Yummy's continuous stripper

[Yummyrum]

OK here's my Continuous stripper build .

It is a work in progress and has gone through many incarnations to get to where it is so far . It is not complete but it is a working model now .
First off , my intentions of building a continuous stripper is not to have it run continuously but rather to embrace some of the advantages these stills have while enjoying my hobby at a new level .

So if you think that I have a still you can switch on and leave to its own devices to run 24/7 , look elsewhere ...that's not what I have made.

But if you want a stripping still that you can turn on and in 10-15 minutes be producing stripped wash with almost no cooling water required , then read on .In fact that's exactly what I have been doing all weekend . A couple of hours here and there , .... 200 liters of wash stripped without need to empty boilers ...and bugger all cooling water needed .

Most Continuous stripper threads start with vague sketches and diagrams and most seem to fizzle out with little coming to completion . I took the other approach and just jumped in and build stuff to see what worked ....and didn't ....and went from there . In saying that , I give thanks to all those that have paved the way in what is a minority part of our hobby .


Here's a few pics of the Chaotic setup this weekend ....but after that I will go back over my story of how I have ended up with what is here today .
Here's the setup .
There is a boiler on a few milk creates to add some height. The boiler has a total of 3300w elements in it ( 2400w + 900w) just because that was the maximum I have ...(got a 1350w but can't find it in my messy shed )
Above that is 1 Meter of 4" packed tightly with Staino scrubbers . That then comes down to the shot gun condenser off my Flute which doubles as the product condenser and the primary wash pre-heater . The 2" dropper down to the Shotty are pipes from my Thumper setup .( That's the only reason they are 2"....just happened to be lying around )

Following the Shotty is my old 3/4" over 1/2" x 36" liebig which functions as an Auxiliary Product condenser .It is just a backup . But I always have the pump running in the 2000 liter water tank while its operating ...just in case .
In normal operation , no cooling water is needed but as these strippers can run out of control , its a good backup ...also , as I found this weekend , running 32 degC wash through it requires a bit more help .

The boiler has an outlet that gravity feeds via a U trap . In my first version , the waste simply exited from here . I have subsequently found the need to use a bottoms pre-heater . In this version you can see I used the Shotty off my old original 2" VM still ( which incidentally is now also used on my VM flute ) .....so as far as new things goes , this whole setup is basically just made out of bits and pieces I have lying around. There most certainly is a lot of miss-matching but timer will tweak all that .

As far as wash goes . I ferment in a 200liter drum and fill a smaller 30 liter fermenter up from it .This gives me a working amount ...lets face it , I only want to run this still for a few hours at a time .
From this I have a Magnetic drive pump that feeds wash via a needle valve to the bottom of the primary pre-heater ( Main product condenser ) I measure the temp of the wash as it exits this pre-heater . This is the main source of heat exchange . As the incoming wash condenses the vapour from the top of the still , it in turn heats up . From here , it goes to the bottoms pre-heater . This is a heat exchanger that takes heat from the waste wash leaving the boiler and adds extra heat to the wash before it is feed to the top of the stripping column . (It actually has a more significant importance that I will talk about later). I have another Thermometer at the feed point in the top of the column so I can monitor how hot the wash feed is .
Thermometer number 3 is in the boiler head space . It shows if wash is entering the boiler still containing alcohol . Normally it sits around 99.5ish -100 degC but can drop down to 98 ish when things are pushed too hard .


I have to admit to being really crappy at taking records but I will try and give as accurate info as I can .

So this wash I run was a sugar wash 30kG in 180 liters ...so round 9.8% AVB ...call it 10%AVB
3.3kW power( 240 v rated elements ) .....it may have been closer to 3.5kW as the power was up to 248v during peak sunshine ( neighbors have grid feed huge solar array)
I monitor Temps at the feed point at the top of the column , the output from the primary pre-heater and in the boiler .

So here are some typical readings today:
Now it is summer here and my wash is around 32degC .

Wash temp leaving the Primary pre-heater 78.7degC

Wash temp at feed point . 81.1deg C

Boiler Temp

and the all important take off rate was 4.6 liters /h

@ 53ish % ABV ( temp not corrected )

And while these results are not to be sneezed at , I have to admit that whacking a keg full of wash on a roaring gas burner with a simple potty head on it gave me superior stripping rates ....but that's simply because I have way more gas power available than electrical .
However... saying that , we have been in drought for many months now with total fire ban on what seems like every other day so to be able to run on mediocre Lecky with hardly any water usage is priceless to me ATM

[acfixer69]

Congratulations yummy, looks very well thought out. I need to study it a bit more to wrap my head around it. I know you have been working on this for awhile now. Nice to see it produce the juice

[Saltbush Bill]

Good to see it working Yummy ....Knew you would get it going eventually

[Yummyrum]

Hey thanks Fellas , know you been watching the torturous evolution on CCSC , Salty sorry I didn't give you the heads up that I was going to run this , this weekend . I know you wanted to come see it working but my wash got flies in it and something nasty on top ...so had to do what I had to do .

Going back to the start. For me , there are two main advantages of a continuous stripper .
1) No water is needed to cool the product ....the wash cools the product .
2)Very quick heat up time before production .....heat-up time is about 5-10 minutes ...enough to boil the 5-10 liters in the re-boiler and heat the packed column .
They are a very energy efficient still

So way before any continuous still
A few years back I was curious to see if a wash could fully condense a wash of the same size . IE , if you stripped 50 liters of wash , could you run another 50 liters through the condenser at the right rate that would fully condense the first wash but both (the stripping run and the coolant ) would finish together . I figured that if you have a 10% AVB wash , you would need to run the coolant at 5 times the rate of production so that both the boiler and the coolant finished the same time .
Turned out that it was indeed possible but it was tricky as hell to get the feed rate right and the flow rate was also critical ...too slow and uncondensed vapour was exited the condenser and the coolant wash was almost boiling , too fast and the product was fully condensed but the wash tank run out too early .

I then had another go at it

This all convinced me that I needed to move to the next step and make a real stripping column .
I didn't want to sink a whole lot of time and money into this as I have pretty much neither ...but as I had been curious about these things for ever , I had to "scratch the itch "
So I decided to just make a feed plate to stick on my 4" packed column and use everything else I had . As far as a first time experiment , I didn't want to make a special boiler with overflow Blaa blaa blaa .....for this first test I would just use my existing electric boiler with enough water to cover the element and run it until the boiler was full .

Now the decision to use my 4" packed column and not my 4 plate flute was based on others that had tried their 4 plate flutes and said the efficiency was poor and the end AVB was really low . Also reading that most commercial stripping columns have at least 20 plates backed up that theory that using 4 plates was just pissing in the wind . I figured that if 20 plates is good enough to strip all the water out and 20 plates is good enough to make a neutral then a 1 meter packed column was going to be pretty much on the money .......so I used my existing Scoria rock packed column .

So I made a feed plate . I claim total artistic licence here. Most just seem to inject wash straight into the top of the column . But I was using packing and I wanted it to me evenly distributed as possible . Having observed how pre-heated wash contains a lot of gas both as CO2 and as partially vaporized Liquid , I also figured that were the wash is injected , there should be a plate that the incoming wash can bubble through as a final exchange stage before either passing out as vapour or re-fluxing back down the column to be further stripped .

The final day came and the lash up was done .

It was a simple affair ...as basic as basic can be .
Wash was pumped up through the product condenser and was heated on the way and injected in the top of the column where it separated . The vapour come out via the pre-heater heating the incoming wash as it did and condensed exiting via the Auxiliary condenser . The wash in the column stripped with the water ending up in the boiler and the stripped alcohol making its way up the column.

All good in theory

But holly shit what a mine field

So You start the boiler running . In the mean time turn on the pump and prime the condenser and load the top plate . Then turn off the pump and wait until the top plate starts to boil. Soon after , product starts to appear so I slowly increase wash feed . But it doesn't take long to realize that wash feed rate is Stupidly ultra-critical . To slow and the Auxiliary condenser is doing all the work and the feed temp is boiling . Too fast and the feed temp starts to dive ....and this is the big issue .
Once the feed temp starts to go down when it hits the column , it basically over powers the upcoming vapour and all the packing cools down , nothing vaporizes , nothing is coming through the product condenser to heat incoming wash and the whole column comes to a crash .
Ain't nothing you can do but kill the feed and wait for it all to heat up again and start all over .

So basically the whole still is by nature unstable . Its either wanting to go up in temp ...or wanting to go down in temp ..... there is no stable point ....well you might be lucky and find it for a few seconds . It's a balance . Heat input verses wash feed rate . Then something comes along like the wind or a piece of shit gets stuck in the needle valve ...don't laugh ...it happens regularly ...a filter in the feed line is essential . Then there is the fluctuations in power supply . Even adding more wash to the bucket will throw it off ...wash temp change
But to make things even more difficult , there was a huge lag time from when you make an adjustment to when things would change the other way ....in the order of 5-10 minuates ...so dialing this bastard in was horrific

I can see why many have thrown their arms up with these stills . They are not set and forget . They are totally hands on .A continuous still needs more monitoring than a normal still

After a few runs with this setup I was about to give it up as a bad joke .

To be continued :

[richard1]

Very interesting. I have a continuous still (100mm diameter packed column …. helical SS coil) that's designed and part built and no further.

Are you able to supply process flow drawings for better understanding of your set up.

[pope]

Looks awesome! I’ve had a similar build in my head but have been trying to keep it there for now, I was thinking of using a water chamber with a float valve for the element to blow steam up the column but this is a little bit simpler. I’m going to have to reread this again to pick up the nuances but very impressive!!

[Yummyrum]

Very interesting. I have a continuous still (100mm diameter packed column …. helical SS coil) that's designed and part built and no further.

Are you able to supply process flow drawings for better understanding of your set up.

Hi Richard , I'm no engineer so I may not be able to give you the diagrams you want but I will try and do sketches of what I am doing through out this topic .

Good Luck with your still . I hope you can get back onto it again .

Looks awesome! I’ve had a similar build in my head but have been trying to keep it there for now, I was thinking of using a water chamber with a float valve for the element to blow steam up the column but this is a little bit simpler. I’m going to have to reread this again to pick up the nuances but very impressive!!

Thanks pope . LOL this bastards been in my head for a long time too ...I totally get what you mean .
With no fixed plans and vague ideas that are floating around , its hard to know where to start .
Steam drive is one of my next missions on this but for now I'm sticking with simple until I nail a few things like Rum wash foam up

If nothing else , I hope you get some ideas and encouragement to get your ideas into the workshop

[Yummyrum]

So before I continue , here is a diagram of the basic test setup I did as requested by Richard ........ It's not the stripper I posted in the initial thread .....I'll get to that later .
The wash is pumped from the bucket via the feed rate valve to the bottom of the Pre-heater . It exits the top of the Pre-heater and and enters the feed plate . Stripped alcohol flows through the Auxiliary Condenser where it is further condensed if necessary them is collected . The water is stripped from the wash in the column and ends up in the boiler .
This is Not a Continuous still as it can only run until the boiler is fill.
Note that the boiler is initially filled with water to cover the element .

Adding an Auto Drain ...and a brief encounter with Rum wash

This was the critical point that distinguishes this as a Continuous stripper . By adding a drain to the boiler , wash can be fed in for ever and waste wash will continue to exit out the drain .
There is pressure in the boiler ...albeit...not very great caused by the vapour rising through the packing . So to add an auto drain , you need to fit a "U" tube so that the "weight" of the liquid in it will be enough to not let the boiler pressure escape out the drain .
The position I fitted the outlet hole on the boiler was not a scientific or strategically placed point ........It just happened to be where I had drilled a hole for a thermometer for an old experiment and I didn't want to make more holes ...and it was roughly in the right place .
Again I mention this to show that there can be a lot of leeway in how one makes there still .It doesn't always come down to exact science or measurement .

This worked quite well....so well that I got all excited and decided to run a Rum wash ...after all that is the reason I want this thing .

Well I quickly discovered that Rum washes foam ...alot ... ....LOL I should have seen that one coming .
There was foaming in the feed plate .

There was puking out the condenser

There was foaming all the way down to the boiler . And suddenly there was Molasses wash spraying up through the anti-siphon vent on my auto drain .

So I quickly grabbed a piece of pipe , wrapped it in teflon and jammed it in the anti-siphon hole .The drain hose was removed and the puke all saved for the spirit run .

By now the boiler was really full of Rum wash and it was puking out the auto drain too ...time to shut down and cry

So that ended my play .Deflated and since then I haven't tried Rum wash again . Got some Silicone anti-foam and waiting on micro- dosing pump .

[Saltbush Bill]

I do enjoy reading your adventures Yummy .....you post them up warts and all...... lol

[cayars]

Bad-Ass Yummy.
Love threads like this!

[Yummyrum]

Pleased you fellas are happy
If I didn't show you the stuff-ups as well as the successes , you might get the miss-guided idea that I knew what I was doing

Anyway ,I thought I would crap on a bit more about this Feed rate issue . I mentioned that if you feed too high , then the column spirals into a cold crash which basically means you have to stop feed and wait until the column heats up again and then start slowly feeding in the wash again until its running .

The other issue is if your wash feed rate is too low .
In this case when you are running wash through the pre-heater at a lower rate , the wash temp leaving the Pre-heater starts to rise , it will get hotter and hotter until it is basically boiling and so instead of a nice constant flow with pockets of vapour , its violently surging into the feed plate . This seems to cause a few issues . As most of it is vapour , instead of doing a phase change on the feed plate it basically all goes up and out and there is not as much liquid to fall back through the packing and be fully stripped . This causes the whole column to heat up more . As there is less feed and therefore less alcohol , the output starts to drop in AVB as there is also steam coming up from the boiler now that is not stripping so much alcohol which also means that the vapour temp entering the pre-heater is higher which escalates the problem of wash boiling .
So instead of a cold crash scenario with too much feed , we now have a run-away over heating column problem .

This is why it is so important to be monitoring the Pre-heater output temp to maintain a correct feedrate .

And herein is the problem with a single Pre-heater .
I concluded that there was not enough heat transfer available from condensing the wash to heat it enough that the still would run stable . I needed extra heat in the wash before it was feed to the column to maintain the correct temps everywhere.

Looking at a few existing strippers , some had harnessed the heat from the waste wash leaving the still ...and why not
The answer is to use a Waste wash heat exchanger or Bottoms pre-heater......or ( Bottoms HX ) .Sorry if I switch these around ...they all refer to the same thing .
Now there seemed to be two diagrams bouncing around the forums . One feed into the Vapour heat exchanger and then into the waste exchanger , the other went into the bottom and then into the vapour exchanger . The former makes more sense to me as it means no cooling water is required in the Auxiliary product condenser. .
There is another method one guy used where he split the wash sending it to both then recombined it before the feed pipe .

I had no idea how much heat was available for transfer from the waste . All I knew was it was pretty much at 100 degC so it was hot .
AS I felt I was close to the desired feed temp I wanted ,I didn't want to have it get way to hot so I remembered years ago I built a contraption to make a Deflagmator for my Flute to have adjustable water level in it ....never used it ...but I figured it was perfect for the job here . I thought I could adjust the level of the waste wash in contact with the shotty tubes and therefore the amount of heat transferred to the wash ...Bloody brilliant

I shagged around and hacked my perfectly good Flute deflag to add the attachment .
If you are wondering why there is so much water on the concrete its because I couldn't get the end plates to seal on the Deflag ( Bottoms HX) and everytime I turned on the pump it leaked and sprayed wash everywhere ....kinda pleased it wasn't a Rum wash

Bottoms heat exchanger set up No 1

I'm collecting the waste in jars ....later I put a hose on it and drained it too the lawn.

You can see the level in the sight glass tube .....it was running full.

This is the diagram of the chaos.

You may have noticed I dropped the Vapour heat exchanger down lower so the interconnecting hoses were closer .
It worked .....well kinda .
Turns out the deflag was a terrible heat exchanger and I had to have the adjuster thingy fully up ...so It was a pointless exercise hacking it
However...there was enough extra heat to get me over the line and for the first time I was able to maintain some sort of equalibrium in the still .

The Bottoms pre-heater turns out to be a rather quirky thing . It seems to average the feed rate temp entering the column in such a way that the upward or downward spiraling effects I experienced with a single pre-heater were significantly reduced. To elaborate , if the temp from the Vapour pre-heater started to drop , the Bottoms pre-heater maintained the temp ....sure it dropped a bit but not so quickly or dramatically that the column would crash.
Likewise when the temp from the vapour pre-heater started to rise , the bottoms pre-heater maintained the temp to the column , rising slowly but not as high .In-fact there were times where the Bottoms pre-heater output temp was lower than what was being fed to it from the vapour pre-heater .

I'm still not 100% sure what the mechanism behind this is. I suspect its a combination of the heat held in the mass of the copper as well as the volume of wash in it ...regardless ...it works

So basically the bottoms pre-heater only seemed to add about 5-10 deg to the wash exiting the vapour HX but it was enough to mean that the column was more stable requiring minimal feed rate tweaks ...but best of all , I could run at a feed rate that ment that the product was fully condensed without need for extra cooling water

Bottoms heat exchanger set up No 2

I was curious about the alternate setup I'd seen . It makes less sense to me . You are running wash through the Bottoms HX which is at 100degC on one side then feeding that through the Vapour HX .
I realized that this setup was going to mean Auxiliary cooling water was required which is one thing I didn't want to have but I sucked it up to try it .

I had to swap over the thermometer to the Bottoms HX but otherwise it was a simple job .

Well this arrangement surprisingly worked well too .

I put a valve in series with my needle valve . This was a good move as I was able to use that to slowly get the wash through the condensers until it had heated up and the needle valve was already set to a good place after last run.

It quickly stabilized and run very well . The wash temp was 25degC , the Bottom pre-heater raised the temp to close to 50DegC .The discharge from the Bottom pre-heater ( waste wash) was around 60degC so this was good because I was discharging cooler waste .

The Vapour pre-heater raised the temp to around 80degC and it happily sat there for a long time . playing with the feed rate didn't seem to be nearly as critical . I think this is a good configuration except you Have to need cooling water

I tried getting greedy with feed rate to see how hard I could push it and ...Bamm ... column crashed . So its not perfect but its a good setup and I think practical .

[Corsaire]

Sooo what would happen if you use wash in the pc?

So wash to pc, from pc to bx, bx to top preheater?

I'm probably missing something very obvious

[LWTCS]

Yummy, need to pump your beer into the low wines HX first, then back to the bottoms HX. Coolest beer needed for best distillate temps.

Edit: Oh I see you have that config further back in the thread.

[LWTCS]

Do you have a way to control pump speed based on a temperature?
Don't focus on beer injection temps. But rather on low wines vapor temps. 93c thru 97c works really well. Pick a temp and see how well she behaves.

[LWTCS]

On your foam up issues, what's worse? Foam up in your reboiler or foam up up top at the injection sight?

Make sure your wash is plenty degassed.

[Yummyrum]

Corsaire , I’m guessing when you say PC you are referring to the Auxilary one I have .
Thats actually a damn good question and one I have pondered on . Why I haven’t tried it is because I want that PC to be a Safety PC independant of whats going on with wash feed .... it runs on recycled water .
In normal operation when the still is functioning correctly , the Vapour HX can fully condense the product so the Auxilary is doing nothing .It’s purely a back-up in the event that the wash feed rate drops causing incomplete condensing of the product in the Vapour HX

If I run wash through it , it would defeat its purpose of being there .

[Yummyrum]

Do you have a way to control pump speed based on a temperature?
Don't focus on beer injection temps. But rather on low wines vapor temps. 93c thru 97c works really well. Pick a temp and see how well she behaves.

Not yet Larry .I’m running this by hand . No controller . I want to get my head around whats doing what why first . I’m using a needle valve and magnetic drive pump ATM but I intend to upgrade to a speed controlled Parasteltic pump soon .

Hadn’t concidered low wines vapour temp . .... will have a look at that
I have been concentrating on the low wines HX output temp however as that seems to be the critical one from my limited experience . It seems to be the fastest changing ( most responsive ) temp and indicative of whats about to happen .
I agree that with the Bottoms HX adding additional heat that monitoring the feed temp is not so important as it seems to hold at a fairly constant temp ( within the bounds of the still operating well) so long as the feed rate is about right .

[Yummyrum]

On your foam up issues, what's worse? Foam up in your reboiler or foam up up top at the injection sight?

Make sure your wash is plenty degassed.

The wash was pretty well degassed Larry . I only tried it once . Need to revisit Molasses again . Regarding what was worse ?
I don’t know .I had puke out the PC and eventually puke out the reboiler .LOL , they both sucked .

[LWTCS]

Yeah, when using the reboiler as the steam source the foam up is a PITA. Also, need more height / distance from the injection elevation to the low wines vapor pipe. This helps to prevent the foam from dragging over into the low wines HX.

[LWTCS]

A bigger low wines HX would minimize the risk of wash boil up in the HX.

[LWTCS]

I'm probably going to abandon using the reboiler as the steam source on my little model because it's a bit too one dimensional.

[LWTCS]

Also Yummy, if your beer has enough dwell time in the column you don't need "20 plates" as such.

10 plates can get the job done. The big unit I have in Illinois only allows .5% alcohol to go down the drain with 10 plates.
So it's not so much about distillation cycles as it is more about giving the alcohol enough time to flash.

[pope]

Was wondering when you’d chime in! Is foaming why your system uses steam injected above the waste exit rather than a boiler at the bottom of the column?

Is there a preference for degassing between agitation, vacuum, and time?

Do you guys have any feedback yet on flavor shift (if any) between identical recipes stripped plated/pot/continuous?

[LWTCS]

Was wondering when you’d chime in! Is foaming why your system uses steam injected above the waste exit rather than a boiler at the bottom of the column?

Is there a preference for degassing between agitation, vacuum, and time?

Do you guys have any feedback yet on flavor shift (if any) between identical recipes stripped plated/pot/continuous?

Using steam injection on the big system because the heat transfer is really efficient.
I was hoping that using a reboiler and spent beer as the steam source would be a good efficient way to preserve as much heat as possible. And theoretically does. But the dern foam up is far more problematic then. The reboiler is fine on BS sugar wash. But as noted the molasses causes problems.

One of the guys does use a vaccum pump set up in a beer feed tube / can to degas. Evidently works well. Anything to degas is going to help. But the molasses appears to be inherently worse for still behavior. Rye can also be a nuisance.

The reboiler will / does work but will compromise the run speed capability because of fighting with foam up control. So far my little one with the reboiler runs at about 20 liters per hour and that's just too slow. It really needs to run at about 40 LPH so that it can process a 1200 tote in a single work day. I'm going to pull the reboiler off and install a small steam generator.

[pope]

Not that this is new info but I've had the worst foam with malt, even 'whole cane syrup' foams like a molasses batch. I've never tried or wanted to try adding fats to the still, I'd rather just have better practices but I can see foaming being a serious issue in continuous. Five Star makes a non-silcone defoamer for distilling but I can't find anything less than a gallon which isn't a huge surprise, do you have any anecdotes about its performance?

I still have my mind on a steam generator build with a float valve for auto-refill, more for batch stripping on the grain, but I'm holding off because I figure once I've built that, tinkering with a continuous isn't far off.

[LWTCS]

Sorry I don't have any first hand knowledge about how well that Five Star product works.

[pope]

No sweat. I figured if anyone knew, or knew anyone who'd used it, it would be you. It's Defoamer 880 and while it's marketed as defoamer for a brewing blow-off, someone here (so this is through-the-grapevine knowledge) said that they recommended it over the 105 defoamer for distilling purposes.

Could adding a small expansion bulb just above the wash injection site help?

[LWTCS]

"Could adding a small expansion bulb just above the wash injection site help?"

I think so. How big? Dunno.

[pope]

Sounds like my internal monologue when facing most distilling problems - 'could x do y? Possibly. What ratio/qty/scale/rate/etc? No clue... let's just give something a shot and see what happens.' Might take me a year but I'll probably get there and have a short list of things that don't work.

[LWTCS]

Sounds like my internal monologue when facing most distilling problems - 'could x do y? Possibly. What ratio/qty/scale/rate/etc? No clue... let's just give something a shot and see what happens.' Might take me a year but I'll probably get there and have a short list of things that don't work.

Haha yeah. That's why taking a bullet for the team is so valuable for the rest of the community.