Riku wrote:Automatic Reflux Control (ARC) The newest gizmo in amateur distillation is ARC, which is generally used to control a regular LM head. The idea in ARC (to my knowledge first introduced by “Farbror Plast”) is to use the temperature at the top third (or so) of packing to control the output/reflux ratio. The actual system is amazingly simple and can be attached to most LM heads. Here are few pictures of building an external pipe version.
PIC – EL-ARC – note, sensor pipe was lengthened later. Note, in the picture I have used silicone hoses. I recommend you use a 5-6mm copper pipe instead.
From LM heads product takeoff you have a copper pipe (8-10mm have been successfully tested) that is sealed at the bottom end, goes down 80% of packed column height (i.e. 80cm pipe for 1 meter column) and is soldered to the outside of column pipe to get good heat transfer. About 20 cm from the top of the pipe is the actual products take off. From the takeoff we have vapor lock and the pipe goes up again to 1-2cm below the liquid surface on LM reservoir and back to collection vessel. On the highest point of this product takeoff is a ventilation hole to prevent siphoning (required for at least smaller pipe diameters).
The principle of operation is that the liquid at the bottom of the 80cm pipe will boil if the temperature at the lower parts of the column where the pipe is attached is higher than the boiling point of liquid inside the sensor pipe (in practice 78.1-2C since it’s the first stuff that reaches the top of the column). When the liquid boils no liquid will enter the product takeoff tube, but all is returned to column via LM head due to pressure difference. When the temperature at the column (in the part where the pipe is soldered to) evens out the boiling stops and liquid will start to flow out of the product takeoff. This causes less reflux and the temperature in the lower part of the column starts to rise. When it’s high enough the output stops and the cycle starts again. In practice this works very well. It adjusts the reflux ratio automatically and stops the collection before any higher boiling point alcohols have a chance to get to the top of the packing.
This system gets rid of the tails but there is still the problem with heads. Several methods to get rid of the heads are researched, but I’ll present here the simplest one.
PIC-heads-collection-bottle
With this system the heads will go to the bottle, and when it’s full to the set limit (ventilation hose/pipe inside the bottle) product will start to flow to the collection vessel. The amount of heads can be adjusted by varying the length of the hose/pipe inside the bottle. The ventilation pipe/hose needs to go above the point where pipe to product vessel leaves, as liquid inside the ventilation pipe will rise to this level. The vaporlock in the tube prevents heads in the bottle from contaminating the product and also enables the adjustment by vent pipe. Otherwise the liquid in the bottle would rise up to the actual feed tube bottom. I recommend this type of system to be built from 8mm pipe as minimum diameter
PIC-Heads collection bottle in practice (built from leftover parts).
So, an example of easy to operate system at the time I write this (actually my current test rig):
1.6 meter column,
40mm wide
700W power (uninsulated fermentation bucket)
1-meter (or a bit more) convector pipe for reflux cooler
ARC to control product takeoff (internal sensor pipe is nice)
Bottle system for heads Fill the boiler, set up heads collection system, turn on the power and let it run. After 14 hours or so (could take a bit more) turn off the power, store heads, empty and clean the still, dilute the ethanol and enjoy the product.
A limitation of external sensor pipe ARC is that it won’t work well with high % (over 20%) mashes. To counter this problem an internal sensor pipe can be used. After some trial and error I constructed following simple to build modular system that has worked very well.
To start with I have a 40cm piece (for 1 meter column, 60cm for 160cm column) of 8mm copper pipe. It’s pressed shut at the lower end, this allows slow dripping of contents meaning that heads won’t stay at bottom of the pipe. At the top I have a double funnel (or funnel with drip ring) where reflux from condenser drips to (you may need a small guide or collar to direct drips to funnel). The ethanol that’s not removed via ARC will overflow and drip from the bottom of the funnel/ drip ring to the packing around the sensor pipe. ~6cm below the funnel top we have a product takeoff from sensor pipe. This is a 6mm pipe that goes through the capillary joint and has a vaporlock and ventilation hole arrangement outside the column, just like with external sensor pipe version. Through the capillary joint you can solder a piece of 8mm pipe. 8mm pipe can be used as a capillary joint for 6mm pipes (seal with PTFE tape) meaning that you can make modular system as shown in picture where the external parts can be removed during disassembly. Another good thing about such setup is that you can use this to adjust the height between LM reservoir and highest point of product takeoff (i.e. actual take off rate). By turning the entire external vaporlock system to an angle in relation to column centerline you can lower the highest point of takeoff tubing and increase the takeoff rate. This is very useful in fine-tuning the system performance. Once it is set you can mark the angle and duplicate the setup for consequent runs. After building this just wrap some mesh/scrubbers around the sensor pipe and insert it into the column. PTFE tape is again used to seal the joint and column is filled from below for the rest of the packing. In my tests I was able to halve the amount of heads by using this version when compared to external pipe version. When tested with 160cm column and 60cm sensor pipe it produced extremely pure ethanol and ceased output before any tails did appear. With 100cm / 40cm combo the ethanol was very good but there was a very faint sweet taste from burned sugar.
PIC – drawing of internal pipe ARC
PIC – internal pipe ARC in practice
PIC – Fine-tuning ARC output
Now the latest innovation for removing heads (pioneered by “Farbror Plast”) is to use separate heads column for that. In practice this means that we have another column that’s closed at the bottom and the bottom of that column goes down to mash in boiler. This gives enough heating for the column to let it operate normally. The product from ARC column goes to top of packing of this heads column. At the top of the column we have condenser as usual, and below the condenser we have a reservoir of 100-150ml where heads will be collected. The pure ethanol is taken out from the bottom of the column. Now the biggest problem with this type of approach is that at the start the column is fed with lots of heads and at the end there’s mostly pure ethanol. This means that we need to give the column some time to stabilize before we start taking product out. Two methods for this have been identified so far:
1. We let the ethanol pool at the bottom of the column and when adequate level is reached it starts to flow out. Finding a proper level requires experimenting but this has been tested to work
2. We collect all the reflux below the packing and feed it back to column via needle valve. When there’s enough reflux flowing the needle valve will restrict the flow and some pure ethanol will start to come out.
These methods are still experimented with and I’ll publish the results when I’ve found a method I’m happy with.
I've been working with SolidEdge the last little bit and it has some neat features that dare I say, exceed SolidWorks.
