Liebig pipe diameters

[StarkBlood80]

Hi
I’m in the research stage of making a still and I’m planning on starting with a pot still made from a keg and 54mm copper column and a Liebig condenser.
What id like to do is reduce from 54mm down to 22mm and carry on the 22mm all the way to the product collection. Would a 28mm diameter water jacket work for the Liebig?
I’ve looked at the easy pot still design on here and that Liebig has a 15mm core and 22mm jacket so there’s 7mm of water surrounding the output and if I went with 22mm core and 28mm jacket I’d have 6mm surrounding the outlet.

[hampk]

Let's look at the cross-sections of the designs

22/15

surface area of water-carrying section is (approx)

(22 * pi) - (15 * pi) = 21.99 mm2

28/22

surface area of water-carrying section is (approx)

(28 * pi) - (22 * pi) = 18.84 mm2

Therefore you have less water flowing per second through your design than through the smaller one. As you also have a somewhat larger space available for vapour, my conclusion is that you will need to up your water flow rates to achieve the same amount of cooling.

Anyone else want to chip in?!

Cheers

Hampk

[Mash]

...and carry on the 22mm all the way to the product collection.

There is some cleaver maths kicking about here somewhere that tells you how long it needs to be - TBH I recall it is more important what power you are putting in the bottom (KW of element).

Will have search when I get a mo..

[Mash]

Ta da.

http://homedistiller.org/calcs/cond_calc is the bit I was thinking of.

This thread is worth a read too.

viewtopic.php?f=12&t=4217&p=38079&hilit=formula#p36573

[StarkBlood80]

Let's look at the cross-sections of the designs

22/15

surface area of water-carrying section is (approx)

(22 * pi) - (15 * pi) = 21.99 mm2

28/22

surface area of water-carrying section is (approx)

(28 * pi) - (22 * pi) = 18.84 mm2

Therefore you have less water flowing per second through your design than through the smaller one. As you also have a somewhat larger space available for vapour, my conclusion is that you will need to up your water flow rates to achieve the same amount of cooling.

Anyone else want to chip in?!

Cheers

Hampk

Hi
That’s a good reply. I did think about soldering a spiral of think copper wire to the outside of the vapour tube before fitting the water jacket with the intention of introducing some turbulence to the coolant which should increase contact time.

[Icefever]

Myles will be here soon... what he doesn't know about condensers is not worth knowing...

[H12rpo]

Sorry, but I think the those spirals around the outside of the tube to ‘introduce turbulence’ are complete bollocks. I cannot see how turbulence would apply more cooling power. The fact that there is marginally more vapour carrying copper exposed to the cooling water might increase the cooling effect but of course that takes up more of the inside jacket space that water would’ve taken up and so any tiny gain would be negated. Same goes for the inner pipes that are crushed alternating by 90degress each time on the foolish premise that ‘it exposes a greater surface area to the water’ ..... pfft!

[Mash]

Myles's will know.

Careful. Turbulence isn't necessarily a good thing, there was some stuff done with crimping and it did not quite deliver as I recall.

[StarkBlood80]

Sorry, but I think the those spirals around the outside of the tube to ‘introduce turbulence’ are complete bollocks. I cannot see how turbulence would apply more cooling power. The fact that there is marginally more vapour carrying copper exposed to the cooling water might increase the cooling effect but of course that takes up more of the inside jacket space that water would’ve taken up and so any tiny gain would be negated. Same goes for the inner pipes that are crushed alternating by 90degress each time on the foolish premise that ‘it exposes a greater surface area to the water’ ..... pfft!

It’s not about increasing the cooling power of the water but increasing the contact time the water has with the copper.
If you induce the water swirl around the vapour pipe then you’re making it travel further than it would if it just traveled bottom to top.

I think for ease seeing as this is my first build I’ll stick with the plans I’ve seen online and go with 15-22 or 15-28.

[H12rpo]

It’s not about increasing the cooling power of the water but increasing the contact time the water has with the copper.
If you induce the water swirl around the vapour pipe then you’re making it travel further than it would if it just traveled bottom to top.

So it’s to save (a tiny amount) of water?? Or to shorten the Liebig ?

[hampk]

I don’t see how turbulence would affect things at all, but equally I don’t think the spiral copper wire concept is about increasing turbulence, it’s about lengthening the path the water takes and hence increasing the cooling effect.

On balance, I’m of the opinion that my simple maths and Mash’s assertion regarding power are probably the key factors - as ever, a balance. For a given power input, you need a equivalent amount of cooling.

Myles? Myles?!

Cheers

Hampk


..

[Easydrinker]

Myles's will know.

Careful. Turbulence isn't necessarily a good thing, there was some stuff done with crimping and it did not quite deliver as I recall.

If you can tempt Myles back here for even the occasional post, then I tip what is left of my xmas cracker hat in your direction.

Robert.

[Myles]

Tada!!!!
Sorry not looked in here for some time.

Start with your maximum power available. I tend to use that same HD calculator as a good guide.
I don't use 22 mm on product condensers. At that size you really need to include vapour turbulators because the centre of the vapour tube is some distance from a cold surface.
For a liebig I would use 15 mm. Above 3 kW and the length gets inconvenient.

That's the point where I switch to multi vapour paths or coil the liebig. For shotgun Pcs these days I use 12 mm vapour tubes.

I don't like crimped inner tubes. In any position other than vertical they can hold little pools of condensate.

As for wire wraps I don't but the theory is sound. Dependent on coolant flow rate you can get HOT spots in any condenser. I tend to run coolant fast so it is less of an issue but I would put in baffle plates to increase turbulence in the water inside the jacket. I doubt the wire wraps do much on a liebig as the coolant is likely to be turbulent except at low flow rates.

[Icefever]

Cheers Myles...

See I said he'd be along.... trust me I'm a soothsayer...oh ya oh ya & thrice ho ya

[StarkBlood80]

Tada!!!!
Sorry not looked in here for some time.

Start with your maximum power available. I tend to use that same HD calculator as a good guide.
I don't use 22 mm on product condensers. At that size you really need to include vapour turbulators because the centre of the vapour tube is some distance from a cold surface.
For a liebig I would use 15 mm. Above 3 kW and the length gets inconvenient.

That's the point where I switch to multi vapour paths or coil the liebig. For shotgun Pcs these days I use 12 mm vapour tubes.

I don't like crimped inner tubes. In any position other than vertical they can hold little pools of condensate.

As for wire wraps I don't but the theory is sound. Dependent on coolant flow rate you can get HOT spots in any condenser. I tend to run coolant fast so it is less of an issue but I would put in baffle plates to increase turbulence in the water inside the jacket. I doubt the wire wraps do much on a liebig as the coolant is likely to be turbulent except at low flow rates.

Thanks for the reply, I have much to think about.
Do you think reducing the vapour stream down from 54-22-15 would increase the velocity much compared to just going from 54-22?