Another continuous stripping still

[Maritimer]

With 80 liters of Birdwatcher's wash to strip, I began to think of a continuous stripping still. This is a popular topic, but I wasn't able to find one like this.

Originally, I thought of a series of four teakettles, each one feeding its overflow into the next:

As the wash is fed into the first, some of the ethanol boils off and as more comes in, the overflow goes to the next kettle. The overflow will be depleted compared to the wash coming in. Then it boils off some ethanol, gets depleted some more, and its overflow goes to the next kettle. The waste coming out will have little ethanol left in it.

Then, to make it simpler, I put the four kettles into a stock pot and took off the vapour from one outlet.

Then I got rid of the kettles and came up with this:

The wash comes in via the lower gold pipe and exits via the upper green pipe. The cylindrical red things in each quadrant are 1000 watt teakettle heating elements. I can use my digitally-controlled-power power supply to tweak the power dissipated in each element.

The only difficulty in building this would be attaching the quadrant.

Would it be OK to use food grade silicone? It could easily withstand the temperature.

[ipee7ABV]

http://homedistiller.org/forum/viewtopi ... =35&t=5090
#8

[Durace11]

What you are doing, whatever the design, is basically a 4 thumper pot still. It could be set up as a continuous system but I think on a hobby scale this type of system is more trouble than it's worth to setup & run. If you want to run 50 gallons at a time just get a SS 55 gallon drum and have at it. Running 50 gallons through a 5 gallon pot still on a continuous basis is not the correct answer IMO.

[Maritimer]

Hi Durace11,

Could you please explain how this is a 4-thumper pot still? I'm not being facecious, I don't understand what you mean.

M

[Durace11]

Ok, just wrote out like 2 paragraphs and it disappeared on me...

I mis-used the term thumper because I didn't have any other name for your unique design pictured above. I believe what you are trying to do is not going to work exactly like you think it will. The boiling liquid will be turblent so it could stay in the vessel for 10 minutes or 10 seconds before moving to the next vessel. Likely your results will be marginal.

I believe you would be much better off using the same parts to create a 4 thumper system. Also, if you go forward with this I would suggest welding everything.

[Maritimer]

The first 20-liter bucket of Birdwatcher's wash was a learning run. The second bucket is coming along very easily. Adjusting the flow rate is critical, and the 3/8 in. gate valve I'm using is too crude. The correct setting is somewhere between 1/2 and 5/8 of the first revolution.

I put thermometers in each of the quadrants, and added a heat exchanger so that the exiting waste heats the incoming wash. It isn't long enough, though, so its effect is wanting.

The criterion for setting the flow rate is to get the last quadrant to as high a temperature as possible. It's running at about 99.5*C, which means that it still contains a fraction of a percent of alcohol.

The liquid temperatures of the quadrants are:
95.2, 97.3, 99.0, 99.4
corresonding to %ABV of:
6.5%, 3.3%, 1.3%, 0.7%

The distillate is 35%.

Here is a picture of the setup:

[Prairiepiss]

How long did it take you to strip 20lt of wash?

[Maritimer]

Sorry, I didn't note the times. Maybe a hour?

I just noticed that each quadrant's %ABV is about half of the one before.

M

[Durace11]

How about using a plate chiller to counter flow the input and output past each other to preheat the input? The heat exchange on those is great. You could also build a counter flow chiller with 2-3 copper coils. 2 smaller tubes inside a larger third tube.

[Prairiepiss]

Well we would need to know time and ABV of collected low wines. Would include heatup times too. Amounts collected and amount/ABV of wash.

[Maritimer]

Yes, Durace11, that sounds like a great idea. I've got about 1 ft of 1/2" pipe inside a 1" pipe. Not adequate.

PP, more characterization is definitely needed. The collection rate is a little slower than runing the pot still with a 20 liter charge, but time is saved in initial heating--3.9 liters instead of 20 liters (and only the first time; each additional 20-liter batch doesn't need to be heated, the heat from the previous batch carries over to the next one), and the time taken to empty and recharge the still for additional batches is omitted. You don't have to deal with the boiling-temperature charge in the pot still that has to be removed.

M

[Maritimer]

So here are the stats. I did a Birdwatcher's wash, which again went over the intended 80 liters. I ended up with four 23 liter pails of 10%ABV beer = 9.2 liters of ethanol. I took two pails and did two batch stripping runs at 4000 watts, then I did a continuous stripping run of two pails, also at 4000 watts.

The batch strips:

First batch
Time 0: turn on power
Time 33 minutes: collection starts
Time 1 hr 31 min: end collection of first batch. Stopping criterion was vapour temperature = 99.7*C

Second batch
Time 0: turn on power
Time 36 minutes: collection starts
Time 1 hr 37 min: end collection of second batch. Stopping criterion was vapour temperature = 99.7*C


Total low wines: 12.2 liters @ 30%ABV = 3.66 liters of ethanol
Total time = 3 hrs 8 minutes

The continuous strip:
Time 0: turn on power
Time 8 minutes: boiling temperature reached. 93*C beer temperature
Time 3 hrs 19 minutes end collection.

Total low wines: 14.7 liters @ 30%ABV = 4.41 liters of ethanol

Total ethanol = 3.66 + 4.41 = 8.07 liters out of the theoretical 9.2 liters in the wash.

Discussion:

The continuous stripper took 3 hrs 19 minutes to do the job.

The batch stripping took 3 hrs 8 minutes of actual stripping time, but the spent wash had to be removed and new wash added. If the boiler were plumbed to make this easy, it would be a negligible chore, amounting to no more than a few minutes.

But in the same time, the continuous stripper gathered 2.5 liters more low wines, which, at 30%ABV, amounts to 0.75 liters of ethanol.

Controlling the flow rate is difficult with the gate valve. The flow rate is 45 liters/199 minutes = 0.226 liters/min = 7.64 fluid ounces/minute. In looking through the HD site, others with this problem have used (or wanted to use) peristaltic pumps. Expensive.

A way to control the flow rate needs to be found. (Maybe the microcontroller could do the trick--bursts of beer through an on-off electrically operated valve?)

M

[rad14701]

A way to control the flow rate needs to be found. (Maybe the microcontroller could do the trick--bursts of beer through an on-off electrically operated valve?)

How about the KISS concept of going with a needle valve...??? Or just about any small fluid regulatory valve for that matter...

[ozone39]

This is an interesting concept.......just wondering how a guy gets the head shots out (methanol) of something like this????

[Maritimer]

Rad,

A problem I had using the gate valve opened so slightly was that at one point it jammed up with some small particle in the beer. My original idea was to use a filter. Alternatively, to make it reliable, the flow path should be wide. Another problem with gravity feed is that the head (pressure) changes as the supply pail empties, changing the flow rate. (BTW, I am a devout disbeliever in the principle of KISS. KISSing is for virgins. Can you imagine the Kama Sutra written by a KISSer?)


Ozone,

This is a stripping still. The foreshots and heads are removed in the spirit run.

M

[1bottler]

Hi Peeps, I've been asked to build a continuous still for a winery. have there been any significant improvements / changes to the one on the parent site? I am going to use a peristaltic pump to regulate the feedstock, 2.5m column.10 ltr boiler. and what ever else it takes,

[Maritimer]

Hi 1bottler,

So now I find out about the continuous stripping still on the parent site! Wondering if the four-section boiler might do instead of trickeling the wash into the column. The vapour coming off is 30% alcohol and condenses crystal-clear, so if instead of a pot still head a reflux column were used, you could get the continuous action without any gunk getting into the packing. Just a thought.

M

[Scrap dogg]

I'm a little confused here. Why don't you just use a 80l boiler and turn your heat up to strip with? Just seems like you are chasing your tail a little. Don't take me as being condescending. I just don't see the advantage of this elaborate set up, which looks to be getting more complicated as you go. I can tell you that with a 22 gallon boiler, you can strip a 18 gallon wash down in less than three hours total if you are only trying to run off an average of about 40% abv.

[Maritimer]

Hi Scrap dogg,

This little stripping still is no more complicated than a reflux still, say. I'm an old retired guy who has been building and inventing stuff since childhood. Doing things the easy way is too boring!

Inventing is like hunting. The problem is the prey. The solution is the kill. But before you can kill the prey, you have to stalk it. I once stalked a problem for five years, and just like in hunting, suddenly the prey was in front of me. I drew my bow and took it down. It was thrilling; the thrill of the hunt. And like hunting, you may be looking for a deer, but by chance a bear is there.

That's what might be happening here. The conventional continuous still has a problem with the packing getting fouled. A solution to one probem could be a solution to another.

M

[Scrap dogg]

Hi Scrap dogg,

This little stripping still is no more complicated than a reflux still, say. I'm an old retired guy who has been building and inventing stuff since childhood. Doing things the easy way is too boring!

Inventing is like hunting. The problem is the prey. The solution is the kill. But before you can kill the prey, you have to stalk it. I once stalked a problem for five years, and just like in hunting, suddenly the prey was in front of me. I drew my bow and took it down. It was thrilling; the thrill of the hunt. And like hunting, you may be looking for a deer, but by chance a bear is there.

That's what might be happening here. The conventional continuous still has a problem with the packing getting fouled. A solution to one probem could be a solution to another.

M

Fair enough.

[1bottler]

I have thought about mounting my 3" boka [1.5m column] on top of my 4" pot [80cm column], spraying the stock feed through a nozzle at the junction of the pot & reflux.

[Lester]

I was studying your data Maritimer and it seems to reinforce a concept I have in my head: It takes a finite amount of heat & time (Energy) to boil off a finite amount of alcohol from the wash. There's no way around this, no matter what still is used, so there's no shortcut to making more alcohol using a given amount of time & heat. Maybe I'm wrong.

I would like to try and build a continuous still but have not so far convinced myself that there would actually be some merit in using one. The concept mentioned above keeps popping back up. Every time. Given the complexity of operating a continuous still, the batch method seems to be the safe (sane?) route to take.

[Cardinalbags]

Maritimer:

I like your veracity in trying new things. Besides, we might not have potato chips had it not been for the time someone was carrying their raw slices of scalloped potatoes past the fat fryer that day....

I get it.

PS: you must spend a lot of time in places like Jentronics....

[Maritimer]

Lester,

I agree. The wash needs to be vapourized. The vapour contains a known fraction of ethanol. But would a vacuum help? Or would the energy required still be the same?

Cardinal,

Never heard of Jentronics. I gave the website a look. Normally, I can get things through McMaster-Carr through the university I used to work for. But that often involves interrogoation because they don't ship to Canada, normally. Do you know of a supply company such as McMaster or Meridian in Canada. There is Granger, but their prices are inflated.

M

[Cardinalbags]

McMaster carr have locations in ontario or at least they used to when i was ordering craploads of stuff from there years ago as part of a previous job. Maybe they no longer have those locations?

There are a lot of hidden gems in the Burnside industrial park, just need to spend a bit of time researching.

Jentronics for relays/switching type gear, oscilloscopes and instrumentation for one. With your penchant for instrumentation, i would have thought it was a second home to you!!

[Lester]

Lester,

I agree. The wash needs to be vapourized. The vapour contains a known fraction of ethanol. But would a vacuum help? Or would the energy required still be the same?

Given the known losses in mechanical systems it would be a stretch to come up with a system that would even come close to something efficient. But I like your idea with the speaker. I have in fact toyed around with the concept because it is easy to implement, putting take-off ports where the nodes would normally be,resonant chambers and such, with no packing to speak of. But I'm not quite there yet.

[Lester]

The continuous stripper took 3 hrs 19 minutes to do the job.

Maritimer,

3hrs 19 minutes to strip 23 liters of wash @ 4kw input power. I wonder how much faster this could have been if you had pre-heated the wash with the vapor?

It takes me roughly the same amount of time (flame on to flame off) to do strip runs but I only use approximately 1.4kw (propane burner). I do not understand why this is the case with your setup. Seems a lot of heat is lost somewhere.

[Maritimer]

Hi Lester,

That 3 hours 19 minutes is for two batches of 23 liters. I used a 90 liter Birdwater's wash and divided it into two, which is four batches, to compare the continuous to the batch pot strippings. They turned out to be almost identical.

I had made a heat exchanger using the waste, and it didn't seem to do anything. It turned out that the outlet for the waste was too low, so the heat wasn't getting transferred. That got fixed for the next try, but I didn't record any numbers.

M

[Lester]

ok got it. Thanks!