My 22mm continuous “Koffie” still with a twist.

We don’t condone the use of Continuous Stripping stills as a method of running 24/7 as this is a commercial setup only .
Home distillers should never leave any still run unattended and Continuous strippers should not be operated for longer periods than a Batch stripping session would typically be run to minimise operator fatigue..

Moderator: Site Moderator

Post Reply
Lafras-H
Novice
Posts: 28
Joined: Mon Jun 13, 2022 3:11 am

My 22mm continuous “Koffie” still with a twist.

Post by Lafras-H »

A few months ago I posted in the Welcome Center about the intrinsic safety features of my 22mm continuous “Koffie” still. As I was preparing a draft post on how it worked and how much I thought it was such an ideal hobby size, liking it to the “Airstill of continuous distillers” when a hell of an idea struck me.

The Airstill is appealing to beginners because it looks safe, it looks and feels like something that is familiar in your kitchen. It does not look like a hill-billy contraption waiting to fall over or explode if you make a mistake. Unfortunately, most beginners that try the Airstill lose interest in the hobby after a few weeks partly because of all the effort it takes to make a little product in such a small still. I believe if there are familiar and convenient kitchen appliances that anyone can use, it will help to normalize home distilling in the eyes of the general public.

So I set my new mission - that home distilling should be just as safe and feel just as familiar and convenient as making bread in a bread maker (since both use yeast).
Regarding a comment on a Pico distiller thread by CatCrap: “It kind of makes a joke out of this craft. It grossly oversimplifies the art and craft of Distillation, and everything that goes along with it“. I do get that some may feel this way but I don’t see why an easy induction to the hobby would detract from the craft, after all, master bakers did not become redundant when bread maker appliances became popular, it just made making your own bread more popular, and likely inspired a new generation to get back into home baking and off shop bought confectioneries.

So next time some wise-ass criticizes our hobby with "Don't blow your house up" or "Don't poison yourself" you could say modern distilling is just safe as making bread. We should want everyone to know someone who distills. Hopefully, that will change the general public's perception and advance the legal status of the hobby.

Well, since that epiphany I have been working frantically to make it into a product and the result is amazing - I still cannot believe what I have made in such a short time.

I do not intend to it sell here on HD… experienced distillers are not my market. There is no way a convenience like this will be a substitution for the art and craft of an experienced distiller, like a master baker won’t use a bread maker I doubt an experienced distiller would use my appliance.

So, I am not going to give a brand or website or any sort of sales pitch – it is not live yet anyway – this if the first public information on this project. Rather my intent is to share with the community a little about what it does and how I dealt with various issues so that if some newbie later sees it in the market and asks a question about it here on HD, the members will know about it.
SidebySideOnBarrel.png
The design is a compromise between form and performance, it processes only about 10 litre of wash a day, and produces just under 1 litre at 90% which is more than a novice distiller for their own use should ever need, so it does not really matter that it runs very slowly...it will slowly work its way through any volume wash.

DiagramText.png
[a simplified diagram]
Internal.png
[Without covers and thermal insulation]

Without the powder-coated aluminum covers and thermal insulation, we can see most of what is going on.

The copper spiral I have dubbed a “trumpet”. Compared to “flutes”, this sort of looks like a trumpet. Only cold low ABV wash gets pumped via a silicone tube into the trumpet, the rest of the path is all copper with stainless steel 316L SPP. There are no plastics anywhere near the path.

It has to be periodically cleaned in place(CIP) using citric acid which is why I chose the packing to be SPP 1mm 316L Stainless steel wire(which is quite big for such a small still). I found the thin strands of copper mesh start to decompose and “turn into mooch” after a just few run/clean cycles but the 1mm thick SS lasts, showing no signs of wear even after dozens of cycles.

The boiler is 50cc, there is a fitting at the bottom to get a brush into the boiler just in case it got scorched. The boiler PTC heaters can deliver 160W, but Vodka is run at about 60W.

I cast molten aluminum in a rectangular block around the 22mm boiler tube to evenly distribute the heat around the tube and have some thermal mass to smooth out the PWM signal that drives the two 80 Watt PTC heaters.

It is a completely open system from the boiler to the vent, there are no valves that could block up the trumpet, and even if the vapor path was blocked the waste path would still be open. (during development and testing when I used higher power heaters and put too much energy into the boiler for all the pressure to leave via the column I got quite violent spurts out of the waste but at least it did not create a pressure hazard and in production units power is limited by the smaller PTC heaters and further by software). Worst case if the boiler control fails it will result in a software shutdown of the pump, a dry boiler, and PTC heaters that just sit there at 160’C.

The boiler waste is a P trap with a vent that runs all the way to the top and then joins with the open top of the column – this is mostly for aesthetic reasons. It could have been two separate vents. When you are done, shut it down, let it cool, then empty it by turning it upside down over a drain so all the liquid will run out of that one vent. Same for cleaning – you block the nozzles with little caps and fill via the vent with citric acid.

The top of the stripping section(22mm tube) has a kind of dephlegmator thermally connected to the back of the case and the case forms an effective radiator. Then the top of the rectifier section (15mm tube) also has a kind of dephlegmator thermally connected to the front of the case, which also acts as the condenser for the hearts weir. Then the heads have a condenser connected to the back of the case and a weir to collect the heads. Both the hearts and heads drain attach to a Nozzle Heatsink that also attaches to the front of the case. And a fan keeps the whole enclosure cool.

A strange feature is that both the dephlegmators have a heater at the interface to the case/heatsink, this means I can independently control the amount of effective reflux on the hearts and the heads independent of the temperature of the case. I control the amount of cooling by slightly heating the thin interface between the dephlegmator and the heat sink. This greatly simplifies the mechanical design. Originally I really really wanted it to be all passive cooling and during our winter the prototypes worked well, but as the days got warmer is was clear I had to add a fan. Having the heaters also means I can completely flatten the reflux gradient effectively running down to just about any ABV.

The tails all leave via the waste and the amount of tails are determined by how close to the pure water boiling point we let the boiler run. Even when I run as close as I can to 0% I do not get the wet dog / wet cardboard flavors of tails, maybe this is due to the short residency of the wash in the boiler, or maybe due to good tails compression.

On a continuous still, some Fusil oils will slowly accumulate below the hearts until there is so much that it will all come out with the hearts(some comes out anyway). I have a simple solution. The software will periodically flush the column and boiler, wasting a little bit of wash.. but it is a really simple solution.

The user interface is a web interface, you connect to it’s Wi-Fi (an ESP32), and you can set it up to connect to your Wi-Fi network –all sorts of remote monitoring/notifications can be done although not implemented yet.

The user selects a “recipe” like Water, Vodka, Rum, Whiskey, Brandy, etc that sets the main process parameters, then there are “heads”, ”Tails” and “Fusil” slider adjustments which will slightly tweak the parameters within range so you can adjust the hearts according to taste, or judge the proportion of hearts/heads that you produce. From making the adjustment it takes about 10 minutes for the new taste to come through. The center setting should work well enough for a novice not to have to worry about adjusting them initially. You can also set it so it will continue with the same recipe after a power outage.

Safety features from my original koffie still post mostly still applies (amended):
Complies with the new safety standards (IEC 62368-1)
All-electric - No open flames.
All copper still, the cold wash is all that comes in contact with the siphon-feed tube.
Totally open with no valves.

The boiler is lined with fireproof thermal insulation
The boiler over temperature is limited by 160'C PTC Heaters.
The boiler over temperature is limited by Software to not exceed 105'C.
The boiler power is limited to 160W, even if the boiler reaches 160’C the appliance enclosure will not rise above 55'C in a 25'C environment.
The boiler is small, and with less than 50ml (2 shot glasses) of liquid at boiling point, so scorching hazards are limited.
The boiler mass is small, limiting the available energy for an unexpected event.
The boiler waste outlet is small to limit sudden spurts but has a large and tall overflow section to contain any sudden spurts.

Software limits a run to 12 hours.
12 Hours gives under 500mL product at 90%+ ABV, so I have to check and clear the container twice daily to carry on. Even if I used a smaller container and it did overflow, it would be limited to a small volume.
The software will shut down the still if the vent temperature indicates the fan has failed.
Possible vapor leaks due to solder joint leaks would be so small that they would never reach LEL.
Possible vapor leaks can be easily checked.

The peristaltic pump tube is expected to last about 25 x 200 liter drums.

That is about it.


All in all, I love the vodka it makes and so do my family and friends, and that's Marvellous... Well, That's What I think.

I will be happy to answer any questions…?

Thanks
Lafras
You do not have the required permissions to view the files attached to this post.

This topic has 19 more replies

You must be a registered member and logged in to view the replies in this topic.


Register Login
 
Post Reply