liebig + 2000w

[Jacksonbrown]

?

I have never understood why some want for example a enormous RC at the top of a refluxstill.

Because it's much more efficient. You'll be able to easily knock down a few thousand watts with a few hundred mls.
The calculator works exactly how I would expect it too.
If water use isn't a concern and you live somewhere cold then put your inlet temp at 10, your outlet at 15 and the thermal coefficient at 2000 due to high flow and extra wire etc for turbulence.
You get 15"
The down side is it'll need nearly 6 l/m to do it.

It sounds like it needs a better description on the page as there's a small amount of assumed knowledge on thermodynamics and fluid mechanics.

I think to use it for a shotgun you multiply the tube diameter by the number of tubes to give an equivalent tube diameter (it'll be the same surface area and length that way although stainless will be a bit off due to a differant conduction value).
That's seems to agree with what I got anyway.

[Bagasso]

Because it's much more efficient. You'll be able to easily knock down a few thousand watts with a few hundred mls.

Everyone is being vague just to be right. Doing that it in a condensor half the size would be more efficient.

The calculator works exactly how I would expect it too.
If water use isn't a concern and you live somewhere cold then put your inlet temp at 10, your outlet at 15 and the thermal coefficient at 2000 due to high flow and extra wire etc for turbulence.
You get 15"
The down side is it'll need nearly 6 l/m to do it.

No need to exaggerate. OP just wants to know if he can use a shorter condensor. The default THE value gives 2m length but

Inlet temp - 15
Outlet - 50
Tube diameter - 1/2"
Heat input - 2Kw
Heat transfer coefficient - 2000
Length - 0.52m
Flowrate - 0.82l/m

Doesn't seem way out there. Even a small aquarium pump can fit that requirement.

It sounds like it needs a better description on the page as there's a small amount of assumed knowledge on thermodynamics and fluid mechanics.

Or they can just up the default THE or tie it to a personalized flow rate value.

[Jacksonbrown]

Do you mean thermal coefficient? With the right conditions 2000 isn't that out there but is 800 ml/min going to be turbulent flow? As I mentioned before you need to work out the Reynolds number to see if it's going to be turbulent flow. A Reynolds number of 4000+ is what you want but there are also rules of thumb that just use flow velocity which are a lot easier to work with.

THE stands for Tubular Heat Exchanger AKA shell and tube heat exchanger. Google those terms and you'll get better calculators.

An efficient condenser is one that has the water coming out near the vapour temp and the distillate coming out near the feed water temp. That's not me talking out my arse, that's fact.

I don't think I'm being vague, I'm happy to further explain anything I've put up.
The OP can easily design a very short condenser.
He needs maximum delta T from water to vapour along the entire length of the tube.
To do that he needs to start with very cold water.
To minimise the deta T from inlet water to outlet water you just need a very high flow rate.
A sketch might help there but so will google.

He can also easily raise the thermal coefficient by adding baffles and increassing turbulence etc.
Bending/crimping won't be so necessary as the high flow will be turbulent anyway.

Using copper also will allow heat to conduct much faster through the wall too but stainless is probably better on the product side for other reasons, a compromise.

It sounds like the the OP had his original question answered and lost interest a while back anyway but it was the original question that was a bit vague, without knowing the coolant temp and flow rate required.

[DAD300]

I have never put packing in the top of a Product condenser. Might be scary...

But I have put a bit of gauze in the end to slow vapor down. There it can be changed between heads and hearts. It will also drop out if there is any back pressure.

[Bagasso]

Do you mean thermal coefficient? With the right conditions 2000 isn't that out there but is 800 ml/min going to be turbulent flow? As I mentioned before you need to work out the Reynolds number to see if it's going to be turbulent flow. A Reynolds number of 4000+ is what you want but there are also rules of thumb that just use flow velocity which are a lot easier to work with.

The calc on the parent site doesn't even allow an input for flow in order to apply those rules of thumb.

An efficient condenser is one that has the water coming out near the vapour temp and the distillate coming out near the feed water temp. That's not me talking out my arse, that's fact.

Which is why posting an example with an inlet of 10 and an outlet of 15 was exaggerating to make your point.

I don't think I'm being vague, I'm happy to further explain anything I've put up.

I meant most members. What is a "huge RC". That is vague. What is "efficient" that is also vague.

ETA: Even the calc page is vague "you can reduce the length if you increase flow". Well, how much? Double the flow half the length, one third, one tenth?

It sounds like the the OP had his original question answered and lost interest a while back anyway but it was the original question that was a bit vague, without knowing the coolant temp and flow rate required.

The coolant temp and/or the flow rate may have been what he was trying to figure out and that is where the calc falls short.

[Jacksonbrown]

Well I tried. I stand by what I posted if you care to reread it.
Like I said I'm happy to answer questions to help you understand but if your just trying to shoot me down I won't bother.

[Bagasso]

Well I tried. I stand by what I posted if you care to reread it.
Like I said I'm happy to answer questions to help you understand but if your just trying to shoot me down I won't bother.

I'm trying too. Pointing out the inherent limitations of the calc isn't shooting you down.

Efficiency isn't black or white. Turbulent flow is more efficient but laminar flow also has its value and it can be above the 850W/m2C used by the parent site calc.

Nobody is disputing what you said. I'm saying that the calc doesn't take any of that into account.

[Jimbo]

Well Jackson, you shot me down for suggesting a copper structure much more suited to vapor cooling than a smooth bore copper structure. Smooth copper pipe for vapor is a piss poor engineered solution to vapor condensing.

[googe]

Yes I was some.what vague in my op, I tried not to be to thorough as I hoped lots f conflicting info would happen, to help me and others get a broader view. If I had a calc that could tell me the answer without knowing the input temp and flow, I would have, but that's not a reality. Thanks for the calcs bagasso, they make a but more sense!. I do.like your theory Jackson, I just get a but lost in it sometimes. I guess I'm asking for a simple answer when when it doesn't exist. I'll start with 70mm 1/2" inner and 50mm 3/4" outer and try different water temps, and go down in length till I hit the sweet spot. Thanks everyone for the input, and no I havnt lost interest .

[Haus]

Googe, if you put a strip of this stuff up in the vapor path it will make it 10x more efficient. I have a leibig with about 40" of working length, and it will knock down 5500W without even breaking a sweat. http://www.brewhaus.com/Copper-Mesh-P947.aspx" onclick="window.open(this.href);return false;" rel="nofollow

This I have done, still do, and Jimbo is spot on. I have a 24" working area liebig and I do a loosely twisted (spiraled end to end) 10" piece of copper mesh on the cool end a little under half way up.

[pulsetech]

Yes I was some.what vague in my op, I tried not to be to thorough as I hoped lots f conflicting info would happen, to help me and others get a broader view. If I had a calc that could tell me the answer without knowing the input temp and flow, I would have, but that's not a reality. Thanks for the calcs bagasso, they make a but more sense!. I do.like your theory Jackson, I just get a but lost in it sometimes. I guess I'm asking for a simple answer when when it doesn't exist. I'll start with 70mm 1/2" inner and 50mm 3/4" outer and try different water temps, and go down in length till I hit the sweet spot. Thanks everyone for the input, and no I havnt lost interest .

hope you mean 700mm inner and 500mm outer.

[googe]

Yes I was some.what vague in my op, I tried not to be to thorough as I hoped lots f conflicting info would happen, to help me and others get a broader view. If I had a calc that could tell me the answer without knowing the input temp and flow, I would have, but that's not a reality. Thanks for the calcs bagasso, they make a but more sense!. I do.like your theory Jackson, I just get a but lost in it sometimes. I guess I'm asking for a simple answer when when it doesn't exist. I'll start with 70mm 1/2" inner and 50mm 3/4" outer and try different water temps, and go down in length till I hit the sweet spot. Thanks everyone for the input, and no I havnt lost interest .

hope you mean 700mm inner and 500mm outer.

Yeah that sounds better

[Bagasso]

Just going from 3/4" over 1/2" to 1" over 3/4" will cut the length by one third.

Also, if you don't like the idea of a scrubbie in the inner tube, an unraveled scrubbie between the inner and outer tube will give you some turbulence to increase the heat transfer coefficient.

[Jacksonbrown]

Not my theory G, its law

Going back to the calculator, to get 0.5m of 1/2" to work you'll need to do a lot get the coefficient (turbulence) right up and it'll use a butt load of water. It'll be interesting to see what real world results you come back with. You may also get vapour pouring out which will need that scrubber in the vapour path to increase contact. This is especially true if you up the tube size to 3/4" or 1".
You can also change the angle to just off horizontal. This will allow the liquid to still drain but will slow the fall of the dense vapour. You'd be surprised how much pointing the PC straight down reduces its effectiveness . This is worse with larger diameter tubes.
If space is a concern you can always just make a heap of 1' condensers and join them in series like switchbacks on a steep road. Google 'holding tube" to see some images. Just keep the fall constant.
That way you can also add and remove them to suit the duty required at the time.

[Jacksonbrown]

Also efficiency isn't vague either. It is a real value with a real equation to calculate it.
There are pluses and minuses to every approach. You just need to way up all your needs and decide what compromises your willing to make. That's real engineering.
It like asking 'how long long is a piece of string'. If you want a real answer you need to ask a few more questions.

[googe]

Thanks bagasso but I want to keep it 1/2" for now. I've always questioned how well 3/4" works without some sort of inner aid like Jackson said. Probably not that big of a deal to wintr about. Weird!, I was looking at something similar last night Jackson, that would work well for space, but Ulf coolant lines would be a bit messy!. Join them with unions and you'd be set, surprised someone hasn't done that here!.

[still_stirrin]

HEX design...I used it a lot when working oil patch production design.

The tube in shell design is most efficient given the smallest foot print. Banks of small tubes have greater surface area exposed to the shell fluid but pose greater flow loss for the heat liberating fluid. But if that fluid is a vapor, then the resulting pressure drop can help condesnsation as well if the vapor is at saturation already. This is the advantage of the shotgun condensers and why you hear of their knockdown power. More involved to build, but compact and efficient.

But even for the concentric (annular) tube in tube flowing in counterflow, turbulence adds heat transfer efficiency. And putting in vanes, or baffles increases contact (conduction) surface area. Increasing contact time improves efficiency as well.

As Jackson pointed out, the differential equations are complicated so its not quite a "punch & crunch" solution. But for the most part, heat transfer can be calculated (with some basic assumptions) to predict average performance. Fortunately, the petro-chemical HEX manufacturers have some emperical charts to aid in design for performance. It simplifies the design a lot.
ss

[Danespirit]

Excellent explanition Still stirrin...

[Jacksonbrown]

S_S, if you needed high efficiency in a small footprint why wouldn't you go plate and frame?
A much better solution IMO.

[Bagasso]

Also efficiency isn't vague either. It is a real value with a real equation to calculate it.

The way you used it was vague because it lacked a real value but it still implied that anything less than that value is not acceptable.

There are pluses and minuses to every approach. You just need to way up all your needs and decide what compromises your willing to make. That's real engineering.

I didn't see the OP ask anyone to engineer a condensor for him to any specific specs. He asked how can you get a 1/2" in 3/4" liebig to handle a 2kw heat source other than increasing length.

If that means using more water then that is up to him to decide if it is practical or not.

@googe another option is to use a coil inside a large sleeve. 2 m of 1/2" coiled in one foot of 3" is a tight little package. You can even use pvc for the outer shell (unless pics are going into the centerfold of stillboy magazine).

[jedneck]

Googe I run a triple wall liebig with 18 inches of cooling. Its a 1 1/4 over 1 with a 3/8 id inter tube. At 2000 watts I had to use a needle valve to slow coolant down enuf. Also use it on a keg at 4500 watts but it take alot of water.

[googe]

Did that with 2" bagasso http://homedistiller.org/forum/viewtopi ... 17&t=49694 , worked well, but want liebig!!!!. Thanks jed, that's cool, I was thinking of the same.design as yours with the way you have the column offset, good way of distributing the weight!.

[Bagasso]

Did that with 2"

Ah, a bug bite. Nothing besides doing that build is going to soothe that itch.

[Jacksonbrown]

What about making 3m, all copper one out of 1/2" over 3/4" with copper wire taking up the free space. Then aneal the lot and wrap it around a keg or something even smaller to make a coil. A bit like the external wort chiller coils you see on some HERM systems that get made out out garden hose.
I would 't do it but it's another option to consider.

[myles]

I like spiral liebigs but wouldn't do it that way.

In the UK the coil inside a domestic hot water tank is a copper coil, readily available as a spare part. Ready built coil jacket - all you need is to fit a suitable inner coil.

Mine is a bit excessive really, 7 foot so that is 21 foot of 3/8 cores, good for over 8kW in a compact 1 cubic foot volume.

[Danespirit]

Myles...i always liked your copperwork...
How did you make the connection for the three pipes inside..?

[Jimbo]

Beautiful work Myles. Why not put the smaller tubings inside first and then bend the whole lot together?

[googe]

Your crazy Myles

[myles]

Beautiful work Myles. Why not put the smaller tubings inside first and then bend the whole lot together?

Easy answer. I already had the outer coil, salvaged out of a domestic hot water cylinder. Have a couple more also that I haven't used yet. Just used it because it was available and I didn't wish to waste a coil of 1" copper tube!!!

[engunear]

My check of the calculator suggests there is a safety factor in there that is not mentioned - around 2X.

You can check the effective length of a leibig by poking a thermocouple up the distillate outlet and seeing how far it has to go before it reaches the area where the temp is the boiling point. You also need to measure the inlet and outlet temperature and the flow rate. When I do this I get values close to the calculator but without the safety margin (this measurement needs repeating btw).

Now maybe I dropped a 2X in my calculation so here it is (in Python):

'''
Performance check for leibig consensors.
'''
import math

To=95
Ti=10
dT=To-Ti
power=2000.0
flow=0.34 # liters per min

length=1.42 # m
dia=18.0/1000 # m

area = length * math.pi * dia
print "area",area

HTC=power/area/dT # Heat transfer coefficient
print "HTC", HTC