3"continuous stripper build

[pipes+hose]

So I've decided to try a continuous stripper. I figured I'd make a build thread and also see if anyone points out any glaring flaws or better ways to do things.

My design goal is to strip a 200L barrel of sugar wash in 1 day (about 10 or 12 hours max), so 20L per hour or better. I also want to do it without getting too much expensive, difficult to manufacture stuff, or significant modifications to my existing still/column. So far I've only ordered the stainless fitting to see if i can get it welded to the keg, i should be able to try that next week.

I currently have a 58L keg boiler with 2 x 1350 W elements and 1 wrap of bubble sheet insulation around it and on top. 59 x 3 inch column with 50 inches packed SPP and 3 wraps of bubble sheet all the way up. Homemade 2 to 3 inch reducer that is about 2.75 inches tall to attach the column to the keg using tri clamps. Boka head, 3/8 inch "double helix" condenser (17 turns, 10 inches long, made from 16+ feet of tube). Recirculated water with a radiator with fans. I have less than 0.5 inch clearance to the ceiling. As an aside i plan up upgrade the rad and add an extra burner to bring me close to 4100 watts in the near future (i hope I'll get around to that, my spirit runs could really use it).

The waste water (ww) will exit the keg from a hole in the side, a couple inches above the top of the elements, go through a U bend/trap, and go through a heat exchanger to preheat the incoming wash. Theoretically the ww should have a little more energy than the incoming wash needs to preheat, but i know the transfer won't be close to perfect so i am going to build it, measure how much more preheat energy i need then do something about getting more wash preheating from the top of the column. The ww preheater will be the second stage (due to the higher available temperature) and the top of the column preheater will be the first stage. The ww preheater will be a very long single wound coil (1/4 inch tube) in a 1.5 inch pipe as a shell. I think I'll run the ww through the shell and the wash through the inside coiled tube. I'll likely put a 3/4 inch sealed pipe down the middle to stop the water in the shell from just going down the middle and not engaging with the coil.

The first stage preheater may be a small coil in the boka section sitting just under the boka plates, because i don't have any height between the main product condenser (pc) and the top of the boka. Maybe 12 inches of 1/4 inch coil that intrudes down into the column, then back up to exit through the boka section. All of the heat transfer here will be stuck as reflux, so i don't want too much, but i do want some in the hopes i can run with my needle valve fully open and not have to regulate output speed vs additional reflux.

I originally planned to bring the preheated wash in through a hole in the boka section so i can avoid drilling a hole in my column and needing to plug it in spirit runs. Then run the wash injection down a tube in the middle of the column to the correct injection height. But now I'm thinking of bringing it in through the 2 to 3 inch coupler between the column and the keg. Not sure yet. I think with either config i will try injecting it with around 25 inches of packing below it. I can change the injection height by cutting the tube or connecting more.



Problems/things that could be better:

1) the boiler is too big, I'll have to put 20+L to get it started, that's more preheating and start time than ideal. But a new small keg with burners and cords would cost around 200 dollars.

2) this all assumes i can weld a stainless coupler to the wall of my key without melting a hole in it. This is yet to be seen but I've ordered a few parts to try.

3) the wash injection going down from the top of the column might lead to the wash getting cooled down by the vapors at that level being only around 78c. I guess it would be better to run up from the bottom. Maybe i could put the injection hole in my 2 to 3 inch adapter and run a pipe up. The 92 degree (best case) wash injection pipe might cause some condensation of the rising water vapour... but I'll have to hope it's not significant. On the plus side this would also help bring the injected wash up to temp before it's free to mix. I'll have to think about all that.

4) i don't have a power controller, so i have to hope 2700w or 4000w works, or i have to make a controller.

I'll get some more pictures once i have anything to show for it.

[Bolverk]

2700w heating 20l to boiling is going to take about 45 min, not ideal in a continuous.

2700w "should" be able to strip 24.9l an hour using the previously discussed rate of 408w pg/ph, but you need to be running pretty efficiently.

The temp of your incoming beer should be as close to boiling (if not actually boiling) before it enters the column (78c or better). If not this is going to be a massive inefficiency, at 92c you should be great but with your current HX plans I don't don't see how you'll get it. You'll need a lot more HX capacity than you currently have. If your system is tuned right, you shouldn't need any external cooling water, the beer should be the cooling media and youre capturing the unwanted heat to preheat the incoming beer.

I can't tell based of the description if you're going to be reboiling the effluent (what you call the WW) or adding new water to the boiler and taking off the effluent before it goes to your boiler. But the reboiler method is the most finicky of the two designs, and your incoming wash will absolutely crash the boiler IF you don't have your feed up to temp before it enters the column. Also reboilers are VERY prone to foaming

This guy's method is far better than the reboiler.

You can see he's setup has the effluent directed off before it hits the boiler and hes using the effluent to heat to incoming water to replenish water taken off in steam.

Your SPP unless properly sized is probably going to choke the column. Consider using marbles in your stripping section, they still have a lot of surface area but allow more bidirectional flow of the beer and vapor.

Additionally, you shouldn't need an element controller. You run the full 2700w and feed your beer in until you're hitting the numbers you want. Ie less than 1% alcohol in the effluent, and your low wines are coming off at about 45-55%. If you have too much alcohol in the effluent your running to fast. If you have too high abv low wines youre running too slow.

[pipes+hose]

I can't tell based of the description if you're going to be reboiling the effluent (what you call the WW) or adding new water to the boiler and taking off the effluent before it goes to your boiler. But the reboiler method is the most finicky of the two designs, and your incoming wash will absolutely crash the boiler IF you don't have your feed up to temp before it enters the column. Also reboilers are VERY prone to foaming

Your SPP unless properly sized is probably going to choke the column. Consider using marbles in your stripping section, they still have a lot of surface area but allow more bidirectional flow of the beer and vapor.

I'm planning to reboil the effluent. That's a good thing for me to keep in mind: that inserting wash that needs too much energy to get up to temperature will crash the column. That guys set up is way more than i can do (space/cost/need) but it's got some good lessons in it for me. I'm hoping my sugar wash will not foam too much, that's another "wait and see how it goes" thing. Thanks for sharing it.

If my SPP isn't performing well I'll consider marbles. My SPP is 7mm od, 4 or 5 mm id and 7 to 14mm long (hand cut). I'm not really sure what level of vapour up and liquid down it can handle, I'll have to watch for pukes as i push it to see if there's a limit. I expect it would do very badly with much sediment or solids.

[Bolverk]

I'd get rid of that LM top and go with more of a pot still design with shotgun condensors as beer heaters, you'll gain a lot more heating capacity and if you find you need more heating/cooling you can just add in another.

Here's a pretty great thread too, lots to be learned here
https://www.stilldragon.org/discussion/ ... er-idea/p1

[Yummyrum]

I think Bolverk has hit all the relevant points .
My own experience with a reboiler agrees . The feed rate is super critical . Feed a nats dick too fast and the column temp collapses and then the boil Slows and stops .

Feed too slow and the column gets too hot . With insufficient flow through the condenser , you get vapour bypass . Then the wash entering the column starts flash boiling so vapour , not liquid hits the top of the column packing . With no liquid hitting the packing , nothing goes down .

Feed rate is a critical balancing act .

From my experience , a PC on its own ( In your case you are using a Reflux coil as a PC) comes close but nit quite close enough to being able to pre-heat the wash but not quite able to safely fully condense the stripped product .
I fought it but in the end had to realise that an auxillary PC needs to be used .
Now that was easy for me because I had the PC ( 1st heat exchanger) connected to a second (Auxillary ) PC which was water cooled …… and that was after re-directing the pre-heated wash from the PC heat exchanger back through the waste effluent heat exchanger to further boost it .

That I see as a challenge in your setup
. If the PC ( in your case , the LM reflux coil) , is unable to fully condense the vapour coming up the column , then vapour will exit the top with no easy way to add an Auxillary PC .

It’s all coming down to that balancing act of will the feed rate flow through the condenser be able to fully condense the vapour at the top without flooding the column and causing it and the reboiler to crash . ?

This is where an -Auxiillary PC gives you some wiggle room .


This was my shits and giggles
viewtopic.php?p=7586167#p7586167

[pipes+hose]

So after hours of thinking and math, and fixing the math, and fixing the math again, and then doing it differently, and then making a post, deleting my post because I'd done some math wrong, and then fixing that... I've come to the conclusion that a simple counter flow heat exchanger can't pull enough heat from the effluent to get the input beer anywhere near a high enough input temp. Even 30+ feet of tubing isn't going to harvest enough energy from the effluent.

So a significant heat exchanger that condenses product is required. I think I'll end up needing to cut some of my column, move it above the boka, and put a condenser coil in it. Then use my existing PC to knock down anything that gets past the preheater product condenser coil.

Stay tuned to find out what conventionally agreed upon theory i take hours of effort to accept next!

[Bolverk]

I feel you man, I've started and stopped over a dozen times because halfway through the build I knew it wasn't going to work.

I've studied every variation of alcohol and petrol column I can find for the last 2 years and I finally feel I have a design that will work

[shadylane]

If my SPP isn't performing well I'll consider marbles. My SPP is 7mm od, 4 or 5 mm id and 7 to 14mm long (hand cut). I'm not really sure what level of vapour up and liquid down it can handle, I'll have to watch for pukes as i push it to see if there's a limit. I expect it would do very badly with much sediment or solids.

Due to sediment and solids, I'm thinking there's better options than SPP for a continuous stripper.
Also keep in mind solids also can cause problem on the bottom of the stripper.

[drmiller100]

I've done this.

I described it at length in the alcohol as fuel forum on this site.
It is is the same work to run 10 percent beer to . 1 percent as it is to take 10 percent up to 95 percent.
Insert the beer mid column. 36 inches down. 36 inches up.

Use the incoming beer to knock down the top of the reflux column then a reverse flow exchanger on the overflow of the boiler to inject boiling beer mid column. I had a 1 gallon boiler.

As a hint for 2 liters per hour of 95 percent 1.5 inch
upper column and 2 inch lowe column.b 2000 watts
Use whatever packing you want but for continuous 12 mm
marbles or so works.

[drmiller100]

For 3 inch upper you need 4 inch lower and 4kw watts. That is 25.galloms or so of beer per hour so 2.to 2.5 gallons of 95 percent per hour.

[pipes+hose]

Today I drilled the waste water/effluent exit hole in my keg and welded the drain on. It looks terrible but it holds water. The down side of that is now I might want to do more welding on the keg, maybe a 3 inch ferrule on top!

Unfortunately the pump i got has too high a flow rate. I wanted 300 to 500 ml/min. It claimed 1.2L/min max and i hoped to run it on lower voltage to get a lower flow but in testing it's 1700 ml/min at 12 volts and 1200ml/min at 9 volts (with out any back pressure). Lower voltage than 9 seems to barely run the diaphragm so i don't think that's an option. I have a tee junction and old needle valve so I'll try that to provide a return path for excess flow, not ready to get the correct flow with that but I'll try it before i but a new pump. If i can't get that working I'll have to look at buying a peristatic. I originally only found that up to 100ml but I've since found 400, 500, 600, 700 and realized that replacement tubing isn't too hard to get. But they still cost more.

I guess the next thing to do is buy some copper and make some heat exchangers. Once those turn out i can cut my column, order new triclamps and maybe a ferrule. .i have some spare 3 inch copper so i can make a short section to take the place of the part I'll need to cut it, and with random packing it's not a big deal.

Thanks for everyone's input by the way. Doug, i hadn't previously seen your topic on the fuel section but i had seen that you've posted a fair bit about continuous and I'd seen your numbers. Do you think it's a somewhat linear relation between column area and power? Do you think 2700 watts, 3 inch, 20-40 L/hour is reasonable? Secondly you mention using 2 column widths, is there any problem with using 1 width the whole way up other than 'wasting' such a large pipe on the upper half?

Shady: SPP definately seems like it would trap solids and silt and plug up, but it's what i have on hand so i try it and see. I'll watch it fairly close and have my temperature probe alarms set up. My elements are a few inches from the bottom of the keg and I'm running SSS so there isn't much solid in that to worry about in the keg.

[drmiller100]

You can underpower a column but it will probably work. 2700 is not enough for a traditional reflux 3 inch column but 2000 is about max for a 3 inch beer stripper.

Don't forget you will lose some pressure as you go clear to the top of the column with your beer to knock down the vapors in the condenser then back down to the bottom for the boiler exchanger then back to the middle to inject into the middle.

I used 8mm tubing for all that.

[greggn]

Today I drilled the waste water/effluent exit hole in my keg and welded the drain on.

Surely, that's soldered and not welded.

[Salt Must Flow]

Today I drilled the waste water/effluent exit hole in my keg and welded the drain on.

Surely, that's soldered and not welded.

Looks welded to me. He said he welded it too. He said, "It looks terrible but it holds water".

[pipes+hose]

Today I drilled the waste water/effluent exit hole in my keg and welded the drain on.

Surely, that's soldered and not welded.

That's the nicest welding I've done all year, and don't call me Surely!

Jokes aside, it's also the only welding I've done all year. I welded one thing (with some practice) last year (which looked better) and nothing in the 15 years before that.

[Bolverk]

304 stainless ain't easy to weld especially considering the beer residue in the inside, different thicknesses of the keg and fitting...

Dude it may be kinda ugly but it works. First time I tried welding on a keg I did nothing but blow holes through it.

[pipes+hose]

I've done some coil bending for the heat exchangers but got blisters on my hands from it so that has slowed me down from completing them.

I cut the 2 inch port off the top of my keg and welded a 3 inch ferrule on. It's slightly off center/tilted but less than the floor is going to sit on, i can compensate with the wood Shim i already use under it.

The real problem is that when i looked in to the keg with a flashlight after i saw that there is rusting around the exit port i welded on previously. The heat must have killed the corrosion resistance and then i got it wet right away to test if it was water tight. I assume it's similarly rusty in the crevices i can't see too. I also expect the same heat issue happened when i did the 3 inch ferrule (although i didn't get that wet yet). I might try putting CLR on it for a bit, then water rinse, then citric acid, then rinse again. I don't know if this is the sort if thing that keeps rusting all the way through or not, is assume all the way. Right now my plan is to keep moving forward till it develops a leak.

Edit: if i get to the point of doing s spirit run or other situation where is have high proof alcohol in the boiler I'll have to reevaluate the risk of leaking flammable liquid.

[pipes+hose]

I have the effluent heat exchangers mostly done. That is a 1.5 inch pipe about 44 inches long that the 1/4 inch coil will go into. The 3/4 inch pipe will be in the centre of the coils to prevent flow through the middle that misses the coils.

The beer will run through the 1/4 inch spiral tubes from left to right while the effluent runs from right to left then out a hose to the drain. The beer will First run through the smaller left coil then up to the heat exchanging product condensing (not pictured) then back down to the right coil, then into the column.

I'll make copper end plates and solder the coils to them (1 coil to each end) and hold them on with 1.5 tri clamps. The end plates will also have ports for the effluent in/out, likely just 1/4 copper that i hose clamp plastic lines to.

[drmiller100]

Way over thinking it.

Do a directional heat exchanger. Set it vertical.
Wash comes out the boiler and into the top of the exchanger. Don't forget it needs an air lock so it doesn't siphon out of the boiler

Warm beer from the top condenser goes in the bottom of one of your coils. Can mes out the top and into the column. .
Do it right and you get boiling fresh beer going into your column

[pipes+hose]

Way over thinking it.

Ya, that's what i do!

Do a directional heat exchanger. Set it vertical.
Wash comes out the boiler and into the top of the exchanger. Don't forget it needs an air lock so it doesn't siphon out of the boiler

Warm beer from the top condenser goes in the bottom of one of your coils. Can mes out the top and into the column. .
Do it right and you get boiling fresh beer going into your column

I do plan an airlock too, i think there may be some playing with things to get the flow thing correctly.


Small update
I've been fairly busy and not had much time to work on things, and the little time I've had has gone into my stupid decision to weld a 3 inch ferrule onto my keg. I hope to have more time for working on stuff in mid November, so maybe progress in late November or early December? then I'm busy for the second half of December, so I expect this to stretch into jan or later.


In the mean time I've still had time that i can research and think, and I've come to somewhat of an epiphany about continuous stills. It's the sort of thing that sounds obvious when you say it but i haven't seen anyone say it. Disclaimer: I haven't physically tested this yet, so don't take the rest of this post as fact, it could be incoherent ramblings.

Continuous stills for 50% abv and continuous reflux stills for 90% are completely different... sort of.

So far I've been reading and researching without distinction and I've come to realize that while there are some similarities, the differences mean I'm not planning correctly. To condense the product (low wines) coming out of a continuous reflux at 90-95 % only gives you about 25-30% of the energy needed to preheat the incoming beer. The rest of the beer heat comes from the relatively stable heat of the effluent. To condense the product (low wines) coming out of a continuous still at around 50% ABV can give you nearly all the energy needed (in fact a little more than needed) to preheat your incoming beer. A lot of people making continuous stills do 50% stills and get a lot of their preheat from the condensate, and I think that's where a lot of their problems come from. I believe such a strong link between the low wines amount/abv and the beer preheater is what causes the feedback loop that pushes things out of control to be way too hot or to crash too cold. The feedback loop looks like this: With a little less heat in the column you get less heat to the condensate (and less volume), which gives less heat to the beer, which needs more heating from the boiler vapors, which leads to less heat in the column, which leads to less heat to the condensate... and so on resulting in a cold collapse of the column. And a similar thing can happen with too much heat running away.

sort of: so they're different in that the 90% reflux version doesn't have much heat to reclaim from the distillate and MUST get a lot of preheat from other sources and should inherently be more stable (and it seems less commonly talked about here on HD). Where as the 50% version has so much more easily reclaimed heat in the condensate that it tempts people to take too much and fall into the feedback loop trap... I think.

The big change for me is that I had previously though that you needed to allow some vapors to pass the HXPC and get to the aux PC (on the theory that that way small changes to the energy of the product vapors would thus not change the beer preheat temp, just the amount that gets to the aux PC) but I think now that the actual way to think about it is that the beer preheat should not be too dependent (and tied to) the product output (low wines) so as not to induce an out of control feedback loop. Which i think should be achievable by having a limited hxpc so you're always pushing heat to the aux pc, or by having a large effluent hx so that the hxpc is only providing a smaller portion.


The change for me is that I had been planning to limit my preheat capture from the condensate to avoid the feedback loop and use an aux condenser. But since I'm only able to get less than 30% of my preheat needs from condensing the product (low wines) I think its no problem that I take it all. I had always planned to run with a heat exchanging PC and an aux PC. But with this new information I think I can run with JUST one heat exchanging PC. Its free to try it out since i already have a pc i can try it with and costs time/effort to make a new heat exchanger PC. So I'm going to try the fabled (and often problematic) no spare condenser water.

[Black Bull]

Whilst I'm no expert you pose some interesting points.
Keen to see how you go mate.

[drmiller100]

Way over thinking it.

Ya, that's what i do!

Do a directional heat exchanger. Set it vertical.
Wash comes out the boiler and into the top of the exchanger. Don't forget it needs an air lock so it doesn't siphon out of the boiler

Warm beer from the top condenser goes in the bottom of one of your coils. Can mes out the top and into the column. .
Do it right and you get boiling fresh beer going into your column

I do plan an airlock too, i think there may be some playing with things to get the flow thing correctly.


Small update
I've been fairly busy and not had much time to work on things, and the little time I've had has gone into my stupid decision to weld a 3 inch ferrule onto my keg. I hope to have more time for working on stuff in mid November, so maybe progress in late November or early December? then I'm busy for the second half of December, so I expect this to stretch into jan or later.


In the mean time I've still had time that i can research and think, and I've come to somewhat of an epiphany about continuous stills. It's the sort of thing that sounds obvious when you say it but i haven't seen anyone say it. Disclaimer: I haven't physically tested this yet, so don't take the rest of this post as fact, it could be incoherent ramblings.

Continuous stills for 50% abv and continuous reflux stills for 90% are completely different... sort of.

So far I've been reading and researching without distinction and I've come to realize that while there are some similarities, the differences mean I'm not planning correctly. To condense the product (low wines) coming out of a continuous reflux at 90-95 % only gives you about 25-30% of the energy needed to preheat the incoming beer. The rest of the beer heat comes from the relatively stable heat of the effluent. To condense the product (low wines) coming out of a continuous still at around 50% ABV can give you nearly all the energy needed (in fact a little more than needed) to preheat your incoming beer. A lot of people making continuous stills do 50% stills and get a lot of their preheat from the condensate, and I think that's where a lot of their problems come from. I believe such a strong link between the low wines amount/abv and the beer preheater is what causes the feedback loop that pushes things out of control to be way too hot or to crash too cold. The feedback loop looks like this: With a little less heat in the column you get less heat to the condensate (and less volume), which gives less heat to the beer, which needs more heating from the boiler vapors, which leads to less heat in the column, which leads to less heat to the condensate... and so on resulting in a cold collapse of the column. And a similar thing can happen with too much heat running away.

sort of: so they're different in that the 90% reflux version doesn't have much heat to reclaim from the distillate and MUST get a lot of preheat from other sources and should inherently be more stable (and it seems less commonly talked about here on HD). Where as the 50% version has so much more easily reclaimed heat in the condensate that it tempts people to take too much and fall into the feedback loop trap... I think.

The big change for me is that I had previously though that you needed to allow some vapors to pass the HXPC and get to the aux PC (on the theory that that way small changes to the energy of the product vapors would thus not change the beer preheat temp, just the amount that gets to the aux PC) but I think now that the actual way to think about it is that the beer preheat should not be too dependent (and tied to) the product output (low wines) so as not to induce an out of control feedback loop. Which i think should be achievable by having a limited hxpc so you're always pushing heat to the aux pc, or by having a large effluent hx so that the hxpc is only providing a smaller portion.


The change for me is that I had been planning to limit my preheat capture from the condensate to avoid the feedback loop and use an aux condenser. But since I'm only able to get less than 30% of my preheat needs from condensing the product (low wines) I think its no problem that I take it all. I had always planned to run with a heat exchanging PC and an aux PC. But with this new information I think I can run with JUST one heat exchanging PC. Its free to try it out since i already have a pc i can try it with and costs time/effort to make a new heat exchanger PC. So I'm going to try the fabled (and often problematic) no spare condenser water.

Yeah. The one condenser raises your beer temp from 80 degrees to about 130 depending on insulation of the column.
The energy at the top of the column is the same whether it is at 59 percent or 95.

[pipes+hose]

I've made some progress. Today i did the first test and cleaning. It produces alcohol. It needs insulation on the various lines, i need to insulate the column (it was off for soldering). And i need to play with it but on the first go it was producing 70%. Beer injection temp was 70c or higher at around 350 or 400 ml/min (initial beer temp was 66f). 1420 watts to the element.

More details to come later once i figure them out and insulate things