(5/90-5/100)*2246=54 Kj (if distilled at 90% weight)
5/100*841 +(5/37-5/100)*2246=233 Kj (if distilled at 37% weight)

The rest is distributed between cooling & reflux. Note: you dont need reflux if you distill at 37 % usually reflux is only used to help reach abvs higher than 85, 1,5m of packed column will be enough. But i wouldnt recommend the pain of building and using this apparatus just for a low strip like that. If just for strippig at 37% a pot still may be enough (no cooling water either).


Of course as you pointed out the reflux is much more limited than by using running water so you have to be careful not to overheat, but if u use gas that will be easy.

I did a run of 85 liters of rhum wash yesterday with a new good coil. the wash goes out of the coil btween 50 and 60C.
When more than 60 C you loose equilibrium and temp will shoot up till alcool vapour leave the top of the column without being condensed (i must increase wash flow then).
If less than 50 the temp will continue falling down to around 35 C and no product is condensed on top even though a lot of cooling is there. That's because the wash flow is too high and kills the rising vapors.

The temp out of the cooling head is important, thanks Nykter for making me pay attention to that. That temps is probably the earliest indicator of what's happening in the column. It's also interesting to note that this equilibrium is like a stone on top of a hill that can roll either side and will keep rolling once started. That's because of the double effect of wash : 1-cooling/reflux on top of the column
2- heat absorb on the bottom of the column,
there is a kind of ping pong/amplifying effect between the two that will make the temp spiral up or down once out of the hill top of 50/60C.


Cheers to that ![/u]

[manu de hanoi]

1868
"Be it known that I, D. SAVALLE, of Paris, France, have invented certain Improvements in Distilling Apparatus; and I do hereby declare the following to be a full, clear, and exact description of the same.
My invention consists of a certain apparatus, constructed as fully described hereafter, whereby crude alcools and other liquors may be quickly distilled and purified, with the use of but little if any water for condensing,
and with a better result than with the ordinary apparatus; the apparatus being simple, inexpensive, not liable to get out of order, and requiring but-little attention during its operation.
"

http://www.google.com/patents/pdf/_SAVA ... pOiUHNrQas" onclick="window.open(this.href);return false;" rel="nofollow

[Nykter]

Ok, I'll _have_ to give it a try!

Seems it is the fact that you distill to 90+ % that gives you the room for reflux. But in order to reach 90% you need reflux...
Maybe its this relation that gives you the instability.

I have noted while stripping in a continuous setup, that its easy to reach 70% ABV whithout reflux or rectifying part in the column. I have introduced cold, non preheated wash into AFC, maybe this have created a reflux in the column by the cooling effect of the AFC tube, or by the cool wash dripping down and cooling the filling.

[manu de hanoi]

Ok, I'll _have_ to give it a try!

Seems it is the fact that you distill to 90+ % that gives you the room for reflux. But in order to reach 90% you need reflux...
Maybe its this relation that gives you the instability.

I have noted while stripping in a continuous setup, that its easy to reach 70% ABV whithout reflux or rectifying part in the column. I have introduced cold, non preheated wash into AFC, maybe this have created a reflux in the column by the cooling effect of the AFC tube, or by the cool wash dripping down and cooling the filling.

yes, perhaps your right about the cause of the instability, on the other hand I was also getting around 70% with a 1,3 meter column and a small coil(see prev posts), and at 70% you already have much more space for reflux than at 44% abv. Increasing the height of the column by 1 meter as i did should help reach 85/90 % right ? The reflux only helps to reach 93%.

In short : use a tall column if u want high abv otherwise using a smaller one for low abv should be ok because you dont need reflux anyway.

[dog1976]

First I would like to say there has been a lot of thought and good discussion in this thread. I like to see that in this forum.

The problem I see with this design. A continuous still adding the wash into a packed column. The boiler will have pure water in it and will have to be run at 212 in order to get heat into the column. If 212 is not maintained the process will stop. With the packing in the lower end of the column always at 212, the only direction the tails can go is up.

All product collected from this still will have to be redistilled.

If I were making a continuous stripping still, I would use a 2-gallon boiler and feed the wash into the boiler. You could control the wash feed rate to maintain what boiler temperature you wanted. You could put a refluxing column on it and run a low reflux rate. Running to much reflux would be a waste of time, you would have to add to water to redistill.