Need feedback on design diagram - my 1st

[lampshade]

See http://homedistiller.org/forum/viewtopic.php?f=15&t=25240&start=45

There Myles mentioned a novel LM product condenser.

The most elegant solution that I have seen is also very simple. The product output tube runs down the inside of the coolant supply hose to the coil - a long, thin, liebig that uses a coolant jacket that would be there anyway. Nice.

[InTheGarage]

Almost forgot... To run this as a pot still, would I just remove the packing and open the LM valve with the VM valve closed?

I wouldn't remove the packing. If you switch off your reflux condenser cooling then the packing will heat up very fast indeed, and behave as if it is not there at all. HOWEVER, you still have the issue of the air vent in the condenser to deal with. If you don't seal it all your vapour will get out there. If you do seal it you have to remember to un-seal it next time you run the condenser.

Far better to build a simple pot still head to replace the LM / VM head.

As for the LM output tube - if the liquid is not hot at that point then you are OVER COOLING the reflux. This is not efficient, but for some it is not a big problem.

Some folks have reported that they get poor flow out of the 1/4" output tube. This could be down to the shape of their output tube, length, bend radius etc etc - but there is some evidence that the liquid "sticks" in the tube. No air can get into the output tube so no liquid can flow out. You can add in a bleed vent between the valve and the condenser but a slight increase in diameter also seems to fix the problem. It might work OK on your rig with 1/4" tube but for the sake of a few inches of 3/8" tube I know what I would do.

Myles, regarding the 1st link you sent of Samohon's Liquid/Vapour Management Still http://homedistiller.org/forum/viewtopic.php?f=1&t=19711#p6886035, couldn't it also be used as a pot still if the reflux condenser was left turned on, the packing or column removed and the take-off left wide open? I'll build the pot still head to replace the LM / VM head if need be, but was hoping to get a best-of-both worlds build if possible. I'll PM Sam if you suggest but would like your feedback.

I'll definitely take yours and Seaguy's advice for a 3/8" output. Makes sense from your explanations.

Thanks again!

[lampshade]

Some folks have reported that they get poor flow out of the 1/4" output tube. This could be down to the shape of their output tube, length, bend radius etc etc - but there is some evidence that the liquid "sticks" in the tube. No air can get into the output tube so no liquid can flow out. You can add in a bleed vent between the valve and the condenser but a slight increase in diameter also seems to fix the problem. It might work OK on your rig with 1/4" tube but for the sake of a few inches of 3/8" tube I know what I would do.

Are those tube dimensions in ID or OD?
.

[myles]

@ lampshade, id or od does not make a huge difference in this case. The problem is simmilar to the concept of a "wine thief" used for testing barrels. A dip tube you put in the barrel and close the open end with a finger - lets you lift out a sample in the tube - and it stays there until you remove your finger.

With the LM output (and I repeat it only seems to affect some folks) if the output tube ends up COMPLETELY full of liquid the output stops. Most of the time the output tube is not actualy completely full, there is some air getting into the condenser from the open end of the output tube. Because no air can get in from the condenser end, your product output can surge and come in pulses, this is why some folks added the breather tube to let air get into the output tube.

I think it is probably easier to just increase the size of the tube, as it is unlikely that you will be able to completely fill it with liquid.

@ InTheGarage, you can do that it just seems counter productive to me. Ideally in pot still mode you want a closed system to build a slight overpressure in the boiler that helps to push all the vapour into the product condenser. If you use your reflux head it has a hole in it - the air vent - so for safety reasons you MUST have your condenser running to prevent vapour escaping to atmosphere. This is a really bad thing and potentially very dangerous. In this situation some of the vapour you produced goes straight back to your boiler to be re-evapourated - just a waste of energy.

You can take the packing in and out but it is a nuisance. Building a modular column so you can remove a packed section and insert an empty one is an option. I have tried to do what you are thinking of before but for different reasons. Early on I had a basic Thors Hammer configuration at the top of the column, but connected the air vent to its own smaller liebig condenser. The idea was to use the reflux condenser in CM mode to split the heads off through the air vent so they never went through the product condenser. When running normally the fact that the air vent was open to atmosphere via a liebig made no difference.

You could do something like this in strip mode but you are adding in a vapour path restriction which is undesireable.

If it was me trying to build a VM / LM head that included an integrated strip option, I would leave the packed column in place and loop round your air vent so that it was plumbed into the VM branch between the valve and the VM product condenser.

In LM reflux mode your air vent still has an open path via the VM product condenser, in VM reflux mode your reflux condenser can still balance vapour pressures, even if the vapour going into the condenser chamber via the air vent is not pure air.
In strip mode with the reflux condenser turned off, the vapour path will split with most going through the VM branch and some going via the reflux condenser chamber - but both vapour paths end up in your VM product condenser. This will give you a faster potential strip rate than trying to take off the product through the smaller LM output tube. Disregard the packed column completely, as without forced reflux the packing has minimal effect on the vapour path once it has heated up.

The LM side of the head is designed to take off product slowly and is not really configured for strip rate production - but the VM side is far more suitable for that application.

[InTheGarage]

@ InTheGarage, you can do that it just seems counter productive to me. Ideally in pot still mode you want a closed system to build a slight overpressure in the boiler that helps to push all the vapour into the product condenser. If you use your reflux head it has a hole in it - the air vent - so for safety reasons you MUST have your condenser running to prevent vapour escaping to atmosphere. This is a really bad thing and potentially very dangerous. In this situation some of the vapour you produced goes straight back to your boiler to be re-evapourated - just a waste of energy.

You can take the packing in and out but it is a nuisance. Building a modular column so you can remove a packed section and insert an empty one is an option. I have tried to do what you are thinking of before but for different reasons. Early on I had a basic Thors Hammer configuration at the top of the column, but connected the air vent to its own smaller liebig condenser. The idea was to use the reflux condenser in CM mode to split the heads off through the air vent so they never went through the product condenser. When running normally the fact that the air vent was open to atmosphere via a liebig made no difference.

You could do something like this in strip mode but you are adding in a vapour path restriction which is undesireable.

If it was me trying to build a VM / LM head that included an integrated strip option, I would leave the packed column in place and loop round your air vent so that it was plumbed into the VM branch between the valve and the VM product condenser.

In LM reflux mode your air vent still has an open path via the VM product condenser, in VM reflux mode your reflux condenser can still balance vapour pressures, even if the vapour going into the condenser chamber via the air vent is not pure air.
In strip mode with the reflux condenser turned off, the vapour path will split with most going through the VM branch and some going via the reflux condenser chamber - but both vapour paths end up in your VM product condenser. This will give you a faster potential strip rate than trying to take off the product through the smaller LM output tube. Disregard the packed column completely, as without forced reflux the packing has minimal effect on the vapour path once it has heated up.

The LM side of the head is designed to take off product slowly and is not really configured for strip rate production - but the VM side is far more suitable for that application.

Myles,

Thanks for taking the time to provide such a thorough explanation. As I'm new to this, it took me a few reads for it to make sense - actually, might just be that I'm getting old and slower - but it does make sense to me.

I'm planning on using a 3"x60" column that will be attached to the keg and the reflux condenser with triclamps. Following your suggestion of plumbing the reflux condensers' air vent to the VM arm, would replacing the 60" column with a shorter one for stripping runs provide any benefit? As an alternative I'm also considering building a separate pot still head (as you've previously suggested) and making the product (Liebig) condenser removable so that it could be used in pot still mode. The more I read, the bigger the project becomes, lol. But I do want to weigh all my options before doing anything. I'm interested in getting feedback... would there be an obvious advantage to a separate pot still head or could similar results be expected using the reflux head (condenser off) piping the rising vapour back to the VM arm?

Thanks again,
ITG

[myles]

Well let's see, this started out as a 3" x 60" VM / LM with an integrated strip option. I wish I used mine enough to justify a 3" column but I don't, but it would be nice.

1. You will need to increase your VM port size. You don't have a packed heads column to aid you, so your vapour split will be dependent on the ratio of cross sectional areas between the entry into the reflux chamber and into the VM branch. Most folks will realise by now that I am not a fan of the inline design, the offset condenser version puts a physical obstruction in the vapour path to aid the split.

If you are going to 3" you clearly wish a greater take off rate - so need to look at the geometry of the VM spliting chamber, otherwise you will not achieve the potential of the 3" column.

2. With a 3" column and 3kW you should be able to get 2-2.5 litres / hour from a 40% charge but your run time will not be super fast. So how quickly do you need to strip The compromise option offered earlier might be enough for some folks. If you WANT to strip faster then a dedicated strip option would suit you better.

Have you started building yet If not, consider designing your reflux condenser so that you can remove it and insert it before your liebig to function as a precondenser in the strip mode. An empty column (riser) will help with speed and the length is determined by where you wish the output of the condenser to be. The dedicated pot still head is a better option but do you actually need it.

[InTheGarage]

Myles,

It seems as though every time you provide answers to my questions, it causes me to come up with new questions which is good, it's how I learn. I also find you discussing concepts that I'm unfamiliar with, or might have read something about in another post earlier in my process of trying to learn about this. In any case, it causes me to look up a lot of what you mention to make sure I understand what you're saying. Bottom line is you're pushing me forward in what I've learned and I thank you for that.

I haven't started building yet...still trying to "feel out" the design that will suit best. I'm involved in this with a group of friends, acting as Project Manager (my regular job should be this fun, lol), so I think the 3" column is justified since there will be many mouths to feed!

Since several people are contributing to the build, I want to make sure this is as flexible as can be so that there will be little compromise to quality whether making a batch of whiskey or vodka. So, I'm still open to ideas and would appreciate if anyone can point to any plans that have been posted that might fit the bill. None of my friends are reading this forum - they're looking to me to figure things out to make the design decisions which is pretty overwhelming with so many ideas available. We're looking at this as a hobby that will grow. I see them becoming more involved in future research after we've all actually experienced how this thing will work, which will give a better understanding of what improvements we might like to add later on.

1. You will need to increase your VM port size. You don't have a packed heads column to aid you, so your vapour split will be dependent on the ratio of cross sectional areas between the entry into the reflux chamber and into the VM branch. Most folks will realise by now that I am not a fan of the inline design, the offset condenser version puts a physical obstruction in the vapour path to aid the split.

If you are going to 3" you clearly wish a greater take off rate - so need to look at the geometry of the VM spliting chamber, otherwise you will not achieve the potential of the 3" column.

I'm not sure what you mean by "You don't have a packed heads column to aid you"? You mentioned the importance of the geometry of the VM splitting chamber. I had found an earlier post where you'd given your thoughts on that as well as a diagram that I found interesting: http://homedistiller.org/forum/viewtopic.php?f=15&t=25720#p6943335 (basically more detail to your "B" diagram on the 1st page of this thread). If there's a definite advantage to building this way we can do it.

2. With a 3" column and 3kW you should be able to get 2-2.5 litres / hour from a 40% charge but your run time will not be super fast.

We'll be using either NG or propane so not yet sure how that will equate to your 3kW example. What do you mean by "but your run time will not be super fast"? Compared to what? Sorry if I'm missing something here.

So how quickly do you need to strip The compromise option offered earlier might be enough for some folks. If you WANT to strip faster then a dedicated strip option would suit you better.

I'm not sure how quickly we need to strip as we've never done this before
Can you provide any examples/references to give me an idea of best/worst case scenarios?

If not, consider designing your reflux condenser so that you can remove it and insert it before your liebig to function as a precondenser in the strip mode. An empty column (riser) will help with speed and the length is determined by where you wish the output of the condenser to be.

Not sure I'm following you here...anything more you can add to this could help.

The dedicated pot still head is a better option but do you actually need it.

Are you referring to having the ability to strip faster or better quality whiskey-type product or both? If given a choice of one or the other, a better product is what we're looking for.

Myles, thanks again for all your help. If you weren't on the other side of the pond, I'd pour you a drink when this thing is done.

[Samohon]

Hi InTheGarage... Sorry I have not posted to this thread, I have family visiting from Ukraine today so it has been all hands on deck...

Anyway, my advice would be to build the Bok first and get some product from it. Then add on the VM with a 2" equal Tee. My original idea was to use 2" to 1" reducers but after some proven advice from the old hands back then, I went with the 2" equal below the LM (Bok). Works flawlessly, done a run of 25L low wines on Sunday and took a wopping 10L @ 96% abv... Very clean neutral, the rest was fores/heads/tails. I collect fores/heads/tails from the LM and the good stuff from my VM. Shit, the VM even turns itself off when the alcohol in the vapour is less than the weight of air. Now you cant beat that.

Build the Bok first... Clean it with vinegar/water + sacrificial wash. Add the VM later and clean it with the alc you collected on your cleaning run from the Bok...

Hope it helps... Gotta git to the airport and taxi them here... Reckon theres gonna be a party tomorrow...

[InTheGarage]

Thanks for the response Samohon, encouraged that your rig is running so well!

I asked about the tee because we're building a 3" column, so the tee is pricey. I wanted to justify it before spending the cash, but you've convinced me. In your opinion would reducing that down to a 1" gate valve defeat the purpose? Going with stainless on all the valves so again I'm trying to justify the cost based on others experience here. The thought would be to add maybe a 6" length of 3" dia. pipe on the VM side of the tee, and then gradually reduce to the 1" valve. Or would I need to transition instead to a 1.5" gate valve (keeping with the proportions in your drawing)? Trying to get my head around the dynamics of the vapor.

Build the Bok first... Clean it with vinegar/water + sacrificial wash. Add the VM later and clean it with the alc you collected on your cleaning run from the Bok...

Curious about your reasoning to add the VM later. Is it simply to use the product created with the Bok for cleaning, or something else?
Thanks again and enjoy yer party!!

[myles]

Can't really comment on the Bok, it is a perfectly good design that works well by all accounts, but I have not built one. There are alternative LM designs, and it boils down to personal choice. Adding a VM later as an alternative head is a fine idea.

On the VM the principle is the vapour splits mostly on the basis of the cross sectional area of the two exit orrifices. One goes to the reflux condenser and one to the product condenser. If they are both the same you get a THEORETICAL 50:50 split. This depends on turbulence, which is why it is common to introduce a reduction first to increase vapour speed, leading to more turbulence. With laminar flow (no turbulence) most of the vapour will go in a straight line and little will take the branch - irrespective of relative areas. Reducing your VM branch orrifice (relative to the reflux condenser one) considerably reduces your MAX POTENTIAL product take off rate.

Packing before the reflux condenser (or a heads column) changes the pressure parameters, this tends to force vapour into the product condenser side - overiding the cross sectional area issue. My preffered option with the offset condenser also FORCES a vapour split due to the physical obstruction. You could probably do without the increase in vapour speed caused by the reduction, but I put it in anyway. In this configuration there will be lots of turbulence

On run times; 15.5 gallon boiler, so lets say for convenience using a charge of 60 litres @ 40%, equivalent to 25 litres @ 95% ABV. 2.5 litres per hour means a run time of 10 hours. PLUS warm up, take down, cleaning and tidying away - probably 12+ hours.

For efficiency, your VM product condenser should be sized appropriately to the product rate during a VM run. By inference, it will be too small for a full blown strip run. With two heads and a modular reflux condenser, you could add the reflux condenser into your strip head before your liebig to work as a precondenser. It means a slight design change to the reflux condenser but it is do-able. Forgive the crude drawing but it serves to give the idea.

[Samohon]

+2, what Myles says...

I was just trying to keep the initial cost of the rig down by building the LM first then adding the VM later, but if your going to build both and dont mind the expense, no problem...

hope it helps...

[myles]

On the run times that 2.5 litres per hour rate is probably a maximum figure. There are lots of other factors to consider and you could be running a LOT slower. For example on a slightly shorter column with conventional packing you could possibly be getting just 1 litre per hour.

As a general rule slower = more purity, but it is determined by your interpretation of flavour.